diff options
Diffstat (limited to 'linux')
43 files changed, 12101 insertions, 0 deletions
diff --git a/linux/bio.c b/linux/bio.c new file mode 100644 index 0000000..966f227 --- /dev/null +++ b/linux/bio.c @@ -0,0 +1,323 @@ +/* + * Copyright (C) 2001 Jens Axboe <axboe@kernel.dk> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + * + */ +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/slab.h> +#include <linux/kernel.h> +#include <linux/export.h> + +void bio_copy_data_iter(struct bio *dst, struct bvec_iter dst_iter, + struct bio *src, struct bvec_iter src_iter) +{ + struct bio_vec src_bv, dst_bv; + void *src_p, *dst_p; + unsigned bytes; + + while (1) { + if (!src_iter.bi_size) { + src = src->bi_next; + if (!src) + break; + + src_iter = src->bi_iter; + } + + if (!dst_iter.bi_size) { + dst = dst->bi_next; + if (!dst) + break; + + dst_iter = dst->bi_iter; + } + + src_bv = bio_iter_iovec(src, src_iter); + dst_bv = bio_iter_iovec(dst, dst_iter); + + bytes = min(src_bv.bv_len, dst_bv.bv_len); + + src_p = kmap_atomic(src_bv.bv_page); + dst_p = kmap_atomic(dst_bv.bv_page); + + memcpy(dst_p + dst_bv.bv_offset, + src_p + src_bv.bv_offset, + bytes); + + kunmap_atomic(dst_p); + kunmap_atomic(src_p); + + bio_advance_iter(src, &src_iter, bytes); + bio_advance_iter(dst, &dst_iter, bytes); + } +} + +void bio_copy_data(struct bio *dst, struct bio *src) +{ + bio_copy_data_iter(dst, dst->bi_iter, + src, src->bi_iter); +} + +void zero_fill_bio_iter(struct bio *bio, struct bvec_iter start) +{ + unsigned long flags; + struct bio_vec bv; + struct bvec_iter iter; + + __bio_for_each_segment(bv, bio, iter, start) { + char *data = bvec_kmap_irq(&bv, &flags); + memset(data, 0, bv.bv_len); + bvec_kunmap_irq(data, &flags); + } +} + +void __bio_clone_fast(struct bio *bio, struct bio *bio_src) +{ + /* + * most users will be overriding ->bi_bdev with a new target, + * so we don't set nor calculate new physical/hw segment counts here + */ + bio->bi_bdev = bio_src->bi_bdev; + bio_set_flag(bio, BIO_CLONED); + bio->bi_opf = bio_src->bi_opf; + bio->bi_iter = bio_src->bi_iter; + bio->bi_io_vec = bio_src->bi_io_vec; +} + +struct bio *bio_clone_fast(struct bio *bio, gfp_t gfp_mask, struct bio_set *bs) +{ + struct bio *b; + + b = bio_alloc_bioset(gfp_mask, 0, bs); + if (!b) + return NULL; + + __bio_clone_fast(b, bio); + return b; +} + +struct bio *bio_split(struct bio *bio, int sectors, + gfp_t gfp, struct bio_set *bs) +{ + struct bio *split = NULL; + + BUG_ON(sectors <= 0); + BUG_ON(sectors >= bio_sectors(bio)); + + /* + * Discards need a mutable bio_vec to accommodate the payload + * required by the DSM TRIM and UNMAP commands. + */ + if (bio_op(bio) == REQ_OP_DISCARD || bio_op(bio) == REQ_OP_SECURE_ERASE) + split = bio_clone_bioset(bio, gfp, bs); + else + split = bio_clone_fast(bio, gfp, bs); + + if (!split) + return NULL; + + split->bi_iter.bi_size = sectors << 9; + + bio_advance(bio, split->bi_iter.bi_size); + + return split; +} + +int bio_alloc_pages(struct bio *bio, gfp_t gfp_mask) +{ + int i; + struct bio_vec *bv; + + bio_for_each_segment_all(bv, bio, i) { + bv->bv_page = alloc_page(gfp_mask); + if (!bv->bv_page) { + while (--bv >= bio->bi_io_vec) + __free_page(bv->bv_page); + return -ENOMEM; + } + } + + return 0; +} + +void bio_advance(struct bio *bio, unsigned bytes) +{ + bio_advance_iter(bio, &bio->bi_iter, bytes); +} + +static void bio_free(struct bio *bio) +{ + unsigned front_pad = bio->bi_pool ? bio->bi_pool->front_pad : 0; + + kfree((void *) bio - front_pad); +} + +void bio_put(struct bio *bio) +{ + if (!bio_flagged(bio, BIO_REFFED)) + bio_free(bio); + else { + BUG_ON(!atomic_read(&bio->__bi_cnt)); + + /* + * last put frees it + */ + if (atomic_dec_and_test(&bio->__bi_cnt)) + bio_free(bio); + } +} + +static inline bool bio_remaining_done(struct bio *bio) +{ + /* + * If we're not chaining, then ->__bi_remaining is always 1 and + * we always end io on the first invocation. + */ + if (!bio_flagged(bio, BIO_CHAIN)) + return true; + + BUG_ON(atomic_read(&bio->__bi_remaining) <= 0); + + if (atomic_dec_and_test(&bio->__bi_remaining)) { + bio_clear_flag(bio, BIO_CHAIN); + return true; + } + + return false; +} + +static struct bio *__bio_chain_endio(struct bio *bio) +{ + struct bio *parent = bio->bi_private; + + if (!parent->bi_error) + parent->bi_error = bio->bi_error; + bio_put(bio); + return parent; +} + +static void bio_chain_endio(struct bio *bio) +{ + bio_endio(__bio_chain_endio(bio)); +} + +void bio_endio(struct bio *bio) +{ +again: + if (!bio_remaining_done(bio)) + return; + + /* + * Need to have a real endio function for chained bios, otherwise + * various corner cases will break (like stacking block devices that + * save/restore bi_end_io) - however, we want to avoid unbounded + * recursion and blowing the stack. Tail call optimization would + * handle this, but compiling with frame pointers also disables + * gcc's sibling call optimization. + */ + if (bio->bi_end_io == bio_chain_endio) { + bio = __bio_chain_endio(bio); + goto again; + } + + if (bio->bi_end_io) + bio->bi_end_io(bio); +} + +void bio_endio_nodec(struct bio *bio) +{ + goto nodec; + + while (bio) { + if (unlikely(!bio_remaining_done(bio))) + break; +nodec: + if (bio->bi_end_io == bio_chain_endio) { + struct bio *parent = bio->bi_private; + parent->bi_error = bio->bi_error; + bio_put(bio); + bio = parent; + } else { + if (bio->bi_end_io) + bio->bi_end_io(bio); + bio = NULL; + } + } +} + +void bio_reset(struct bio *bio) +{ + unsigned long flags = bio->bi_flags & (~0UL << BIO_RESET_BITS); + + memset(bio, 0, BIO_RESET_BYTES); + bio->bi_flags = flags; + atomic_set(&bio->__bi_remaining, 1); +} + +struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) +{ + unsigned front_pad = bs ? bs->front_pad : 0; + struct bio *bio; + void *p; + + p = kmalloc(front_pad + + sizeof(struct bio) + + nr_iovecs * sizeof(struct bio_vec), + gfp_mask); + + if (unlikely(!p)) + return NULL; + + bio = p + front_pad; + bio_init(bio); + bio->bi_pool = bs; + bio->bi_max_vecs = nr_iovecs; + bio->bi_io_vec = bio->bi_inline_vecs; + + return bio; +} + +struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, + struct bio_set *bs) +{ + struct bvec_iter iter; + struct bio_vec bv; + struct bio *bio; + + bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); + if (!bio) + return NULL; + + bio->bi_bdev = bio_src->bi_bdev; + bio->bi_opf = bio_src->bi_opf; + bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; + bio->bi_iter.bi_size = bio_src->bi_iter.bi_size; + + switch (bio_op(bio)) { + case REQ_OP_DISCARD: + case REQ_OP_SECURE_ERASE: + break; + case REQ_OP_WRITE_SAME: + bio->bi_io_vec[bio->bi_vcnt++] = bio_src->bi_io_vec[0]; + break; + default: + bio_for_each_segment(bv, bio_src, iter) + bio->bi_io_vec[bio->bi_vcnt++] = bv; + break; + } + + return bio; +} diff --git a/linux/bitrev.c b/linux/bitrev.c new file mode 100644 index 0000000..61207bb --- /dev/null +++ b/linux/bitrev.c @@ -0,0 +1,37 @@ +#include <linux/types.h> +#include <linux/bitrev.h> + +const u8 byte_rev_table[256] = { + 0x00, 0x80, 0x40, 0xc0, 0x20, 0xa0, 0x60, 0xe0, + 0x10, 0x90, 0x50, 0xd0, 0x30, 0xb0, 0x70, 0xf0, + 0x08, 0x88, 0x48, 0xc8, 0x28, 0xa8, 0x68, 0xe8, + 0x18, 0x98, 0x58, 0xd8, 0x38, 0xb8, 0x78, 0xf8, + 0x04, 0x84, 0x44, 0xc4, 0x24, 0xa4, 0x64, 0xe4, + 0x14, 0x94, 0x54, 0xd4, 0x34, 0xb4, 0x74, 0xf4, + 0x0c, 0x8c, 0x4c, 0xcc, 0x2c, 0xac, 0x6c, 0xec, + 0x1c, 0x9c, 0x5c, 0xdc, 0x3c, 0xbc, 0x7c, 0xfc, + 0x02, 0x82, 0x42, 0xc2, 0x22, 0xa2, 0x62, 0xe2, + 0x12, 0x92, 0x52, 0xd2, 0x32, 0xb2, 0x72, 0xf2, + 0x0a, 0x8a, 0x4a, 0xca, 0x2a, 0xaa, 0x6a, 0xea, + 0x1a, 0x9a, 0x5a, 0xda, 0x3a, 0xba, 0x7a, 0xfa, + 0x06, 0x86, 0x46, 0xc6, 0x26, 0xa6, 0x66, 0xe6, + 0x16, 0x96, 0x56, 0xd6, 0x36, 0xb6, 0x76, 0xf6, + 0x0e, 0x8e, 0x4e, 0xce, 0x2e, 0xae, 0x6e, 0xee, + 0x1e, 0x9e, 0x5e, 0xde, 0x3e, 0xbe, 0x7e, 0xfe, + 0x01, 0x81, 0x41, 0xc1, 0x21, 0xa1, 0x61, 0xe1, + 0x11, 0x91, 0x51, 0xd1, 0x31, 0xb1, 0x71, 0xf1, + 0x09, 0x89, 0x49, 0xc9, 0x29, 0xa9, 0x69, 0xe9, + 0x19, 0x99, 0x59, 0xd9, 0x39, 0xb9, 0x79, 0xf9, + 0x05, 0x85, 0x45, 0xc5, 0x25, 0xa5, 0x65, 0xe5, + 0x15, 0x95, 0x55, 0xd5, 0x35, 0xb5, 0x75, 0xf5, + 0x0d, 0x8d, 0x4d, 0xcd, 0x2d, 0xad, 0x6d, 0xed, + 0x1d, 0x9d, 0x5d, 0xdd, 0x3d, 0xbd, 0x7d, 0xfd, + 0x03, 0x83, 0x43, 0xc3, 0x23, 0xa3, 0x63, 0xe3, + 0x13, 0x93, 0x53, 0xd3, 0x33, 0xb3, 0x73, 0xf3, + 0x0b, 0x8b, 0x4b, 0xcb, 0x2b, 0xab, 0x6b, 0xeb, + 0x1b, 0x9b, 0x5b, 0xdb, 0x3b, 0xbb, 0x7b, 0xfb, + 0x07, 0x87, 0x47, 0xc7, 0x27, 0xa7, 0x67, 0xe7, + 0x17, 0x97, 0x57, 0xd7, 0x37, 0xb7, 0x77, 0xf7, + 0x0f, 0x8f, 0x4f, 0xcf, 0x2f, 0xaf, 0x6f, 0xef, + 0x1f, 0x9f, 0x5f, 0xdf, 0x3f, 0xbf, 0x7f, 0xff, +}; diff --git a/linux/blkdev.c b/linux/blkdev.c new file mode 100644 index 0000000..b4ff451 --- /dev/null +++ b/linux/blkdev.c @@ -0,0 +1,155 @@ + +#include <alloca.h> +#include <errno.h> +#include <fcntl.h> +#include <sys/ioctl.h> +#include <sys/stat.h> +#include <sys/types.h> +#include <sys/uio.h> +#include <unistd.h> + +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/fs.h> + +int submit_bio_wait(struct bio *bio) +{ + struct iovec *iov; + struct bvec_iter iter; + struct bio_vec bv; + ssize_t ret; + unsigned i; + + if (bio->bi_opf & REQ_PREFLUSH) + fdatasync(bio->bi_bdev->bd_fd); + + i = 0; + bio_for_each_segment(bv, bio, iter) + i++; + + iov = alloca(sizeof(*iov) * i); + + i = 0; + bio_for_each_segment(bv, bio, iter) + iov[i++] = (struct iovec) { + .iov_base = page_address(bv.bv_page) + bv.bv_offset, + .iov_len = bv.bv_len, + }; + + switch (bio_op(bio)) { + case REQ_OP_READ: + ret = preadv(bio->bi_bdev->bd_fd, iov, i, + bio->bi_iter.bi_sector << 9); + break; + case REQ_OP_WRITE: + ret = pwritev(bio->bi_bdev->bd_fd, iov, i, + bio->bi_iter.bi_sector << 9); + break; + default: + BUG(); + } + + if (bio->bi_opf & REQ_FUA) + fdatasync(bio->bi_bdev->bd_fd); + + return ret == bio->bi_iter.bi_size ? 0 : -EIO; +} + +void generic_make_request(struct bio *bio) +{ + bio->bi_error = submit_bio_wait(bio); + bio_endio(bio); +} + +int blkdev_issue_discard(struct block_device *bdev, + sector_t sector, sector_t nr_sects, + gfp_t gfp_mask, unsigned long flags) +{ + return 0; +} + +unsigned bdev_logical_block_size(struct block_device *bdev) +{ + struct stat statbuf; + unsigned blksize; + int ret; + + ret = fstat(bdev->bd_fd, &statbuf); + BUG_ON(ret); + + if (!S_ISBLK(statbuf.st_mode)) + return statbuf.st_blksize >> 9; + + ret = ioctl(bdev->bd_fd, BLKPBSZGET, &blksize); + BUG_ON(ret); + + return blksize >> 9; +} + +sector_t get_capacity(struct gendisk *disk) +{ + struct block_device *bdev = + container_of(disk, struct block_device, __bd_disk); + struct stat statbuf; + u64 bytes; + int ret; + + ret = fstat(bdev->bd_fd, &statbuf); + BUG_ON(ret); + + if (!S_ISBLK(statbuf.st_mode)) + return statbuf.st_size >> 9; + + ret = ioctl(bdev->bd_fd, BLKGETSIZE64, &bytes); + BUG_ON(ret); + + return bytes >> 9; +} + +void blkdev_put(struct block_device *bdev, fmode_t mode) +{ + fdatasync(bdev->bd_fd); + close(bdev->bd_fd); + free(bdev); +} + +struct block_device *blkdev_get_by_path(const char *path, fmode_t mode, + void *holder) +{ + struct block_device *bdev; + int flags = O_DIRECT; + + if ((mode & (FMODE_READ|FMODE_WRITE)) == (FMODE_READ|FMODE_WRITE)) + flags = O_RDWR; + else if (mode & FMODE_READ) + flags = O_RDONLY; + else if (mode & FMODE_WRITE) + flags = O_WRONLY; + + if (mode & FMODE_EXCL) + flags |= O_EXCL; + + bdev = malloc(sizeof(*bdev)); + memset(bdev, 0, sizeof(*bdev)); + + strncpy(bdev->name, path, sizeof(bdev->name)); + bdev->name[sizeof(bdev->name) - 1] = '\0'; + + bdev->bd_fd = open(path, flags); + bdev->bd_holder = holder; + bdev->bd_disk = &bdev->__bd_disk; + + BUG_ON(bdev->bd_fd < 0); + + return bdev; +} + +void bdput(struct block_device *bdev) +{ + BUG(); +} + +struct block_device *lookup_bdev(const char *path) +{ + return ERR_PTR(-EINVAL); +} diff --git a/linux/completion.c b/linux/completion.c new file mode 100644 index 0000000..fcc77a0 --- /dev/null +++ b/linux/completion.c @@ -0,0 +1,311 @@ +/* + * Generic wait-for-completion handler; + * + * It differs from semaphores in that their default case is the opposite, + * wait_for_completion default blocks whereas semaphore default non-block. The + * interface also makes it easy to 'complete' multiple waiting threads, + * something which isn't entirely natural for semaphores. + * + * But more importantly, the primitive documents the usage. Semaphores would + * typically be used for exclusion which gives rise to priority inversion. + * Waiting for completion is a typically sync point, but not an exclusion point. + */ + +#include <linux/sched.h> +#include <linux/completion.h> + +/** + * complete: - signals a single thread waiting on this completion + * @x: holds the state of this particular completion + * + * This will wake up a single thread waiting on this completion. Threads will be + * awakened in the same order in which they were queued. + * + * See also complete_all(), wait_for_completion() and related routines. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. + */ +void complete(struct completion *x) +{ + unsigned long flags; + + spin_lock_irqsave(&x->wait.lock, flags); + x->done++; + __wake_up_locked(&x->wait, TASK_NORMAL, 1); + spin_unlock_irqrestore(&x->wait.lock, flags); +} +EXPORT_SYMBOL(complete); + +/** + * complete_all: - signals all threads waiting on this completion + * @x: holds the state of this particular completion + * + * This will wake up all threads waiting on this particular completion event. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. + */ +void complete_all(struct completion *x) +{ + unsigned long flags; + + spin_lock_irqsave(&x->wait.lock, flags); + x->done += UINT_MAX/2; + __wake_up_locked(&x->wait, TASK_NORMAL, 0); + spin_unlock_irqrestore(&x->wait.lock, flags); +} +EXPORT_SYMBOL(complete_all); + +static inline long __sched +do_wait_for_common(struct completion *x, + long (*action)(long), long timeout, int state) +{ + if (!x->done) { + DECLARE_WAITQUEUE(wait, current); + + __add_wait_queue_tail_exclusive(&x->wait, &wait); + do { + __set_current_state(state); + spin_unlock_irq(&x->wait.lock); + timeout = action(timeout); + spin_lock_irq(&x->wait.lock); + } while (!x->done && timeout); + __remove_wait_queue(&x->wait, &wait); + if (!x->done) + return timeout; + } + x->done--; + return timeout ?: 1; +} + +static inline long __sched +__wait_for_common(struct completion *x, + long (*action)(long), long timeout, int state) +{ + might_sleep(); + + spin_lock_irq(&x->wait.lock); + timeout = do_wait_for_common(x, action, timeout, state); + spin_unlock_irq(&x->wait.lock); + return timeout; +} + +static long __sched +wait_for_common(struct completion *x, long timeout, int state) +{ + return __wait_for_common(x, schedule_timeout, timeout, state); +} + +static long __sched +wait_for_common_io(struct completion *x, long timeout, int state) +{ + return __wait_for_common(x, io_schedule_timeout, timeout, state); +} + +/** + * wait_for_completion: - waits for completion of a task + * @x: holds the state of this particular completion + * + * This waits to be signaled for completion of a specific task. It is NOT + * interruptible and there is no timeout. + * + * See also similar routines (i.e. wait_for_completion_timeout()) with timeout + * and interrupt capability. Also see complete(). + */ +void __sched wait_for_completion(struct completion *x) +{ + wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(wait_for_completion); + +/** + * wait_for_completion_timeout: - waits for completion of a task (w/timeout) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be signaled or for a + * specified timeout to expire. The timeout is in jiffies. It is not + * interruptible. + * + * Return: 0 if timed out, and positive (at least 1, or number of jiffies left + * till timeout) if completed. + */ +unsigned long __sched +wait_for_completion_timeout(struct completion *x, unsigned long timeout) +{ + return wait_for_common(x, timeout, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(wait_for_completion_timeout); + +/** + * wait_for_completion_io: - waits for completion of a task + * @x: holds the state of this particular completion + * + * This waits to be signaled for completion of a specific task. It is NOT + * interruptible and there is no timeout. The caller is accounted as waiting + * for IO (which traditionally means blkio only). + */ +void __sched wait_for_completion_io(struct completion *x) +{ + wait_for_common_io(x, MAX_SCHEDULE_TIMEOUT, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(wait_for_completion_io); + +/** + * wait_for_completion_io_timeout: - waits for completion of a task (w/timeout) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be signaled or for a + * specified timeout to expire. The timeout is in jiffies. It is not + * interruptible. The caller is accounted as waiting for IO (which traditionally + * means blkio only). + * + * Return: 0 if timed out, and positive (at least 1, or number of jiffies left + * till timeout) if completed. + */ +unsigned long __sched +wait_for_completion_io_timeout(struct completion *x, unsigned long timeout) +{ + return wait_for_common_io(x, timeout, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(wait_for_completion_io_timeout); + +/** + * wait_for_completion_interruptible: - waits for completion of a task (w/intr) + * @x: holds the state of this particular completion + * + * This waits for completion of a specific task to be signaled. It is + * interruptible. + * + * Return: -ERESTARTSYS if interrupted, 0 if completed. + */ +int __sched wait_for_completion_interruptible(struct completion *x) +{ + wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_INTERRUPTIBLE); + return 0; +} +EXPORT_SYMBOL(wait_for_completion_interruptible); + +/** + * wait_for_completion_interruptible_timeout: - waits for completion (w/(to,intr)) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be signaled or for a + * specified timeout to expire. It is interruptible. The timeout is in jiffies. + * + * Return: -ERESTARTSYS if interrupted, 0 if timed out, positive (at least 1, + * or number of jiffies left till timeout) if completed. + */ +long __sched +wait_for_completion_interruptible_timeout(struct completion *x, + unsigned long timeout) +{ + return wait_for_common(x, timeout, TASK_INTERRUPTIBLE); +} +EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); + +/** + * wait_for_completion_killable: - waits for completion of a task (killable) + * @x: holds the state of this particular completion + * + * This waits to be signaled for completion of a specific task. It can be + * interrupted by a kill signal. + * + * Return: -ERESTARTSYS if interrupted, 0 if completed. + */ +int __sched wait_for_completion_killable(struct completion *x) +{ + wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_KILLABLE); + return 0; +} +EXPORT_SYMBOL(wait_for_completion_killable); + +/** + * wait_for_completion_killable_timeout: - waits for completion of a task (w/(to,killable)) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be + * signaled or for a specified timeout to expire. It can be + * interrupted by a kill signal. The timeout is in jiffies. + * + * Return: -ERESTARTSYS if interrupted, 0 if timed out, positive (at least 1, + * or number of jiffies left till timeout) if completed. + */ +long __sched +wait_for_completion_killable_timeout(struct completion *x, + unsigned long timeout) +{ + return wait_for_common(x, timeout, TASK_KILLABLE); +} +EXPORT_SYMBOL(wait_for_completion_killable_timeout); + +/** + * try_wait_for_completion - try to decrement a completion without blocking + * @x: completion structure + * + * Return: 0 if a decrement cannot be done without blocking + * 1 if a decrement succeeded. + * + * If a completion is being used as a counting completion, + * attempt to decrement the counter without blocking. This + * enables us to avoid waiting if the resource the completion + * is protecting is not available. + */ +bool try_wait_for_completion(struct completion *x) +{ + unsigned long flags; + int ret = 1; + + /* + * Since x->done will need to be locked only + * in the non-blocking case, we check x->done + * first without taking the lock so we can + * return early in the blocking case. + */ + if (!READ_ONCE(x->done)) + return 0; + + spin_lock_irqsave(&x->wait.lock, flags); + if (!x->done) + ret = 0; + else + x->done--; + spin_unlock_irqrestore(&x->wait.lock, flags); + return ret; +} +EXPORT_SYMBOL(try_wait_for_completion); + +/** + * completion_done - Test to see if a completion has any waiters + * @x: completion structure + * + * Return: 0 if there are waiters (wait_for_completion() in progress) + * 1 if there are no waiters. + * + */ +bool completion_done(struct completion *x) +{ + if (!READ_ONCE(x->done)) + return false; + + /* + * If ->done, we need to wait for complete() to release ->wait.lock + * otherwise we can end up freeing the completion before complete() + * is done referencing it. + * + * The RMB pairs with complete()'s RELEASE of ->wait.lock and orders + * the loads of ->done and ->wait.lock such that we cannot observe + * the lock before complete() acquires it while observing the ->done + * after it's acquired the lock. + */ + smp_rmb(); + //spin_unlock_wait(&x->wait.lock); + spin_lock(&x->wait.lock); + spin_unlock(&x->wait.lock); + return true; +} +EXPORT_SYMBOL(completion_done); diff --git a/linux/crypto/algapi.c b/linux/crypto/algapi.c new file mode 100644 index 0000000..5e8e97b --- /dev/null +++ b/linux/crypto/algapi.c @@ -0,0 +1,315 @@ +/* + * Cryptographic API for algorithms (i.e., low-level API). + * + * Copyright (c) 2006 Herbert Xu <herbert@gondor.apana.org.au> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ + +#include <linux/byteorder.h> +#include <linux/err.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/printk.h> +#include <linux/rtnetlink.h> +#include <linux/slab.h> +#include <linux/string.h> + +#include "internal.h" + +static inline int crypto_set_driver_name(struct crypto_alg *alg) +{ + static const char suffix[] = "-generic"; + char *driver_name = alg->cra_driver_name; + int len; + + if (*driver_name) + return 0; + + len = strlcpy(driver_name, alg->cra_name, CRYPTO_MAX_ALG_NAME); + if (len + sizeof(suffix) > CRYPTO_MAX_ALG_NAME) + return -ENAMETOOLONG; + + memcpy(driver_name + len, suffix, sizeof(suffix)); + return 0; +} + +static int crypto_check_alg(struct crypto_alg *alg) +{ + if (alg->cra_alignmask & (alg->cra_alignmask + 1)) + return -EINVAL; + + if (alg->cra_blocksize > PAGE_SIZE / 8) + return -EINVAL; + + if (alg->cra_priority < 0) + return -EINVAL; + + atomic_set(&alg->cra_refcnt, 1); + + return crypto_set_driver_name(alg); +} + +static int __crypto_register_alg(struct crypto_alg *alg) +{ + struct crypto_alg *q; + int ret = -EAGAIN; + + INIT_LIST_HEAD(&alg->cra_users); + + ret = -EEXIST; + + list_for_each_entry(q, &crypto_alg_list, cra_list) { + if (q == alg) + goto err; + + if (!strcmp(q->cra_driver_name, alg->cra_name) || + !strcmp(q->cra_name, alg->cra_driver_name)) + goto err; + } + + list_add(&alg->cra_list, &crypto_alg_list); + return 0; +err: + return ret; +} + +void crypto_remove_final(struct list_head *list) +{ + struct crypto_alg *alg; + struct crypto_alg *n; + + list_for_each_entry_safe(alg, n, list, cra_list) { + list_del_init(&alg->cra_list); + crypto_alg_put(alg); + } +} + +int crypto_register_alg(struct crypto_alg *alg) +{ + int err; + + err = crypto_check_alg(alg); + if (err) + return err; + + down_write(&crypto_alg_sem); + err = __crypto_register_alg(alg); + up_write(&crypto_alg_sem); + + return err; +} + +static int crypto_remove_alg(struct crypto_alg *alg, struct list_head *list) +{ + if (unlikely(list_empty(&alg->cra_list))) + return -ENOENT; + + list_del_init(&alg->cra_list); + return 0; +} + +int crypto_unregister_alg(struct crypto_alg *alg) +{ + int ret; + LIST_HEAD(list); + + down_write(&crypto_alg_sem); + ret = crypto_remove_alg(alg, &list); + up_write(&crypto_alg_sem); + + if (ret) + return ret; + + BUG_ON(atomic_read(&alg->cra_refcnt) != 1); + if (alg->cra_destroy) + alg->cra_destroy(alg); + + crypto_remove_final(&list); + return 0; +} + +int crypto_register_algs(struct crypto_alg *algs, int count) +{ + int i, ret; + + for (i = 0; i < count; i++) { + ret = crypto_register_alg(&algs[i]); + if (ret) + goto err; + } + + return 0; + +err: + for (--i; i >= 0; --i) + crypto_unregister_alg(&algs[i]); + + return ret; +} + +int crypto_unregister_algs(struct crypto_alg *algs, int count) +{ + int i, ret; + + for (i = 0; i < count; i++) { + ret = crypto_unregister_alg(&algs[i]); + if (ret) + pr_err("Failed to unregister %s %s: %d\n", + algs[i].cra_driver_name, algs[i].cra_name, ret); + } + + return 0; +} + +struct crypto_attr_type *crypto_get_attr_type(struct rtattr **tb) +{ + struct rtattr *rta = tb[0]; + struct crypto_attr_type *algt; + + if (!rta) + return ERR_PTR(-ENOENT); + if (RTA_PAYLOAD(rta) < sizeof(*algt)) + return ERR_PTR(-EINVAL); + if (rta->rta_type != CRYPTOA_TYPE) + return ERR_PTR(-EINVAL); + + algt = RTA_DATA(rta); + + return algt; +} + +int crypto_check_attr_type(struct rtattr **tb, u32 type) +{ + struct crypto_attr_type *algt; + + algt = crypto_get_attr_type(tb); + if (IS_ERR(algt)) + return PTR_ERR(algt); + + if ((algt->type ^ type) & algt->mask) + return -EINVAL; + + return 0; +} + +const char *crypto_attr_alg_name(struct rtattr *rta) +{ + struct crypto_attr_alg *alga; + + if (!rta) + return ERR_PTR(-ENOENT); + if (RTA_PAYLOAD(rta) < sizeof(*alga)) + return ERR_PTR(-EINVAL); + if (rta->rta_type != CRYPTOA_ALG) + return ERR_PTR(-EINVAL); + + alga = RTA_DATA(rta); + alga->name[CRYPTO_MAX_ALG_NAME - 1] = 0; + + return alga->name; +} + +struct crypto_alg *crypto_attr_alg2(struct rtattr *rta, + const struct crypto_type *frontend, + u32 type, u32 mask) +{ + const char *name; + + name = crypto_attr_alg_name(rta); + if (IS_ERR(name)) + return ERR_CAST(name); + + return crypto_find_alg(name, frontend, type, mask); +} + +int crypto_attr_u32(struct rtattr *rta, u32 *num) +{ + struct crypto_attr_u32 *nu32; + + if (!rta) + return -ENOENT; + if (RTA_PAYLOAD(rta) < sizeof(*nu32)) + return -EINVAL; + if (rta->rta_type != CRYPTOA_U32) + return -EINVAL; + + nu32 = RTA_DATA(rta); + *num = nu32->num; + + return 0; +} + +static inline void crypto_inc_byte(u8 *a, unsigned int size) +{ + u8 *b = (a + size); + u8 c; + + for (; size; size--) { + c = *--b + 1; + *b = c; + if (c) + break; + } +} + +void crypto_inc(u8 *a, unsigned int size) +{ + __be32 *b = (__be32 *)(a + size); + u32 c; + + for (; size >= 4; size -= 4) { + c = be32_to_cpu(*--b) + 1; + *b = cpu_to_be32(c); + if (c) + return; + } + + crypto_inc_byte(a, size); +} + +static inline void crypto_xor_byte(u8 *a, const u8 *b, unsigned int size) +{ + for (; size; size--) + *a++ ^= *b++; +} + +void crypto_xor(u8 *dst, const u8 *src, unsigned int size) +{ + u32 *a = (u32 *)dst; + u32 *b = (u32 *)src; + + for (; size >= 4; size -= 4) + *a++ ^= *b++; + + crypto_xor_byte((u8 *)a, (u8 *)b, size); +} + +unsigned int crypto_alg_extsize(struct crypto_alg *alg) +{ + return alg->cra_ctxsize + + (alg->cra_alignmask & ~(crypto_tfm_ctx_alignment() - 1)); +} + +int crypto_type_has_alg(const char *name, const struct crypto_type *frontend, + u32 type, u32 mask) +{ + int ret = 0; + struct crypto_alg *alg = crypto_find_alg(name, frontend, type, mask); + + if (!IS_ERR(alg)) { + crypto_alg_put(alg); + ret = 1; + } + + return ret; +} + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Cryptographic algorithms API"); diff --git a/linux/crypto/api.c b/linux/crypto/api.c new file mode 100644 index 0000000..513a48a --- /dev/null +++ b/linux/crypto/api.c @@ -0,0 +1,326 @@ +/* + * Scatterlist Cryptographic API. + * + * Copyright (c) 2002 James Morris <jmorris@intercode.com.au> + * Copyright (c) 2002 David S. Miller (davem@redhat.com) + * Copyright (c) 2005 Herbert Xu <herbert@gondor.apana.org.au> + * + * Portions derived from Cryptoapi, by Alexander Kjeldaas <astor@fast.no> + * and Nettle, by Niels Möller. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ + +#include <linux/device.h> +#include <linux/err.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/param.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/string.h> +#include "internal.h" + +LIST_HEAD(crypto_alg_list); +DECLARE_RWSEM(crypto_alg_sem); + +static struct crypto_alg *__crypto_alg_lookup(const char *name, u32 type, + u32 mask) +{ + struct crypto_alg *q, *alg = NULL; + int best = -2; + + list_for_each_entry(q, &crypto_alg_list, cra_list) { + int exact, fuzzy; + + if ((q->cra_flags ^ type) & mask) + continue; + + exact = !strcmp(q->cra_driver_name, name); + fuzzy = !strcmp(q->cra_name, name); + if (!exact && !(fuzzy && q->cra_priority > best)) + continue; + + if (unlikely(!crypto_alg_get(q))) + continue; + + best = q->cra_priority; + if (alg) + crypto_alg_put(alg); + alg = q; + + if (exact) + break; + } + + return alg; +} + +struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask) +{ + struct crypto_alg *alg; + + /* + * If the internal flag is set for a cipher, require a caller to + * to invoke the cipher with the internal flag to use that cipher. + * Also, if a caller wants to allocate a cipher that may or may + * not be an internal cipher, use type | CRYPTO_ALG_INTERNAL and + * !(mask & CRYPTO_ALG_INTERNAL). + */ + if (!((type | mask) & CRYPTO_ALG_INTERNAL)) + mask |= CRYPTO_ALG_INTERNAL; + + down_read(&crypto_alg_sem); + alg = __crypto_alg_lookup(name, type, mask); + up_read(&crypto_alg_sem); + + return alg ?: ERR_PTR(-ENOENT); +} + +static int crypto_init_ops(struct crypto_tfm *tfm, u32 type, u32 mask) +{ + const struct crypto_type *type_obj = tfm->__crt_alg->cra_type; + + if (type_obj) + return type_obj->init(tfm, type, mask); + + switch (crypto_tfm_alg_type(tfm)) { + case CRYPTO_ALG_TYPE_CIPHER: + return crypto_init_cipher_ops(tfm); + default: + break; + } + + BUG(); + return -EINVAL; +} + +static void crypto_exit_ops(struct crypto_tfm *tfm) +{ + const struct crypto_type *type = tfm->__crt_alg->cra_type; + + if (type) { + if (tfm->exit) + tfm->exit(tfm); + return; + } + + switch (crypto_tfm_alg_type(tfm)) { + case CRYPTO_ALG_TYPE_CIPHER: + crypto_exit_cipher_ops(tfm); + break; + + default: + BUG(); + } +} + +static unsigned int crypto_ctxsize(struct crypto_alg *alg, u32 type, u32 mask) +{ + const struct crypto_type *type_obj = alg->cra_type; + unsigned int len; + + len = alg->cra_alignmask & ~(crypto_tfm_ctx_alignment() - 1); + if (type_obj) + return len + type_obj->ctxsize(alg, type, mask); + + switch (alg->cra_flags & CRYPTO_ALG_TYPE_MASK) { + default: + BUG(); + + case CRYPTO_ALG_TYPE_CIPHER: + len += crypto_cipher_ctxsize(alg); + break; + } + + return len; +} + +struct crypto_tfm *__crypto_alloc_tfm(struct crypto_alg *alg, u32 type, + u32 mask) +{ + struct crypto_tfm *tfm = NULL; + unsigned int tfm_size; + int err = -ENOMEM; + + tfm_size = sizeof(*tfm) + crypto_ctxsize(alg, type, mask); + tfm = kzalloc(tfm_size, GFP_KERNEL); + if (tfm == NULL) + goto out_err; + + tfm->__crt_alg = alg; + + err = crypto_init_ops(tfm, type, mask); + if (err) + goto out_free_tfm; + + if (!tfm->exit && alg->cra_init && (err = alg->cra_init(tfm))) + goto cra_init_failed; + + goto out; + +cra_init_failed: + crypto_exit_ops(tfm); +out_free_tfm: + kfree(tfm); +out_err: + tfm = ERR_PTR(err); +out: + return tfm; +} + +/* + * crypto_alloc_base - Locate algorithm and allocate transform + * @alg_name: Name of algorithm + * @type: Type of algorithm + * @mask: Mask for type comparison + * + * This function should not be used by new algorithm types. + * Please use crypto_alloc_tfm instead. + * + * crypto_alloc_base() will first attempt to locate an already loaded + * algorithm. If that fails and the kernel supports dynamically loadable + * modules, it will then attempt to load a module of the same name or + * alias. If that fails it will send a query to any loaded crypto manager + * to construct an algorithm on the fly. A refcount is grabbed on the + * algorithm which is then associated with the new transform. + * + * The returned transform is of a non-determinate type. Most people + * should use one of the more specific allocation functions such as + * crypto_alloc_blkcipher. + * + * In case of error the return value is an error pointer. + */ +struct crypto_tfm *crypto_alloc_base(const char *alg_name, u32 type, u32 mask) +{ + struct crypto_alg *alg; + struct crypto_tfm *tfm; + + alg = crypto_alg_mod_lookup(alg_name, type, mask); + if (IS_ERR(alg)) { + fprintf(stderr, "unknown cipher %s\n", alg_name); + return ERR_CAST(alg); + } + + tfm = __crypto_alloc_tfm(alg, type, mask); + if (IS_ERR(tfm)) { + crypto_alg_put(alg); + return tfm; + } + + return tfm; +} + +void *crypto_create_tfm(struct crypto_alg *alg, + const struct crypto_type *frontend) +{ + char *mem; + struct crypto_tfm *tfm = NULL; + unsigned int tfmsize; + unsigned int total; + int err = -ENOMEM; + + tfmsize = frontend->tfmsize; + total = tfmsize + sizeof(*tfm) + frontend->extsize(alg); + + mem = kzalloc(total, GFP_KERNEL); + if (mem == NULL) + goto out_err; + + tfm = (struct crypto_tfm *)(mem + tfmsize); + tfm->__crt_alg = alg; + + err = frontend->init_tfm(tfm); + if (err) + goto out_free_tfm; + + if (!tfm->exit && alg->cra_init && (err = alg->cra_init(tfm))) + goto cra_init_failed; + + goto out; + +cra_init_failed: + crypto_exit_ops(tfm); +out_free_tfm: + kfree(mem); +out_err: + mem = ERR_PTR(err); +out: + return mem; +} + +struct crypto_alg *crypto_find_alg(const char *alg_name, + const struct crypto_type *frontend, + u32 type, u32 mask) +{ + struct crypto_alg *(*lookup)(const char *name, u32 type, u32 mask) = + crypto_alg_mod_lookup; + + if (frontend) { + type &= frontend->maskclear; + mask &= frontend->maskclear; + type |= frontend->type; + mask |= frontend->maskset; + + if (frontend->lookup) + lookup = frontend->lookup; + } + + return lookup(alg_name, type, mask); +} + +void *crypto_alloc_tfm(const char *alg_name, + const struct crypto_type *frontend, u32 type, u32 mask) +{ + struct crypto_alg *alg; + void *tfm; + + alg = crypto_find_alg(alg_name, frontend, type, mask); + if (IS_ERR(alg)) + return ERR_CAST(alg); + + tfm = crypto_create_tfm(alg, frontend); + if (IS_ERR(tfm)) { + crypto_alg_put(alg); + return tfm; + } + + return tfm; +} + +void crypto_destroy_tfm(void *mem, struct crypto_tfm *tfm) +{ + struct crypto_alg *alg; + + if (unlikely(!mem)) + return; + + alg = tfm->__crt_alg; + + if (!tfm->exit && alg->cra_exit) + alg->cra_exit(tfm); + crypto_exit_ops(tfm); + crypto_alg_put(alg); + kzfree(mem); +} + +int crypto_has_alg(const char *name, u32 type, u32 mask) +{ + int ret = 0; + struct crypto_alg *alg = crypto_alg_mod_lookup(name, type, mask); + + if (!IS_ERR(alg)) { + crypto_alg_put(alg); + ret = 1; + } + + return ret; +} + +MODULE_DESCRIPTION("Cryptographic core API"); +MODULE_LICENSE("GPL"); diff --git a/linux/crypto/cipher.c b/linux/crypto/cipher.c new file mode 100644 index 0000000..6f47ac6 --- /dev/null +++ b/linux/crypto/cipher.c @@ -0,0 +1,123 @@ +/* + * Cryptographic API. + * + * Cipher operations. + * + * Copyright (c) 2002 James Morris <jmorris@intercode.com.au> + * Copyright (c) 2005 Herbert Xu <herbert@gondor.apana.org.au> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ + +#include <linux/kernel.h> +#include <linux/crypto.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/string.h> +#include "internal.h" + +static int setkey_unaligned(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct cipher_alg *cia = &tfm->__crt_alg->cra_cipher; + unsigned long alignmask = crypto_tfm_alg_alignmask(tfm); + int ret; + u8 *buffer, *alignbuffer; + unsigned long absize; + + absize = keylen + alignmask; + buffer = kmalloc(absize, GFP_ATOMIC); + if (!buffer) + return -ENOMEM; + + alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); + memcpy(alignbuffer, key, keylen); + ret = cia->cia_setkey(tfm, alignbuffer, keylen); + memset(alignbuffer, 0, keylen); + kfree(buffer); + return ret; + +} + +static int setkey_default(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct cipher_alg *cia = &tfm->__crt_alg->cra_cipher; + unsigned long alignmask = crypto_tfm_alg_alignmask(tfm); + + tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK; + if (keylen < cia->cia_min_keysize || keylen > cia->cia_max_keysize) { + tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + if ((unsigned long)key & alignmask) + return setkey_unaligned(tfm, key, keylen); + + return cia->cia_setkey(tfm, key, keylen); +} + +static void cipher_crypt_unaligned(void (*fn)(struct crypto_tfm *, u8 *, + const u8 *), + struct crypto_tfm *tfm, + u8 *dst, const u8 *src) +{ + unsigned long alignmask = crypto_tfm_alg_alignmask(tfm); + unsigned int size = crypto_tfm_alg_blocksize(tfm); + u8 buffer[size + alignmask]; + u8 *tmp = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); + + memcpy(tmp, src, size); + fn(tfm, tmp, tmp); + memcpy(dst, tmp, size); +} + +static void cipher_encrypt_unaligned(struct crypto_tfm *tfm, + u8 *dst, const u8 *src) +{ + unsigned long alignmask = crypto_tfm_alg_alignmask(tfm); + struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher; + + if (unlikely(((unsigned long)dst | (unsigned long)src) & alignmask)) { + cipher_crypt_unaligned(cipher->cia_encrypt, tfm, dst, src); + return; + } + + cipher->cia_encrypt(tfm, dst, src); +} + +static void cipher_decrypt_unaligned(struct crypto_tfm *tfm, + u8 *dst, const u8 *src) +{ + unsigned long alignmask = crypto_tfm_alg_alignmask(tfm); + struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher; + + if (unlikely(((unsigned long)dst | (unsigned long)src) & alignmask)) { + cipher_crypt_unaligned(cipher->cia_decrypt, tfm, dst, src); + return; + } + + cipher->cia_decrypt(tfm, dst, src); +} + +int crypto_init_cipher_ops(struct crypto_tfm *tfm) +{ + struct cipher_tfm *ops = &tfm->crt_cipher; + struct cipher_alg *cipher = &tfm->__crt_alg->cra_cipher; + + ops->cit_setkey = setkey_default; + ops->cit_encrypt_one = crypto_tfm_alg_alignmask(tfm) ? + cipher_encrypt_unaligned : cipher->cia_encrypt; + ops->cit_decrypt_one = crypto_tfm_alg_alignmask(tfm) ? + cipher_decrypt_unaligned : cipher->cia_decrypt; + + return 0; +} + +void crypto_exit_cipher_ops(struct crypto_tfm *tfm) +{ +} diff --git a/linux/crypto/internal.h b/linux/crypto/internal.h new file mode 100644 index 0000000..b00dcea --- /dev/null +++ b/linux/crypto/internal.h @@ -0,0 +1,78 @@ +/* + * Cryptographic API. + * + * Copyright (c) 2002 James Morris <jmorris@intercode.com.au> + * Copyright (c) 2005 Herbert Xu <herbert@gondor.apana.org.au> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ +#ifndef _CRYPTO_INTERNAL_H +#define _CRYPTO_INTERNAL_H + +#include <crypto/algapi.h> +#include <linux/completion.h> +#include <linux/mm.h> +#include <linux/list.h> +#include <linux/kernel.h> +#include <linux/notifier.h> +#include <linux/rwsem.h> +#include <linux/slab.h> + +struct crypto_instance; +struct crypto_template; + +struct crypto_larval { + struct crypto_alg alg; + struct crypto_alg *adult; + struct completion completion; + u32 mask; +}; + +extern struct list_head crypto_alg_list; +extern struct rw_semaphore crypto_alg_sem; + +static inline unsigned int crypto_cipher_ctxsize(struct crypto_alg *alg) +{ + return alg->cra_ctxsize; +} + +int crypto_init_cipher_ops(struct crypto_tfm *tfm); +void crypto_exit_cipher_ops(struct crypto_tfm *tfm); + +void crypto_remove_final(struct list_head *list); +struct crypto_tfm *__crypto_alloc_tfm(struct crypto_alg *alg, u32 type, + u32 mask); +void *crypto_create_tfm(struct crypto_alg *alg, + const struct crypto_type *frontend); +struct crypto_alg *crypto_find_alg(const char *alg_name, + const struct crypto_type *frontend, + u32 type, u32 mask); +void *crypto_alloc_tfm(const char *alg_name, + const struct crypto_type *frontend, u32 type, u32 mask); + +int crypto_register_notifier(struct notifier_block *nb); +int crypto_unregister_notifier(struct notifier_block *nb); + +unsigned int crypto_alg_extsize(struct crypto_alg *alg); + +int crypto_type_has_alg(const char *name, const struct crypto_type *frontend, + u32 type, u32 mask); + +static inline struct crypto_alg *crypto_alg_get(struct crypto_alg *alg) +{ + atomic_inc(&alg->cra_refcnt); + return alg; +} + +static inline void crypto_alg_put(struct crypto_alg *alg) +{ + if (atomic_dec_and_test(&alg->cra_refcnt) && alg->cra_destroy) + alg->cra_destroy(alg); +} + +#endif /* _CRYPTO_INTERNAL_H */ + diff --git a/linux/crypto/sha1_generic.c b/linux/crypto/sha1_generic.c new file mode 100644 index 0000000..b0b9cd1 --- /dev/null +++ b/linux/crypto/sha1_generic.c @@ -0,0 +1,92 @@ +/* + * Cryptographic API. + * + * SHA1 Secure Hash Algorithm. + * + * Derived from cryptoapi implementation, adapted for in-place + * scatterlist interface. + * + * Copyright (c) Alan Smithee. + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) Jean-Francois Dive <jef@linuxbe.org> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ +#include <crypto/internal/hash.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <crypto/sha.h> +#include <crypto/sha1_base.h> +#include <asm/byteorder.h> + +const u8 sha1_zero_message_hash[SHA1_DIGEST_SIZE] = { + 0xda, 0x39, 0xa3, 0xee, 0x5e, 0x6b, 0x4b, 0x0d, + 0x32, 0x55, 0xbf, 0xef, 0x95, 0x60, 0x18, 0x90, + 0xaf, 0xd8, 0x07, 0x09 +}; + +static void sha1_generic_block_fn(struct sha1_state *sst, u8 const *src, + int blocks) +{ + u32 temp[SHA_WORKSPACE_WORDS]; + + while (blocks--) { + sha_transform(sst->state, src, temp); + src += SHA1_BLOCK_SIZE; + } + memzero_explicit(temp, sizeof(temp)); +} + +int crypto_sha1_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha1_base_do_update(desc, data, len, sha1_generic_block_fn); +} + +static int sha1_final(struct shash_desc *desc, u8 *out) +{ + sha1_base_do_finalize(desc, sha1_generic_block_fn); + return sha1_base_finish(desc, out); +} + +int crypto_sha1_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + sha1_base_do_update(desc, data, len, sha1_generic_block_fn); + return sha1_final(desc, out); +} + +static struct shash_alg alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_base_init, + .update = crypto_sha1_update, + .final = sha1_final, + .finup = crypto_sha1_finup, + .descsize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name= "sha1-generic", + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int __init sha1_generic_mod_init(void) +{ + return crypto_register_shash(&alg); +} + +static void __exit sha1_generic_mod_fini(void) +{ + crypto_unregister_shash(&alg); +} + +module_init(sha1_generic_mod_init); +module_exit(sha1_generic_mod_fini); diff --git a/linux/crypto/shash.c b/linux/crypto/shash.c new file mode 100644 index 0000000..406ddfe --- /dev/null +++ b/linux/crypto/shash.c @@ -0,0 +1,294 @@ +/* + * Synchronous Cryptographic Hash operations. + * + * Copyright (c) 2008 Herbert Xu <herbert@gondor.apana.org.au> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ + +#include <crypto/internal/hash.h> +#include <linux/err.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/printk.h> +#include <linux/slab.h> + +#include "internal.h" + +static int shash_no_setkey(struct crypto_shash *tfm, const u8 *key, + unsigned int keylen) +{ + return -ENOSYS; +} + +static int shash_setkey_unaligned(struct crypto_shash *tfm, const u8 *key, + unsigned int keylen) +{ + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + unsigned long absize; + u8 *buffer, *alignbuffer; + int err; + + absize = keylen + (alignmask & ~(crypto_tfm_ctx_alignment() - 1)); + buffer = kmalloc(absize, GFP_KERNEL); + if (!buffer) + return -ENOMEM; + + alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); + memcpy(alignbuffer, key, keylen); + err = shash->setkey(tfm, alignbuffer, keylen); + kzfree(buffer); + return err; +} + +int crypto_shash_setkey(struct crypto_shash *tfm, const u8 *key, + unsigned int keylen) +{ + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + + if ((unsigned long)key & alignmask) + return shash_setkey_unaligned(tfm, key, keylen); + + return shash->setkey(tfm, key, keylen); +} + +static inline unsigned int shash_align_buffer_size(unsigned len, + unsigned long mask) +{ + typedef u8 __attribute__ ((aligned)) u8_aligned; + return len + (mask & ~(__alignof__(u8_aligned) - 1)); +} + +static int shash_update_unaligned(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct crypto_shash *tfm = desc->tfm; + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + unsigned int unaligned_len = alignmask + 1 - + ((unsigned long)data & alignmask); + u8 ubuf[shash_align_buffer_size(unaligned_len, alignmask)] + __attribute__ ((aligned)); + u8 *buf = PTR_ALIGN(&ubuf[0], alignmask + 1); + int err; + + if (unaligned_len > len) + unaligned_len = len; + + memcpy(buf, data, unaligned_len); + err = shash->update(desc, buf, unaligned_len); + memset(buf, 0, unaligned_len); + + return err ?: + shash->update(desc, data + unaligned_len, len - unaligned_len); +} + +int crypto_shash_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct crypto_shash *tfm = desc->tfm; + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + + if ((unsigned long)data & alignmask) + return shash_update_unaligned(desc, data, len); + + return shash->update(desc, data, len); +} + +static int shash_final_unaligned(struct shash_desc *desc, u8 *out) +{ + struct crypto_shash *tfm = desc->tfm; + unsigned long alignmask = crypto_shash_alignmask(tfm); + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned int ds = crypto_shash_digestsize(tfm); + u8 ubuf[shash_align_buffer_size(ds, alignmask)] + __attribute__ ((aligned)); + u8 *buf = PTR_ALIGN(&ubuf[0], alignmask + 1); + int err; + + err = shash->final(desc, buf); + if (err) + goto out; + + memcpy(out, buf, ds); + +out: + memset(buf, 0, ds); + return err; +} + +int crypto_shash_final(struct shash_desc *desc, u8 *out) +{ + struct crypto_shash *tfm = desc->tfm; + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + + if ((unsigned long)out & alignmask) + return shash_final_unaligned(desc, out); + + return shash->final(desc, out); +} + +static int shash_finup_unaligned(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return crypto_shash_update(desc, data, len) ?: + crypto_shash_final(desc, out); +} + +int crypto_shash_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + struct crypto_shash *tfm = desc->tfm; + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + + if (((unsigned long)data | (unsigned long)out) & alignmask) + return shash_finup_unaligned(desc, data, len, out); + + return shash->finup(desc, data, len, out); +} + +static int shash_digest_unaligned(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return crypto_shash_init(desc) ?: + crypto_shash_finup(desc, data, len, out); +} + +int crypto_shash_digest(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + struct crypto_shash *tfm = desc->tfm; + struct shash_alg *shash = crypto_shash_alg(tfm); + unsigned long alignmask = crypto_shash_alignmask(tfm); + + if (((unsigned long)data | (unsigned long)out) & alignmask) + return shash_digest_unaligned(desc, data, len, out); + + return shash->digest(desc, data, len, out); +} + +static int shash_default_export(struct shash_desc *desc, void *out) +{ + memcpy(out, shash_desc_ctx(desc), crypto_shash_descsize(desc->tfm)); + return 0; +} + +static int shash_default_import(struct shash_desc *desc, const void *in) +{ + memcpy(shash_desc_ctx(desc), in, crypto_shash_descsize(desc->tfm)); + return 0; +} + +static int crypto_shash_init_tfm(struct crypto_tfm *tfm) +{ + struct crypto_shash *hash = __crypto_shash_cast(tfm); + + hash->descsize = crypto_shash_alg(hash)->descsize; + return 0; +} + +static const struct crypto_type crypto_shash_type = { + .extsize = crypto_alg_extsize, + .init_tfm = crypto_shash_init_tfm, + .maskclear = ~CRYPTO_ALG_TYPE_MASK, + .maskset = CRYPTO_ALG_TYPE_MASK, + .type = CRYPTO_ALG_TYPE_SHASH, + .tfmsize = offsetof(struct crypto_shash, base), +}; + +struct crypto_shash *crypto_alloc_shash(const char *alg_name, u32 type, + u32 mask) +{ + return crypto_alloc_tfm(alg_name, &crypto_shash_type, type, mask); +} + +static int shash_prepare_alg(struct shash_alg *alg) +{ + struct crypto_alg *base = &alg->base; + + if (alg->digestsize > PAGE_SIZE / 8 || + alg->descsize > PAGE_SIZE / 8 || + alg->statesize > PAGE_SIZE / 8) + return -EINVAL; + + base->cra_type = &crypto_shash_type; + base->cra_flags &= ~CRYPTO_ALG_TYPE_MASK; + base->cra_flags |= CRYPTO_ALG_TYPE_SHASH; + + if (!alg->finup) + alg->finup = shash_finup_unaligned; + if (!alg->digest) + alg->digest = shash_digest_unaligned; + if (!alg->export) { + alg->export = shash_default_export; + alg->import = shash_default_import; + alg->statesize = alg->descsize; + } + if (!alg->setkey) + alg->setkey = shash_no_setkey; + + return 0; +} + +int crypto_register_shash(struct shash_alg *alg) +{ + struct crypto_alg *base = &alg->base; + int err; + + err = shash_prepare_alg(alg); + if (err) + return err; + + return crypto_register_alg(base); +} + +int crypto_unregister_shash(struct shash_alg *alg) +{ + return crypto_unregister_alg(&alg->base); +} + +int crypto_register_shashes(struct shash_alg *algs, int count) +{ + int i, ret; + + for (i = 0; i < count; i++) { + ret = crypto_register_shash(&algs[i]); + if (ret) + goto err; + } + + return 0; + +err: + for (--i; i >= 0; --i) + crypto_unregister_shash(&algs[i]); + + return ret; +} + +int crypto_unregister_shashes(struct shash_alg *algs, int count) +{ + int i, ret; + + for (i = count - 1; i >= 0; --i) { + ret = crypto_unregister_shash(&algs[i]); + if (ret) + pr_err("Failed to unregister %s %s: %d\n", + algs[i].base.cra_driver_name, + algs[i].base.cra_name, ret); + } + + return 0; +} + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Synchronous cryptographic hash type"); diff --git a/linux/fs.c b/linux/fs.c new file mode 100644 index 0000000..0002846 --- /dev/null +++ b/linux/fs.c @@ -0,0 +1,14 @@ +#include <linux/fs.h> +#include <linux/posix_acl.h> +#include <linux/posix_acl_xattr.h> +#include <linux/xattr.h> + +const struct xattr_handler posix_acl_access_xattr_handler = { + .name = XATTR_NAME_POSIX_ACL_ACCESS, + .flags = ACL_TYPE_ACCESS, +}; + +const struct xattr_handler posix_acl_default_xattr_handler = { + .name = XATTR_NAME_POSIX_ACL_DEFAULT, + .flags = ACL_TYPE_DEFAULT, +}; diff --git a/linux/generic-radix-tree.c b/linux/generic-radix-tree.c new file mode 100644 index 0000000..5c4a275 --- /dev/null +++ b/linux/generic-radix-tree.c @@ -0,0 +1,167 @@ + +#include <linux/export.h> +#include <linux/generic-radix-tree.h> +#include <linux/gfp.h> + +#define GENRADIX_ARY (PAGE_SIZE / sizeof(struct genradix_node *)) +#define GENRADIX_ARY_SHIFT ilog2(GENRADIX_ARY) + +struct genradix_node { + union { + /* Interior node: */ + struct genradix_node *children[GENRADIX_ARY]; + + /* Leaf: */ + u8 data[PAGE_SIZE]; + }; +}; + +static inline unsigned genradix_depth_shift(unsigned depth) +{ + return PAGE_SHIFT + GENRADIX_ARY_SHIFT * depth; +} + +/* + * Returns size (of data, in bytes) that a tree of a given depth holds: + */ +static inline size_t genradix_depth_size(unsigned depth) +{ + return 1UL << genradix_depth_shift(depth); +} + +/* + * Returns pointer to the specified byte @offset within @radix, or NULL if not + * allocated + */ +void *__genradix_ptr(struct __genradix *radix, size_t offset) +{ + size_t level = radix->depth; + struct genradix_node *n = radix->root; + + if (offset >= genradix_depth_size(radix->depth)) + return NULL; + + while (1) { + if (!n) + return NULL; + if (!level) + break; + + level--; + + n = n->children[offset >> genradix_depth_shift(level)]; + offset &= genradix_depth_size(level) - 1; + } + + return &n->data[offset]; +} +EXPORT_SYMBOL(__genradix_ptr); + +/* + * Returns pointer to the specified byte @offset within @radix, allocating it if + * necessary - newly allocated slots are always zeroed out: + */ +void *__genradix_ptr_alloc(struct __genradix *radix, size_t offset, + gfp_t gfp_mask) +{ + struct genradix_node **n; + size_t level; + + /* Increase tree depth if necessary: */ + + while (offset >= genradix_depth_size(radix->depth)) { + struct genradix_node *new_root = + (void *) __get_free_page(gfp_mask|__GFP_ZERO); + + if (!new_root) + return NULL; + + new_root->children[0] = radix->root; + radix->root = new_root; + radix->depth++; + } + + n = &radix->root; + level = radix->depth; + + while (1) { + if (!*n) { + *n = (void *) __get_free_page(gfp_mask|__GFP_ZERO); + if (!*n) + return NULL; + } + + if (!level) + break; + + level--; + + n = &(*n)->children[offset >> genradix_depth_shift(level)]; + offset &= genradix_depth_size(level) - 1; + } + + return &(*n)->data[offset]; +} +EXPORT_SYMBOL(__genradix_ptr_alloc); + +void *__genradix_iter_peek(struct genradix_iter *iter, + struct __genradix *radix, + size_t objs_per_page) +{ + struct genradix_node *n; + size_t level, i; + + if (!radix->root) + return NULL; +restart: + if (iter->offset >= genradix_depth_size(radix->depth)) + return NULL; + + n = radix->root; + level = radix->depth; + + while (level) { + level--; + + i = (iter->offset >> genradix_depth_shift(level)) & + (GENRADIX_ARY - 1); + + while (!n->children[i]) { + i++; + iter->offset = round_down(iter->offset + + genradix_depth_size(level), + genradix_depth_size(level)); + iter->pos = (iter->offset >> PAGE_SHIFT) * + objs_per_page; + if (i == GENRADIX_ARY) + goto restart; + } + + n = n->children[i]; + } + + return &n->data[iter->offset & (PAGE_SIZE - 1)]; +} +EXPORT_SYMBOL(__genradix_iter_peek); + +static void genradix_free_recurse(struct genradix_node *n, unsigned level) +{ + if (level) { + unsigned i; + + for (i = 0; i < GENRADIX_ARY; i++) + if (n->children[i]) + genradix_free_recurse(n->children[i], level - 1); + } + + free_page((unsigned long) n); +} + +void __genradix_free(struct __genradix *radix) +{ + genradix_free_recurse(radix->root, radix->depth); + + radix->root = NULL; + radix->depth = 0; +} +EXPORT_SYMBOL(__genradix_free); diff --git a/linux/kstrtox.c b/linux/kstrtox.c new file mode 100644 index 0000000..af6b222 --- /dev/null +++ b/linux/kstrtox.c @@ -0,0 +1,368 @@ +/* + * Convert integer string representation to an integer. + * If an integer doesn't fit into specified type, -E is returned. + * + * Integer starts with optional sign. + * kstrtou*() functions do not accept sign "-". + * + * Radix 0 means autodetection: leading "0x" implies radix 16, + * leading "0" implies radix 8, otherwise radix is 10. + * Autodetection hints work after optional sign, but not before. + * + * If -E is returned, result is not touched. + */ +#include <linux/ctype.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/math64.h> +#include <linux/export.h> +#include <linux/types.h> +#include "kstrtox.h" + +const char *_parse_integer_fixup_radix(const char *s, unsigned int *base) +{ + if (*base == 0) { + if (s[0] == '0') { + if (_tolower(s[1]) == 'x' && isxdigit(s[2])) + *base = 16; + else + *base = 8; + } else + *base = 10; + } + if (*base == 16 && s[0] == '0' && _tolower(s[1]) == 'x') + s += 2; + return s; +} + +/* + * Convert non-negative integer string representation in explicitly given radix + * to an integer. + * Return number of characters consumed maybe or-ed with overflow bit. + * If overflow occurs, result integer (incorrect) is still returned. + * + * Don't you dare use this function. + */ +unsigned int _parse_integer(const char *s, unsigned int base, unsigned long long *p) +{ + unsigned long long res; + unsigned int rv; + int overflow; + + res = 0; + rv = 0; + overflow = 0; + while (*s) { + unsigned int val; + + if ('0' <= *s && *s <= '9') + val = *s - '0'; + else if ('a' <= _tolower(*s) && _tolower(*s) <= 'f') + val = _tolower(*s) - 'a' + 10; + else + break; + + if (val >= base) + break; + /* + * Check for overflow only if we are within range of + * it in the max base we support (16) + */ + if (unlikely(res & (~0ull << 60))) { + if (res > div_u64(ULLONG_MAX - val, base)) + overflow = 1; + } + res = res * base + val; + rv++; + s++; + } + *p = res; + if (overflow) + rv |= KSTRTOX_OVERFLOW; + return rv; +} + +static int _kstrtoull(const char *s, unsigned int base, unsigned long long *res) +{ + unsigned long long _res; + unsigned int rv; + + s = _parse_integer_fixup_radix(s, &base); + rv = _parse_integer(s, base, &_res); + if (rv & KSTRTOX_OVERFLOW) + return -ERANGE; + if (rv == 0) + return -EINVAL; + s += rv; + if (*s == '\n') + s++; + if (*s) + return -EINVAL; + *res = _res; + return 0; +} + +/** + * kstrtoull - convert a string to an unsigned long long + * @s: The start of the string. The string must be null-terminated, and may also + * include a single newline before its terminating null. The first character + * may also be a plus sign, but not a minus sign. + * @base: The number base to use. The maximum supported base is 16. If base is + * given as 0, then the base of the string is automatically detected with the + * conventional semantics - If it begins with 0x the number will be parsed as a + * hexadecimal (case insensitive), if it otherwise begins with 0, it will be + * parsed as an octal number. Otherwise it will be parsed as a decimal. + * @res: Where to write the result of the conversion on success. + * + * Returns 0 on success, -ERANGE on overflow and -EINVAL on parsing error. + * Used as a replacement for the obsolete simple_strtoull. Return code must + * be checked. + */ +int kstrtoull(const char *s, unsigned int base, unsigned long long *res) +{ + if (s[0] == '+') + s++; + return _kstrtoull(s, base, res); +} +EXPORT_SYMBOL(kstrtoull); + +/** + * kstrtoll - convert a string to a long long + * @s: The start of the string. The string must be null-terminated, and may also + * include a single newline before its terminating null. The first character + * may also be a plus sign or a minus sign. + * @base: The number base to use. The maximum supported base is 16. If base is + * given as 0, then the base of the string is automatically detected with the + * conventional semantics - If it begins with 0x the number will be parsed as a + * hexadecimal (case insensitive), if it otherwise begins with 0, it will be + * parsed as an octal number. Otherwise it will be parsed as a decimal. + * @res: Where to write the result of the conversion on success. + * + * Returns 0 on success, -ERANGE on overflow and -EINVAL on parsing error. + * Used as a replacement for the obsolete simple_strtoull. Return code must + * be checked. + */ +int kstrtoll(const char *s, unsigned int base, long long *res) +{ + unsigned long long tmp; + int rv; + + if (s[0] == '-') { + rv = _kstrtoull(s + 1, base, &tmp); + if (rv < 0) + return rv; + if ((long long)-tmp > 0) + return -ERANGE; + *res = -tmp; + } else { + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if ((long long)tmp < 0) + return -ERANGE; + *res = tmp; + } + return 0; +} +EXPORT_SYMBOL(kstrtoll); + +/* Internal, do not use. */ +int _kstrtoul(const char *s, unsigned int base, unsigned long *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(unsigned long)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(_kstrtoul); + +/* Internal, do not use. */ +int _kstrtol(const char *s, unsigned int base, long *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(long)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(_kstrtol); + +/** + * kstrtouint - convert a string to an unsigned int + * @s: The start of the string. The string must be null-terminated, and may also + * include a single newline before its terminating null. The first character + * may also be a plus sign, but not a minus sign. + * @base: The number base to use. The maximum supported base is 16. If base is + * given as 0, then the base of the string is automatically detected with the + * conventional semantics - If it begins with 0x the number will be parsed as a + * hexadecimal (case insensitive), if it otherwise begins with 0, it will be + * parsed as an octal number. Otherwise it will be parsed as a decimal. + * @res: Where to write the result of the conversion on success. + * + * Returns 0 on success, -ERANGE on overflow and -EINVAL on parsing error. + * Used as a replacement for the obsolete simple_strtoull. Return code must + * be checked. + */ +int kstrtouint(const char *s, unsigned int base, unsigned int *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(unsigned int)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtouint); + +/** + * kstrtoint - convert a string to an int + * @s: The start of the string. The string must be null-terminated, and may also + * include a single newline before its terminating null. The first character + * may also be a plus sign or a minus sign. + * @base: The number base to use. The maximum supported base is 16. If base is + * given as 0, then the base of the string is automatically detected with the + * conventional semantics - If it begins with 0x the number will be parsed as a + * hexadecimal (case insensitive), if it otherwise begins with 0, it will be + * parsed as an octal number. Otherwise it will be parsed as a decimal. + * @res: Where to write the result of the conversion on success. + * + * Returns 0 on success, -ERANGE on overflow and -EINVAL on parsing error. + * Used as a replacement for the obsolete simple_strtoull. Return code must + * be checked. + */ +int kstrtoint(const char *s, unsigned int base, int *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(int)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtoint); + +int kstrtou16(const char *s, unsigned int base, u16 *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(u16)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtou16); + +int kstrtos16(const char *s, unsigned int base, s16 *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(s16)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtos16); + +int kstrtou8(const char *s, unsigned int base, u8 *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(u8)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtou8); + +int kstrtos8(const char *s, unsigned int base, s8 *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(s8)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtos8); + +/** + * kstrtobool - convert common user inputs into boolean values + * @s: input string + * @res: result + * + * This routine returns 0 iff the first character is one of 'Yy1Nn0', or + * [oO][NnFf] for "on" and "off". Otherwise it will return -EINVAL. Value + * pointed to by res is updated upon finding a match. + */ +int kstrtobool(const char *s, bool *res) +{ + if (!s) + return -EINVAL; + + switch (s[0]) { + case 'y': + case 'Y': + case '1': + *res = true; + return 0; + case 'n': + case 'N': + case '0': + *res = false; + return 0; + case 'o': + case 'O': + switch (s[1]) { + case 'n': + case 'N': + *res = true; + return 0; + case 'f': + case 'F': + *res = false; + return 0; + default: + break; + } + default: + break; + } + + return -EINVAL; +} +EXPORT_SYMBOL(kstrtobool); diff --git a/linux/kstrtox.h b/linux/kstrtox.h new file mode 100644 index 0000000..f13eeea --- /dev/null +++ b/linux/kstrtox.h @@ -0,0 +1,8 @@ +#ifndef _LIB_KSTRTOX_H +#define _LIB_KSTRTOX_H + +#define KSTRTOX_OVERFLOW (1U << 31) +const char *_parse_integer_fixup_radix(const char *s, unsigned int *base); +unsigned int _parse_integer(const char *s, unsigned int base, unsigned long long *res); + +#endif diff --git a/linux/kthread.c b/linux/kthread.c new file mode 100644 index 0000000..0f4b571 --- /dev/null +++ b/linux/kthread.c @@ -0,0 +1,117 @@ +#include <pthread.h> +#include <stdlib.h> +#include <string.h> + +#include <linux/bitops.h> +#include <linux/kthread.h> +#include <linux/rcupdate.h> +#include <linux/sched.h> + +enum KTHREAD_BITS { + KTHREAD_IS_PER_CPU = 0, + KTHREAD_SHOULD_STOP, + KTHREAD_SHOULD_PARK, + KTHREAD_IS_PARKED, +}; + +static void *kthread_start_fn(void *data) +{ + rcu_register_thread(); + + current = data; + schedule(); + current->thread_fn(current->thread_data); + + complete(¤t->exited); + put_task_struct(current); + rcu_unregister_thread(); + return NULL; +} + +/** + * kthread_create_on_node - create a kthread. + * @threadfn: the function to run until signal_pending(current). + * @data: data ptr for @threadfn. + * @node: task and thread structures for the thread are allocated on this node + * @namefmt: printf-style name for the thread. + * + * Description: This helper function creates and names a kernel + * thread. The thread will be stopped: use wake_up_process() to start + * it. See also kthread_run(). The new thread has SCHED_NORMAL policy and + * is affine to all CPUs. + * + * If thread is going to be bound on a particular cpu, give its node + * in @node, to get NUMA affinity for kthread stack, or else give NUMA_NO_NODE. + * When woken, the thread will run @threadfn() with @data as its + * argument. @threadfn() can either call do_exit() directly if it is a + * standalone thread for which no one will call kthread_stop(), or + * return when 'kthread_should_stop()' is true (which means + * kthread_stop() has been called). The return value should be zero + * or a negative error number; it will be passed to kthread_stop(). + * + * Returns a task_struct or ERR_PTR(-ENOMEM) or ERR_PTR(-EINTR). + */ +struct task_struct *kthread_create(int (*thread_fn)(void *data), + void *thread_data, + const char namefmt[], ...) +{ + va_list args; + struct task_struct *p = malloc(sizeof(*p)); + + memset(p, 0, sizeof(*p)); + + va_start(args, namefmt); + vsnprintf(p->comm, sizeof(p->comm), namefmt, args); + va_end(args); + + p->thread_fn = thread_fn; + p->thread_data = thread_data; + p->state = TASK_UNINTERRUPTIBLE; + pthread_mutex_init(&p->lock, NULL); + pthread_cond_init(&p->wait, NULL); + atomic_set(&p->usage, 1); + init_completion(&p->exited); + + pthread_create(&p->thread, NULL, kthread_start_fn, p); + return p; +} + +/** + * kthread_should_stop - should this kthread return now? + * + * When someone calls kthread_stop() on your kthread, it will be woken + * and this will return true. You should then return, and your return + * value will be passed through to kthread_stop(). + */ +bool kthread_should_stop(void) +{ + return test_bit(KTHREAD_SHOULD_STOP, ¤t->kthread_flags); +} + +/** + * kthread_stop - stop a thread created by kthread_create(). + * @k: thread created by kthread_create(). + * + * Sets kthread_should_stop() for @k to return true, wakes it, and + * waits for it to exit. This can also be called after kthread_create() + * instead of calling wake_up_process(): the thread will exit without + * calling threadfn(). + * + * If threadfn() may call do_exit() itself, the caller must ensure + * task_struct can't go away. + * + * Returns the result of threadfn(), or %-EINTR if wake_up_process() + * was never called. + */ +int kthread_stop(struct task_struct *p) +{ + get_task_struct(p); + + set_bit(KTHREAD_SHOULD_STOP, &p->kthread_flags); + wake_up_process(p); + wait_for_completion(&p->exited); + + put_task_struct(p); + + return 0; +} diff --git a/linux/llist.c b/linux/llist.c new file mode 100644 index 0000000..ae5872b --- /dev/null +++ b/linux/llist.c @@ -0,0 +1,104 @@ +/* + * Lock-less NULL terminated single linked list + * + * The basic atomic operation of this list is cmpxchg on long. On + * architectures that don't have NMI-safe cmpxchg implementation, the + * list can NOT be used in NMI handlers. So code that uses the list in + * an NMI handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. + * + * Copyright 2010,2011 Intel Corp. + * Author: Huang Ying <ying.huang@intel.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version + * 2 as published by the Free Software Foundation; + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ +#include <linux/kernel.h> +#include <linux/export.h> +#include <linux/llist.h> + + +/** + * llist_add_batch - add several linked entries in batch + * @new_first: first entry in batch to be added + * @new_last: last entry in batch to be added + * @head: the head for your lock-less list + * + * Return whether list is empty before adding. + */ +bool llist_add_batch(struct llist_node *new_first, struct llist_node *new_last, + struct llist_head *head) +{ + struct llist_node *first; + + do { + new_last->next = first = ACCESS_ONCE(head->first); + } while (cmpxchg(&head->first, first, new_first) != first); + + return !first; +} +EXPORT_SYMBOL_GPL(llist_add_batch); + +/** + * llist_del_first - delete the first entry of lock-less list + * @head: the head for your lock-less list + * + * If list is empty, return NULL, otherwise, return the first entry + * deleted, this is the newest added one. + * + * Only one llist_del_first user can be used simultaneously with + * multiple llist_add users without lock. Because otherwise + * llist_del_first, llist_add, llist_add (or llist_del_all, llist_add, + * llist_add) sequence in another user may change @head->first->next, + * but keep @head->first. If multiple consumers are needed, please + * use llist_del_all or use lock between consumers. + */ +struct llist_node *llist_del_first(struct llist_head *head) +{ + struct llist_node *entry, *old_entry, *next; + + entry = smp_load_acquire(&head->first); + for (;;) { + if (entry == NULL) + return NULL; + old_entry = entry; + next = READ_ONCE(entry->next); + entry = cmpxchg(&head->first, old_entry, next); + if (entry == old_entry) + break; + } + + return entry; +} +EXPORT_SYMBOL_GPL(llist_del_first); + +/** + * llist_reverse_order - reverse order of a llist chain + * @head: first item of the list to be reversed + * + * Reverse the order of a chain of llist entries and return the + * new first entry. + */ +struct llist_node *llist_reverse_order(struct llist_node *head) +{ + struct llist_node *new_head = NULL; + + while (head) { + struct llist_node *tmp = head; + head = head->next; + tmp->next = new_head; + new_head = tmp; + } + + return new_head; +} +EXPORT_SYMBOL_GPL(llist_reverse_order); diff --git a/linux/lz4_compress.c b/linux/lz4_compress.c new file mode 100644 index 0000000..65243c7 --- /dev/null +++ b/linux/lz4_compress.c @@ -0,0 +1,258 @@ +/* + * LZ4 - Fast LZ compression algorithm + * Copyright (C) 2011-2012, Yann Collet. + * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are + * met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following disclaimer + * in the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You can contact the author at : + * - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html + * - LZ4 source repository : http://code.google.com/p/lz4/ + * + * Changed for kernel use by: + * Chanho Min <chanho.min@lge.com> + */ + +#include <linux/log2.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/lz4.h> +#include <asm/unaligned.h> +#include "lz4defs.h" + +#define LZ4_HASH_VALUE(p, _table) \ + __HASH_VALUE(p, MEMORY_USAGE - ilog2(sizeof(_table[0]))) + +struct lz4_hash_table { + const u8 *(*add)(const struct lz4_hash_table, const u8 *); + void *ctx; + const u8 *base; +}; + +#if __SIZEOF_POINTER__ == 4 +static inline const u8 *hash_table_add32(const struct lz4_hash_table hash, + const u8 *ip) +{ + const u8 **table = hash.ctx; + + swap(table[LZ4_HASH_VALUE(ip, table)], ip); + return ip; +} +#else +static inline const u8 *hash_table_add32(const struct lz4_hash_table hash, + const u8 *ip) +{ + u32 *table = hash.ctx; + size_t offset = ip - hash.base; + + swap(table[LZ4_HASH_VALUE(ip, table)], offset); + return hash.base + offset; +} +#endif + +static inline const u8 *hash_table_add16(const struct lz4_hash_table hash, + const u8 *ip) +{ + u16 *table = hash.ctx; + size_t offset = ip - hash.base; + + swap(table[LZ4_HASH_VALUE(ip, table)], offset); + return hash.base + offset; +} + +static inline const u8 *try_match(const struct lz4_hash_table hash, + const u8 *ip) +{ + const u8 *ref = hash.add(hash, ip); + + return ref >= ip - MAX_DISTANCE && + A32(ref) == A32(ip) ? ref : NULL; +} + +static inline const u8 *find_match(const struct lz4_hash_table hash, + const u8 **ip, const u8 *anchor, + const u8 *start, const u8 *end) +{ + + int findmatchattempts = (1U << SKIPSTRENGTH) + 3; + const u8 *next_ip = *ip, *ref; + + do { + *ip = next_ip; + next_ip += findmatchattempts++ >> SKIPSTRENGTH; + + if (unlikely(next_ip > end)) + return NULL; + } while (!(ref = try_match(hash, *ip))); + + /* Catch up */ + while (*ip > anchor && + ref > start && + unlikely((*ip)[-1] == ref[-1])) { + (*ip)--; + ref--; + } + + return ref; +} + +/* + * LZ4_compressCtx : + * ----------------- + * Compress 'isize' bytes from 'source' into an output buffer 'dest' of + * maximum size 'maxOutputSize'. * If it cannot achieve it, compression + * will stop, and result of the function will be zero. + * return : the number of bytes written in buffer 'dest', or 0 if the + * compression fails + */ +static inline int lz4_compressctx(const struct lz4_hash_table hash, + const u8 *src, size_t src_len, + u8 *dst, size_t *dst_len) +{ + const u8 *ip = src; + const u8 *anchor = ip, *ref; + const u8 *const iend = ip + src_len; + const u8 *const mflimit = iend - MFLIMIT; + const u8 *const matchlimit = iend - LASTLITERALS; + size_t maxoutputsize = *dst_len; + u8 *op = dst; + u8 *const oend = op + maxoutputsize; + int length; + u8 *token; + + /* Init */ + if (src_len < MINLENGTH) + goto _last_literals; + + memset(hash.ctx, 0, LZ4_MEM_COMPRESS); + hash.add(hash, ip); + + /* Main Loop */ + while (1) { + /* Starting a literal: */ + anchor = ip++; + ref = find_match(hash, &ip, anchor, src, mflimit); + if (!ref) + goto _last_literals; + + /* + * We found a match; @ip now points to the match and @ref points + * to the prior part of the input we matched with. Everything up + * to @anchor has been encoded; the range from @anchor to @ip + * didn't match and now has to be encoded as a literal: + */ + length = ip - anchor; + token = op++; + + /* check output limit */ + if (unlikely(op + length + (2 + 1 + LASTLITERALS) + + (length >> 8) > oend)) + return -(ip - src); + + *token = encode_length(&op, length) << ML_BITS; + + /* Copy Literals */ + MEMCPY_ADVANCE_CHUNKED(op, anchor, length); + + /* Encode matches: */ + while (1) { + /* Match offset: */ + PUT_LE16_ADVANCE(op, ip - ref); + + /* MINMATCH bytes already matched from find_match(): */ + ip += MINMATCH; + ref += MINMATCH; + + length = common_length(ip, ref, matchlimit); + + /* Check output limit */ + if (unlikely(op + (1 + LASTLITERALS) + + (length >> 8) > oend)) + return -(ip - src); + + ip += length; + + *token += encode_length(&op, length); + + /* Test end of chunk */ + if (ip > mflimit) { + anchor = ip; + break; + } + + /* Fill table */ + hash.add(hash, ip - 2); + + /* Test next position */ + ref = try_match(hash, ip); + if (!ref) + break; + + token = op++; + *token = 0; + } + } + +_last_literals: + /* Encode Last Literals */ + length = iend - anchor; + if ((op - dst) + length + 1 + + ((length + 255 - RUN_MASK) / 255) > (u32)maxoutputsize) + return -(ip - src); + + token = op++; + *token = encode_length(&op, length) << ML_BITS; + MEMCPY_ADVANCE(op, anchor, iend - anchor); + + /* End */ + *dst_len = op - dst; + return 0; +} + +__attribute__((flatten)) +int lz4_compress(const unsigned char *src, size_t src_len, + unsigned char *dst, size_t *dst_len, void *wrkmem) +{ + if (src_len < LZ4_64KLIMIT) { + const struct lz4_hash_table hash = { + .add = hash_table_add16, + .ctx = wrkmem, + .base = src, + }; + + return lz4_compressctx(hash, src, src_len, dst, dst_len); + } else { + const struct lz4_hash_table hash = { + .add = hash_table_add32, + .ctx = wrkmem, + .base = src, + }; + + return lz4_compressctx(hash, src, src_len, dst, dst_len); + } +} +EXPORT_SYMBOL(lz4_compress); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("LZ4 compressor"); diff --git a/linux/lz4_decompress.c b/linux/lz4_decompress.c new file mode 100644 index 0000000..0f3e42d --- /dev/null +++ b/linux/lz4_decompress.c @@ -0,0 +1,316 @@ +/* + * LZ4 Decompressor for Linux kernel + * + * Copyright (C) 2013, LG Electronics, Kyungsik Lee <kyungsik.lee@lge.com> + * + * Based on LZ4 implementation by Yann Collet. + * + * LZ4 - Fast LZ compression algorithm + * Copyright (C) 2011-2012, Yann Collet. + * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are + * met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following disclaimer + * in the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You can contact the author at : + * - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html + * - LZ4 source repository : http://code.google.com/p/lz4/ + */ + +#ifndef STATIC +#include <linux/module.h> +#include <linux/kernel.h> +#endif +#include <linux/lz4.h> + +#include "lz4defs.h" + +static const int dec32table[8] = {0, 3, 2, 3, 0, 0, 0, 0}; +#if LZ4_ARCH64 +static const int dec64table[8] = {0, 0, 0, -1, 0, 1, 2, 3}; +#else +static const int dec64table[8] = {0, 0, 0, 0, 0, 0, 0, 0}; +#endif + +static inline size_t get_length(const u8 **ip, size_t length) +{ + if (length == LENGTH_LONG) { + size_t len; + + do { + length += (len = *(*ip)++); + } while (len == 255); + } + + return length; +} + +static int lz4_uncompress(const u8 *source, u8 *dest, int osize) +{ + const u8 *ip = source; + const u8 *ref; + u8 *op = dest; + u8 * const oend = op + osize; + u8 *cpy; + unsigned token, offset; + ssize_t length; + + while (1) { + /* get runlength */ + token = *ip++; + length = get_length(&ip, token >> ML_BITS); + + /* copy literals */ + if (unlikely(op + length > oend - COPYLENGTH)) { + /* + * Error: not enough place for another match + * (min 4) + 5 literals + */ + if (op + length != oend) + goto _output_error; + + MEMCPY_ADVANCE(op, ip, length); + break; /* EOF */ + } + MEMCPY_ADVANCE_CHUNKED(op, ip, length); + + /* get match offset */ + offset = GET_LE16_ADVANCE(ip); + ref = op - offset; + + /* Error: offset create reference outside destination buffer */ + if (unlikely(ref < (u8 *const) dest)) + goto _output_error; + + /* get match length */ + length = get_length(&ip, token & ML_MASK); + length += MINMATCH; + + /* copy first STEPSIZE bytes of match: */ + if (unlikely(offset < STEPSIZE)) { + MEMCPY_ADVANCE_BYTES(op, ref, 4); + ref -= dec32table[offset]; + + memcpy(op, ref, 4); + op += STEPSIZE - 4; + ref -= dec64table[offset]; + } else { + MEMCPY_ADVANCE(op, ref, STEPSIZE); + } + length -= STEPSIZE; + /* + * Note - length could have been < STEPSIZE; that's ok, length + * will now be negative and we'll just end up rewinding op: + */ + + /* copy rest of match: */ + cpy = op + length; + if (cpy > oend - COPYLENGTH) { + /* Error: request to write beyond destination buffer */ + if (cpy > oend || + ref + COPYLENGTH > oend) + goto _output_error; +#if !LZ4_ARCH64 + if (op + COPYLENGTH > oend) + goto _output_error; +#endif + MEMCPY_ADVANCE_CHUNKED_NOFIXUP(op, ref, oend - COPYLENGTH); + /* op could be > cpy here */ + while (op < cpy) + *op++ = *ref++; + op = cpy; + /* + * Check EOF (should never happen, since last 5 bytes + * are supposed to be literals) + */ + if (op == oend) + goto _output_error; + } else { + MEMCPY_ADVANCE_CHUNKED(op, ref, length); + } + } + /* end of decoding */ + return ip - source; + + /* write overflow error detected */ +_output_error: + return -1; +} + +static inline ssize_t get_length_safe(const u8 **ip, ssize_t length) +{ + if (length == 15) { + size_t len; + + do { + length += (len = *(*ip)++); + if (unlikely((ssize_t) length < 0)) + return -1; + + length += len; + } while (len == 255); + } + + return length; +} + +static int lz4_uncompress_unknownoutputsize(const u8 *source, u8 *dest, + int isize, size_t maxoutputsize) +{ + const u8 *ip = source; + const u8 *const iend = ip + isize; + const u8 *ref; + u8 *op = dest; + u8 * const oend = op + maxoutputsize; + u8 *cpy; + unsigned token, offset; + size_t length; + + /* Main Loop */ + while (ip < iend) { + /* get runlength */ + token = *ip++; + length = get_length_safe(&ip, token >> ML_BITS); + if (unlikely((ssize_t) length < 0)) + goto _output_error; + + /* copy literals */ + if ((op + length > oend - COPYLENGTH) || + (ip + length > iend - COPYLENGTH)) { + + if (op + length > oend) + goto _output_error;/* writes beyond buffer */ + + if (ip + length != iend) + goto _output_error;/* + * Error: LZ4 format requires + * to consume all input + * at this stage + */ + MEMCPY_ADVANCE(op, ip, length); + break;/* Necessarily EOF, due to parsing restrictions */ + } + MEMCPY_ADVANCE_CHUNKED(op, ip, length); + + /* get match offset */ + offset = GET_LE16_ADVANCE(ip); + ref = op - offset; + + /* Error: offset create reference outside destination buffer */ + if (ref < (u8 * const) dest) + goto _output_error; + + /* get match length */ + length = get_length_safe(&ip, token & ML_MASK); + if (unlikely((ssize_t) length < 0)) + goto _output_error; + + length += MINMATCH; + + /* copy first STEPSIZE bytes of match: */ + if (unlikely(offset < STEPSIZE)) { + MEMCPY_ADVANCE_BYTES(op, ref, 4); + ref -= dec32table[offset]; + + memcpy(op, ref, 4); + op += STEPSIZE - 4; + ref -= dec64table[offset]; + } else { + MEMCPY_ADVANCE(op, ref, STEPSIZE); + } + length -= STEPSIZE; + + /* copy rest of match: */ + cpy = op + length; + if (cpy > oend - COPYLENGTH) { + /* Error: request to write beyond destination buffer */ + if (cpy > oend || + ref + COPYLENGTH > oend) + goto _output_error; +#if !LZ4_ARCH64 + if (op + COPYLENGTH > oend) + goto _output_error; +#endif + MEMCPY_ADVANCE_CHUNKED_NOFIXUP(op, ref, oend - COPYLENGTH); + while (op < cpy) + *op++ = *ref++; + op = cpy; + /* + * Check EOF (should never happen, since last 5 bytes + * are supposed to be literals) + */ + if (op == oend) + goto _output_error; + } else { + MEMCPY_ADVANCE_CHUNKED(op, ref, length); + } + } + /* end of decoding */ + return op - dest; + + /* write overflow error detected */ +_output_error: + return -1; +} + +int lz4_decompress(const unsigned char *src, size_t *src_len, + unsigned char *dest, size_t actual_dest_len) +{ + int ret = -1; + int input_len = 0; + + input_len = lz4_uncompress(src, dest, actual_dest_len); + if (input_len < 0) + goto exit_0; + *src_len = input_len; + + return 0; +exit_0: + return ret; +} +#ifndef STATIC +EXPORT_SYMBOL(lz4_decompress); +#endif + +int lz4_decompress_unknownoutputsize(const unsigned char *src, size_t src_len, + unsigned char *dest, size_t *dest_len) +{ + int ret = -1; + int out_len = 0; + + out_len = lz4_uncompress_unknownoutputsize(src, dest, src_len, + *dest_len); + if (out_len < 0) + goto exit_0; + *dest_len = out_len; + + return 0; +exit_0: + return ret; +} +#ifndef STATIC +EXPORT_SYMBOL(lz4_decompress_unknownoutputsize); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("LZ4 Decompressor"); +#endif diff --git a/linux/lz4defs.h b/linux/lz4defs.h new file mode 100644 index 0000000..586b217 --- /dev/null +++ b/linux/lz4defs.h @@ -0,0 +1,181 @@ +/* + * lz4defs.h -- architecture specific defines + * + * Copyright (C) 2013, LG Electronics, Kyungsik Lee <kyungsik.lee@lge.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +/* + * Detects 64 bits mode + */ +#if __SIZEOF_POINTER__ == 8 +#define LZ4_ARCH64 1 +#else +#define LZ4_ARCH64 0 +#endif + +#include <linux/string.h> +#include <asm/unaligned.h> + +#define A32(_p) get_unaligned((u32 *) (_p)) +#define A16(_p) get_unaligned((u16 *) (_p)) + +#define GET_LE16_ADVANCE(_src) \ +({ \ + u16 _r = get_unaligned_le16(_src); \ + (_src) += 2; \ + _r; \ +}) + +#define PUT_LE16_ADVANCE(_dst, _v) \ +do { \ + put_unaligned_le16((_v), (_dst)); \ + (_dst) += 2; \ +} while (0) + +#define LENGTH_LONG 15 +#define COPYLENGTH 8 +#define ML_BITS 4 +#define ML_MASK ((1U << ML_BITS) - 1) +#define RUN_BITS (8 - ML_BITS) +#define RUN_MASK ((1U << RUN_BITS) - 1) +#define MEMORY_USAGE 14 +#define MINMATCH 4 +#define SKIPSTRENGTH 6 +#define LASTLITERALS 5 +#define MFLIMIT (COPYLENGTH + MINMATCH) +#define MINLENGTH (MFLIMIT + 1) +#define MAXD_LOG 16 +#define MAXD (1 << MAXD_LOG) +#define MAXD_MASK (u32)(MAXD - 1) +#define MAX_DISTANCE (MAXD - 1) +#define HASH_LOG (MAXD_LOG - 1) +#define HASHTABLESIZE (1 << HASH_LOG) +#define MAX_NB_ATTEMPTS 256 +#define OPTIMAL_ML (int)((ML_MASK-1)+MINMATCH) +#define LZ4_64KLIMIT ((1<<16) + (MFLIMIT - 1)) + +#define __HASH_VALUE(p, bits) \ + (((A32(p)) * 2654435761U) >> (32 - (bits))) + +#define HASH_VALUE(p) __HASH_VALUE(p, HASH_LOG) + +#define MEMCPY_ADVANCE(_dst, _src, length) \ +do { \ + typeof(length) _length = (length); \ + memcpy(_dst, _src, _length); \ + _src += _length; \ + _dst += _length; \ +} while (0) + +#define MEMCPY_ADVANCE_BYTES(_dst, _src, _length) \ +do { \ + const u8 *_end = (_src) + (_length); \ + while ((_src) < _end) \ + *_dst++ = *_src++; \ +} while (0) + +#define STEPSIZE __SIZEOF_LONG__ + +#define LZ4_COPYPACKET(_src, _dst) \ +do { \ + MEMCPY_ADVANCE(_dst, _src, STEPSIZE); \ + MEMCPY_ADVANCE(_dst, _src, COPYLENGTH - STEPSIZE);\ +} while (0) + +/* + * Equivalent to MEMCPY_ADVANCE - except may overrun @_dst and @_src by + * COPYLENGTH: + * + * Note: src and dst may overlap (with src < dst) - we must do the copy in + * STEPSIZE chunks for correctness + * + * Note also: length may be negative - we must not call memcpy if length is + * negative, but still adjust dst and src by length + */ +#define MEMCPY_ADVANCE_CHUNKED(_dst, _src, _length) \ +do { \ + u8 *_end = (_dst) + (_length); \ + while ((_dst) < _end) \ + LZ4_COPYPACKET(_src, _dst); \ + _src -= (_dst) - _end; \ + _dst = _end; \ +} while (0) + +#define MEMCPY_ADVANCE_CHUNKED_NOFIXUP(_dst, _src, _end)\ +do { \ + while ((_dst) < (_end)) \ + LZ4_COPYPACKET((_src), (_dst)); \ +} while (0) + +struct lz4_hashtable { +#if LZ4_ARCH64 + const u8 * const base; + u32 *table; +#else + const int base; + const u8 *table; +#endif +}; + +#if LZ4_ARCH64 +#define HTYPE u32 +#else /* 32-bit */ +#define HTYPE const u8* +#endif + +#ifdef __BIG_ENDIAN +#define LZ4_NBCOMMONBYTES(val) (__builtin_clzl(val) >> 3) +#else +#define LZ4_NBCOMMONBYTES(val) (__builtin_ctzl(val) >> 3) +#endif + +static inline unsigned common_length(const u8 *l, const u8 *r, + const u8 *const l_end) +{ + const u8 *l_start = l; + + while (likely(l <= l_end - sizeof(long))) { + unsigned long diff = + get_unaligned((unsigned long *) l) ^ + get_unaligned((unsigned long *) r); + + if (diff) + return l + LZ4_NBCOMMONBYTES(diff) - l_start; + + l += sizeof(long); + r += sizeof(long); + } +#if LZ4_ARCH64 + if (l <= l_end - 4 && A32(r) == A32(l)) { + l += 4; + r += 4; + } +#endif + if (l <= l_end - 2 && A16(r) == A16(l)) { + l += 2; + r += 2; + } + if (l <= l_end - 1 && *r == *l) { + l++; + r++; + } + + return l - l_start; +} + +static inline unsigned encode_length(u8 **op, unsigned length) +{ + if (length >= LENGTH_LONG) { + length -= LENGTH_LONG; + + for (; length > 254 ; length -= 255) + *(*op)++ = 255; + *(*op)++ = length; + return LENGTH_LONG; + } else + return length; +} diff --git a/linux/lz4hc_compress.c b/linux/lz4hc_compress.c new file mode 100644 index 0000000..b64ded0 --- /dev/null +++ b/linux/lz4hc_compress.c @@ -0,0 +1,454 @@ +/* + * LZ4 HC - High Compression Mode of LZ4 + * Copyright (C) 2011-2012, Yann Collet. + * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are + * met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following disclaimer + * in the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You can contact the author at : + * - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html + * - LZ4 source repository : http://code.google.com/p/lz4/ + * + * Changed for kernel use by: + * Chanho Min <chanho.min@lge.com> + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/lz4.h> +#include <asm/unaligned.h> +#include "lz4defs.h" + +struct lz4hc_data { + const u8 *base; + HTYPE hashtable[HASHTABLESIZE]; + u16 chaintable[MAXD]; + const u8 *nexttoupdate; +} __attribute__((__packed__)); + +static inline int lz4hc_init(struct lz4hc_data *hc4, const u8 *base) +{ + memset((void *)hc4->hashtable, 0, sizeof(hc4->hashtable)); + memset(hc4->chaintable, 0xFF, sizeof(hc4->chaintable)); + +#if LZ4_ARCH64 + hc4->nexttoupdate = base + 1; +#else + hc4->nexttoupdate = base; +#endif + hc4->base = base; + return 1; +} + +/* Update chains up to ip (excluded) */ +static inline void lz4hc_insert(struct lz4hc_data *hc4, const u8 *ip) +{ + u16 *chaintable = hc4->chaintable; + HTYPE *hashtable = hc4->hashtable; +#if LZ4_ARCH64 + const u8 * const base = hc4->base; +#else + const int base = 0; +#endif + + while (hc4->nexttoupdate < ip) { + const u8 *p = hc4->nexttoupdate; + size_t delta = p - (hashtable[HASH_VALUE(p)] + base); + if (delta > MAX_DISTANCE) + delta = MAX_DISTANCE; + chaintable[(size_t)(p) & MAXD_MASK] = (u16)delta; + hashtable[HASH_VALUE(p)] = (p) - base; + hc4->nexttoupdate++; + } +} + +static inline int lz4hc_insertandfindbestmatch(struct lz4hc_data *hc4, + const u8 *ip, const u8 *const matchlimit, const u8 **matchpos) +{ + u16 *const chaintable = hc4->chaintable; + HTYPE *const hashtable = hc4->hashtable; + const u8 *ref; +#if LZ4_ARCH64 + const u8 * const base = hc4->base; +#else + const int base = 0; +#endif + int nbattempts = MAX_NB_ATTEMPTS; + size_t repl = 0, ml = 0; + u16 delta; + + /* HC4 match finder */ + lz4hc_insert(hc4, ip); + ref = hashtable[HASH_VALUE(ip)] + base; + + /* potential repetition */ + if (ref >= ip-4) { + /* confirmed */ + if (A32(ref) == A32(ip)) { + delta = (u16)(ip-ref); + repl = ml = common_length(ip + MINMATCH, + ref + MINMATCH, matchlimit) + MINMATCH; + *matchpos = ref; + } + ref -= (size_t)chaintable[(size_t)(ref) & MAXD_MASK]; + } + + while ((ref >= ip - MAX_DISTANCE) && nbattempts) { + nbattempts--; + if (*(ref + ml) == *(ip + ml)) { + if (A32(ref) == A32(ip)) { + size_t mlt = + common_length(ip + MINMATCH, + ref + MINMATCH, matchlimit) + MINMATCH; + if (mlt > ml) { + ml = mlt; + *matchpos = ref; + } + } + } + ref -= (size_t)chaintable[(size_t)(ref) & MAXD_MASK]; + } + + /* Complete table */ + if (repl) { + const u8 *ptr = ip; + const u8 *end; + end = ip + repl - (MINMATCH-1); + /* Pre-Load */ + while (ptr < end - delta) { + chaintable[(size_t)(ptr) & MAXD_MASK] = delta; + ptr++; + } + do { + chaintable[(size_t)(ptr) & MAXD_MASK] = delta; + /* Head of chain */ + hashtable[HASH_VALUE(ptr)] = (ptr) - base; + ptr++; + } while (ptr < end); + hc4->nexttoupdate = end; + } + + return (int)ml; +} + +static inline int lz4hc_insertandgetwidermatch(struct lz4hc_data *hc4, + const u8 *ip, const u8 *startlimit, const u8 *matchlimit, int longest, + const u8 **matchpos, const u8 **startpos) +{ + u16 *const chaintable = hc4->chaintable; + HTYPE *const hashtable = hc4->hashtable; +#if LZ4_ARCH64 + const u8 * const base = hc4->base; +#else + const int base = 0; +#endif + const u8 *ref; + int nbattempts = MAX_NB_ATTEMPTS; + int delta = (int)(ip - startlimit); + + /* First Match */ + lz4hc_insert(hc4, ip); + ref = hashtable[HASH_VALUE(ip)] + base; + + while ((ref >= ip - MAX_DISTANCE) && (ref >= hc4->base) + && (nbattempts)) { + nbattempts--; + if (*(startlimit + longest) == *(ref - delta + longest)) { + if (A32(ref) == A32(ip)) { + const u8 *reft = ref; + const u8 *startt = ip; + unsigned length = + common_length(ip + MINMATCH, + ref + MINMATCH, + matchlimit); + + while ((startt > startlimit) + && (reft > hc4->base) + && (startt[-1] == reft[-1])) { + startt--; + reft--; + length++; + } + + if (length > longest) { + longest = length; + *matchpos = reft; + *startpos = startt; + } + } + } + ref -= (size_t)chaintable[(size_t)(ref) & MAXD_MASK]; + } + return longest; +} + +static inline int lz4_encodesequence(const u8 **ip, u8 **op, const u8 **anchor, + int ml, const u8 *ref) +{ + unsigned length; + u8 *token; + + /* Encode Literal length */ + length = *ip - *anchor; + token = (*op)++; + *token = encode_length(op, length) << ML_BITS; + + /* Copy Literals */ + MEMCPY_ADVANCE_CHUNKED(*op, *anchor, length); + + /* Encode Offset */ + PUT_LE16_ADVANCE(*op, (u16)(*ip - ref)); + + *token += encode_length(op, ml - MINMATCH); + + /* Prepare next loop */ + *ip += ml; + *anchor = *ip; + + return 0; +} + +static int lz4_compresshcctx(struct lz4hc_data *ctx, + const char *source, + char *dest, + int isize) +{ + const u8 *ip = (const u8 *)source; + const u8 *anchor = ip; + const u8 *const iend = ip + isize; + const u8 *const mflimit = iend - MFLIMIT; + const u8 *const matchlimit = (iend - LASTLITERALS); + + u8 *op = (u8 *)dest; + + int ml, ml2, ml3, ml0; + const u8 *ref = NULL; + const u8 *start2 = NULL; + const u8 *ref2 = NULL; + const u8 *start3 = NULL; + const u8 *ref3 = NULL; + const u8 *start0; + const u8 *ref0; + int lastrun; + + ip++; + + /* Main Loop */ + while (ip < mflimit) { + ml = lz4hc_insertandfindbestmatch(ctx, ip, matchlimit, (&ref)); + if (!ml) { + ip++; + continue; + } + + /* saved, in case we would skip too much */ + start0 = ip; + ref0 = ref; + ml0 = ml; +_search2: + if (ip+ml < mflimit) + ml2 = lz4hc_insertandgetwidermatch(ctx, ip + ml - 2, + ip + 1, matchlimit, ml, &ref2, &start2); + else + ml2 = ml; + /* No better match */ + if (ml2 == ml) { + lz4_encodesequence(&ip, &op, &anchor, ml, ref); + continue; + } + + if (start0 < ip) { + /* empirical */ + if (start2 < ip + ml0) { + ip = start0; + ref = ref0; + ml = ml0; + } + } + /* + * Here, start0==ip + * First Match too small : removed + */ + if ((start2 - ip) < 3) { + ml = ml2; + ip = start2; + ref = ref2; + goto _search2; + } + +_search3: + /* + * Currently we have : + * ml2 > ml1, and + * ip1+3 <= ip2 (usually < ip1+ml1) + */ + if ((start2 - ip) < OPTIMAL_ML) { + int correction; + int new_ml = ml; + if (new_ml > OPTIMAL_ML) + new_ml = OPTIMAL_ML; + if (ip + new_ml > start2 + ml2 - MINMATCH) + new_ml = (int)(start2 - ip) + ml2 - MINMATCH; + correction = new_ml - (int)(start2 - ip); + if (correction > 0) { + start2 += correction; + ref2 += correction; + ml2 -= correction; + } + } + /* + * Now, we have start2 = ip+new_ml, + * with new_ml=min(ml, OPTIMAL_ML=18) + */ + if (start2 + ml2 < mflimit) + ml3 = lz4hc_insertandgetwidermatch(ctx, + start2 + ml2 - 3, start2, matchlimit, + ml2, &ref3, &start3); + else + ml3 = ml2; + + /* No better match : 2 sequences to encode */ + if (ml3 == ml2) { + /* ip & ref are known; Now for ml */ + if (start2 < ip+ml) + ml = (int)(start2 - ip); + + /* Now, encode 2 sequences */ + lz4_encodesequence(&ip, &op, &anchor, ml, ref); + ip = start2; + lz4_encodesequence(&ip, &op, &anchor, ml2, ref2); + continue; + } + + /* Not enough space for match 2 : remove it */ + if (start3 < ip + ml + 3) { + /* + * can write Seq1 immediately ==> Seq2 is removed, + * so Seq3 becomes Seq1 + */ + if (start3 >= (ip + ml)) { + if (start2 < ip + ml) { + int correction = + (int)(ip + ml - start2); + start2 += correction; + ref2 += correction; + ml2 -= correction; + if (ml2 < MINMATCH) { + start2 = start3; + ref2 = ref3; + ml2 = ml3; + } + } + + lz4_encodesequence(&ip, &op, &anchor, ml, ref); + ip = start3; + ref = ref3; + ml = ml3; + + start0 = start2; + ref0 = ref2; + ml0 = ml2; + goto _search2; + } + + start2 = start3; + ref2 = ref3; + ml2 = ml3; + goto _search3; + } + + /* + * OK, now we have 3 ascending matches; let's write at least + * the first one ip & ref are known; Now for ml + */ + if (start2 < ip + ml) { + if ((start2 - ip) < (int)ML_MASK) { + int correction; + if (ml > OPTIMAL_ML) + ml = OPTIMAL_ML; + if (ip + ml > start2 + ml2 - MINMATCH) + ml = (int)(start2 - ip) + ml2 + - MINMATCH; + correction = ml - (int)(start2 - ip); + if (correction > 0) { + start2 += correction; + ref2 += correction; + ml2 -= correction; + } + } else + ml = (int)(start2 - ip); + } + lz4_encodesequence(&ip, &op, &anchor, ml, ref); + + ip = start2; + ref = ref2; + ml = ml2; + + start2 = start3; + ref2 = ref3; + ml2 = ml3; + + goto _search3; + } + + /* Encode Last Literals */ + lastrun = (int)(iend - anchor); + if (lastrun >= (int)RUN_MASK) { + *op++ = (RUN_MASK << ML_BITS); + lastrun -= RUN_MASK; + for (; lastrun > 254 ; lastrun -= 255) + *op++ = 255; + *op++ = (u8) lastrun; + } else + *op++ = (lastrun << ML_BITS); + memcpy(op, anchor, iend - anchor); + op += iend - anchor; + /* End */ + return (int) (((char *)op) - dest); +} + +int lz4hc_compress(const unsigned char *src, size_t src_len, + unsigned char *dst, size_t *dst_len, void *wrkmem) +{ + int ret = -1; + int out_len = 0; + + struct lz4hc_data *hc4 = (struct lz4hc_data *)wrkmem; + lz4hc_init(hc4, (const u8 *)src); + out_len = lz4_compresshcctx((struct lz4hc_data *)hc4, (const u8 *)src, + (char *)dst, (int)src_len); + + if (out_len < 0) + goto exit; + + *dst_len = out_len; + return 0; + +exit: + return ret; +} +EXPORT_SYMBOL(lz4hc_compress); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("LZ4HC compressor"); diff --git a/linux/rbtree.c b/linux/rbtree.c new file mode 100644 index 0000000..d0e3cbf --- /dev/null +++ b/linux/rbtree.c @@ -0,0 +1,615 @@ +/* + Red Black Trees + (C) 1999 Andrea Arcangeli <andrea@suse.de> + (C) 2002 David Woodhouse <dwmw2@infradead.org> + (C) 2012 Michel Lespinasse <walken@google.com> + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + + linux/lib/rbtree.c +*/ + +#include <linux/atomic.h> +#include <linux/rbtree_augmented.h> +#include <linux/export.h> + +/* + * red-black trees properties: http://en.wikipedia.org/wiki/Rbtree + * + * 1) A node is either red or black + * 2) The root is black + * 3) All leaves (NULL) are black + * 4) Both children of every red node are black + * 5) Every simple path from root to leaves contains the same number + * of black nodes. + * + * 4 and 5 give the O(log n) guarantee, since 4 implies you cannot have two + * consecutive red nodes in a path and every red node is therefore followed by + * a black. So if B is the number of black nodes on every simple path (as per + * 5), then the longest possible path due to 4 is 2B. + * + * We shall indicate color with case, where black nodes are uppercase and red + * nodes will be lowercase. Unknown color nodes shall be drawn as red within + * parentheses and have some accompanying text comment. + */ + +/* + * Notes on lockless lookups: + * + * All stores to the tree structure (rb_left and rb_right) must be done using + * WRITE_ONCE(). And we must not inadvertently cause (temporary) loops in the + * tree structure as seen in program order. + * + * These two requirements will allow lockless iteration of the tree -- not + * correct iteration mind you, tree rotations are not atomic so a lookup might + * miss entire subtrees. + * + * But they do guarantee that any such traversal will only see valid elements + * and that it will indeed complete -- does not get stuck in a loop. + * + * It also guarantees that if the lookup returns an element it is the 'correct' + * one. But not returning an element does _NOT_ mean it's not present. + * + * NOTE: + * + * Stores to __rb_parent_color are not important for simple lookups so those + * are left undone as of now. Nor did I check for loops involving parent + * pointers. + */ + +static inline void rb_set_black(struct rb_node *rb) +{ + rb->__rb_parent_color |= RB_BLACK; +} + +static inline struct rb_node *rb_red_parent(struct rb_node *red) +{ + return (struct rb_node *)red->__rb_parent_color; +} + +/* + * Helper function for rotations: + * - old's parent and color get assigned to new + * - old gets assigned new as a parent and 'color' as a color. + */ +static inline void +__rb_rotate_set_parents(struct rb_node *old, struct rb_node *new, + struct rb_root *root, int color) +{ + struct rb_node *parent = rb_parent(old); + new->__rb_parent_color = old->__rb_parent_color; + rb_set_parent_color(old, new, color); + __rb_change_child(old, new, parent, root); +} + +static __always_inline void +__rb_insert(struct rb_node *node, struct rb_root *root, + void (*augment_rotate)(struct rb_node *old, struct rb_node *new)) +{ + struct rb_node *parent = rb_red_parent(node), *gparent, *tmp; + + while (true) { + /* + * Loop invariant: node is red + * + * If there is a black parent, we are done. + * Otherwise, take some corrective action as we don't + * want a red root or two consecutive red nodes. + */ + if (!parent) { + rb_set_parent_color(node, NULL, RB_BLACK); + break; + } else if (rb_is_black(parent)) + break; + + gparent = rb_red_parent(parent); + + tmp = gparent->rb_right; + if (parent != tmp) { /* parent == gparent->rb_left */ + if (tmp && rb_is_red(tmp)) { + /* + * Case 1 - color flips + * + * G g + * / \ / \ + * p u --> P U + * / / + * n n + * + * However, since g's parent might be red, and + * 4) does not allow this, we need to recurse + * at g. + */ + rb_set_parent_color(tmp, gparent, RB_BLACK); + rb_set_parent_color(parent, gparent, RB_BLACK); + node = gparent; + parent = rb_parent(node); + rb_set_parent_color(node, parent, RB_RED); + continue; + } + + tmp = parent->rb_right; + if (node == tmp) { + /* + * Case 2 - left rotate at parent + * + * G G + * / \ / \ + * p U --> n U + * \ / + * n p + * + * This still leaves us in violation of 4), the + * continuation into Case 3 will fix that. + */ + tmp = node->rb_left; + WRITE_ONCE(parent->rb_right, tmp); + WRITE_ONCE(node->rb_left, parent); + if (tmp) + rb_set_parent_color(tmp, parent, + RB_BLACK); + rb_set_parent_color(parent, node, RB_RED); + augment_rotate(parent, node); + parent = node; + tmp = node->rb_right; + } + + /* + * Case 3 - right rotate at gparent + * + * G P + * / \ / \ + * p U --> n g + * / \ + * n U + */ + WRITE_ONCE(gparent->rb_left, tmp); /* == parent->rb_right */ + WRITE_ONCE(parent->rb_right, gparent); + if (tmp) + rb_set_parent_color(tmp, gparent, RB_BLACK); + __rb_rotate_set_parents(gparent, parent, root, RB_RED); + augment_rotate(gparent, parent); + break; + } else { + tmp = gparent->rb_left; + if (tmp && rb_is_red(tmp)) { + /* Case 1 - color flips */ + rb_set_parent_color(tmp, gparent, RB_BLACK); + rb_set_parent_color(parent, gparent, RB_BLACK); + node = gparent; + parent = rb_parent(node); + rb_set_parent_color(node, parent, RB_RED); + continue; + } + + tmp = parent->rb_left; + if (node == tmp) { + /* Case 2 - right rotate at parent */ + tmp = node->rb_right; + WRITE_ONCE(parent->rb_left, tmp); + WRITE_ONCE(node->rb_right, parent); + if (tmp) + rb_set_parent_color(tmp, parent, + RB_BLACK); + rb_set_parent_color(parent, node, RB_RED); + augment_rotate(parent, node); + parent = node; + tmp = node->rb_left; + } + + /* Case 3 - left rotate at gparent */ + WRITE_ONCE(gparent->rb_right, tmp); /* == parent->rb_left */ + WRITE_ONCE(parent->rb_left, gparent); + if (tmp) + rb_set_parent_color(tmp, gparent, RB_BLACK); + __rb_rotate_set_parents(gparent, parent, root, RB_RED); + augment_rotate(gparent, parent); + break; + } + } +} + +/* + * Inline version for rb_erase() use - we want to be able to inline + * and eliminate the dummy_rotate callback there + */ +static __always_inline void +____rb_erase_color(struct rb_node *parent, struct rb_root *root, + void (*augment_rotate)(struct rb_node *old, struct rb_node *new)) +{ + struct rb_node *node = NULL, *sibling, *tmp1, *tmp2; + + while (true) { + /* + * Loop invariants: + * - node is black (or NULL on first iteration) + * - node is not the root (parent is not NULL) + * - All leaf paths going through parent and node have a + * black node count that is 1 lower than other leaf paths. + */ + sibling = parent->rb_right; + if (node != sibling) { /* node == parent->rb_left */ + if (rb_is_red(sibling)) { + /* + * Case 1 - left rotate at parent + * + * P S + * / \ / \ + * N s --> p Sr + * / \ / \ + * Sl Sr N Sl + */ + tmp1 = sibling->rb_left; + WRITE_ONCE(parent->rb_right, tmp1); + WRITE_ONCE(sibling->rb_left, parent); + rb_set_parent_color(tmp1, parent, RB_BLACK); + __rb_rotate_set_parents(parent, sibling, root, + RB_RED); + augment_rotate(parent, sibling); + sibling = tmp1; + } + tmp1 = sibling->rb_right; + if (!tmp1 || rb_is_black(tmp1)) { + tmp2 = sibling->rb_left; + if (!tmp2 || rb_is_black(tmp2)) { + /* + * Case 2 - sibling color flip + * (p could be either color here) + * + * (p) (p) + * / \ / \ + * N S --> N s + * / \ / \ + * Sl Sr Sl Sr + * + * This leaves us violating 5) which + * can be fixed by flipping p to black + * if it was red, or by recursing at p. + * p is red when coming from Case 1. + */ + rb_set_parent_color(sibling, parent, + RB_RED); + if (rb_is_red(parent)) + rb_set_black(parent); + else { + node = parent; + parent = rb_parent(node); + if (parent) + continue; + } + break; + } + /* + * Case 3 - right rotate at sibling + * (p could be either color here) + * + * (p) (p) + * / \ / \ + * N S --> N Sl + * / \ \ + * sl Sr s + * \ + * Sr + */ + tmp1 = tmp2->rb_right; + WRITE_ONCE(sibling->rb_left, tmp1); + WRITE_ONCE(tmp2->rb_right, sibling); + WRITE_ONCE(parent->rb_right, tmp2); + if (tmp1) + rb_set_parent_color(tmp1, sibling, + RB_BLACK); + augment_rotate(sibling, tmp2); + tmp1 = sibling; + sibling = tmp2; + } + /* + * Case 4 - left rotate at parent + color flips + * (p and sl could be either color here. + * After rotation, p becomes black, s acquires + * p's color, and sl keeps its color) + * + * (p) (s) + * / \ / \ + * N S --> P Sr + * / \ / \ + * (sl) sr N (sl) + */ + tmp2 = sibling->rb_left; + WRITE_ONCE(parent->rb_right, tmp2); + WRITE_ONCE(sibling->rb_left, parent); + rb_set_parent_color(tmp1, sibling, RB_BLACK); + if (tmp2) + rb_set_parent(tmp2, parent); + __rb_rotate_set_parents(parent, sibling, root, + RB_BLACK); + augment_rotate(parent, sibling); + break; + } else { + sibling = parent->rb_left; + if (rb_is_red(sibling)) { + /* Case 1 - right rotate at parent */ + tmp1 = sibling->rb_right; + WRITE_ONCE(parent->rb_left, tmp1); + WRITE_ONCE(sibling->rb_right, parent); + rb_set_parent_color(tmp1, parent, RB_BLACK); + __rb_rotate_set_parents(parent, sibling, root, + RB_RED); + augment_rotate(parent, sibling); + sibling = tmp1; + } + tmp1 = sibling->rb_left; + if (!tmp1 || rb_is_black(tmp1)) { + tmp2 = sibling->rb_right; + if (!tmp2 || rb_is_black(tmp2)) { + /* Case 2 - sibling color flip */ + rb_set_parent_color(sibling, parent, + RB_RED); + if (rb_is_red(parent)) + rb_set_black(parent); + else { + node = parent; + parent = rb_parent(node); + if (parent) + continue; + } + break; + } + /* Case 3 - right rotate at sibling */ + tmp1 = tmp2->rb_left; + WRITE_ONCE(sibling->rb_right, tmp1); + WRITE_ONCE(tmp2->rb_left, sibling); + WRITE_ONCE(parent->rb_left, tmp2); + if (tmp1) + rb_set_parent_color(tmp1, sibling, + RB_BLACK); + augment_rotate(sibling, tmp2); + tmp1 = sibling; + sibling = tmp2; + } + /* Case 4 - left rotate at parent + color flips */ + tmp2 = sibling->rb_right; + WRITE_ONCE(parent->rb_left, tmp2); + WRITE_ONCE(sibling->rb_right, parent); + rb_set_parent_color(tmp1, sibling, RB_BLACK); + if (tmp2) + rb_set_parent(tmp2, parent); + __rb_rotate_set_parents(parent, sibling, root, + RB_BLACK); + augment_rotate(parent, sibling); + break; + } + } +} + +/* Non-inline version for rb_erase_augmented() use */ +void __rb_erase_color(struct rb_node *parent, struct rb_root *root, + void (*augment_rotate)(struct rb_node *old, struct rb_node *new)) +{ + ____rb_erase_color(parent, root, augment_rotate); +} +EXPORT_SYMBOL(__rb_erase_color); + +/* + * Non-augmented rbtree manipulation functions. + * + * We use dummy augmented callbacks here, and have the compiler optimize them + * out of the rb_insert_color() and rb_erase() function definitions. + */ + +static inline void dummy_propagate(struct rb_node *node, struct rb_node *stop) {} +static inline void dummy_copy(struct rb_node *old, struct rb_node *new) {} +static inline void dummy_rotate(struct rb_node *old, struct rb_node *new) {} + +static const struct rb_augment_callbacks dummy_callbacks = { + dummy_propagate, dummy_copy, dummy_rotate +}; + +void rb_insert_color(struct rb_node *node, struct rb_root *root) +{ + __rb_insert(node, root, dummy_rotate); +} +EXPORT_SYMBOL(rb_insert_color); + +void rb_erase(struct rb_node *node, struct rb_root *root) +{ + struct rb_node *rebalance; + rebalance = __rb_erase_augmented(node, root, &dummy_callbacks); + if (rebalance) + ____rb_erase_color(rebalance, root, dummy_rotate); +} +EXPORT_SYMBOL(rb_erase); + +/* + * Augmented rbtree manipulation functions. + * + * This instantiates the same __always_inline functions as in the non-augmented + * case, but this time with user-defined callbacks. + */ + +void __rb_insert_augmented(struct rb_node *node, struct rb_root *root, + void (*augment_rotate)(struct rb_node *old, struct rb_node *new)) +{ + __rb_insert(node, root, augment_rotate); +} +EXPORT_SYMBOL(__rb_insert_augmented); + +/* + * This function returns the first node (in sort order) of the tree. + */ +struct rb_node *rb_first(const struct rb_root *root) +{ + struct rb_node *n; + + n = root->rb_node; + if (!n) + return NULL; + while (n->rb_left) + n = n->rb_left; + return n; +} +EXPORT_SYMBOL(rb_first); + +struct rb_node *rb_last(const struct rb_root *root) +{ + struct rb_node *n; + + n = root->rb_node; + if (!n) + return NULL; + while (n->rb_right) + n = n->rb_right; + return n; +} +EXPORT_SYMBOL(rb_last); + +struct rb_node *rb_next(const struct rb_node *node) +{ + struct rb_node *parent; + + if (RB_EMPTY_NODE(node)) + return NULL; + + /* + * If we have a right-hand child, go down and then left as far + * as we can. + */ + if (node->rb_right) { + node = node->rb_right; + while (node->rb_left) + node=node->rb_left; + return (struct rb_node *)node; + } + + /* + * No right-hand children. Everything down and left is smaller than us, + * so any 'next' node must be in the general direction of our parent. + * Go up the tree; any time the ancestor is a right-hand child of its + * parent, keep going up. First time it's a left-hand child of its + * parent, said parent is our 'next' node. + */ + while ((parent = rb_parent(node)) && node == parent->rb_right) + node = parent; + + return parent; +} +EXPORT_SYMBOL(rb_next); + +struct rb_node *rb_prev(const struct rb_node *node) +{ + struct rb_node *parent; + + if (RB_EMPTY_NODE(node)) + return NULL; + + /* + * If we have a left-hand child, go down and then right as far + * as we can. + */ + if (node->rb_left) { + node = node->rb_left; + while (node->rb_right) + node=node->rb_right; + return (struct rb_node *)node; + } + + /* + * No left-hand children. Go up till we find an ancestor which + * is a right-hand child of its parent. + */ + while ((parent = rb_parent(node)) && node == parent->rb_left) + node = parent; + + return parent; +} +EXPORT_SYMBOL(rb_prev); + +void rb_replace_node(struct rb_node *victim, struct rb_node *new, + struct rb_root *root) +{ + struct rb_node *parent = rb_parent(victim); + + /* Copy the pointers/colour from the victim to the replacement */ + *new = *victim; + + /* Set the surrounding nodes to point to the replacement */ + if (victim->rb_left) + rb_set_parent(victim->rb_left, new); + if (victim->rb_right) + rb_set_parent(victim->rb_right, new); + __rb_change_child(victim, new, parent, root); +} +EXPORT_SYMBOL(rb_replace_node); + +void rb_replace_node_rcu(struct rb_node *victim, struct rb_node *new, + struct rb_root *root) +{ + struct rb_node *parent = rb_parent(victim); + + /* Copy the pointers/colour from the victim to the replacement */ + *new = *victim; + + /* Set the surrounding nodes to point to the replacement */ + if (victim->rb_left) + rb_set_parent(victim->rb_left, new); + if (victim->rb_right) + rb_set_parent(victim->rb_right, new); + + /* Set the parent's pointer to the new node last after an RCU barrier + * so that the pointers onwards are seen to be set correctly when doing + * an RCU walk over the tree. + */ + __rb_change_child_rcu(victim, new, parent, root); +} +EXPORT_SYMBOL(rb_replace_node_rcu); + +static struct rb_node *rb_left_deepest_node(const struct rb_node *node) +{ + for (;;) { + if (node->rb_left) + node = node->rb_left; + else if (node->rb_right) + node = node->rb_right; + else + return (struct rb_node *)node; + } +} + +struct rb_node *rb_next_postorder(const struct rb_node *node) +{ + const struct rb_node *parent; + if (!node) + return NULL; + parent = rb_parent(node); + + /* If we're sitting on node, we've already seen our children */ + if (parent && node == parent->rb_left && parent->rb_right) { + /* If we are the parent's left node, go to the parent's right + * node then all the way down to the left */ + return rb_left_deepest_node(parent->rb_right); + } else + /* Otherwise we are the parent's right node, and the parent + * should be next */ + return (struct rb_node *)parent; +} +EXPORT_SYMBOL(rb_next_postorder); + +struct rb_node *rb_first_postorder(const struct rb_root *root) +{ + if (!root->rb_node) + return NULL; + + return rb_left_deepest_node(root->rb_node); +} +EXPORT_SYMBOL(rb_first_postorder); diff --git a/linux/rhashtable.c b/linux/rhashtable.c new file mode 100644 index 0000000..035d82a --- /dev/null +++ b/linux/rhashtable.c @@ -0,0 +1,860 @@ +/* + * Resizable, Scalable, Concurrent Hash Table + * + * Copyright (c) 2015 Herbert Xu <herbert@gondor.apana.org.au> + * Copyright (c) 2014-2015 Thomas Graf <tgraf@suug.ch> + * Copyright (c) 2008-2014 Patrick McHardy <kaber@trash.net> + * + * Code partially derived from nft_hash + * Rewritten with rehash code from br_multicast plus single list + * pointer as suggested by Josh Triplett + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/atomic.h> +#include <linux/kernel.h> +#include <linux/log2.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <linux/mm.h> +#include <linux/jhash.h> +#include <linux/random.h> +#include <linux/rhashtable.h> +#include <linux/err.h> +#include <linux/export.h> + +#define HASH_DEFAULT_SIZE 64UL +#define HASH_MIN_SIZE 4U +#define BUCKET_LOCKS_PER_CPU 32UL + +static u32 head_hashfn(struct rhashtable *ht, + const struct bucket_table *tbl, + const struct rhash_head *he) +{ + return rht_head_hashfn(ht, tbl, he, ht->p); +} + +#ifdef CONFIG_PROVE_LOCKING +#define ASSERT_RHT_MUTEX(HT) BUG_ON(!lockdep_rht_mutex_is_held(HT)) + +int lockdep_rht_mutex_is_held(struct rhashtable *ht) +{ + return (debug_locks) ? lockdep_is_held(&ht->mutex) : 1; +} +EXPORT_SYMBOL_GPL(lockdep_rht_mutex_is_held); + +int lockdep_rht_bucket_is_held(const struct bucket_table *tbl, u32 hash) +{ + spinlock_t *lock = rht_bucket_lock(tbl, hash); + + return (debug_locks) ? lockdep_is_held(lock) : 1; +} +EXPORT_SYMBOL_GPL(lockdep_rht_bucket_is_held); +#else +#define ASSERT_RHT_MUTEX(HT) +#endif + + +static int alloc_bucket_locks(struct rhashtable *ht, struct bucket_table *tbl, + gfp_t gfp) +{ + unsigned int i, size; +#if defined(CONFIG_PROVE_LOCKING) + unsigned int nr_pcpus = 2; +#else + unsigned int nr_pcpus = num_possible_cpus(); +#endif + + nr_pcpus = min_t(unsigned int, nr_pcpus, 64UL); + size = roundup_pow_of_two(nr_pcpus * ht->p.locks_mul); + + /* Never allocate more than 0.5 locks per bucket */ + size = min_t(unsigned int, size, tbl->size >> 1); + + if (sizeof(spinlock_t) != 0) { + tbl->locks = NULL; +#ifdef CONFIG_NUMA + if (size * sizeof(spinlock_t) > PAGE_SIZE && + gfp == GFP_KERNEL) + tbl->locks = vmalloc(size * sizeof(spinlock_t)); +#endif + if (gfp != GFP_KERNEL) + gfp |= __GFP_NOWARN | __GFP_NORETRY; + + if (!tbl->locks) + tbl->locks = kmalloc_array(size, sizeof(spinlock_t), + gfp); + if (!tbl->locks) + return -ENOMEM; + for (i = 0; i < size; i++) + spin_lock_init(&tbl->locks[i]); + } + tbl->locks_mask = size - 1; + + return 0; +} + +static void bucket_table_free(struct bucket_table *tbl) +{ + if (tbl) + kvfree(tbl->locks); + + kvfree(tbl); +} + +static void bucket_table_free_rcu(struct rcu_head *head) +{ + bucket_table_free(container_of(head, struct bucket_table, rcu)); +} + +static struct bucket_table *bucket_table_alloc(struct rhashtable *ht, + size_t nbuckets, + gfp_t gfp) +{ + struct bucket_table *tbl = NULL; + size_t size; + int i; + + size = sizeof(*tbl) + nbuckets * sizeof(tbl->buckets[0]); + if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER) || + gfp != GFP_KERNEL) + tbl = kzalloc(size, gfp | __GFP_NOWARN | __GFP_NORETRY); + if (tbl == NULL && gfp == GFP_KERNEL) + tbl = vzalloc(size); + if (tbl == NULL) + return NULL; + + tbl->size = nbuckets; + + if (alloc_bucket_locks(ht, tbl, gfp) < 0) { + bucket_table_free(tbl); + return NULL; + } + + INIT_LIST_HEAD(&tbl->walkers); + + get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd)); + + for (i = 0; i < nbuckets; i++) + INIT_RHT_NULLS_HEAD(tbl->buckets[i], ht, i); + + return tbl; +} + +static struct bucket_table *rhashtable_last_table(struct rhashtable *ht, + struct bucket_table *tbl) +{ + struct bucket_table *new_tbl; + + do { + new_tbl = tbl; + tbl = rht_dereference_rcu(tbl->future_tbl, ht); + } while (tbl); + + return new_tbl; +} + +static int rhashtable_rehash_one(struct rhashtable *ht, unsigned int old_hash) +{ + struct bucket_table *old_tbl = rht_dereference(ht->tbl, ht); + struct bucket_table *new_tbl = rhashtable_last_table(ht, + rht_dereference_rcu(old_tbl->future_tbl, ht)); + struct rhash_head __rcu **pprev = &old_tbl->buckets[old_hash]; + int err = -ENOENT; + struct rhash_head *head, *next, *entry; + spinlock_t *new_bucket_lock; + unsigned int new_hash; + + rht_for_each(entry, old_tbl, old_hash) { + err = 0; + next = rht_dereference_bucket(entry->next, old_tbl, old_hash); + + if (rht_is_a_nulls(next)) + break; + + pprev = &entry->next; + } + + if (err) + goto out; + + new_hash = head_hashfn(ht, new_tbl, entry); + + new_bucket_lock = rht_bucket_lock(new_tbl, new_hash); + + spin_lock_nested(new_bucket_lock, SINGLE_DEPTH_NESTING); + head = rht_dereference_bucket(new_tbl->buckets[new_hash], + new_tbl, new_hash); + + RCU_INIT_POINTER(entry->next, head); + + rcu_assign_pointer(new_tbl->buckets[new_hash], entry); + spin_unlock(new_bucket_lock); + + rcu_assign_pointer(*pprev, next); + +out: + return err; +} + +static void rhashtable_rehash_chain(struct rhashtable *ht, + unsigned int old_hash) +{ + struct bucket_table *old_tbl = rht_dereference(ht->tbl, ht); + spinlock_t *old_bucket_lock; + + old_bucket_lock = rht_bucket_lock(old_tbl, old_hash); + + spin_lock_bh(old_bucket_lock); + while (!rhashtable_rehash_one(ht, old_hash)) + ; + old_tbl->rehash++; + spin_unlock_bh(old_bucket_lock); +} + +static int rhashtable_rehash_attach(struct rhashtable *ht, + struct bucket_table *old_tbl, + struct bucket_table *new_tbl) +{ + /* Protect future_tbl using the first bucket lock. */ + spin_lock_bh(old_tbl->locks); + + /* Did somebody beat us to it? */ + if (rcu_access_pointer(old_tbl->future_tbl)) { + spin_unlock_bh(old_tbl->locks); + return -EEXIST; + } + + /* Make insertions go into the new, empty table right away. Deletions + * and lookups will be attempted in both tables until we synchronize. + */ + rcu_assign_pointer(old_tbl->future_tbl, new_tbl); + + spin_unlock_bh(old_tbl->locks); + + return 0; +} + +static int rhashtable_rehash_table(struct rhashtable *ht) +{ + struct bucket_table *old_tbl = rht_dereference(ht->tbl, ht); + struct bucket_table *new_tbl; + struct rhashtable_walker *walker; + unsigned int old_hash; + + new_tbl = rht_dereference(old_tbl->future_tbl, ht); + if (!new_tbl) + return 0; + + for (old_hash = 0; old_hash < old_tbl->size; old_hash++) + rhashtable_rehash_chain(ht, old_hash); + + /* Publish the new table pointer. */ + rcu_assign_pointer(ht->tbl, new_tbl); + + spin_lock(&ht->lock); + list_for_each_entry(walker, &old_tbl->walkers, list) + walker->tbl = NULL; + spin_unlock(&ht->lock); + + /* Wait for readers. All new readers will see the new + * table, and thus no references to the old table will + * remain. + */ + call_rcu(&old_tbl->rcu, bucket_table_free_rcu); + + return rht_dereference(new_tbl->future_tbl, ht) ? -EAGAIN : 0; +} + +/** + * rhashtable_expand - Expand hash table while allowing concurrent lookups + * @ht: the hash table to expand + * + * A secondary bucket array is allocated and the hash entries are migrated. + * + * This function may only be called in a context where it is safe to call + * synchronize_rcu(), e.g. not within a rcu_read_lock() section. + * + * The caller must ensure that no concurrent resizing occurs by holding + * ht->mutex. + * + * It is valid to have concurrent insertions and deletions protected by per + * bucket locks or concurrent RCU protected lookups and traversals. + */ +static int rhashtable_expand(struct rhashtable *ht) +{ + struct bucket_table *new_tbl, *old_tbl = rht_dereference(ht->tbl, ht); + int err; + + ASSERT_RHT_MUTEX(ht); + + old_tbl = rhashtable_last_table(ht, old_tbl); + + new_tbl = bucket_table_alloc(ht, old_tbl->size * 2, GFP_KERNEL); + if (new_tbl == NULL) + return -ENOMEM; + + err = rhashtable_rehash_attach(ht, old_tbl, new_tbl); + if (err) + bucket_table_free(new_tbl); + + return err; +} + +/** + * rhashtable_shrink - Shrink hash table while allowing concurrent lookups + * @ht: the hash table to shrink + * + * This function shrinks the hash table to fit, i.e., the smallest + * size would not cause it to expand right away automatically. + * + * The caller must ensure that no concurrent resizing occurs by holding + * ht->mutex. + * + * The caller must ensure that no concurrent table mutations take place. + * It is however valid to have concurrent lookups if they are RCU protected. + * + * It is valid to have concurrent insertions and deletions protected by per + * bucket locks or concurrent RCU protected lookups and traversals. + */ +static int rhashtable_shrink(struct rhashtable *ht) +{ + struct bucket_table *new_tbl, *old_tbl = rht_dereference(ht->tbl, ht); + unsigned int nelems = atomic_read(&ht->nelems); + unsigned int size = 0; + int err; + + ASSERT_RHT_MUTEX(ht); + + if (nelems) + size = roundup_pow_of_two(nelems * 3 / 2); + if (size < ht->p.min_size) + size = ht->p.min_size; + + if (old_tbl->size <= size) + return 0; + + if (rht_dereference(old_tbl->future_tbl, ht)) + return -EEXIST; + + new_tbl = bucket_table_alloc(ht, size, GFP_KERNEL); + if (new_tbl == NULL) + return -ENOMEM; + + err = rhashtable_rehash_attach(ht, old_tbl, new_tbl); + if (err) + bucket_table_free(new_tbl); + + return err; +} + +static void rht_deferred_worker(struct work_struct *work) +{ + struct rhashtable *ht; + struct bucket_table *tbl; + int err = 0; + + ht = container_of(work, struct rhashtable, run_work); + mutex_lock(&ht->mutex); + + tbl = rht_dereference(ht->tbl, ht); + tbl = rhashtable_last_table(ht, tbl); + + if (rht_grow_above_75(ht, tbl)) + rhashtable_expand(ht); + else if (ht->p.automatic_shrinking && rht_shrink_below_30(ht, tbl)) + rhashtable_shrink(ht); + + err = rhashtable_rehash_table(ht); + + mutex_unlock(&ht->mutex); + + if (err) + schedule_work(&ht->run_work); +} + +static bool rhashtable_check_elasticity(struct rhashtable *ht, + struct bucket_table *tbl, + unsigned int hash) +{ + unsigned int elasticity = ht->elasticity; + struct rhash_head *head; + + rht_for_each(head, tbl, hash) + if (!--elasticity) + return true; + + return false; +} + +int rhashtable_insert_rehash(struct rhashtable *ht, + struct bucket_table *tbl) +{ + struct bucket_table *old_tbl; + struct bucket_table *new_tbl; + unsigned int size; + int err; + + old_tbl = rht_dereference_rcu(ht->tbl, ht); + + size = tbl->size; + + err = -EBUSY; + + if (rht_grow_above_75(ht, tbl)) + size *= 2; + /* Do not schedule more than one rehash */ + else if (old_tbl != tbl) + goto fail; + + err = -ENOMEM; + + new_tbl = bucket_table_alloc(ht, size, GFP_ATOMIC); + if (new_tbl == NULL) + goto fail; + + err = rhashtable_rehash_attach(ht, tbl, new_tbl); + if (err) { + bucket_table_free(new_tbl); + if (err == -EEXIST) + err = 0; + } else + schedule_work(&ht->run_work); + + return err; + +fail: + /* Do not fail the insert if someone else did a rehash. */ + if (likely(rcu_dereference_raw(tbl->future_tbl))) + return 0; + + /* Schedule async rehash to retry allocation in process context. */ + if (err == -ENOMEM) + schedule_work(&ht->run_work); + + return err; +} +EXPORT_SYMBOL_GPL(rhashtable_insert_rehash); + +struct bucket_table *rhashtable_insert_slow(struct rhashtable *ht, + const void *key, + struct rhash_head *obj, + struct bucket_table *tbl) +{ + struct rhash_head *head; + unsigned int hash; + int err; + + tbl = rhashtable_last_table(ht, tbl); + hash = head_hashfn(ht, tbl, obj); + spin_lock_nested(rht_bucket_lock(tbl, hash), SINGLE_DEPTH_NESTING); + + err = -EEXIST; + if (key && rhashtable_lookup_fast(ht, key, ht->p)) + goto exit; + + err = -E2BIG; + if (unlikely(rht_grow_above_max(ht, tbl))) + goto exit; + + err = -EAGAIN; + if (rhashtable_check_elasticity(ht, tbl, hash) || + rht_grow_above_100(ht, tbl)) + goto exit; + + err = 0; + + head = rht_dereference_bucket(tbl->buckets[hash], tbl, hash); + + RCU_INIT_POINTER(obj->next, head); + + rcu_assign_pointer(tbl->buckets[hash], obj); + + atomic_inc(&ht->nelems); + +exit: + spin_unlock(rht_bucket_lock(tbl, hash)); + + if (err == 0) + return NULL; + else if (err == -EAGAIN) + return tbl; + else + return ERR_PTR(err); +} +EXPORT_SYMBOL_GPL(rhashtable_insert_slow); + +/** + * rhashtable_walk_init - Initialise an iterator + * @ht: Table to walk over + * @iter: Hash table Iterator + * @gfp: GFP flags for allocations + * + * This function prepares a hash table walk. + * + * Note that if you restart a walk after rhashtable_walk_stop you + * may see the same object twice. Also, you may miss objects if + * there are removals in between rhashtable_walk_stop and the next + * call to rhashtable_walk_start. + * + * For a completely stable walk you should construct your own data + * structure outside the hash table. + * + * This function may sleep so you must not call it from interrupt + * context or with spin locks held. + * + * You must call rhashtable_walk_exit if this function returns + * successfully. + */ +int rhashtable_walk_init(struct rhashtable *ht, struct rhashtable_iter *iter, + gfp_t gfp) +{ + iter->ht = ht; + iter->p = NULL; + iter->slot = 0; + iter->skip = 0; + + iter->walker = kmalloc(sizeof(*iter->walker), gfp); + if (!iter->walker) + return -ENOMEM; + + spin_lock(&ht->lock); + iter->walker->tbl = + rcu_dereference_protected(ht->tbl, lockdep_is_held(&ht->lock)); + list_add(&iter->walker->list, &iter->walker->tbl->walkers); + spin_unlock(&ht->lock); + + return 0; +} +EXPORT_SYMBOL_GPL(rhashtable_walk_init); + +/** + * rhashtable_walk_exit - Free an iterator + * @iter: Hash table Iterator + * + * This function frees resources allocated by rhashtable_walk_init. + */ +void rhashtable_walk_exit(struct rhashtable_iter *iter) +{ + spin_lock(&iter->ht->lock); + if (iter->walker->tbl) + list_del(&iter->walker->list); + spin_unlock(&iter->ht->lock); + kfree(iter->walker); +} +EXPORT_SYMBOL_GPL(rhashtable_walk_exit); + +/** + * rhashtable_walk_start - Start a hash table walk + * @iter: Hash table iterator + * + * Start a hash table walk. Note that we take the RCU lock in all + * cases including when we return an error. So you must always call + * rhashtable_walk_stop to clean up. + * + * Returns zero if successful. + * + * Returns -EAGAIN if resize event occured. Note that the iterator + * will rewind back to the beginning and you may use it immediately + * by calling rhashtable_walk_next. + */ +int rhashtable_walk_start(struct rhashtable_iter *iter) + __acquires(RCU) +{ + struct rhashtable *ht = iter->ht; + + rcu_read_lock(); + + spin_lock(&ht->lock); + if (iter->walker->tbl) + list_del(&iter->walker->list); + spin_unlock(&ht->lock); + + if (!iter->walker->tbl) { + iter->walker->tbl = rht_dereference_rcu(ht->tbl, ht); + return -EAGAIN; + } + + return 0; +} +EXPORT_SYMBOL_GPL(rhashtable_walk_start); + +/** + * rhashtable_walk_next - Return the next object and advance the iterator + * @iter: Hash table iterator + * + * Note that you must call rhashtable_walk_stop when you are finished + * with the walk. + * + * Returns the next object or NULL when the end of the table is reached. + * + * Returns -EAGAIN if resize event occured. Note that the iterator + * will rewind back to the beginning and you may continue to use it. + */ +void *rhashtable_walk_next(struct rhashtable_iter *iter) +{ + struct bucket_table *tbl = iter->walker->tbl; + struct rhashtable *ht = iter->ht; + struct rhash_head *p = iter->p; + + if (p) { + p = rht_dereference_bucket_rcu(p->next, tbl, iter->slot); + goto next; + } + + for (; iter->slot < tbl->size; iter->slot++) { + int skip = iter->skip; + + rht_for_each_rcu(p, tbl, iter->slot) { + if (!skip) + break; + skip--; + } + +next: + if (!rht_is_a_nulls(p)) { + iter->skip++; + iter->p = p; + return rht_obj(ht, p); + } + + iter->skip = 0; + } + + iter->p = NULL; + + /* Ensure we see any new tables. */ + smp_rmb(); + + iter->walker->tbl = rht_dereference_rcu(tbl->future_tbl, ht); + if (iter->walker->tbl) { + iter->slot = 0; + iter->skip = 0; + return ERR_PTR(-EAGAIN); + } + + return NULL; +} +EXPORT_SYMBOL_GPL(rhashtable_walk_next); + +/** + * rhashtable_walk_stop - Finish a hash table walk + * @iter: Hash table iterator + * + * Finish a hash table walk. + */ +void rhashtable_walk_stop(struct rhashtable_iter *iter) + __releases(RCU) +{ + struct rhashtable *ht; + struct bucket_table *tbl = iter->walker->tbl; + + if (!tbl) + goto out; + + ht = iter->ht; + + spin_lock(&ht->lock); + if (tbl->rehash < tbl->size) + list_add(&iter->walker->list, &tbl->walkers); + else + iter->walker->tbl = NULL; + spin_unlock(&ht->lock); + + iter->p = NULL; + +out: + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(rhashtable_walk_stop); + +static size_t rounded_hashtable_size(const struct rhashtable_params *params) +{ + return max(roundup_pow_of_two(params->nelem_hint * 4 / 3), + (unsigned long)params->min_size); +} + +static u32 rhashtable_jhash2(const void *key, u32 length, u32 seed) +{ + return jhash2(key, length, seed); +} + +/** + * rhashtable_init - initialize a new hash table + * @ht: hash table to be initialized + * @params: configuration parameters + * + * Initializes a new hash table based on the provided configuration + * parameters. A table can be configured either with a variable or + * fixed length key: + * + * Configuration Example 1: Fixed length keys + * struct test_obj { + * int key; + * void * my_member; + * struct rhash_head node; + * }; + * + * struct rhashtable_params params = { + * .head_offset = offsetof(struct test_obj, node), + * .key_offset = offsetof(struct test_obj, key), + * .key_len = sizeof(int), + * .hashfn = jhash, + * .nulls_base = (1U << RHT_BASE_SHIFT), + * }; + * + * Configuration Example 2: Variable length keys + * struct test_obj { + * [...] + * struct rhash_head node; + * }; + * + * u32 my_hash_fn(const void *data, u32 len, u32 seed) + * { + * struct test_obj *obj = data; + * + * return [... hash ...]; + * } + * + * struct rhashtable_params params = { + * .head_offset = offsetof(struct test_obj, node), + * .hashfn = jhash, + * .obj_hashfn = my_hash_fn, + * }; + */ +int rhashtable_init(struct rhashtable *ht, + const struct rhashtable_params *params) +{ + struct bucket_table *tbl; + size_t size; + + size = HASH_DEFAULT_SIZE; + + if ((!params->key_len && !params->obj_hashfn) || + (params->obj_hashfn && !params->obj_cmpfn)) + return -EINVAL; + + if (params->nulls_base && params->nulls_base < (1U << RHT_BASE_SHIFT)) + return -EINVAL; + + memset(ht, 0, sizeof(*ht)); + mutex_init(&ht->mutex); + spin_lock_init(&ht->lock); + memcpy(&ht->p, params, sizeof(*params)); + + if (params->min_size) + ht->p.min_size = roundup_pow_of_two(params->min_size); + + if (params->max_size) + ht->p.max_size = rounddown_pow_of_two(params->max_size); + + if (params->insecure_max_entries) + ht->p.insecure_max_entries = + rounddown_pow_of_two(params->insecure_max_entries); + else + ht->p.insecure_max_entries = ht->p.max_size * 2; + + ht->p.min_size = max(ht->p.min_size, HASH_MIN_SIZE); + + if (params->nelem_hint) + size = rounded_hashtable_size(&ht->p); + + /* The maximum (not average) chain length grows with the + * size of the hash table, at a rate of (log N)/(log log N). + * The value of 16 is selected so that even if the hash + * table grew to 2^32 you would not expect the maximum + * chain length to exceed it unless we are under attack + * (or extremely unlucky). + * + * As this limit is only to detect attacks, we don't need + * to set it to a lower value as you'd need the chain + * length to vastly exceed 16 to have any real effect + * on the system. + */ + if (!params->insecure_elasticity) + ht->elasticity = 16; + + if (params->locks_mul) + ht->p.locks_mul = roundup_pow_of_two(params->locks_mul); + else + ht->p.locks_mul = BUCKET_LOCKS_PER_CPU; + + ht->key_len = ht->p.key_len; + if (!params->hashfn) { + ht->p.hashfn = jhash; + + if (!(ht->key_len & (sizeof(u32) - 1))) { + ht->key_len /= sizeof(u32); + ht->p.hashfn = rhashtable_jhash2; + } + } + + tbl = bucket_table_alloc(ht, size, GFP_KERNEL); + if (tbl == NULL) + return -ENOMEM; + + atomic_set(&ht->nelems, 0); + + RCU_INIT_POINTER(ht->tbl, tbl); + + INIT_WORK(&ht->run_work, rht_deferred_worker); + + return 0; +} +EXPORT_SYMBOL_GPL(rhashtable_init); + +/** + * rhashtable_free_and_destroy - free elements and destroy hash table + * @ht: the hash table to destroy + * @free_fn: callback to release resources of element + * @arg: pointer passed to free_fn + * + * Stops an eventual async resize. If defined, invokes free_fn for each + * element to releasal resources. Please note that RCU protected + * readers may still be accessing the elements. Releasing of resources + * must occur in a compatible manner. Then frees the bucket array. + * + * This function will eventually sleep to wait for an async resize + * to complete. The caller is responsible that no further write operations + * occurs in parallel. + */ +void rhashtable_free_and_destroy(struct rhashtable *ht, + void (*free_fn)(void *ptr, void *arg), + void *arg) +{ + struct bucket_table *tbl; + unsigned int i; + + cancel_work_sync(&ht->run_work); + + mutex_lock(&ht->mutex); + tbl = rht_dereference(ht->tbl, ht); + if (free_fn) { + for (i = 0; i < tbl->size; i++) { + struct rhash_head *pos, *next; + + for (pos = rht_dereference(tbl->buckets[i], ht), + next = !rht_is_a_nulls(pos) ? + rht_dereference(pos->next, ht) : NULL; + !rht_is_a_nulls(pos); + pos = next, + next = !rht_is_a_nulls(pos) ? + rht_dereference(pos->next, ht) : NULL) + free_fn(rht_obj(ht, pos), arg); + } + } + + bucket_table_free(tbl); + mutex_unlock(&ht->mutex); +} +EXPORT_SYMBOL_GPL(rhashtable_free_and_destroy); + +void rhashtable_destroy(struct rhashtable *ht) +{ + return rhashtable_free_and_destroy(ht, NULL, NULL); +} +EXPORT_SYMBOL_GPL(rhashtable_destroy); diff --git a/linux/sched.c b/linux/sched.c new file mode 100644 index 0000000..11480f3 --- /dev/null +++ b/linux/sched.c @@ -0,0 +1,178 @@ + +#include <string.h> + +#include <linux/math64.h> +#include <linux/printk.h> +#include <linux/rcupdate.h> +#include <linux/sched.h> +#include <linux/timer.h> + +__thread struct task_struct *current; + +void __put_task_struct(struct task_struct *t) +{ + pthread_join(t->thread, NULL); + free(t); +} + +/* returns true if process was woken up, false if it was already running */ +int wake_up_process(struct task_struct *p) +{ + int ret; + + pthread_mutex_lock(&p->lock); + ret = p->state != TASK_RUNNING; + p->state = TASK_RUNNING; + + pthread_cond_signal(&p->wait); + pthread_mutex_unlock(&p->lock); + + return ret; +} + +void schedule(void) +{ + rcu_quiescent_state(); + + pthread_mutex_lock(¤t->lock); + + while (current->state != TASK_RUNNING) + pthread_cond_wait(¤t->wait, ¤t->lock); + + pthread_mutex_unlock(¤t->lock); +} + +static void process_timeout(unsigned long __data) +{ + wake_up_process((struct task_struct *)__data); +} + +long schedule_timeout(long timeout) +{ + struct timer_list timer; + unsigned long expire; + + switch (timeout) + { + case MAX_SCHEDULE_TIMEOUT: + /* + * These two special cases are useful to be comfortable + * in the caller. Nothing more. We could take + * MAX_SCHEDULE_TIMEOUT from one of the negative value + * but I' d like to return a valid offset (>=0) to allow + * the caller to do everything it want with the retval. + */ + schedule(); + goto out; + default: + /* + * Another bit of PARANOID. Note that the retval will be + * 0 since no piece of kernel is supposed to do a check + * for a negative retval of schedule_timeout() (since it + * should never happens anyway). You just have the printk() + * that will tell you if something is gone wrong and where. + */ + if (timeout < 0) { + printk(KERN_ERR "schedule_timeout: wrong timeout " + "value %lx\n", timeout); + current->state = TASK_RUNNING; + goto out; + } + } + + expire = timeout + jiffies; + + setup_timer(&timer, process_timeout, (unsigned long)current); + mod_timer(&timer, expire); + schedule(); + del_timer_sync(&timer); + + timeout = expire - jiffies; +out: + return timeout < 0 ? 0 : timeout; +} + +unsigned long __msecs_to_jiffies(const unsigned int m) +{ + /* + * Negative value, means infinite timeout: + */ + if ((int)m < 0) + return MAX_JIFFY_OFFSET; + return _msecs_to_jiffies(m); +} + +u64 nsecs_to_jiffies64(u64 n) +{ +#if (NSEC_PER_SEC % HZ) == 0 + /* Common case, HZ = 100, 128, 200, 250, 256, 500, 512, 1000 etc. */ + return div_u64(n, NSEC_PER_SEC / HZ); +#elif (HZ % 512) == 0 + /* overflow after 292 years if HZ = 1024 */ + return div_u64(n * HZ / 512, NSEC_PER_SEC / 512); +#else + /* + * Generic case - optimized for cases where HZ is a multiple of 3. + * overflow after 64.99 years, exact for HZ = 60, 72, 90, 120 etc. + */ + return div_u64(n * 9, (9ull * NSEC_PER_SEC + HZ / 2) / HZ); +#endif +} + +unsigned long nsecs_to_jiffies(u64 n) +{ + return (unsigned long)nsecs_to_jiffies64(n); +} + +unsigned int jiffies_to_msecs(const unsigned long j) +{ +#if HZ <= MSEC_PER_SEC && !(MSEC_PER_SEC % HZ) + return (MSEC_PER_SEC / HZ) * j; +#elif HZ > MSEC_PER_SEC && !(HZ % MSEC_PER_SEC) + return (j + (HZ / MSEC_PER_SEC) - 1)/(HZ / MSEC_PER_SEC); +#else +# if BITS_PER_LONG == 32 + return (HZ_TO_MSEC_MUL32 * j) >> HZ_TO_MSEC_SHR32; +# else + return (j * HZ_TO_MSEC_NUM) / HZ_TO_MSEC_DEN; +# endif +#endif +} + +unsigned int jiffies_to_usecs(const unsigned long j) +{ + /* + * Hz usually doesn't go much further MSEC_PER_SEC. + * jiffies_to_usecs() and usecs_to_jiffies() depend on that. + */ + BUILD_BUG_ON(HZ > USEC_PER_SEC); + +#if !(USEC_PER_SEC % HZ) + return (USEC_PER_SEC / HZ) * j; +#else +# if BITS_PER_LONG == 32 + return (HZ_TO_USEC_MUL32 * j) >> HZ_TO_USEC_SHR32; +# else + return (j * HZ_TO_USEC_NUM) / HZ_TO_USEC_DEN; +# endif +#endif +} + +__attribute__((constructor(101))) +static void sched_init(void) +{ + struct task_struct *p = malloc(sizeof(*p)); + + memset(p, 0, sizeof(*p)); + + p->state = TASK_RUNNING; + pthread_mutex_init(&p->lock, NULL); + pthread_cond_init(&p->wait, NULL); + atomic_set(&p->usage, 1); + init_completion(&p->exited); + + current = p; + + rcu_init(); + rcu_register_thread(); +} diff --git a/linux/semaphore.c b/linux/semaphore.c new file mode 100644 index 0000000..6561dd2 --- /dev/null +++ b/linux/semaphore.c @@ -0,0 +1,256 @@ +/* + * Copyright (c) 2008 Intel Corporation + * Author: Matthew Wilcox <willy@linux.intel.com> + * + * Distributed under the terms of the GNU GPL, version 2 + * + * This file implements counting semaphores. + * A counting semaphore may be acquired 'n' times before sleeping. + * See mutex.c for single-acquisition sleeping locks which enforce + * rules which allow code to be debugged more easily. + */ + +/* + * Some notes on the implementation: + * + * The spinlock controls access to the other members of the semaphore. + * down_trylock() and up() can be called from interrupt context, so we + * have to disable interrupts when taking the lock. It turns out various + * parts of the kernel expect to be able to use down() on a semaphore in + * interrupt context when they know it will succeed, so we have to use + * irqsave variants for down(), down_interruptible() and down_killable() + * too. + * + * The ->count variable represents how many more tasks can acquire this + * semaphore. If it's zero, there may be tasks waiting on the wait_list. + */ + +#include <linux/compiler.h> +#include <linux/kernel.h> +#include <linux/export.h> +#include <linux/sched.h> +#include <linux/semaphore.h> +#include <linux/spinlock.h> + +static noinline void __down(struct semaphore *sem); +static noinline int __down_interruptible(struct semaphore *sem); +static noinline int __down_killable(struct semaphore *sem); +static noinline int __down_timeout(struct semaphore *sem, long timeout); +static noinline void __up(struct semaphore *sem); + +/** + * down - acquire the semaphore + * @sem: the semaphore to be acquired + * + * Acquires the semaphore. If no more tasks are allowed to acquire the + * semaphore, calling this function will put the task to sleep until the + * semaphore is released. + * + * Use of this function is deprecated, please use down_interruptible() or + * down_killable() instead. + */ +void down(struct semaphore *sem) +{ + unsigned long flags; + + raw_spin_lock_irqsave(&sem->lock, flags); + if (likely(sem->count > 0)) + sem->count--; + else + __down(sem); + raw_spin_unlock_irqrestore(&sem->lock, flags); +} +EXPORT_SYMBOL(down); + +/** + * down_interruptible - acquire the semaphore unless interrupted + * @sem: the semaphore to be acquired + * + * Attempts to acquire the semaphore. If no more tasks are allowed to + * acquire the semaphore, calling this function will put the task to sleep. + * If the sleep is interrupted by a signal, this function will return -EINTR. + * If the semaphore is successfully acquired, this function returns 0. + */ +int down_interruptible(struct semaphore *sem) +{ + unsigned long flags; + int result = 0; + + raw_spin_lock_irqsave(&sem->lock, flags); + if (likely(sem->count > 0)) + sem->count--; + else + result = __down_interruptible(sem); + raw_spin_unlock_irqrestore(&sem->lock, flags); + + return result; +} +EXPORT_SYMBOL(down_interruptible); + +/** + * down_killable - acquire the semaphore unless killed + * @sem: the semaphore to be acquired + * + * Attempts to acquire the semaphore. If no more tasks are allowed to + * acquire the semaphore, calling this function will put the task to sleep. + * If the sleep is interrupted by a fatal signal, this function will return + * -EINTR. If the semaphore is successfully acquired, this function returns + * 0. + */ +int down_killable(struct semaphore *sem) +{ + unsigned long flags; + int result = 0; + + raw_spin_lock_irqsave(&sem->lock, flags); + if (likely(sem->count > 0)) + sem->count--; + else + result = __down_killable(sem); + raw_spin_unlock_irqrestore(&sem->lock, flags); + + return result; +} +EXPORT_SYMBOL(down_killable); + +/** + * down_trylock - try to acquire the semaphore, without waiting + * @sem: the semaphore to be acquired + * + * Try to acquire the semaphore atomically. Returns 0 if the semaphore has + * been acquired successfully or 1 if it it cannot be acquired. + * + * NOTE: This return value is inverted from both spin_trylock and + * mutex_trylock! Be careful about this when converting code. + * + * Unlike mutex_trylock, this function can be used from interrupt context, + * and the semaphore can be released by any task or interrupt. + */ +int down_trylock(struct semaphore *sem) +{ + unsigned long flags; + int count; + + raw_spin_lock_irqsave(&sem->lock, flags); + count = sem->count - 1; + if (likely(count >= 0)) + sem->count = count; + raw_spin_unlock_irqrestore(&sem->lock, flags); + + return (count < 0); +} +EXPORT_SYMBOL(down_trylock); + +/** + * down_timeout - acquire the semaphore within a specified time + * @sem: the semaphore to be acquired + * @timeout: how long to wait before failing + * + * Attempts to acquire the semaphore. If no more tasks are allowed to + * acquire the semaphore, calling this function will put the task to sleep. + * If the semaphore is not released within the specified number of jiffies, + * this function returns -ETIME. It returns 0 if the semaphore was acquired. + */ +int down_timeout(struct semaphore *sem, long timeout) +{ + unsigned long flags; + int result = 0; + + raw_spin_lock_irqsave(&sem->lock, flags); + if (likely(sem->count > 0)) + sem->count--; + else + result = __down_timeout(sem, timeout); + raw_spin_unlock_irqrestore(&sem->lock, flags); + + return result; +} +EXPORT_SYMBOL(down_timeout); + +/** + * up - release the semaphore + * @sem: the semaphore to release + * + * Release the semaphore. Unlike mutexes, up() may be called from any + * context and even by tasks which have never called down(). + */ +void up(struct semaphore *sem) +{ + unsigned long flags; + + raw_spin_lock_irqsave(&sem->lock, flags); + if (likely(list_empty(&sem->wait_list))) + sem->count++; + else + __up(sem); + raw_spin_unlock_irqrestore(&sem->lock, flags); +} +EXPORT_SYMBOL(up); + +/* Functions for the contended case */ + +struct semaphore_waiter { + struct list_head list; + struct task_struct *task; + bool up; +}; + +/* + * Because this function is inlined, the 'state' parameter will be + * constant, and thus optimised away by the compiler. Likewise the + * 'timeout' parameter for the cases without timeouts. + */ +static inline int __sched __down_common(struct semaphore *sem, long state, + long timeout) +{ + struct task_struct *task = current; + struct semaphore_waiter waiter; + + list_add_tail(&waiter.list, &sem->wait_list); + waiter.task = task; + waiter.up = false; + + for (;;) { + if (unlikely(timeout <= 0)) + goto timed_out; + __set_task_state(task, state); + raw_spin_unlock_irq(&sem->lock); + timeout = schedule_timeout(timeout); + raw_spin_lock_irq(&sem->lock); + if (waiter.up) + return 0; + } + + timed_out: + list_del(&waiter.list); + return -1; +} + +static noinline void __sched __down(struct semaphore *sem) +{ + __down_common(sem, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT); +} + +static noinline int __sched __down_interruptible(struct semaphore *sem) +{ + return __down_common(sem, TASK_INTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT); +} + +static noinline int __sched __down_killable(struct semaphore *sem) +{ + return __down_common(sem, TASK_KILLABLE, MAX_SCHEDULE_TIMEOUT); +} + +static noinline int __sched __down_timeout(struct semaphore *sem, long timeout) +{ + return __down_common(sem, TASK_UNINTERRUPTIBLE, timeout); +} + +static noinline void __sched __up(struct semaphore *sem) +{ + struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list, + struct semaphore_waiter, list); + list_del(&waiter->list); + waiter->up = true; + wake_up_process(waiter->task); +} diff --git a/linux/sha1.c b/linux/sha1.c new file mode 100644 index 0000000..5a56dfd --- /dev/null +++ b/linux/sha1.c @@ -0,0 +1,201 @@ +/* + * SHA1 routine optimized to do word accesses rather than byte accesses, + * and to avoid unnecessary copies into the context array. + * + * This was based on the git SHA1 implementation. + */ + +#include <linux/kernel.h> +#include <linux/export.h> +#include <linux/bitops.h> +#include <linux/cryptohash.h> +#include <asm/unaligned.h> + +/* + * If you have 32 registers or more, the compiler can (and should) + * try to change the array[] accesses into registers. However, on + * machines with less than ~25 registers, that won't really work, + * and at least gcc will make an unholy mess of it. + * + * So to avoid that mess which just slows things down, we force + * the stores to memory to actually happen (we might be better off + * with a 'W(t)=(val);asm("":"+m" (W(t))' there instead, as + * suggested by Artur Skawina - that will also make gcc unable to + * try to do the silly "optimize away loads" part because it won't + * see what the value will be). + * + * Ben Herrenschmidt reports that on PPC, the C version comes close + * to the optimized asm with this (ie on PPC you don't want that + * 'volatile', since there are lots of registers). + * + * On ARM we get the best code generation by forcing a full memory barrier + * between each SHA_ROUND, otherwise gcc happily get wild with spilling and + * the stack frame size simply explode and performance goes down the drain. + */ + +#ifdef CONFIG_X86 + #define setW(x, val) (*(volatile __u32 *)&W(x) = (val)) +#elif defined(CONFIG_ARM) + #define setW(x, val) do { W(x) = (val); __asm__("":::"memory"); } while (0) +#else + #define setW(x, val) (W(x) = (val)) +#endif + +/* This "rolls" over the 512-bit array */ +#define W(x) (array[(x)&15]) + +/* + * Where do we get the source from? The first 16 iterations get it from + * the input data, the next mix it from the 512-bit array. + */ +#define SHA_SRC(t) get_unaligned_be32((__u32 *)data + t) +#define SHA_MIX(t) rol32(W(t+13) ^ W(t+8) ^ W(t+2) ^ W(t), 1) + +#define SHA_ROUND(t, input, fn, constant, A, B, C, D, E) do { \ + __u32 TEMP = input(t); setW(t, TEMP); \ + E += TEMP + rol32(A,5) + (fn) + (constant); \ + B = ror32(B, 2); } while (0) + +#define T_0_15(t, A, B, C, D, E) SHA_ROUND(t, SHA_SRC, (((C^D)&B)^D) , 0x5a827999, A, B, C, D, E ) +#define T_16_19(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, (((C^D)&B)^D) , 0x5a827999, A, B, C, D, E ) +#define T_20_39(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, (B^C^D) , 0x6ed9eba1, A, B, C, D, E ) +#define T_40_59(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, ((B&C)+(D&(B^C))) , 0x8f1bbcdc, A, B, C, D, E ) +#define T_60_79(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, (B^C^D) , 0xca62c1d6, A, B, C, D, E ) + +/** + * sha_transform - single block SHA1 transform + * + * @digest: 160 bit digest to update + * @data: 512 bits of data to hash + * @array: 16 words of workspace (see note) + * + * This function generates a SHA1 digest for a single 512-bit block. + * Be warned, it does not handle padding and message digest, do not + * confuse it with the full FIPS 180-1 digest algorithm for variable + * length messages. + * + * Note: If the hash is security sensitive, the caller should be sure + * to clear the workspace. This is left to the caller to avoid + * unnecessary clears between chained hashing operations. + */ +void sha_transform(__u32 *digest, const char *data, __u32 *array) +{ + __u32 A, B, C, D, E; + + A = digest[0]; + B = digest[1]; + C = digest[2]; + D = digest[3]; + E = digest[4]; + + /* Round 1 - iterations 0-16 take their input from 'data' */ + T_0_15( 0, A, B, C, D, E); + T_0_15( 1, E, A, B, C, D); + T_0_15( 2, D, E, A, B, C); + T_0_15( 3, C, D, E, A, B); + T_0_15( 4, B, C, D, E, A); + T_0_15( 5, A, B, C, D, E); + T_0_15( 6, E, A, B, C, D); + T_0_15( 7, D, E, A, B, C); + T_0_15( 8, C, D, E, A, B); + T_0_15( 9, B, C, D, E, A); + T_0_15(10, A, B, C, D, E); + T_0_15(11, E, A, B, C, D); + T_0_15(12, D, E, A, B, C); + T_0_15(13, C, D, E, A, B); + T_0_15(14, B, C, D, E, A); + T_0_15(15, A, B, C, D, E); + + /* Round 1 - tail. Input from 512-bit mixing array */ + T_16_19(16, E, A, B, C, D); + T_16_19(17, D, E, A, B, C); + T_16_19(18, C, D, E, A, B); + T_16_19(19, B, C, D, E, A); + + /* Round 2 */ + T_20_39(20, A, B, C, D, E); + T_20_39(21, E, A, B, C, D); + T_20_39(22, D, E, A, B, C); + T_20_39(23, C, D, E, A, B); + T_20_39(24, B, C, D, E, A); + T_20_39(25, A, B, C, D, E); + T_20_39(26, E, A, B, C, D); + T_20_39(27, D, E, A, B, C); + T_20_39(28, C, D, E, A, B); + T_20_39(29, B, C, D, E, A); + T_20_39(30, A, B, C, D, E); + T_20_39(31, E, A, B, C, D); + T_20_39(32, D, E, A, B, C); + T_20_39(33, C, D, E, A, B); + T_20_39(34, B, C, D, E, A); + T_20_39(35, A, B, C, D, E); + T_20_39(36, E, A, B, C, D); + T_20_39(37, D, E, A, B, C); + T_20_39(38, C, D, E, A, B); + T_20_39(39, B, C, D, E, A); + + /* Round 3 */ + T_40_59(40, A, B, C, D, E); + T_40_59(41, E, A, B, C, D); + T_40_59(42, D, E, A, B, C); + T_40_59(43, C, D, E, A, B); + T_40_59(44, B, C, D, E, A); + T_40_59(45, A, B, C, D, E); + T_40_59(46, E, A, B, C, D); + T_40_59(47, D, E, A, B, C); + T_40_59(48, C, D, E, A, B); + T_40_59(49, B, C, D, E, A); + T_40_59(50, A, B, C, D, E); + T_40_59(51, E, A, B, C, D); + T_40_59(52, D, E, A, B, C); + T_40_59(53, C, D, E, A, B); + T_40_59(54, B, C, D, E, A); + T_40_59(55, A, B, C, D, E); + T_40_59(56, E, A, B, C, D); + T_40_59(57, D, E, A, B, C); + T_40_59(58, C, D, E, A, B); + T_40_59(59, B, C, D, E, A); + + /* Round 4 */ + T_60_79(60, A, B, C, D, E); + T_60_79(61, E, A, B, C, D); + T_60_79(62, D, E, A, B, C); + T_60_79(63, C, D, E, A, B); + T_60_79(64, B, C, D, E, A); + T_60_79(65, A, B, C, D, E); + T_60_79(66, E, A, B, C, D); + T_60_79(67, D, E, A, B, C); + T_60_79(68, C, D, E, A, B); + T_60_79(69, B, C, D, E, A); + T_60_79(70, A, B, C, D, E); + T_60_79(71, E, A, B, C, D); + T_60_79(72, D, E, A, B, C); + T_60_79(73, C, D, E, A, B); + T_60_79(74, B, C, D, E, A); + T_60_79(75, A, B, C, D, E); + T_60_79(76, E, A, B, C, D); + T_60_79(77, D, E, A, B, C); + T_60_79(78, C, D, E, A, B); + T_60_79(79, B, C, D, E, A); + + digest[0] += A; + digest[1] += B; + digest[2] += C; + digest[3] += D; + digest[4] += E; +} +EXPORT_SYMBOL(sha_transform); + +/** + * sha_init - initialize the vectors for a SHA1 digest + * @buf: vector to initialize + */ +void sha_init(__u32 *buf) +{ + buf[0] = 0x67452301; + buf[1] = 0xefcdab89; + buf[2] = 0x98badcfe; + buf[3] = 0x10325476; + buf[4] = 0xc3d2e1f0; +} +EXPORT_SYMBOL(sha_init); diff --git a/linux/sort.c b/linux/sort.c new file mode 100644 index 0000000..15e8d11 --- /dev/null +++ b/linux/sort.c @@ -0,0 +1,143 @@ +/* + * A fast, small, non-recursive O(nlog n) sort for the Linux kernel + * + * Jan 23 2005 Matt Mackall <mpm@selenic.com> + */ + +#include <linux/types.h> +#include <linux/export.h> +#include <linux/kernel.h> +#include <linux/sort.h> + +static int alignment_ok(const void *base, int align) +{ + return IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) || + ((unsigned long)base & (align - 1)) == 0; +} + +static void u32_swap(void *a, void *b, int size) +{ + u32 t = *(u32 *)a; + *(u32 *)a = *(u32 *)b; + *(u32 *)b = t; +} + +static void u64_swap(void *a, void *b, int size) +{ + u64 t = *(u64 *)a; + *(u64 *)a = *(u64 *)b; + *(u64 *)b = t; +} + +static void generic_swap(void *a, void *b, int size) +{ + char t; + + do { + t = *(char *)a; + *(char *)a++ = *(char *)b; + *(char *)b++ = t; + } while (--size > 0); +} + +/** + * sort - sort an array of elements + * @base: pointer to data to sort + * @num: number of elements + * @size: size of each element + * @cmp_func: pointer to comparison function + * @swap_func: pointer to swap function or NULL + * + * This function does a heapsort on the given array. You may provide a + * swap_func function optimized to your element type. + * + * Sorting time is O(n log n) both on average and worst-case. While + * qsort is about 20% faster on average, it suffers from exploitable + * O(n*n) worst-case behavior and extra memory requirements that make + * it less suitable for kernel use. + */ + +void sort(void *base, size_t num, size_t size, + int (*cmp_func)(const void *, const void *), + void (*swap_func)(void *, void *, int size)) +{ + /* pre-scale counters for performance */ + int i = (num/2 - 1) * size, n = num * size, c, r; + + if (!swap_func) { + if (size == 4 && alignment_ok(base, 4)) + swap_func = u32_swap; + else if (size == 8 && alignment_ok(base, 8)) + swap_func = u64_swap; + else + swap_func = generic_swap; + } + + /* heapify */ + for ( ; i >= 0; i -= size) { + for (r = i; r * 2 + size < n; r = c) { + c = r * 2 + size; + if (c < n - size && + cmp_func(base + c, base + c + size) < 0) + c += size; + if (cmp_func(base + r, base + c) >= 0) + break; + swap_func(base + r, base + c, size); + } + } + + /* sort */ + for (i = n - size; i > 0; i -= size) { + swap_func(base, base + i, size); + for (r = 0; r * 2 + size < i; r = c) { + c = r * 2 + size; + if (c < i - size && + cmp_func(base + c, base + c + size) < 0) + c += size; + if (cmp_func(base + r, base + c) >= 0) + break; + swap_func(base + r, base + c, size); + } + } +} + +EXPORT_SYMBOL(sort); + +#if 0 +#include <linux/slab.h> +/* a simple boot-time regression test */ + +int cmpint(const void *a, const void *b) +{ + return *(int *)a - *(int *)b; +} + +static int sort_test(void) +{ + int *a, i, r = 1; + + a = kmalloc(1000 * sizeof(int), GFP_KERNEL); + BUG_ON(!a); + + printk("testing sort()\n"); + + for (i = 0; i < 1000; i++) { + r = (r * 725861) % 6599; + a[i] = r; + } + + sort(a, 1000, sizeof(int), cmpint, NULL); + + for (i = 0; i < 999; i++) + if (a[i] > a[i+1]) { + printk("sort() failed!\n"); + break; + } + + kfree(a); + + return 0; +} + +module_init(sort_test); +#endif diff --git a/linux/string.c b/linux/string.c new file mode 100644 index 0000000..0f23f07 --- /dev/null +++ b/linux/string.c @@ -0,0 +1,97 @@ +/* + * linux/lib/string.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +/* + * stupid library routines.. The optimized versions should generally be found + * as inline code in <asm-xx/string.h> + * + * These are buggy as well.. + * + * * Fri Jun 25 1999, Ingo Oeser <ioe@informatik.tu-chemnitz.de> + * - Added strsep() which will replace strtok() soon (because strsep() is + * reentrant and should be faster). Use only strsep() in new code, please. + * + * * Sat Feb 09 2002, Jason Thomas <jason@topic.com.au>, + * Matthew Hawkins <matt@mh.dropbear.id.au> + * - Kissed strtok() goodbye + */ + +#include <linux/types.h> +#include <linux/string.h> +#include <linux/ctype.h> +#include <linux/kernel.h> +#include <linux/export.h> +#include <linux/bug.h> +#include <linux/errno.h> + +#include <string.h> + +/** + * skip_spaces - Removes leading whitespace from @str. + * @str: The string to be stripped. + * + * Returns a pointer to the first non-whitespace character in @str. + */ +char *skip_spaces(const char *str) +{ + while (isspace(*str)) + ++str; + return (char *)str; +} + +/** + * strim - Removes leading and trailing whitespace from @s. + * @s: The string to be stripped. + * + * Note that the first trailing whitespace is replaced with a %NUL-terminator + * in the given string @s. Returns a pointer to the first non-whitespace + * character in @s. + */ +char *strim(char *s) +{ + size_t size; + char *end; + + size = strlen(s); + if (!size) + return s; + + end = s + size - 1; + while (end >= s && isspace(*end)) + end--; + *(end + 1) = '\0'; + + return skip_spaces(s); +} + +/** + * strlcpy - Copy a C-string into a sized buffer + * @dest: Where to copy the string to + * @src: Where to copy the string from + * @size: size of destination buffer + * + * Compatible with *BSD: the result is always a valid + * NUL-terminated string that fits in the buffer (unless, + * of course, the buffer size is zero). It does not pad + * out the result like strncpy() does. + */ +size_t strlcpy(char *dest, const char *src, size_t size) +{ + size_t ret = strlen(src); + + if (size) { + size_t len = (ret >= size) ? size - 1 : ret; + memcpy(dest, src, len); + dest[len] = '\0'; + } + return ret; +} + +void memzero_explicit(void *s, size_t count) +{ + memset(s, 0, count); + barrier_data(s); +} diff --git a/linux/timer.c b/linux/timer.c new file mode 100644 index 0000000..557af09 --- /dev/null +++ b/linux/timer.c @@ -0,0 +1,311 @@ + +#include <pthread.h> +#include <signal.h> +#include <time.h> + +#include <linux/kernel.h> +#include <linux/kthread.h> +#include <linux/timer.h> + +/** + * timespec_add_ns - Adds nanoseconds to a timespec + * @a: pointer to timespec to be incremented + * @ns: unsigned nanoseconds value to be added + * + * This must always be inlined because its used from the x86-64 vdso, + * which cannot call other kernel functions. + */ +static struct timespec timespec_add_ns(struct timespec a, u64 ns) +{ + a.tv_nsec += ns; + a.tv_sec += a.tv_nsec / NSEC_PER_SEC; + a.tv_nsec %= NSEC_PER_SEC; + return a; +} + +#define DECLARE_HEAP(type) \ +struct { \ + size_t size, used; \ + type *data; \ +} + +#define heap_init(heap, _size) \ +({ \ + size_t _bytes; \ + (heap)->used = 0; \ + (heap)->size = (_size); \ + _bytes = (heap)->size * sizeof(*(heap)->data); \ + (heap)->data = malloc(_bytes); \ + (heap)->data; \ +}) + +#define heap_free(heap) \ +do { \ + kvfree((heap)->data); \ + (heap)->data = NULL; \ +} while (0) + +#define heap_swap(h, i, j) swap((h)->data[i], (h)->data[j]) + +#define heap_sift(h, i, cmp) \ +do { \ + size_t _r, _j = i; \ + \ + for (; _j * 2 + 1 < (h)->used; _j = _r) { \ + _r = _j * 2 + 1; \ + if (_r + 1 < (h)->used && \ + cmp((h)->data[_r], (h)->data[_r + 1])) \ + _r++; \ + \ + if (cmp((h)->data[_r], (h)->data[_j])) \ + break; \ + heap_swap(h, _r, _j); \ + } \ +} while (0) + +#define heap_sift_down(h, i, cmp) \ +do { \ + while (i) { \ + size_t p = (i - 1) / 2; \ + if (cmp((h)->data[i], (h)->data[p])) \ + break; \ + heap_swap(h, i, p); \ + i = p; \ + } \ +} while (0) + +#define heap_add(h, d, cmp) \ +({ \ + bool _r = !heap_full(h); \ + if (_r) { \ + size_t _i = (h)->used++; \ + (h)->data[_i] = d; \ + \ + heap_sift_down(h, _i, cmp); \ + heap_sift(h, _i, cmp); \ + } \ + _r; \ +}) + +#define heap_del(h, i, cmp) \ +do { \ + size_t _i = (i); \ + \ + BUG_ON(_i >= (h)->used); \ + (h)->used--; \ + heap_swap(h, _i, (h)->used); \ + heap_sift_down(h, _i, cmp); \ + heap_sift(h, _i, cmp); \ +} while (0) + +#define heap_pop(h, d, cmp) \ +({ \ + bool _r = (h)->used; \ + if (_r) { \ + (d) = (h)->data[0]; \ + heap_del(h, 0, cmp); \ + } \ + _r; \ +}) + +#define heap_peek(h) ((h)->used ? &(h)->data[0] : NULL) +#define heap_full(h) ((h)->used == (h)->size) +#define heap_empty(h) ((h)->used == 0) + +#define heap_resort(heap, cmp) \ +do { \ + ssize_t _i; \ + for (_i = (ssize_t) (heap)->used / 2 - 1; _i >= 0; --_i) \ + heap_sift(heap, _i, cmp); \ +} while (0) + +struct pending_timer { + struct timer_list *timer; + unsigned long expires; +}; + +static inline bool pending_timer_cmp(struct pending_timer a, + struct pending_timer b) +{ + return a.expires < b.expires; +} + +static DECLARE_HEAP(struct pending_timer) pending_timers; + +static pthread_mutex_t timer_lock = PTHREAD_MUTEX_INITIALIZER; +static pthread_cond_t timer_cond = PTHREAD_COND_INITIALIZER; +static pthread_cond_t timer_running_cond = PTHREAD_COND_INITIALIZER; +static unsigned long timer_seq; + +static inline bool timer_running(void) +{ + return timer_seq & 1; +} + +static size_t timer_idx(struct timer_list *timer) +{ + size_t i; + + for (i = 0; i < pending_timers.size; i++) + if (pending_timers.data[i].timer == timer) + return i; + BUG(); +} + +int del_timer(struct timer_list *timer) +{ + int pending; + + pthread_mutex_lock(&timer_lock); + pending = timer_pending(timer); + timer->pending = false; + + if (pending) + heap_del(&pending_timers, timer_idx(timer), pending_timer_cmp); + + pthread_mutex_unlock(&timer_lock); + + return pending; +} + +void flush_timers(void) +{ + unsigned long seq; + + pthread_mutex_lock(&timer_lock); + seq = timer_seq; + while (timer_running() && seq == timer_seq) + pthread_cond_wait(&timer_running_cond, &timer_lock); + + pthread_mutex_unlock(&timer_lock); +} + +int del_timer_sync(struct timer_list *timer) +{ + unsigned long seq; + int pending; + + pthread_mutex_lock(&timer_lock); + pending = timer_pending(timer); + timer->pending = false; + + if (pending) + heap_del(&pending_timers, timer_idx(timer), pending_timer_cmp); + + seq = timer_seq; + while (timer_running() && seq == timer_seq) + pthread_cond_wait(&timer_running_cond, &timer_lock); + + pthread_mutex_unlock(&timer_lock); + + return pending; +} + +int mod_timer(struct timer_list *timer, unsigned long expires) +{ + int pending; + size_t i; + + pthread_mutex_lock(&timer_lock); + pending = timer_pending(timer); + + if (pending && timer->expires == expires) + goto out; + + timer->expires = expires; + timer->pending = true; + + if (pending) { + i = timer_idx(timer); + pending_timers.data[i].expires = expires; + + heap_sift_down(&pending_timers, i, pending_timer_cmp); + heap_sift(&pending_timers, i, pending_timer_cmp); + } else { + if (heap_full(&pending_timers)) { + pending_timers.size *= 2; + pending_timers.data = + realloc(pending_timers.data, + pending_timers.size * + sizeof(struct pending_timer)); + + BUG_ON(!pending_timers.data); + } + + heap_add(&pending_timers, + ((struct pending_timer) { + .timer = timer, + .expires = expires, + }), + pending_timer_cmp); + } + + pthread_cond_signal(&timer_cond); +out: + pthread_mutex_unlock(&timer_lock); + + return pending; +} + +static int timer_thread(void *arg) +{ + struct pending_timer *p; + struct timespec ts; + unsigned long now; + int ret; + + pthread_mutex_lock(&timer_lock); + + while (1) { + now = jiffies; + p = heap_peek(&pending_timers); + + if (!p) { + pthread_cond_wait(&timer_cond, &timer_lock); + continue; + } + + if (time_after_eq(now, p->expires)) { + struct timer_list *timer = p->timer; + + heap_del(&pending_timers, 0, pending_timer_cmp); + BUG_ON(!timer_pending(timer)); + timer->pending = false; + + timer_seq++; + BUG_ON(!timer_running()); + + pthread_mutex_unlock(&timer_lock); + timer->function(timer->data); + pthread_mutex_lock(&timer_lock); + + timer_seq++; + pthread_cond_broadcast(&timer_running_cond); + continue; + } + + + ret = clock_gettime(CLOCK_REALTIME, &ts); + BUG_ON(ret); + + ts = timespec_add_ns(ts, jiffies_to_nsecs(p->expires - now)); + + pthread_cond_timedwait(&timer_cond, &timer_lock, &ts); + } + + pthread_mutex_unlock(&timer_lock); + + return 0; +} + +__attribute__((constructor(103))) +static void timers_init(void) +{ + struct task_struct *p; + + heap_init(&pending_timers, 64); + BUG_ON(!pending_timers.data); + + p = kthread_run(timer_thread, NULL, "timers"); + BUG_ON(IS_ERR(p)); +} diff --git a/linux/vsprintf.c b/linux/vsprintf.c new file mode 100644 index 0000000..3642200 --- /dev/null +++ b/linux/vsprintf.c @@ -0,0 +1,75 @@ +#include <linux/kernel.h> +#include "kstrtox.h" + +/** + * simple_strtoull - convert a string to an unsigned long long + * @cp: The start of the string + * @endp: A pointer to the end of the parsed string will be placed here + * @base: The number base to use + * + * This function is obsolete. Please use kstrtoull instead. + */ +unsigned long long simple_strtoull(const char *cp, char **endp, unsigned int base) +{ + unsigned long long result; + unsigned int rv; + + cp = _parse_integer_fixup_radix(cp, &base); + rv = _parse_integer(cp, base, &result); + /* FIXME */ + cp += (rv & ~KSTRTOX_OVERFLOW); + + if (endp) + *endp = (char *)cp; + + return result; +} +EXPORT_SYMBOL(simple_strtoull); + +/** + * simple_strtoul - convert a string to an unsigned long + * @cp: The start of the string + * @endp: A pointer to the end of the parsed string will be placed here + * @base: The number base to use + * + * This function is obsolete. Please use kstrtoul instead. + */ +unsigned long simple_strtoul(const char *cp, char **endp, unsigned int base) +{ + return simple_strtoull(cp, endp, base); +} +EXPORT_SYMBOL(simple_strtoul); + +/** + * simple_strtol - convert a string to a signed long + * @cp: The start of the string + * @endp: A pointer to the end of the parsed string will be placed here + * @base: The number base to use + * + * This function is obsolete. Please use kstrtol instead. + */ +long simple_strtol(const char *cp, char **endp, unsigned int base) +{ + if (*cp == '-') + return -simple_strtoul(cp + 1, endp, base); + + return simple_strtoul(cp, endp, base); +} +EXPORT_SYMBOL(simple_strtol); + +/** + * simple_strtoll - convert a string to a signed long long + * @cp: The start of the string + * @endp: A pointer to the end of the parsed string will be placed here + * @base: The number base to use + * + * This function is obsolete. Please use kstrtoll instead. + */ +long long simple_strtoll(const char *cp, char **endp, unsigned int base) +{ + if (*cp == '-') + return -simple_strtoull(cp + 1, endp, base); + + return simple_strtoull(cp, endp, base); +} +EXPORT_SYMBOL(simple_strtoll); diff --git a/linux/wait.c b/linux/wait.c new file mode 100644 index 0000000..83f4e85 --- /dev/null +++ b/linux/wait.c @@ -0,0 +1,616 @@ +/* + * Generic waiting primitives. + * + * (C) 2004 Nadia Yvette Chambers, Oracle + */ +#include <linux/export.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/wait.h> +#include <linux/hash.h> +#include <linux/kthread.h> + +void __init_waitqueue_head(wait_queue_head_t *q, const char *name, struct lock_class_key *key) +{ + spin_lock_init(&q->lock); + lockdep_set_class_and_name(&q->lock, key, name); + INIT_LIST_HEAD(&q->task_list); +} + +EXPORT_SYMBOL(__init_waitqueue_head); + +void add_wait_queue(wait_queue_head_t *q, wait_queue_t *wait) +{ + unsigned long flags; + + wait->flags &= ~WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&q->lock, flags); + __add_wait_queue(q, wait); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(add_wait_queue); + +void add_wait_queue_exclusive(wait_queue_head_t *q, wait_queue_t *wait) +{ + unsigned long flags; + + wait->flags |= WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&q->lock, flags); + __add_wait_queue_tail(q, wait); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(add_wait_queue_exclusive); + +void remove_wait_queue(wait_queue_head_t *q, wait_queue_t *wait) +{ + unsigned long flags; + + spin_lock_irqsave(&q->lock, flags); + __remove_wait_queue(q, wait); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(remove_wait_queue); + + +/* + * The core wakeup function. Non-exclusive wakeups (nr_exclusive == 0) just + * wake everything up. If it's an exclusive wakeup (nr_exclusive == small +ve + * number) then we wake all the non-exclusive tasks and one exclusive task. + * + * There are circumstances in which we can try to wake a task which has already + * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns + * zero in this (rare) case, and we handle it by continuing to scan the queue. + */ +static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, int wake_flags, void *key) +{ + wait_queue_t *curr, *next; + + list_for_each_entry_safe(curr, next, &q->task_list, task_list) { + unsigned flags = curr->flags; + + if (curr->func(curr, mode, wake_flags, key) && + (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) + break; + } +} + +/** + * __wake_up - wake up threads blocked on a waitqueue. + * @q: the waitqueue + * @mode: which threads + * @nr_exclusive: how many wake-one or wake-many threads to wake up + * @key: is directly passed to the wakeup function + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. + */ +void __wake_up(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, void *key) +{ + unsigned long flags; + + spin_lock_irqsave(&q->lock, flags); + __wake_up_common(q, mode, nr_exclusive, 0, key); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(__wake_up); + +/* + * Same as __wake_up but called with the spinlock in wait_queue_head_t held. + */ +void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr) +{ + __wake_up_common(q, mode, nr, 0, NULL); +} +EXPORT_SYMBOL_GPL(__wake_up_locked); + +void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key) +{ + __wake_up_common(q, mode, 1, 0, key); +} +EXPORT_SYMBOL_GPL(__wake_up_locked_key); + +/** + * __wake_up_sync_key - wake up threads blocked on a waitqueue. + * @q: the waitqueue + * @mode: which threads + * @nr_exclusive: how many wake-one or wake-many threads to wake up + * @key: opaque value to be passed to wakeup targets + * + * The sync wakeup differs that the waker knows that it will schedule + * away soon, so while the target thread will be woken up, it will not + * be migrated to another CPU - ie. the two threads are 'synchronized' + * with each other. This can prevent needless bouncing between CPUs. + * + * On UP it can prevent extra preemption. + * + * It may be assumed that this function implies a write memory barrier before + * changing the task state if and only if any tasks are woken up. + */ +void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, void *key) +{ + unsigned long flags; + int wake_flags = 1; /* XXX WF_SYNC */ + + if (unlikely(!q)) + return; + + if (unlikely(nr_exclusive != 1)) + wake_flags = 0; + + spin_lock_irqsave(&q->lock, flags); + __wake_up_common(q, mode, nr_exclusive, wake_flags, key); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL_GPL(__wake_up_sync_key); + +/* + * __wake_up_sync - see __wake_up_sync_key() + */ +void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive) +{ + __wake_up_sync_key(q, mode, nr_exclusive, NULL); +} +EXPORT_SYMBOL_GPL(__wake_up_sync); /* For internal use only */ + +/* + * Note: we use "set_current_state()" _after_ the wait-queue add, + * because we need a memory barrier there on SMP, so that any + * wake-function that tests for the wait-queue being active + * will be guaranteed to see waitqueue addition _or_ subsequent + * tests in this thread will see the wakeup having taken place. + * + * The spin_unlock() itself is semi-permeable and only protects + * one way (it only protects stuff inside the critical region and + * stops them from bleeding out - it would still allow subsequent + * loads to move into the critical region). + */ +void +prepare_to_wait(wait_queue_head_t *q, wait_queue_t *wait, int state) +{ + unsigned long flags; + + wait->flags &= ~WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&q->lock, flags); + if (list_empty(&wait->task_list)) + __add_wait_queue(q, wait); + set_current_state(state); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(prepare_to_wait); + +void +prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state) +{ + unsigned long flags; + + wait->flags |= WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&q->lock, flags); + if (list_empty(&wait->task_list)) + __add_wait_queue_tail(q, wait); + set_current_state(state); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(prepare_to_wait_exclusive); + +long prepare_to_wait_event(wait_queue_head_t *q, wait_queue_t *wait, int state) +{ + unsigned long flags; + + wait->private = current; + wait->func = autoremove_wake_function; + + spin_lock_irqsave(&q->lock, flags); + if (list_empty(&wait->task_list)) { + if (wait->flags & WQ_FLAG_EXCLUSIVE) + __add_wait_queue_tail(q, wait); + else + __add_wait_queue(q, wait); + } + set_current_state(state); + spin_unlock_irqrestore(&q->lock, flags); + + return 0; +} +EXPORT_SYMBOL(prepare_to_wait_event); + +/** + * finish_wait - clean up after waiting in a queue + * @q: waitqueue waited on + * @wait: wait descriptor + * + * Sets current thread back to running state and removes + * the wait descriptor from the given waitqueue if still + * queued. + */ +void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) +{ + unsigned long flags; + + __set_current_state(TASK_RUNNING); + /* + * We can check for list emptiness outside the lock + * IFF: + * - we use the "careful" check that verifies both + * the next and prev pointers, so that there cannot + * be any half-pending updates in progress on other + * CPU's that we haven't seen yet (and that might + * still change the stack area. + * and + * - all other users take the lock (ie we can only + * have _one_ other CPU that looks at or modifies + * the list). + */ + if (!list_empty_careful(&wait->task_list)) { + spin_lock_irqsave(&q->lock, flags); + list_del_init(&wait->task_list); + spin_unlock_irqrestore(&q->lock, flags); + } +} +EXPORT_SYMBOL(finish_wait); + +/** + * abort_exclusive_wait - abort exclusive waiting in a queue + * @q: waitqueue waited on + * @wait: wait descriptor + * @mode: runstate of the waiter to be woken + * @key: key to identify a wait bit queue or %NULL + * + * Sets current thread back to running state and removes + * the wait descriptor from the given waitqueue if still + * queued. + * + * Wakes up the next waiter if the caller is concurrently + * woken up through the queue. + * + * This prevents waiter starvation where an exclusive waiter + * aborts and is woken up concurrently and no one wakes up + * the next waiter. + */ +void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, + unsigned int mode, void *key) +{ + unsigned long flags; + + __set_current_state(TASK_RUNNING); + spin_lock_irqsave(&q->lock, flags); + if (!list_empty(&wait->task_list)) + list_del_init(&wait->task_list); + else if (waitqueue_active(q)) + __wake_up_locked_key(q, mode, key); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(abort_exclusive_wait); + +int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags, + void *key) +{ + return wake_up_process(curr->private); +} + +int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) +{ + int ret = default_wake_function(wait, mode, sync, key); + + if (ret) + list_del_init(&wait->task_list); + return ret; +} +EXPORT_SYMBOL(autoremove_wake_function); + +static inline bool is_kthread_should_stop(void) +{ + return (current->flags & PF_KTHREAD) && kthread_should_stop(); +} + +/* + * DEFINE_WAIT_FUNC(wait, woken_wake_func); + * + * add_wait_queue(&wq, &wait); + * for (;;) { + * if (condition) + * break; + * + * p->state = mode; condition = true; + * smp_mb(); // A smp_wmb(); // C + * if (!wait->flags & WQ_FLAG_WOKEN) wait->flags |= WQ_FLAG_WOKEN; + * schedule() try_to_wake_up(); + * p->state = TASK_RUNNING; ~~~~~~~~~~~~~~~~~~ + * wait->flags &= ~WQ_FLAG_WOKEN; condition = true; + * smp_mb() // B smp_wmb(); // C + * wait->flags |= WQ_FLAG_WOKEN; + * } + * remove_wait_queue(&wq, &wait); + * + */ +long wait_woken(wait_queue_t *wait, unsigned mode, long timeout) +{ + set_current_state(mode); /* A */ + /* + * The above implies an smp_mb(), which matches with the smp_wmb() from + * woken_wake_function() such that if we observe WQ_FLAG_WOKEN we must + * also observe all state before the wakeup. + */ + if (!(wait->flags & WQ_FLAG_WOKEN) && !is_kthread_should_stop()) + timeout = schedule_timeout(timeout); + __set_current_state(TASK_RUNNING); + + /* + * The below implies an smp_mb(), it too pairs with the smp_wmb() from + * woken_wake_function() such that we must either observe the wait + * condition being true _OR_ WQ_FLAG_WOKEN such that we will not miss + * an event. + */ + smp_store_mb(wait->flags, wait->flags & ~WQ_FLAG_WOKEN); /* B */ + + return timeout; +} +EXPORT_SYMBOL(wait_woken); + +int woken_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) +{ + /* + * Although this function is called under waitqueue lock, LOCK + * doesn't imply write barrier and the users expects write + * barrier semantics on wakeup functions. The following + * smp_wmb() is equivalent to smp_wmb() in try_to_wake_up() + * and is paired with smp_store_mb() in wait_woken(). + */ + smp_wmb(); /* C */ + wait->flags |= WQ_FLAG_WOKEN; + + return default_wake_function(wait, mode, sync, key); +} +EXPORT_SYMBOL(woken_wake_function); + +int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *arg) +{ + struct wait_bit_key *key = arg; + struct wait_bit_queue *wait_bit + = container_of(wait, struct wait_bit_queue, wait); + + if (wait_bit->key.flags != key->flags || + wait_bit->key.bit_nr != key->bit_nr || + test_bit(key->bit_nr, key->flags)) + return 0; + else + return autoremove_wake_function(wait, mode, sync, key); +} +EXPORT_SYMBOL(wake_bit_function); + +/* + * To allow interruptible waiting and asynchronous (i.e. nonblocking) + * waiting, the actions of __wait_on_bit() and __wait_on_bit_lock() are + * permitted return codes. Nonzero return codes halt waiting and return. + */ +int __sched +__wait_on_bit(wait_queue_head_t *wq, struct wait_bit_queue *q, + wait_bit_action_f *action, unsigned mode) +{ + int ret = 0; + + do { + prepare_to_wait(wq, &q->wait, mode); + if (test_bit(q->key.bit_nr, q->key.flags)) + ret = (*action)(&q->key, mode); + } while (test_bit(q->key.bit_nr, q->key.flags) && !ret); + finish_wait(wq, &q->wait); + return ret; +} +EXPORT_SYMBOL(__wait_on_bit); + +int __sched out_of_line_wait_on_bit(void *word, int bit, + wait_bit_action_f *action, unsigned mode) +{ + wait_queue_head_t *wq = bit_waitqueue(word, bit); + DEFINE_WAIT_BIT(wait, word, bit); + + return __wait_on_bit(wq, &wait, action, mode); +} +EXPORT_SYMBOL(out_of_line_wait_on_bit); + +int __sched out_of_line_wait_on_bit_timeout( + void *word, int bit, wait_bit_action_f *action, + unsigned mode, unsigned long timeout) +{ + wait_queue_head_t *wq = bit_waitqueue(word, bit); + DEFINE_WAIT_BIT(wait, word, bit); + + wait.key.timeout = jiffies + timeout; + return __wait_on_bit(wq, &wait, action, mode); +} +EXPORT_SYMBOL_GPL(out_of_line_wait_on_bit_timeout); + +int __sched +__wait_on_bit_lock(wait_queue_head_t *wq, struct wait_bit_queue *q, + wait_bit_action_f *action, unsigned mode) +{ + do { + int ret; + + prepare_to_wait_exclusive(wq, &q->wait, mode); + if (!test_bit(q->key.bit_nr, q->key.flags)) + continue; + ret = action(&q->key, mode); + if (!ret) + continue; + abort_exclusive_wait(wq, &q->wait, mode, &q->key); + return ret; + } while (test_and_set_bit(q->key.bit_nr, q->key.flags)); + finish_wait(wq, &q->wait); + return 0; +} +EXPORT_SYMBOL(__wait_on_bit_lock); + +int __sched out_of_line_wait_on_bit_lock(void *word, int bit, + wait_bit_action_f *action, unsigned mode) +{ + wait_queue_head_t *wq = bit_waitqueue(word, bit); + DEFINE_WAIT_BIT(wait, word, bit); + + return __wait_on_bit_lock(wq, &wait, action, mode); +} +EXPORT_SYMBOL(out_of_line_wait_on_bit_lock); + +void __wake_up_bit(wait_queue_head_t *wq, void *word, int bit) +{ + struct wait_bit_key key = __WAIT_BIT_KEY_INITIALIZER(word, bit); + if (waitqueue_active(wq)) + __wake_up(wq, TASK_NORMAL, 1, &key); +} +EXPORT_SYMBOL(__wake_up_bit); + +/** + * wake_up_bit - wake up a waiter on a bit + * @word: the word being waited on, a kernel virtual address + * @bit: the bit of the word being waited on + * + * There is a standard hashed waitqueue table for generic use. This + * is the part of the hashtable's accessor API that wakes up waiters + * on a bit. For instance, if one were to have waiters on a bitflag, + * one would call wake_up_bit() after clearing the bit. + * + * In order for this to function properly, as it uses waitqueue_active() + * internally, some kind of memory barrier must be done prior to calling + * this. Typically, this will be smp_mb__after_atomic(), but in some + * cases where bitflags are manipulated non-atomically under a lock, one + * may need to use a less regular barrier, such fs/inode.c's smp_mb(), + * because spin_unlock() does not guarantee a memory barrier. + */ +void wake_up_bit(void *word, int bit) +{ + __wake_up_bit(bit_waitqueue(word, bit), word, bit); +} +EXPORT_SYMBOL(wake_up_bit); + +static DECLARE_WAIT_QUEUE_HEAD(__bit_waitqueue); + +wait_queue_head_t *bit_waitqueue(void *word, int bit) +{ + return &__bit_waitqueue; +} +EXPORT_SYMBOL(bit_waitqueue); + +/* + * Manipulate the atomic_t address to produce a better bit waitqueue table hash + * index (we're keying off bit -1, but that would produce a horrible hash + * value). + */ +static inline wait_queue_head_t *atomic_t_waitqueue(atomic_t *p) +{ + if (BITS_PER_LONG == 64) { + unsigned long q = (unsigned long)p; + return bit_waitqueue((void *)(q & ~1), q & 1); + } + return bit_waitqueue(p, 0); +} + +static int wake_atomic_t_function(wait_queue_t *wait, unsigned mode, int sync, + void *arg) +{ + struct wait_bit_key *key = arg; + struct wait_bit_queue *wait_bit + = container_of(wait, struct wait_bit_queue, wait); + atomic_t *val = key->flags; + + if (wait_bit->key.flags != key->flags || + wait_bit->key.bit_nr != key->bit_nr || + atomic_read(val) != 0) + return 0; + return autoremove_wake_function(wait, mode, sync, key); +} + +/* + * To allow interruptible waiting and asynchronous (i.e. nonblocking) waiting, + * the actions of __wait_on_atomic_t() are permitted return codes. Nonzero + * return codes halt waiting and return. + */ +static __sched +int __wait_on_atomic_t(wait_queue_head_t *wq, struct wait_bit_queue *q, + int (*action)(atomic_t *), unsigned mode) +{ + atomic_t *val; + int ret = 0; + + do { + prepare_to_wait(wq, &q->wait, mode); + val = q->key.flags; + if (atomic_read(val) == 0) + break; + ret = (*action)(val); + } while (!ret && atomic_read(val) != 0); + finish_wait(wq, &q->wait); + return ret; +} + +#define DEFINE_WAIT_ATOMIC_T(name, p) \ + struct wait_bit_queue name = { \ + .key = __WAIT_ATOMIC_T_KEY_INITIALIZER(p), \ + .wait = { \ + .private = current, \ + .func = wake_atomic_t_function, \ + .task_list = \ + LIST_HEAD_INIT((name).wait.task_list), \ + }, \ + } + +__sched int out_of_line_wait_on_atomic_t(atomic_t *p, int (*action)(atomic_t *), + unsigned mode) +{ + wait_queue_head_t *wq = atomic_t_waitqueue(p); + DEFINE_WAIT_ATOMIC_T(wait, p); + + return __wait_on_atomic_t(wq, &wait, action, mode); +} +EXPORT_SYMBOL(out_of_line_wait_on_atomic_t); + +/** + * wake_up_atomic_t - Wake up a waiter on a atomic_t + * @p: The atomic_t being waited on, a kernel virtual address + * + * Wake up anyone waiting for the atomic_t to go to zero. + * + * Abuse the bit-waker function and its waitqueue hash table set (the atomic_t + * check is done by the waiter's wake function, not the by the waker itself). + */ +void wake_up_atomic_t(atomic_t *p) +{ + __wake_up_bit(atomic_t_waitqueue(p), p, WAIT_ATOMIC_T_BIT_NR); +} +EXPORT_SYMBOL(wake_up_atomic_t); + +__sched int bit_wait(struct wait_bit_key *word, int mode) +{ + schedule(); + return 0; +} +EXPORT_SYMBOL(bit_wait); + +__sched int bit_wait_io(struct wait_bit_key *word, int mode) +{ + io_schedule(); + return 0; +} +EXPORT_SYMBOL(bit_wait_io); + +__sched int bit_wait_timeout(struct wait_bit_key *word, int mode) +{ + unsigned long now = jiffies; + if (time_after_eq(now, word->timeout)) + return -EAGAIN; + schedule_timeout(word->timeout - now); + return 0; +} +EXPORT_SYMBOL_GPL(bit_wait_timeout); + +__sched int bit_wait_io_timeout(struct wait_bit_key *word, int mode) +{ + unsigned long now = jiffies; + if (time_after_eq(now, word->timeout)) + return -EAGAIN; + io_schedule_timeout(word->timeout - now); + return 0; +} +EXPORT_SYMBOL_GPL(bit_wait_io_timeout); diff --git a/linux/workqueue.c b/linux/workqueue.c new file mode 100644 index 0000000..b25e7a4 --- /dev/null +++ b/linux/workqueue.c @@ -0,0 +1,318 @@ +#include <pthread.h> + +#include <linux/kthread.h> +#include <linux/slab.h> +#include <linux/workqueue.h> + +static pthread_mutex_t wq_lock = PTHREAD_MUTEX_INITIALIZER; +static LIST_HEAD(wq_list); + +struct workqueue_struct { + struct list_head list; + + struct work_struct *current_work; + struct list_head pending_work; + + pthread_cond_t work_finished; + + struct task_struct *worker; + char name[24]; +}; + +enum { + WORK_PENDING_BIT, +}; + +static void clear_work_pending(struct work_struct *work) +{ + clear_bit(WORK_PENDING_BIT, work_data_bits(work)); +} + +static bool set_work_pending(struct work_struct *work) +{ + return !test_and_set_bit(WORK_PENDING_BIT, work_data_bits(work)); +} + +static void __queue_work(struct workqueue_struct *wq, + struct work_struct *work) +{ + BUG_ON(!test_bit(WORK_PENDING_BIT, work_data_bits(work))); + BUG_ON(!list_empty(&work->entry)); + + list_add_tail(&work->entry, &wq->pending_work); + wake_up_process(wq->worker); +} + +bool queue_work(struct workqueue_struct *wq, struct work_struct *work) +{ + bool ret; + + pthread_mutex_lock(&wq_lock); + if ((ret = set_work_pending(work))) + __queue_work(wq, work); + pthread_mutex_unlock(&wq_lock); + + return ret; +} + +void delayed_work_timer_fn(unsigned long __data) +{ + struct delayed_work *dwork = (struct delayed_work *) __data; + + pthread_mutex_lock(&wq_lock); + __queue_work(dwork->wq, &dwork->work); + pthread_mutex_unlock(&wq_lock); +} + +static void __queue_delayed_work(struct workqueue_struct *wq, + struct delayed_work *dwork, + unsigned long delay) +{ + struct timer_list *timer = &dwork->timer; + struct work_struct *work = &dwork->work; + + BUG_ON(timer->function != delayed_work_timer_fn || + timer->data != (unsigned long)dwork); + BUG_ON(timer_pending(timer)); + BUG_ON(!list_empty(&work->entry)); + + if (!delay) { + __queue_work(wq, &dwork->work); + } else { + dwork->wq = wq; + timer->expires = jiffies + delay; + add_timer(timer); + } +} + +bool queue_delayed_work(struct workqueue_struct *wq, + struct delayed_work *dwork, + unsigned long delay) +{ + struct work_struct *work = &dwork->work; + bool ret; + + pthread_mutex_lock(&wq_lock); + if ((ret = set_work_pending(work))) + __queue_delayed_work(wq, dwork, delay); + pthread_mutex_unlock(&wq_lock); + + return ret; +} + +static bool grab_pending(struct work_struct *work, bool is_dwork) +{ +retry: + if (set_work_pending(work)) { + BUG_ON(!list_empty(&work->entry)); + return false; + } + + if (is_dwork) { + struct delayed_work *dwork = to_delayed_work(work); + + if (likely(del_timer(&dwork->timer))) { + BUG_ON(!list_empty(&work->entry)); + return true; + } + } + + if (!list_empty(&work->entry)) { + list_del_init(&work->entry); + return true; + } + + BUG_ON(!is_dwork); + + pthread_mutex_unlock(&wq_lock); + flush_timers(); + pthread_mutex_lock(&wq_lock); + goto retry; +} + +static bool __flush_work(struct work_struct *work) +{ + struct workqueue_struct *wq; + bool ret = false; +retry: + list_for_each_entry(wq, &wq_list, list) + if (wq->current_work == work) { + pthread_cond_wait(&wq->work_finished, &wq_lock); + ret = true; + goto retry; + } + + return ret; +} + +bool cancel_work_sync(struct work_struct *work) +{ + bool ret; + + pthread_mutex_lock(&wq_lock); + ret = grab_pending(work, false); + + __flush_work(work); + clear_work_pending(work); + pthread_mutex_unlock(&wq_lock); + + return ret; +} + +bool mod_delayed_work(struct workqueue_struct *wq, + struct delayed_work *dwork, + unsigned long delay) +{ + struct work_struct *work = &dwork->work; + bool ret; + + pthread_mutex_lock(&wq_lock); + ret = grab_pending(work, true); + + __queue_delayed_work(wq, dwork, delay); + pthread_mutex_unlock(&wq_lock); + + return ret; +} + +bool cancel_delayed_work(struct delayed_work *dwork) +{ + struct work_struct *work = &dwork->work; + bool ret; + + pthread_mutex_lock(&wq_lock); + ret = grab_pending(work, true); + + clear_work_pending(&dwork->work); + pthread_mutex_unlock(&wq_lock); + + return ret; +} + +bool cancel_delayed_work_sync(struct delayed_work *dwork) +{ + struct work_struct *work = &dwork->work; + bool ret; + + pthread_mutex_lock(&wq_lock); + ret = grab_pending(work, true); + + __flush_work(work); + clear_work_pending(work); + pthread_mutex_unlock(&wq_lock); + + return ret; +} + +static int worker_thread(void *arg) +{ + struct workqueue_struct *wq = arg; + struct work_struct *work; + + pthread_mutex_lock(&wq_lock); + while (1) { + __set_current_state(TASK_INTERRUPTIBLE); + work = list_first_entry_or_null(&wq->pending_work, + struct work_struct, entry); + wq->current_work = work; + + if (kthread_should_stop()) { + BUG_ON(wq->current_work); + break; + } + + if (!work) { + pthread_mutex_unlock(&wq_lock); + schedule(); + pthread_mutex_lock(&wq_lock); + continue; + } + + BUG_ON(!test_bit(WORK_PENDING_BIT, work_data_bits(work))); + list_del_init(&work->entry); + clear_work_pending(work); + + pthread_mutex_unlock(&wq_lock); + work->func(work); + pthread_mutex_lock(&wq_lock); + + pthread_cond_broadcast(&wq->work_finished); + } + pthread_mutex_unlock(&wq_lock); + + return 0; +} + +void destroy_workqueue(struct workqueue_struct *wq) +{ + kthread_stop(wq->worker); + + pthread_mutex_lock(&wq_lock); + list_del(&wq->list); + pthread_mutex_unlock(&wq_lock); + + kfree(wq); +} + +struct workqueue_struct *alloc_workqueue(const char *fmt, + unsigned flags, + int max_active, + ...) +{ + va_list args; + struct workqueue_struct *wq; + + wq = kzalloc(sizeof(*wq), GFP_KERNEL); + if (!wq) + return NULL; + + INIT_LIST_HEAD(&wq->list); + INIT_LIST_HEAD(&wq->pending_work); + + pthread_cond_init(&wq->work_finished, NULL); + + va_start(args, max_active); + vsnprintf(wq->name, sizeof(wq->name), fmt, args); + va_end(args); + + wq->worker = kthread_run(worker_thread, wq, "%s", wq->name); + if (IS_ERR(wq->worker)) { + kfree(wq); + return NULL; + } + + pthread_mutex_lock(&wq_lock); + list_add(&wq->list, &wq_list); + pthread_mutex_unlock(&wq_lock); + + return wq; +} + +struct workqueue_struct *system_wq; +struct workqueue_struct *system_highpri_wq; +struct workqueue_struct *system_long_wq; +struct workqueue_struct *system_unbound_wq; +struct workqueue_struct *system_freezable_wq; +struct workqueue_struct *system_power_efficient_wq; +struct workqueue_struct *system_freezable_power_efficient_wq; + +__attribute__((constructor(102))) +static void wq_init(void) +{ + system_wq = alloc_workqueue("events", 0, 0); + system_highpri_wq = alloc_workqueue("events_highpri", WQ_HIGHPRI, 0); + system_long_wq = alloc_workqueue("events_long", 0, 0); + system_unbound_wq = alloc_workqueue("events_unbound", WQ_UNBOUND, + WQ_UNBOUND_MAX_ACTIVE); + system_freezable_wq = alloc_workqueue("events_freezable", + WQ_FREEZABLE, 0); + system_power_efficient_wq = alloc_workqueue("events_power_efficient", + WQ_POWER_EFFICIENT, 0); + system_freezable_power_efficient_wq = alloc_workqueue("events_freezable_power_efficient", + WQ_FREEZABLE | WQ_POWER_EFFICIENT, + 0); + BUG_ON(!system_wq || !system_highpri_wq || !system_long_wq || + !system_unbound_wq || !system_freezable_wq || + !system_power_efficient_wq || + !system_freezable_power_efficient_wq); +} diff --git a/linux/zlib_deflate/deflate.c b/linux/zlib_deflate/deflate.c new file mode 100644 index 0000000..d20ef45 --- /dev/null +++ b/linux/zlib_deflate/deflate.c @@ -0,0 +1,1137 @@ +/* +++ deflate.c */ +/* deflate.c -- compress data using the deflation algorithm + * Copyright (C) 1995-1996 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* + * ALGORITHM + * + * The "deflation" process depends on being able to identify portions + * of the input text which are identical to earlier input (within a + * sliding window trailing behind the input currently being processed). + * + * The most straightforward technique turns out to be the fastest for + * most input files: try all possible matches and select the longest. + * The key feature of this algorithm is that insertions into the string + * dictionary are very simple and thus fast, and deletions are avoided + * completely. Insertions are performed at each input character, whereas + * string matches are performed only when the previous match ends. So it + * is preferable to spend more time in matches to allow very fast string + * insertions and avoid deletions. The matching algorithm for small + * strings is inspired from that of Rabin & Karp. A brute force approach + * is used to find longer strings when a small match has been found. + * A similar algorithm is used in comic (by Jan-Mark Wams) and freeze + * (by Leonid Broukhis). + * A previous version of this file used a more sophisticated algorithm + * (by Fiala and Greene) which is guaranteed to run in linear amortized + * time, but has a larger average cost, uses more memory and is patented. + * However the F&G algorithm may be faster for some highly redundant + * files if the parameter max_chain_length (described below) is too large. + * + * ACKNOWLEDGEMENTS + * + * The idea of lazy evaluation of matches is due to Jan-Mark Wams, and + * I found it in 'freeze' written by Leonid Broukhis. + * Thanks to many people for bug reports and testing. + * + * REFERENCES + * + * Deutsch, L.P.,"DEFLATE Compressed Data Format Specification". + * Available in ftp://ds.internic.net/rfc/rfc1951.txt + * + * A description of the Rabin and Karp algorithm is given in the book + * "Algorithms" by R. Sedgewick, Addison-Wesley, p252. + * + * Fiala,E.R., and Greene,D.H. + * Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595 + * + */ + +#include <linux/module.h> +#include <linux/zutil.h> +#include "defutil.h" + + +/* =========================================================================== + * Function prototypes. + */ +typedef enum { + need_more, /* block not completed, need more input or more output */ + block_done, /* block flush performed */ + finish_started, /* finish started, need only more output at next deflate */ + finish_done /* finish done, accept no more input or output */ +} block_state; + +typedef block_state (*compress_func) (deflate_state *s, int flush); +/* Compression function. Returns the block state after the call. */ + +static void fill_window (deflate_state *s); +static block_state deflate_stored (deflate_state *s, int flush); +static block_state deflate_fast (deflate_state *s, int flush); +static block_state deflate_slow (deflate_state *s, int flush); +static void lm_init (deflate_state *s); +static void putShortMSB (deflate_state *s, uInt b); +static void flush_pending (z_streamp strm); +static int read_buf (z_streamp strm, Byte *buf, unsigned size); +static uInt longest_match (deflate_state *s, IPos cur_match); + +#ifdef DEBUG_ZLIB +static void check_match (deflate_state *s, IPos start, IPos match, + int length); +#endif + +/* =========================================================================== + * Local data + */ + +#define NIL 0 +/* Tail of hash chains */ + +#ifndef TOO_FAR +# define TOO_FAR 4096 +#endif +/* Matches of length 3 are discarded if their distance exceeds TOO_FAR */ + +#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1) +/* Minimum amount of lookahead, except at the end of the input file. + * See deflate.c for comments about the MIN_MATCH+1. + */ + +/* Values for max_lazy_match, good_match and max_chain_length, depending on + * the desired pack level (0..9). The values given below have been tuned to + * exclude worst case performance for pathological files. Better values may be + * found for specific files. + */ +typedef struct config_s { + ush good_length; /* reduce lazy search above this match length */ + ush max_lazy; /* do not perform lazy search above this match length */ + ush nice_length; /* quit search above this match length */ + ush max_chain; + compress_func func; +} config; + +static const config configuration_table[10] = { +/* good lazy nice chain */ +/* 0 */ {0, 0, 0, 0, deflate_stored}, /* store only */ +/* 1 */ {4, 4, 8, 4, deflate_fast}, /* maximum speed, no lazy matches */ +/* 2 */ {4, 5, 16, 8, deflate_fast}, +/* 3 */ {4, 6, 32, 32, deflate_fast}, + +/* 4 */ {4, 4, 16, 16, deflate_slow}, /* lazy matches */ +/* 5 */ {8, 16, 32, 32, deflate_slow}, +/* 6 */ {8, 16, 128, 128, deflate_slow}, +/* 7 */ {8, 32, 128, 256, deflate_slow}, +/* 8 */ {32, 128, 258, 1024, deflate_slow}, +/* 9 */ {32, 258, 258, 4096, deflate_slow}}; /* maximum compression */ + +/* Note: the deflate() code requires max_lazy >= MIN_MATCH and max_chain >= 4 + * For deflate_fast() (levels <= 3) good is ignored and lazy has a different + * meaning. + */ + +#define EQUAL 0 +/* result of memcmp for equal strings */ + +/* =========================================================================== + * Update a hash value with the given input byte + * IN assertion: all calls to UPDATE_HASH are made with consecutive + * input characters, so that a running hash key can be computed from the + * previous key instead of complete recalculation each time. + */ +#define UPDATE_HASH(s,h,c) (h = (((h)<<s->hash_shift) ^ (c)) & s->hash_mask) + + +/* =========================================================================== + * Insert string str in the dictionary and set match_head to the previous head + * of the hash chain (the most recent string with same hash key). Return + * the previous length of the hash chain. + * IN assertion: all calls to INSERT_STRING are made with consecutive + * input characters and the first MIN_MATCH bytes of str are valid + * (except for the last MIN_MATCH-1 bytes of the input file). + */ +#define INSERT_STRING(s, str, match_head) \ + (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ + s->prev[(str) & s->w_mask] = match_head = s->head[s->ins_h], \ + s->head[s->ins_h] = (Pos)(str)) + +/* =========================================================================== + * Initialize the hash table (avoiding 64K overflow for 16 bit systems). + * prev[] will be initialized on the fly. + */ +#define CLEAR_HASH(s) \ + s->head[s->hash_size-1] = NIL; \ + memset((char *)s->head, 0, (unsigned)(s->hash_size-1)*sizeof(*s->head)); + +/* ========================================================================= */ +int zlib_deflateInit2( + z_streamp strm, + int level, + int method, + int windowBits, + int memLevel, + int strategy +) +{ + deflate_state *s; + int noheader = 0; + deflate_workspace *mem; + char *next; + + ush *overlay; + /* We overlay pending_buf and d_buf+l_buf. This works since the average + * output size for (length,distance) codes is <= 24 bits. + */ + + if (strm == NULL) return Z_STREAM_ERROR; + + strm->msg = NULL; + + if (level == Z_DEFAULT_COMPRESSION) level = 6; + + mem = (deflate_workspace *) strm->workspace; + + if (windowBits < 0) { /* undocumented feature: suppress zlib header */ + noheader = 1; + windowBits = -windowBits; + } + if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED || + windowBits < 9 || windowBits > 15 || level < 0 || level > 9 || + strategy < 0 || strategy > Z_HUFFMAN_ONLY) { + return Z_STREAM_ERROR; + } + + /* + * Direct the workspace's pointers to the chunks that were allocated + * along with the deflate_workspace struct. + */ + next = (char *) mem; + next += sizeof(*mem); + mem->window_memory = (Byte *) next; + next += zlib_deflate_window_memsize(windowBits); + mem->prev_memory = (Pos *) next; + next += zlib_deflate_prev_memsize(windowBits); + mem->head_memory = (Pos *) next; + next += zlib_deflate_head_memsize(memLevel); + mem->overlay_memory = next; + + s = (deflate_state *) &(mem->deflate_memory); + strm->state = (struct internal_state *)s; + s->strm = strm; + + s->noheader = noheader; + s->w_bits = windowBits; + s->w_size = 1 << s->w_bits; + s->w_mask = s->w_size - 1; + + s->hash_bits = memLevel + 7; + s->hash_size = 1 << s->hash_bits; + s->hash_mask = s->hash_size - 1; + s->hash_shift = ((s->hash_bits+MIN_MATCH-1)/MIN_MATCH); + + s->window = (Byte *) mem->window_memory; + s->prev = (Pos *) mem->prev_memory; + s->head = (Pos *) mem->head_memory; + + s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */ + + overlay = (ush *) mem->overlay_memory; + s->pending_buf = (uch *) overlay; + s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L); + + s->d_buf = overlay + s->lit_bufsize/sizeof(ush); + s->l_buf = s->pending_buf + (1+sizeof(ush))*s->lit_bufsize; + + s->level = level; + s->strategy = strategy; + s->method = (Byte)method; + + return zlib_deflateReset(strm); +} + +/* ========================================================================= */ +int zlib_deflateReset( + z_streamp strm +) +{ + deflate_state *s; + + if (strm == NULL || strm->state == NULL) + return Z_STREAM_ERROR; + + strm->total_in = strm->total_out = 0; + strm->msg = NULL; + strm->data_type = Z_UNKNOWN; + + s = (deflate_state *)strm->state; + s->pending = 0; + s->pending_out = s->pending_buf; + + if (s->noheader < 0) { + s->noheader = 0; /* was set to -1 by deflate(..., Z_FINISH); */ + } + s->status = s->noheader ? BUSY_STATE : INIT_STATE; + strm->adler = 1; + s->last_flush = Z_NO_FLUSH; + + zlib_tr_init(s); + lm_init(s); + + return Z_OK; +} + +/* ========================================================================= + * Put a short in the pending buffer. The 16-bit value is put in MSB order. + * IN assertion: the stream state is correct and there is enough room in + * pending_buf. + */ +static void putShortMSB( + deflate_state *s, + uInt b +) +{ + put_byte(s, (Byte)(b >> 8)); + put_byte(s, (Byte)(b & 0xff)); +} + +/* ========================================================================= + * Flush as much pending output as possible. All deflate() output goes + * through this function so some applications may wish to modify it + * to avoid allocating a large strm->next_out buffer and copying into it. + * (See also read_buf()). + */ +static void flush_pending( + z_streamp strm +) +{ + deflate_state *s = (deflate_state *) strm->state; + unsigned len = s->pending; + + if (len > strm->avail_out) len = strm->avail_out; + if (len == 0) return; + + if (strm->next_out != NULL) { + memcpy(strm->next_out, s->pending_out, len); + strm->next_out += len; + } + s->pending_out += len; + strm->total_out += len; + strm->avail_out -= len; + s->pending -= len; + if (s->pending == 0) { + s->pending_out = s->pending_buf; + } +} + +/* ========================================================================= */ +int zlib_deflate( + z_streamp strm, + int flush +) +{ + int old_flush; /* value of flush param for previous deflate call */ + deflate_state *s; + + if (strm == NULL || strm->state == NULL || + flush > Z_FINISH || flush < 0) { + return Z_STREAM_ERROR; + } + s = (deflate_state *) strm->state; + + if ((strm->next_in == NULL && strm->avail_in != 0) || + (s->status == FINISH_STATE && flush != Z_FINISH)) { + return Z_STREAM_ERROR; + } + if (strm->avail_out == 0) return Z_BUF_ERROR; + + s->strm = strm; /* just in case */ + old_flush = s->last_flush; + s->last_flush = flush; + + /* Write the zlib header */ + if (s->status == INIT_STATE) { + + uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8; + uInt level_flags = (s->level-1) >> 1; + + if (level_flags > 3) level_flags = 3; + header |= (level_flags << 6); + if (s->strstart != 0) header |= PRESET_DICT; + header += 31 - (header % 31); + + s->status = BUSY_STATE; + putShortMSB(s, header); + + /* Save the adler32 of the preset dictionary: */ + if (s->strstart != 0) { + putShortMSB(s, (uInt)(strm->adler >> 16)); + putShortMSB(s, (uInt)(strm->adler & 0xffff)); + } + strm->adler = 1L; + } + + /* Flush as much pending output as possible */ + if (s->pending != 0) { + flush_pending(strm); + if (strm->avail_out == 0) { + /* Since avail_out is 0, deflate will be called again with + * more output space, but possibly with both pending and + * avail_in equal to zero. There won't be anything to do, + * but this is not an error situation so make sure we + * return OK instead of BUF_ERROR at next call of deflate: + */ + s->last_flush = -1; + return Z_OK; + } + + /* Make sure there is something to do and avoid duplicate consecutive + * flushes. For repeated and useless calls with Z_FINISH, we keep + * returning Z_STREAM_END instead of Z_BUFF_ERROR. + */ + } else if (strm->avail_in == 0 && flush <= old_flush && + flush != Z_FINISH) { + return Z_BUF_ERROR; + } + + /* User must not provide more input after the first FINISH: */ + if (s->status == FINISH_STATE && strm->avail_in != 0) { + return Z_BUF_ERROR; + } + + /* Start a new block or continue the current one. + */ + if (strm->avail_in != 0 || s->lookahead != 0 || + (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) { + block_state bstate; + + bstate = (*(configuration_table[s->level].func))(s, flush); + + if (bstate == finish_started || bstate == finish_done) { + s->status = FINISH_STATE; + } + if (bstate == need_more || bstate == finish_started) { + if (strm->avail_out == 0) { + s->last_flush = -1; /* avoid BUF_ERROR next call, see above */ + } + return Z_OK; + /* If flush != Z_NO_FLUSH && avail_out == 0, the next call + * of deflate should use the same flush parameter to make sure + * that the flush is complete. So we don't have to output an + * empty block here, this will be done at next call. This also + * ensures that for a very small output buffer, we emit at most + * one empty block. + */ + } + if (bstate == block_done) { + if (flush == Z_PARTIAL_FLUSH) { + zlib_tr_align(s); + } else if (flush == Z_PACKET_FLUSH) { + /* Output just the 3-bit `stored' block type value, + but not a zero length. */ + zlib_tr_stored_type_only(s); + } else { /* FULL_FLUSH or SYNC_FLUSH */ + zlib_tr_stored_block(s, (char*)0, 0L, 0); + /* For a full flush, this empty block will be recognized + * as a special marker by inflate_sync(). + */ + if (flush == Z_FULL_FLUSH) { + CLEAR_HASH(s); /* forget history */ + } + } + flush_pending(strm); + if (strm->avail_out == 0) { + s->last_flush = -1; /* avoid BUF_ERROR at next call, see above */ + return Z_OK; + } + } + } + Assert(strm->avail_out > 0, "bug2"); + + if (flush != Z_FINISH) return Z_OK; + if (s->noheader) return Z_STREAM_END; + + /* Write the zlib trailer (adler32) */ + putShortMSB(s, (uInt)(strm->adler >> 16)); + putShortMSB(s, (uInt)(strm->adler & 0xffff)); + flush_pending(strm); + /* If avail_out is zero, the application will call deflate again + * to flush the rest. + */ + s->noheader = -1; /* write the trailer only once! */ + return s->pending != 0 ? Z_OK : Z_STREAM_END; +} + +/* ========================================================================= */ +int zlib_deflateEnd( + z_streamp strm +) +{ + int status; + deflate_state *s; + + if (strm == NULL || strm->state == NULL) return Z_STREAM_ERROR; + s = (deflate_state *) strm->state; + + status = s->status; + if (status != INIT_STATE && status != BUSY_STATE && + status != FINISH_STATE) { + return Z_STREAM_ERROR; + } + + strm->state = NULL; + + return status == BUSY_STATE ? Z_DATA_ERROR : Z_OK; +} + +/* =========================================================================== + * Read a new buffer from the current input stream, update the adler32 + * and total number of bytes read. All deflate() input goes through + * this function so some applications may wish to modify it to avoid + * allocating a large strm->next_in buffer and copying from it. + * (See also flush_pending()). + */ +static int read_buf( + z_streamp strm, + Byte *buf, + unsigned size +) +{ + unsigned len = strm->avail_in; + + if (len > size) len = size; + if (len == 0) return 0; + + strm->avail_in -= len; + + if (!((deflate_state *)(strm->state))->noheader) { + strm->adler = zlib_adler32(strm->adler, strm->next_in, len); + } + memcpy(buf, strm->next_in, len); + strm->next_in += len; + strm->total_in += len; + + return (int)len; +} + +/* =========================================================================== + * Initialize the "longest match" routines for a new zlib stream + */ +static void lm_init( + deflate_state *s +) +{ + s->window_size = (ulg)2L*s->w_size; + + CLEAR_HASH(s); + + /* Set the default configuration parameters: + */ + s->max_lazy_match = configuration_table[s->level].max_lazy; + s->good_match = configuration_table[s->level].good_length; + s->nice_match = configuration_table[s->level].nice_length; + s->max_chain_length = configuration_table[s->level].max_chain; + + s->strstart = 0; + s->block_start = 0L; + s->lookahead = 0; + s->match_length = s->prev_length = MIN_MATCH-1; + s->match_available = 0; + s->ins_h = 0; +} + +/* =========================================================================== + * Set match_start to the longest match starting at the given string and + * return its length. Matches shorter or equal to prev_length are discarded, + * in which case the result is equal to prev_length and match_start is + * garbage. + * IN assertions: cur_match is the head of the hash chain for the current + * string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1 + * OUT assertion: the match length is not greater than s->lookahead. + */ +/* For 80x86 and 680x0, an optimized version will be provided in match.asm or + * match.S. The code will be functionally equivalent. + */ +static uInt longest_match( + deflate_state *s, + IPos cur_match /* current match */ +) +{ + unsigned chain_length = s->max_chain_length;/* max hash chain length */ + register Byte *scan = s->window + s->strstart; /* current string */ + register Byte *match; /* matched string */ + register int len; /* length of current match */ + int best_len = s->prev_length; /* best match length so far */ + int nice_match = s->nice_match; /* stop if match long enough */ + IPos limit = s->strstart > (IPos)MAX_DIST(s) ? + s->strstart - (IPos)MAX_DIST(s) : NIL; + /* Stop when cur_match becomes <= limit. To simplify the code, + * we prevent matches with the string of window index 0. + */ + Pos *prev = s->prev; + uInt wmask = s->w_mask; + +#ifdef UNALIGNED_OK + /* Compare two bytes at a time. Note: this is not always beneficial. + * Try with and without -DUNALIGNED_OK to check. + */ + register Byte *strend = s->window + s->strstart + MAX_MATCH - 1; + register ush scan_start = *(ush*)scan; + register ush scan_end = *(ush*)(scan+best_len-1); +#else + register Byte *strend = s->window + s->strstart + MAX_MATCH; + register Byte scan_end1 = scan[best_len-1]; + register Byte scan_end = scan[best_len]; +#endif + + /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16. + * It is easy to get rid of this optimization if necessary. + */ + Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever"); + + /* Do not waste too much time if we already have a good match: */ + if (s->prev_length >= s->good_match) { + chain_length >>= 2; + } + /* Do not look for matches beyond the end of the input. This is necessary + * to make deflate deterministic. + */ + if ((uInt)nice_match > s->lookahead) nice_match = s->lookahead; + + Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead"); + + do { + Assert(cur_match < s->strstart, "no future"); + match = s->window + cur_match; + + /* Skip to next match if the match length cannot increase + * or if the match length is less than 2: + */ +#if (defined(UNALIGNED_OK) && MAX_MATCH == 258) + /* This code assumes sizeof(unsigned short) == 2. Do not use + * UNALIGNED_OK if your compiler uses a different size. + */ + if (*(ush*)(match+best_len-1) != scan_end || + *(ush*)match != scan_start) continue; + + /* It is not necessary to compare scan[2] and match[2] since they are + * always equal when the other bytes match, given that the hash keys + * are equal and that HASH_BITS >= 8. Compare 2 bytes at a time at + * strstart+3, +5, ... up to strstart+257. We check for insufficient + * lookahead only every 4th comparison; the 128th check will be made + * at strstart+257. If MAX_MATCH-2 is not a multiple of 8, it is + * necessary to put more guard bytes at the end of the window, or + * to check more often for insufficient lookahead. + */ + Assert(scan[2] == match[2], "scan[2]?"); + scan++, match++; + do { + } while (*(ush*)(scan+=2) == *(ush*)(match+=2) && + *(ush*)(scan+=2) == *(ush*)(match+=2) && + *(ush*)(scan+=2) == *(ush*)(match+=2) && + *(ush*)(scan+=2) == *(ush*)(match+=2) && + scan < strend); + /* The funny "do {}" generates better code on most compilers */ + + /* Here, scan <= window+strstart+257 */ + Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); + if (*scan == *match) scan++; + + len = (MAX_MATCH - 1) - (int)(strend-scan); + scan = strend - (MAX_MATCH-1); + +#else /* UNALIGNED_OK */ + + if (match[best_len] != scan_end || + match[best_len-1] != scan_end1 || + *match != *scan || + *++match != scan[1]) continue; + + /* The check at best_len-1 can be removed because it will be made + * again later. (This heuristic is not always a win.) + * It is not necessary to compare scan[2] and match[2] since they + * are always equal when the other bytes match, given that + * the hash keys are equal and that HASH_BITS >= 8. + */ + scan += 2, match++; + Assert(*scan == *match, "match[2]?"); + + /* We check for insufficient lookahead only every 8th comparison; + * the 256th check will be made at strstart+258. + */ + do { + } while (*++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + scan < strend); + + Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); + + len = MAX_MATCH - (int)(strend - scan); + scan = strend - MAX_MATCH; + +#endif /* UNALIGNED_OK */ + + if (len > best_len) { + s->match_start = cur_match; + best_len = len; + if (len >= nice_match) break; +#ifdef UNALIGNED_OK + scan_end = *(ush*)(scan+best_len-1); +#else + scan_end1 = scan[best_len-1]; + scan_end = scan[best_len]; +#endif + } + } while ((cur_match = prev[cur_match & wmask]) > limit + && --chain_length != 0); + + if ((uInt)best_len <= s->lookahead) return best_len; + return s->lookahead; +} + +#ifdef DEBUG_ZLIB +/* =========================================================================== + * Check that the match at match_start is indeed a match. + */ +static void check_match( + deflate_state *s, + IPos start, + IPos match, + int length +) +{ + /* check that the match is indeed a match */ + if (memcmp((char *)s->window + match, + (char *)s->window + start, length) != EQUAL) { + fprintf(stderr, " start %u, match %u, length %d\n", + start, match, length); + do { + fprintf(stderr, "%c%c", s->window[match++], s->window[start++]); + } while (--length != 0); + z_error("invalid match"); + } + if (z_verbose > 1) { + fprintf(stderr,"\\[%d,%d]", start-match, length); + do { putc(s->window[start++], stderr); } while (--length != 0); + } +} +#else +# define check_match(s, start, match, length) +#endif + +/* =========================================================================== + * Fill the window when the lookahead becomes insufficient. + * Updates strstart and lookahead. + * + * IN assertion: lookahead < MIN_LOOKAHEAD + * OUT assertions: strstart <= window_size-MIN_LOOKAHEAD + * At least one byte has been read, or avail_in == 0; reads are + * performed for at least two bytes (required for the zip translate_eol + * option -- not supported here). + */ +static void fill_window( + deflate_state *s +) +{ + register unsigned n, m; + register Pos *p; + unsigned more; /* Amount of free space at the end of the window. */ + uInt wsize = s->w_size; + + do { + more = (unsigned)(s->window_size -(ulg)s->lookahead -(ulg)s->strstart); + + /* Deal with !@#$% 64K limit: */ + if (more == 0 && s->strstart == 0 && s->lookahead == 0) { + more = wsize; + + } else if (more == (unsigned)(-1)) { + /* Very unlikely, but possible on 16 bit machine if strstart == 0 + * and lookahead == 1 (input done one byte at time) + */ + more--; + + /* If the window is almost full and there is insufficient lookahead, + * move the upper half to the lower one to make room in the upper half. + */ + } else if (s->strstart >= wsize+MAX_DIST(s)) { + + memcpy((char *)s->window, (char *)s->window+wsize, + (unsigned)wsize); + s->match_start -= wsize; + s->strstart -= wsize; /* we now have strstart >= MAX_DIST */ + s->block_start -= (long) wsize; + + /* Slide the hash table (could be avoided with 32 bit values + at the expense of memory usage). We slide even when level == 0 + to keep the hash table consistent if we switch back to level > 0 + later. (Using level 0 permanently is not an optimal usage of + zlib, so we don't care about this pathological case.) + */ + n = s->hash_size; + p = &s->head[n]; + do { + m = *--p; + *p = (Pos)(m >= wsize ? m-wsize : NIL); + } while (--n); + + n = wsize; + p = &s->prev[n]; + do { + m = *--p; + *p = (Pos)(m >= wsize ? m-wsize : NIL); + /* If n is not on any hash chain, prev[n] is garbage but + * its value will never be used. + */ + } while (--n); + more += wsize; + } + if (s->strm->avail_in == 0) return; + + /* If there was no sliding: + * strstart <= WSIZE+MAX_DIST-1 && lookahead <= MIN_LOOKAHEAD - 1 && + * more == window_size - lookahead - strstart + * => more >= window_size - (MIN_LOOKAHEAD-1 + WSIZE + MAX_DIST-1) + * => more >= window_size - 2*WSIZE + 2 + * In the BIG_MEM or MMAP case (not yet supported), + * window_size == input_size + MIN_LOOKAHEAD && + * strstart + s->lookahead <= input_size => more >= MIN_LOOKAHEAD. + * Otherwise, window_size == 2*WSIZE so more >= 2. + * If there was sliding, more >= WSIZE. So in all cases, more >= 2. + */ + Assert(more >= 2, "more < 2"); + + n = read_buf(s->strm, s->window + s->strstart + s->lookahead, more); + s->lookahead += n; + + /* Initialize the hash value now that we have some input: */ + if (s->lookahead >= MIN_MATCH) { + s->ins_h = s->window[s->strstart]; + UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]); +#if MIN_MATCH != 3 + Call UPDATE_HASH() MIN_MATCH-3 more times +#endif + } + /* If the whole input has less than MIN_MATCH bytes, ins_h is garbage, + * but this is not important since only literal bytes will be emitted. + */ + + } while (s->lookahead < MIN_LOOKAHEAD && s->strm->avail_in != 0); +} + +/* =========================================================================== + * Flush the current block, with given end-of-file flag. + * IN assertion: strstart is set to the end of the current match. + */ +#define FLUSH_BLOCK_ONLY(s, eof) { \ + zlib_tr_flush_block(s, (s->block_start >= 0L ? \ + (char *)&s->window[(unsigned)s->block_start] : \ + NULL), \ + (ulg)((long)s->strstart - s->block_start), \ + (eof)); \ + s->block_start = s->strstart; \ + flush_pending(s->strm); \ + Tracev((stderr,"[FLUSH]")); \ +} + +/* Same but force premature exit if necessary. */ +#define FLUSH_BLOCK(s, eof) { \ + FLUSH_BLOCK_ONLY(s, eof); \ + if (s->strm->avail_out == 0) return (eof) ? finish_started : need_more; \ +} + +/* =========================================================================== + * Copy without compression as much as possible from the input stream, return + * the current block state. + * This function does not insert new strings in the dictionary since + * uncompressible data is probably not useful. This function is used + * only for the level=0 compression option. + * NOTE: this function should be optimized to avoid extra copying from + * window to pending_buf. + */ +static block_state deflate_stored( + deflate_state *s, + int flush +) +{ + /* Stored blocks are limited to 0xffff bytes, pending_buf is limited + * to pending_buf_size, and each stored block has a 5 byte header: + */ + ulg max_block_size = 0xffff; + ulg max_start; + + if (max_block_size > s->pending_buf_size - 5) { + max_block_size = s->pending_buf_size - 5; + } + + /* Copy as much as possible from input to output: */ + for (;;) { + /* Fill the window as much as possible: */ + if (s->lookahead <= 1) { + + Assert(s->strstart < s->w_size+MAX_DIST(s) || + s->block_start >= (long)s->w_size, "slide too late"); + + fill_window(s); + if (s->lookahead == 0 && flush == Z_NO_FLUSH) return need_more; + + if (s->lookahead == 0) break; /* flush the current block */ + } + Assert(s->block_start >= 0L, "block gone"); + + s->strstart += s->lookahead; + s->lookahead = 0; + + /* Emit a stored block if pending_buf will be full: */ + max_start = s->block_start + max_block_size; + if (s->strstart == 0 || (ulg)s->strstart >= max_start) { + /* strstart == 0 is possible when wraparound on 16-bit machine */ + s->lookahead = (uInt)(s->strstart - max_start); + s->strstart = (uInt)max_start; + FLUSH_BLOCK(s, 0); + } + /* Flush if we may have to slide, otherwise block_start may become + * negative and the data will be gone: + */ + if (s->strstart - (uInt)s->block_start >= MAX_DIST(s)) { + FLUSH_BLOCK(s, 0); + } + } + FLUSH_BLOCK(s, flush == Z_FINISH); + return flush == Z_FINISH ? finish_done : block_done; +} + +/* =========================================================================== + * Compress as much as possible from the input stream, return the current + * block state. + * This function does not perform lazy evaluation of matches and inserts + * new strings in the dictionary only for unmatched strings or for short + * matches. It is used only for the fast compression options. + */ +static block_state deflate_fast( + deflate_state *s, + int flush +) +{ + IPos hash_head = NIL; /* head of the hash chain */ + int bflush; /* set if current block must be flushed */ + + for (;;) { + /* Make sure that we always have enough lookahead, except + * at the end of the input file. We need MAX_MATCH bytes + * for the next match, plus MIN_MATCH bytes to insert the + * string following the next match. + */ + if (s->lookahead < MIN_LOOKAHEAD) { + fill_window(s); + if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { + return need_more; + } + if (s->lookahead == 0) break; /* flush the current block */ + } + + /* Insert the string window[strstart .. strstart+2] in the + * dictionary, and set hash_head to the head of the hash chain: + */ + if (s->lookahead >= MIN_MATCH) { + INSERT_STRING(s, s->strstart, hash_head); + } + + /* Find the longest match, discarding those <= prev_length. + * At this point we have always match_length < MIN_MATCH + */ + if (hash_head != NIL && s->strstart - hash_head <= MAX_DIST(s)) { + /* To simplify the code, we prevent matches with the string + * of window index 0 (in particular we have to avoid a match + * of the string with itself at the start of the input file). + */ + if (s->strategy != Z_HUFFMAN_ONLY) { + s->match_length = longest_match (s, hash_head); + } + /* longest_match() sets match_start */ + } + if (s->match_length >= MIN_MATCH) { + check_match(s, s->strstart, s->match_start, s->match_length); + + bflush = zlib_tr_tally(s, s->strstart - s->match_start, + s->match_length - MIN_MATCH); + + s->lookahead -= s->match_length; + + /* Insert new strings in the hash table only if the match length + * is not too large. This saves time but degrades compression. + */ + if (s->match_length <= s->max_insert_length && + s->lookahead >= MIN_MATCH) { + s->match_length--; /* string at strstart already in hash table */ + do { + s->strstart++; + INSERT_STRING(s, s->strstart, hash_head); + /* strstart never exceeds WSIZE-MAX_MATCH, so there are + * always MIN_MATCH bytes ahead. + */ + } while (--s->match_length != 0); + s->strstart++; + } else { + s->strstart += s->match_length; + s->match_length = 0; + s->ins_h = s->window[s->strstart]; + UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]); +#if MIN_MATCH != 3 + Call UPDATE_HASH() MIN_MATCH-3 more times +#endif + /* If lookahead < MIN_MATCH, ins_h is garbage, but it does not + * matter since it will be recomputed at next deflate call. + */ + } + } else { + /* No match, output a literal byte */ + Tracevv((stderr,"%c", s->window[s->strstart])); + bflush = zlib_tr_tally (s, 0, s->window[s->strstart]); + s->lookahead--; + s->strstart++; + } + if (bflush) FLUSH_BLOCK(s, 0); + } + FLUSH_BLOCK(s, flush == Z_FINISH); + return flush == Z_FINISH ? finish_done : block_done; +} + +/* =========================================================================== + * Same as above, but achieves better compression. We use a lazy + * evaluation for matches: a match is finally adopted only if there is + * no better match at the next window position. + */ +static block_state deflate_slow( + deflate_state *s, + int flush +) +{ + IPos hash_head = NIL; /* head of hash chain */ + int bflush; /* set if current block must be flushed */ + + /* Process the input block. */ + for (;;) { + /* Make sure that we always have enough lookahead, except + * at the end of the input file. We need MAX_MATCH bytes + * for the next match, plus MIN_MATCH bytes to insert the + * string following the next match. + */ + if (s->lookahead < MIN_LOOKAHEAD) { + fill_window(s); + if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { + return need_more; + } + if (s->lookahead == 0) break; /* flush the current block */ + } + + /* Insert the string window[strstart .. strstart+2] in the + * dictionary, and set hash_head to the head of the hash chain: + */ + if (s->lookahead >= MIN_MATCH) { + INSERT_STRING(s, s->strstart, hash_head); + } + + /* Find the longest match, discarding those <= prev_length. + */ + s->prev_length = s->match_length, s->prev_match = s->match_start; + s->match_length = MIN_MATCH-1; + + if (hash_head != NIL && s->prev_length < s->max_lazy_match && + s->strstart - hash_head <= MAX_DIST(s)) { + /* To simplify the code, we prevent matches with the string + * of window index 0 (in particular we have to avoid a match + * of the string with itself at the start of the input file). + */ + if (s->strategy != Z_HUFFMAN_ONLY) { + s->match_length = longest_match (s, hash_head); + } + /* longest_match() sets match_start */ + + if (s->match_length <= 5 && (s->strategy == Z_FILTERED || + (s->match_length == MIN_MATCH && + s->strstart - s->match_start > TOO_FAR))) { + + /* If prev_match is also MIN_MATCH, match_start is garbage + * but we will ignore the current match anyway. + */ + s->match_length = MIN_MATCH-1; + } + } + /* If there was a match at the previous step and the current + * match is not better, output the previous match: + */ + if (s->prev_length >= MIN_MATCH && s->match_length <= s->prev_length) { + uInt max_insert = s->strstart + s->lookahead - MIN_MATCH; + /* Do not insert strings in hash table beyond this. */ + + check_match(s, s->strstart-1, s->prev_match, s->prev_length); + + bflush = zlib_tr_tally(s, s->strstart -1 - s->prev_match, + s->prev_length - MIN_MATCH); + + /* Insert in hash table all strings up to the end of the match. + * strstart-1 and strstart are already inserted. If there is not + * enough lookahead, the last two strings are not inserted in + * the hash table. + */ + s->lookahead -= s->prev_length-1; + s->prev_length -= 2; + do { + if (++s->strstart <= max_insert) { + INSERT_STRING(s, s->strstart, hash_head); + } + } while (--s->prev_length != 0); + s->match_available = 0; + s->match_length = MIN_MATCH-1; + s->strstart++; + + if (bflush) FLUSH_BLOCK(s, 0); + + } else if (s->match_available) { + /* If there was no match at the previous position, output a + * single literal. If there was a match but the current match + * is longer, truncate the previous match to a single literal. + */ + Tracevv((stderr,"%c", s->window[s->strstart-1])); + if (zlib_tr_tally (s, 0, s->window[s->strstart-1])) { + FLUSH_BLOCK_ONLY(s, 0); + } + s->strstart++; + s->lookahead--; + if (s->strm->avail_out == 0) return need_more; + } else { + /* There is no previous match to compare with, wait for + * the next step to decide. + */ + s->match_available = 1; + s->strstart++; + s->lookahead--; + } + } + Assert (flush != Z_NO_FLUSH, "no flush?"); + if (s->match_available) { + Tracevv((stderr,"%c", s->window[s->strstart-1])); + zlib_tr_tally (s, 0, s->window[s->strstart-1]); + s->match_available = 0; + } + FLUSH_BLOCK(s, flush == Z_FINISH); + return flush == Z_FINISH ? finish_done : block_done; +} + +int zlib_deflate_workspacesize(int windowBits, int memLevel) +{ + if (windowBits < 0) /* undocumented feature: suppress zlib header */ + windowBits = -windowBits; + + /* Since the return value is typically passed to vmalloc() unchecked... */ + BUG_ON(memLevel < 1 || memLevel > MAX_MEM_LEVEL || windowBits < 9 || + windowBits > 15); + + return sizeof(deflate_workspace) + + zlib_deflate_window_memsize(windowBits) + + zlib_deflate_prev_memsize(windowBits) + + zlib_deflate_head_memsize(memLevel) + + zlib_deflate_overlay_memsize(memLevel); +} diff --git a/linux/zlib_deflate/deftree.c b/linux/zlib_deflate/deftree.c new file mode 100644 index 0000000..9b1756b --- /dev/null +++ b/linux/zlib_deflate/deftree.c @@ -0,0 +1,1113 @@ +/* +++ trees.c */ +/* trees.c -- output deflated data using Huffman coding + * Copyright (C) 1995-1996 Jean-loup Gailly + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* + * ALGORITHM + * + * The "deflation" process uses several Huffman trees. The more + * common source values are represented by shorter bit sequences. + * + * Each code tree is stored in a compressed form which is itself + * a Huffman encoding of the lengths of all the code strings (in + * ascending order by source values). The actual code strings are + * reconstructed from the lengths in the inflate process, as described + * in the deflate specification. + * + * REFERENCES + * + * Deutsch, L.P.,"'Deflate' Compressed Data Format Specification". + * Available in ftp.uu.net:/pub/archiving/zip/doc/deflate-1.1.doc + * + * Storer, James A. + * Data Compression: Methods and Theory, pp. 49-50. + * Computer Science Press, 1988. ISBN 0-7167-8156-5. + * + * Sedgewick, R. + * Algorithms, p290. + * Addison-Wesley, 1983. ISBN 0-201-06672-6. + */ + +/* From: trees.c,v 1.11 1996/07/24 13:41:06 me Exp $ */ + +/* #include "deflate.h" */ + +#include <linux/zutil.h> +#include <linux/bitrev.h> +#include "defutil.h" + +#ifdef DEBUG_ZLIB +# include <ctype.h> +#endif + +/* =========================================================================== + * Constants + */ + +#define MAX_BL_BITS 7 +/* Bit length codes must not exceed MAX_BL_BITS bits */ + +#define END_BLOCK 256 +/* end of block literal code */ + +#define REP_3_6 16 +/* repeat previous bit length 3-6 times (2 bits of repeat count) */ + +#define REPZ_3_10 17 +/* repeat a zero length 3-10 times (3 bits of repeat count) */ + +#define REPZ_11_138 18 +/* repeat a zero length 11-138 times (7 bits of repeat count) */ + +static const int extra_lbits[LENGTH_CODES] /* extra bits for each length code */ + = {0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0}; + +static const int extra_dbits[D_CODES] /* extra bits for each distance code */ + = {0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; + +static const int extra_blbits[BL_CODES]/* extra bits for each bit length code */ + = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,3,7}; + +static const uch bl_order[BL_CODES] + = {16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15}; +/* The lengths of the bit length codes are sent in order of decreasing + * probability, to avoid transmitting the lengths for unused bit length codes. + */ + +#define Buf_size (8 * 2*sizeof(char)) +/* Number of bits used within bi_buf. (bi_buf might be implemented on + * more than 16 bits on some systems.) + */ + +/* =========================================================================== + * Local data. These are initialized only once. + */ + +static ct_data static_ltree[L_CODES+2]; +/* The static literal tree. Since the bit lengths are imposed, there is no + * need for the L_CODES extra codes used during heap construction. However + * The codes 286 and 287 are needed to build a canonical tree (see zlib_tr_init + * below). + */ + +static ct_data static_dtree[D_CODES]; +/* The static distance tree. (Actually a trivial tree since all codes use + * 5 bits.) + */ + +static uch dist_code[512]; +/* distance codes. The first 256 values correspond to the distances + * 3 .. 258, the last 256 values correspond to the top 8 bits of + * the 15 bit distances. + */ + +static uch length_code[MAX_MATCH-MIN_MATCH+1]; +/* length code for each normalized match length (0 == MIN_MATCH) */ + +static int base_length[LENGTH_CODES]; +/* First normalized length for each code (0 = MIN_MATCH) */ + +static int base_dist[D_CODES]; +/* First normalized distance for each code (0 = distance of 1) */ + +struct static_tree_desc_s { + const ct_data *static_tree; /* static tree or NULL */ + const int *extra_bits; /* extra bits for each code or NULL */ + int extra_base; /* base index for extra_bits */ + int elems; /* max number of elements in the tree */ + int max_length; /* max bit length for the codes */ +}; + +static static_tree_desc static_l_desc = +{static_ltree, extra_lbits, LITERALS+1, L_CODES, MAX_BITS}; + +static static_tree_desc static_d_desc = +{static_dtree, extra_dbits, 0, D_CODES, MAX_BITS}; + +static static_tree_desc static_bl_desc = +{(const ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS}; + +/* =========================================================================== + * Local (static) routines in this file. + */ + +static void tr_static_init (void); +static void init_block (deflate_state *s); +static void pqdownheap (deflate_state *s, ct_data *tree, int k); +static void gen_bitlen (deflate_state *s, tree_desc *desc); +static void gen_codes (ct_data *tree, int max_code, ush *bl_count); +static void build_tree (deflate_state *s, tree_desc *desc); +static void scan_tree (deflate_state *s, ct_data *tree, int max_code); +static void send_tree (deflate_state *s, ct_data *tree, int max_code); +static int build_bl_tree (deflate_state *s); +static void send_all_trees (deflate_state *s, int lcodes, int dcodes, + int blcodes); +static void compress_block (deflate_state *s, ct_data *ltree, + ct_data *dtree); +static void set_data_type (deflate_state *s); +static void bi_windup (deflate_state *s); +static void bi_flush (deflate_state *s); +static void copy_block (deflate_state *s, char *buf, unsigned len, + int header); + +#ifndef DEBUG_ZLIB +# define send_code(s, c, tree) send_bits(s, tree[c].Code, tree[c].Len) + /* Send a code of the given tree. c and tree must not have side effects */ + +#else /* DEBUG_ZLIB */ +# define send_code(s, c, tree) \ + { if (z_verbose>2) fprintf(stderr,"\ncd %3d ",(c)); \ + send_bits(s, tree[c].Code, tree[c].Len); } +#endif + +#define d_code(dist) \ + ((dist) < 256 ? dist_code[dist] : dist_code[256+((dist)>>7)]) +/* Mapping from a distance to a distance code. dist is the distance - 1 and + * must not have side effects. dist_code[256] and dist_code[257] are never + * used. + */ + +/* =========================================================================== + * Send a value on a given number of bits. + * IN assertion: length <= 16 and value fits in length bits. + */ +#ifdef DEBUG_ZLIB +static void send_bits (deflate_state *s, int value, int length); + +static void send_bits( + deflate_state *s, + int value, /* value to send */ + int length /* number of bits */ +) +{ + Tracevv((stderr," l %2d v %4x ", length, value)); + Assert(length > 0 && length <= 15, "invalid length"); + s->bits_sent += (ulg)length; + + /* If not enough room in bi_buf, use (valid) bits from bi_buf and + * (16 - bi_valid) bits from value, leaving (width - (16-bi_valid)) + * unused bits in value. + */ + if (s->bi_valid > (int)Buf_size - length) { + s->bi_buf |= (value << s->bi_valid); + put_short(s, s->bi_buf); + s->bi_buf = (ush)value >> (Buf_size - s->bi_valid); + s->bi_valid += length - Buf_size; + } else { + s->bi_buf |= value << s->bi_valid; + s->bi_valid += length; + } +} +#else /* !DEBUG_ZLIB */ + +#define send_bits(s, value, length) \ +{ int len = length;\ + if (s->bi_valid > (int)Buf_size - len) {\ + int val = value;\ + s->bi_buf |= (val << s->bi_valid);\ + put_short(s, s->bi_buf);\ + s->bi_buf = (ush)val >> (Buf_size - s->bi_valid);\ + s->bi_valid += len - Buf_size;\ + } else {\ + s->bi_buf |= (value) << s->bi_valid;\ + s->bi_valid += len;\ + }\ +} +#endif /* DEBUG_ZLIB */ + +/* =========================================================================== + * Initialize the various 'constant' tables. In a multi-threaded environment, + * this function may be called by two threads concurrently, but this is + * harmless since both invocations do exactly the same thing. + */ +static void tr_static_init(void) +{ + static int static_init_done; + int n; /* iterates over tree elements */ + int bits; /* bit counter */ + int length; /* length value */ + int code; /* code value */ + int dist; /* distance index */ + ush bl_count[MAX_BITS+1]; + /* number of codes at each bit length for an optimal tree */ + + if (static_init_done) return; + + /* Initialize the mapping length (0..255) -> length code (0..28) */ + length = 0; + for (code = 0; code < LENGTH_CODES-1; code++) { + base_length[code] = length; + for (n = 0; n < (1<<extra_lbits[code]); n++) { + length_code[length++] = (uch)code; + } + } + Assert (length == 256, "tr_static_init: length != 256"); + /* Note that the length 255 (match length 258) can be represented + * in two different ways: code 284 + 5 bits or code 285, so we + * overwrite length_code[255] to use the best encoding: + */ + length_code[length-1] = (uch)code; + + /* Initialize the mapping dist (0..32K) -> dist code (0..29) */ + dist = 0; + for (code = 0 ; code < 16; code++) { + base_dist[code] = dist; + for (n = 0; n < (1<<extra_dbits[code]); n++) { + dist_code[dist++] = (uch)code; + } + } + Assert (dist == 256, "tr_static_init: dist != 256"); + dist >>= 7; /* from now on, all distances are divided by 128 */ + for ( ; code < D_CODES; code++) { + base_dist[code] = dist << 7; + for (n = 0; n < (1<<(extra_dbits[code]-7)); n++) { + dist_code[256 + dist++] = (uch)code; + } + } + Assert (dist == 256, "tr_static_init: 256+dist != 512"); + + /* Construct the codes of the static literal tree */ + for (bits = 0; bits <= MAX_BITS; bits++) bl_count[bits] = 0; + n = 0; + while (n <= 143) static_ltree[n++].Len = 8, bl_count[8]++; + while (n <= 255) static_ltree[n++].Len = 9, bl_count[9]++; + while (n <= 279) static_ltree[n++].Len = 7, bl_count[7]++; + while (n <= 287) static_ltree[n++].Len = 8, bl_count[8]++; + /* Codes 286 and 287 do not exist, but we must include them in the + * tree construction to get a canonical Huffman tree (longest code + * all ones) + */ + gen_codes((ct_data *)static_ltree, L_CODES+1, bl_count); + + /* The static distance tree is trivial: */ + for (n = 0; n < D_CODES; n++) { + static_dtree[n].Len = 5; + static_dtree[n].Code = bitrev32((u32)n) >> (32 - 5); + } + static_init_done = 1; +} + +/* =========================================================================== + * Initialize the tree data structures for a new zlib stream. + */ +void zlib_tr_init( + deflate_state *s +) +{ + tr_static_init(); + + s->compressed_len = 0L; + + s->l_desc.dyn_tree = s->dyn_ltree; + s->l_desc.stat_desc = &static_l_desc; + + s->d_desc.dyn_tree = s->dyn_dtree; + s->d_desc.stat_desc = &static_d_desc; + + s->bl_desc.dyn_tree = s->bl_tree; + s->bl_desc.stat_desc = &static_bl_desc; + + s->bi_buf = 0; + s->bi_valid = 0; + s->last_eob_len = 8; /* enough lookahead for inflate */ +#ifdef DEBUG_ZLIB + s->bits_sent = 0L; +#endif + + /* Initialize the first block of the first file: */ + init_block(s); +} + +/* =========================================================================== + * Initialize a new block. + */ +static void init_block( + deflate_state *s +) +{ + int n; /* iterates over tree elements */ + + /* Initialize the trees. */ + for (n = 0; n < L_CODES; n++) s->dyn_ltree[n].Freq = 0; + for (n = 0; n < D_CODES; n++) s->dyn_dtree[n].Freq = 0; + for (n = 0; n < BL_CODES; n++) s->bl_tree[n].Freq = 0; + + s->dyn_ltree[END_BLOCK].Freq = 1; + s->opt_len = s->static_len = 0L; + s->last_lit = s->matches = 0; +} + +#define SMALLEST 1 +/* Index within the heap array of least frequent node in the Huffman tree */ + + +/* =========================================================================== + * Remove the smallest element from the heap and recreate the heap with + * one less element. Updates heap and heap_len. + */ +#define pqremove(s, tree, top) \ +{\ + top = s->heap[SMALLEST]; \ + s->heap[SMALLEST] = s->heap[s->heap_len--]; \ + pqdownheap(s, tree, SMALLEST); \ +} + +/* =========================================================================== + * Compares to subtrees, using the tree depth as tie breaker when + * the subtrees have equal frequency. This minimizes the worst case length. + */ +#define smaller(tree, n, m, depth) \ + (tree[n].Freq < tree[m].Freq || \ + (tree[n].Freq == tree[m].Freq && depth[n] <= depth[m])) + +/* =========================================================================== + * Restore the heap property by moving down the tree starting at node k, + * exchanging a node with the smallest of its two sons if necessary, stopping + * when the heap property is re-established (each father smaller than its + * two sons). + */ +static void pqdownheap( + deflate_state *s, + ct_data *tree, /* the tree to restore */ + int k /* node to move down */ +) +{ + int v = s->heap[k]; + int j = k << 1; /* left son of k */ + while (j <= s->heap_len) { + /* Set j to the smallest of the two sons: */ + if (j < s->heap_len && + smaller(tree, s->heap[j+1], s->heap[j], s->depth)) { + j++; + } + /* Exit if v is smaller than both sons */ + if (smaller(tree, v, s->heap[j], s->depth)) break; + + /* Exchange v with the smallest son */ + s->heap[k] = s->heap[j]; k = j; + + /* And continue down the tree, setting j to the left son of k */ + j <<= 1; + } + s->heap[k] = v; +} + +/* =========================================================================== + * Compute the optimal bit lengths for a tree and update the total bit length + * for the current block. + * IN assertion: the fields freq and dad are set, heap[heap_max] and + * above are the tree nodes sorted by increasing frequency. + * OUT assertions: the field len is set to the optimal bit length, the + * array bl_count contains the frequencies for each bit length. + * The length opt_len is updated; static_len is also updated if stree is + * not null. + */ +static void gen_bitlen( + deflate_state *s, + tree_desc *desc /* the tree descriptor */ +) +{ + ct_data *tree = desc->dyn_tree; + int max_code = desc->max_code; + const ct_data *stree = desc->stat_desc->static_tree; + const int *extra = desc->stat_desc->extra_bits; + int base = desc->stat_desc->extra_base; + int max_length = desc->stat_desc->max_length; + int h; /* heap index */ + int n, m; /* iterate over the tree elements */ + int bits; /* bit length */ + int xbits; /* extra bits */ + ush f; /* frequency */ + int overflow = 0; /* number of elements with bit length too large */ + + for (bits = 0; bits <= MAX_BITS; bits++) s->bl_count[bits] = 0; + + /* In a first pass, compute the optimal bit lengths (which may + * overflow in the case of the bit length tree). + */ + tree[s->heap[s->heap_max]].Len = 0; /* root of the heap */ + + for (h = s->heap_max+1; h < HEAP_SIZE; h++) { + n = s->heap[h]; + bits = tree[tree[n].Dad].Len + 1; + if (bits > max_length) bits = max_length, overflow++; + tree[n].Len = (ush)bits; + /* We overwrite tree[n].Dad which is no longer needed */ + + if (n > max_code) continue; /* not a leaf node */ + + s->bl_count[bits]++; + xbits = 0; + if (n >= base) xbits = extra[n-base]; + f = tree[n].Freq; + s->opt_len += (ulg)f * (bits + xbits); + if (stree) s->static_len += (ulg)f * (stree[n].Len + xbits); + } + if (overflow == 0) return; + + Trace((stderr,"\nbit length overflow\n")); + /* This happens for example on obj2 and pic of the Calgary corpus */ + + /* Find the first bit length which could increase: */ + do { + bits = max_length-1; + while (s->bl_count[bits] == 0) bits--; + s->bl_count[bits]--; /* move one leaf down the tree */ + s->bl_count[bits+1] += 2; /* move one overflow item as its brother */ + s->bl_count[max_length]--; + /* The brother of the overflow item also moves one step up, + * but this does not affect bl_count[max_length] + */ + overflow -= 2; + } while (overflow > 0); + + /* Now recompute all bit lengths, scanning in increasing frequency. + * h is still equal to HEAP_SIZE. (It is simpler to reconstruct all + * lengths instead of fixing only the wrong ones. This idea is taken + * from 'ar' written by Haruhiko Okumura.) + */ + for (bits = max_length; bits != 0; bits--) { + n = s->bl_count[bits]; + while (n != 0) { + m = s->heap[--h]; + if (m > max_code) continue; + if (tree[m].Len != (unsigned) bits) { + Trace((stderr,"code %d bits %d->%d\n", m, tree[m].Len, bits)); + s->opt_len += ((long)bits - (long)tree[m].Len) + *(long)tree[m].Freq; + tree[m].Len = (ush)bits; + } + n--; + } + } +} + +/* =========================================================================== + * Generate the codes for a given tree and bit counts (which need not be + * optimal). + * IN assertion: the array bl_count contains the bit length statistics for + * the given tree and the field len is set for all tree elements. + * OUT assertion: the field code is set for all tree elements of non + * zero code length. + */ +static void gen_codes( + ct_data *tree, /* the tree to decorate */ + int max_code, /* largest code with non zero frequency */ + ush *bl_count /* number of codes at each bit length */ +) +{ + ush next_code[MAX_BITS+1]; /* next code value for each bit length */ + ush code = 0; /* running code value */ + int bits; /* bit index */ + int n; /* code index */ + + /* The distribution counts are first used to generate the code values + * without bit reversal. + */ + for (bits = 1; bits <= MAX_BITS; bits++) { + next_code[bits] = code = (code + bl_count[bits-1]) << 1; + } + /* Check that the bit counts in bl_count are consistent. The last code + * must be all ones. + */ + Assert (code + bl_count[MAX_BITS]-1 == (1<<MAX_BITS)-1, + "inconsistent bit counts"); + Tracev((stderr,"\ngen_codes: max_code %d ", max_code)); + + for (n = 0; n <= max_code; n++) { + int len = tree[n].Len; + if (len == 0) continue; + /* Now reverse the bits */ + tree[n].Code = bitrev32((u32)(next_code[len]++)) >> (32 - len); + + Tracecv(tree != static_ltree, (stderr,"\nn %3d %c l %2d c %4x (%x) ", + n, (isgraph(n) ? n : ' '), len, tree[n].Code, next_code[len]-1)); + } +} + +/* =========================================================================== + * Construct one Huffman tree and assigns the code bit strings and lengths. + * Update the total bit length for the current block. + * IN assertion: the field freq is set for all tree elements. + * OUT assertions: the fields len and code are set to the optimal bit length + * and corresponding code. The length opt_len is updated; static_len is + * also updated if stree is not null. The field max_code is set. + */ +static void build_tree( + deflate_state *s, + tree_desc *desc /* the tree descriptor */ +) +{ + ct_data *tree = desc->dyn_tree; + const ct_data *stree = desc->stat_desc->static_tree; + int elems = desc->stat_desc->elems; + int n, m; /* iterate over heap elements */ + int max_code = -1; /* largest code with non zero frequency */ + int node; /* new node being created */ + + /* Construct the initial heap, with least frequent element in + * heap[SMALLEST]. The sons of heap[n] are heap[2*n] and heap[2*n+1]. + * heap[0] is not used. + */ + s->heap_len = 0, s->heap_max = HEAP_SIZE; + + for (n = 0; n < elems; n++) { + if (tree[n].Freq != 0) { + s->heap[++(s->heap_len)] = max_code = n; + s->depth[n] = 0; + } else { + tree[n].Len = 0; + } + } + + /* The pkzip format requires that at least one distance code exists, + * and that at least one bit should be sent even if there is only one + * possible code. So to avoid special checks later on we force at least + * two codes of non zero frequency. + */ + while (s->heap_len < 2) { + node = s->heap[++(s->heap_len)] = (max_code < 2 ? ++max_code : 0); + tree[node].Freq = 1; + s->depth[node] = 0; + s->opt_len--; if (stree) s->static_len -= stree[node].Len; + /* node is 0 or 1 so it does not have extra bits */ + } + desc->max_code = max_code; + + /* The elements heap[heap_len/2+1 .. heap_len] are leaves of the tree, + * establish sub-heaps of increasing lengths: + */ + for (n = s->heap_len/2; n >= 1; n--) pqdownheap(s, tree, n); + + /* Construct the Huffman tree by repeatedly combining the least two + * frequent nodes. + */ + node = elems; /* next internal node of the tree */ + do { + pqremove(s, tree, n); /* n = node of least frequency */ + m = s->heap[SMALLEST]; /* m = node of next least frequency */ + + s->heap[--(s->heap_max)] = n; /* keep the nodes sorted by frequency */ + s->heap[--(s->heap_max)] = m; + + /* Create a new node father of n and m */ + tree[node].Freq = tree[n].Freq + tree[m].Freq; + s->depth[node] = (uch) (max(s->depth[n], s->depth[m]) + 1); + tree[n].Dad = tree[m].Dad = (ush)node; +#ifdef DUMP_BL_TREE + if (tree == s->bl_tree) { + fprintf(stderr,"\nnode %d(%d), sons %d(%d) %d(%d)", + node, tree[node].Freq, n, tree[n].Freq, m, tree[m].Freq); + } +#endif + /* and insert the new node in the heap */ + s->heap[SMALLEST] = node++; + pqdownheap(s, tree, SMALLEST); + + } while (s->heap_len >= 2); + + s->heap[--(s->heap_max)] = s->heap[SMALLEST]; + + /* At this point, the fields freq and dad are set. We can now + * generate the bit lengths. + */ + gen_bitlen(s, (tree_desc *)desc); + + /* The field len is now set, we can generate the bit codes */ + gen_codes ((ct_data *)tree, max_code, s->bl_count); +} + +/* =========================================================================== + * Scan a literal or distance tree to determine the frequencies of the codes + * in the bit length tree. + */ +static void scan_tree( + deflate_state *s, + ct_data *tree, /* the tree to be scanned */ + int max_code /* and its largest code of non zero frequency */ +) +{ + int n; /* iterates over all tree elements */ + int prevlen = -1; /* last emitted length */ + int curlen; /* length of current code */ + int nextlen = tree[0].Len; /* length of next code */ + int count = 0; /* repeat count of the current code */ + int max_count = 7; /* max repeat count */ + int min_count = 4; /* min repeat count */ + + if (nextlen == 0) max_count = 138, min_count = 3; + tree[max_code+1].Len = (ush)0xffff; /* guard */ + + for (n = 0; n <= max_code; n++) { + curlen = nextlen; nextlen = tree[n+1].Len; + if (++count < max_count && curlen == nextlen) { + continue; + } else if (count < min_count) { + s->bl_tree[curlen].Freq += count; + } else if (curlen != 0) { + if (curlen != prevlen) s->bl_tree[curlen].Freq++; + s->bl_tree[REP_3_6].Freq++; + } else if (count <= 10) { + s->bl_tree[REPZ_3_10].Freq++; + } else { + s->bl_tree[REPZ_11_138].Freq++; + } + count = 0; prevlen = curlen; + if (nextlen == 0) { + max_count = 138, min_count = 3; + } else if (curlen == nextlen) { + max_count = 6, min_count = 3; + } else { + max_count = 7, min_count = 4; + } + } +} + +/* =========================================================================== + * Send a literal or distance tree in compressed form, using the codes in + * bl_tree. + */ +static void send_tree( + deflate_state *s, + ct_data *tree, /* the tree to be scanned */ + int max_code /* and its largest code of non zero frequency */ +) +{ + int n; /* iterates over all tree elements */ + int prevlen = -1; /* last emitted length */ + int curlen; /* length of current code */ + int nextlen = tree[0].Len; /* length of next code */ + int count = 0; /* repeat count of the current code */ + int max_count = 7; /* max repeat count */ + int min_count = 4; /* min repeat count */ + + /* tree[max_code+1].Len = -1; */ /* guard already set */ + if (nextlen == 0) max_count = 138, min_count = 3; + + for (n = 0; n <= max_code; n++) { + curlen = nextlen; nextlen = tree[n+1].Len; + if (++count < max_count && curlen == nextlen) { + continue; + } else if (count < min_count) { + do { send_code(s, curlen, s->bl_tree); } while (--count != 0); + + } else if (curlen != 0) { + if (curlen != prevlen) { + send_code(s, curlen, s->bl_tree); count--; + } + Assert(count >= 3 && count <= 6, " 3_6?"); + send_code(s, REP_3_6, s->bl_tree); send_bits(s, count-3, 2); + + } else if (count <= 10) { + send_code(s, REPZ_3_10, s->bl_tree); send_bits(s, count-3, 3); + + } else { + send_code(s, REPZ_11_138, s->bl_tree); send_bits(s, count-11, 7); + } + count = 0; prevlen = curlen; + if (nextlen == 0) { + max_count = 138, min_count = 3; + } else if (curlen == nextlen) { + max_count = 6, min_count = 3; + } else { + max_count = 7, min_count = 4; + } + } +} + +/* =========================================================================== + * Construct the Huffman tree for the bit lengths and return the index in + * bl_order of the last bit length code to send. + */ +static int build_bl_tree( + deflate_state *s +) +{ + int max_blindex; /* index of last bit length code of non zero freq */ + + /* Determine the bit length frequencies for literal and distance trees */ + scan_tree(s, (ct_data *)s->dyn_ltree, s->l_desc.max_code); + scan_tree(s, (ct_data *)s->dyn_dtree, s->d_desc.max_code); + + /* Build the bit length tree: */ + build_tree(s, (tree_desc *)(&(s->bl_desc))); + /* opt_len now includes the length of the tree representations, except + * the lengths of the bit lengths codes and the 5+5+4 bits for the counts. + */ + + /* Determine the number of bit length codes to send. The pkzip format + * requires that at least 4 bit length codes be sent. (appnote.txt says + * 3 but the actual value used is 4.) + */ + for (max_blindex = BL_CODES-1; max_blindex >= 3; max_blindex--) { + if (s->bl_tree[bl_order[max_blindex]].Len != 0) break; + } + /* Update opt_len to include the bit length tree and counts */ + s->opt_len += 3*(max_blindex+1) + 5+5+4; + Tracev((stderr, "\ndyn trees: dyn %ld, stat %ld", + s->opt_len, s->static_len)); + + return max_blindex; +} + +/* =========================================================================== + * Send the header for a block using dynamic Huffman trees: the counts, the + * lengths of the bit length codes, the literal tree and the distance tree. + * IN assertion: lcodes >= 257, dcodes >= 1, blcodes >= 4. + */ +static void send_all_trees( + deflate_state *s, + int lcodes, /* number of codes for each tree */ + int dcodes, /* number of codes for each tree */ + int blcodes /* number of codes for each tree */ +) +{ + int rank; /* index in bl_order */ + + Assert (lcodes >= 257 && dcodes >= 1 && blcodes >= 4, "not enough codes"); + Assert (lcodes <= L_CODES && dcodes <= D_CODES && blcodes <= BL_CODES, + "too many codes"); + Tracev((stderr, "\nbl counts: ")); + send_bits(s, lcodes-257, 5); /* not +255 as stated in appnote.txt */ + send_bits(s, dcodes-1, 5); + send_bits(s, blcodes-4, 4); /* not -3 as stated in appnote.txt */ + for (rank = 0; rank < blcodes; rank++) { + Tracev((stderr, "\nbl code %2d ", bl_order[rank])); + send_bits(s, s->bl_tree[bl_order[rank]].Len, 3); + } + Tracev((stderr, "\nbl tree: sent %ld", s->bits_sent)); + + send_tree(s, (ct_data *)s->dyn_ltree, lcodes-1); /* literal tree */ + Tracev((stderr, "\nlit tree: sent %ld", s->bits_sent)); + + send_tree(s, (ct_data *)s->dyn_dtree, dcodes-1); /* distance tree */ + Tracev((stderr, "\ndist tree: sent %ld", s->bits_sent)); +} + +/* =========================================================================== + * Send a stored block + */ +void zlib_tr_stored_block( + deflate_state *s, + char *buf, /* input block */ + ulg stored_len, /* length of input block */ + int eof /* true if this is the last block for a file */ +) +{ + send_bits(s, (STORED_BLOCK<<1)+eof, 3); /* send block type */ + s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L; + s->compressed_len += (stored_len + 4) << 3; + + copy_block(s, buf, (unsigned)stored_len, 1); /* with header */ +} + +/* Send just the `stored block' type code without any length bytes or data. + */ +void zlib_tr_stored_type_only( + deflate_state *s +) +{ + send_bits(s, (STORED_BLOCK << 1), 3); + bi_windup(s); + s->compressed_len = (s->compressed_len + 3) & ~7L; +} + + +/* =========================================================================== + * Send one empty static block to give enough lookahead for inflate. + * This takes 10 bits, of which 7 may remain in the bit buffer. + * The current inflate code requires 9 bits of lookahead. If the + * last two codes for the previous block (real code plus EOB) were coded + * on 5 bits or less, inflate may have only 5+3 bits of lookahead to decode + * the last real code. In this case we send two empty static blocks instead + * of one. (There are no problems if the previous block is stored or fixed.) + * To simplify the code, we assume the worst case of last real code encoded + * on one bit only. + */ +void zlib_tr_align( + deflate_state *s +) +{ + send_bits(s, STATIC_TREES<<1, 3); + send_code(s, END_BLOCK, static_ltree); + s->compressed_len += 10L; /* 3 for block type, 7 for EOB */ + bi_flush(s); + /* Of the 10 bits for the empty block, we have already sent + * (10 - bi_valid) bits. The lookahead for the last real code (before + * the EOB of the previous block) was thus at least one plus the length + * of the EOB plus what we have just sent of the empty static block. + */ + if (1 + s->last_eob_len + 10 - s->bi_valid < 9) { + send_bits(s, STATIC_TREES<<1, 3); + send_code(s, END_BLOCK, static_ltree); + s->compressed_len += 10L; + bi_flush(s); + } + s->last_eob_len = 7; +} + +/* =========================================================================== + * Determine the best encoding for the current block: dynamic trees, static + * trees or store, and output the encoded block to the zip file. This function + * returns the total compressed length for the file so far. + */ +ulg zlib_tr_flush_block( + deflate_state *s, + char *buf, /* input block, or NULL if too old */ + ulg stored_len, /* length of input block */ + int eof /* true if this is the last block for a file */ +) +{ + ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */ + int max_blindex = 0; /* index of last bit length code of non zero freq */ + + /* Build the Huffman trees unless a stored block is forced */ + if (s->level > 0) { + + /* Check if the file is ascii or binary */ + if (s->data_type == Z_UNKNOWN) set_data_type(s); + + /* Construct the literal and distance trees */ + build_tree(s, (tree_desc *)(&(s->l_desc))); + Tracev((stderr, "\nlit data: dyn %ld, stat %ld", s->opt_len, + s->static_len)); + + build_tree(s, (tree_desc *)(&(s->d_desc))); + Tracev((stderr, "\ndist data: dyn %ld, stat %ld", s->opt_len, + s->static_len)); + /* At this point, opt_len and static_len are the total bit lengths of + * the compressed block data, excluding the tree representations. + */ + + /* Build the bit length tree for the above two trees, and get the index + * in bl_order of the last bit length code to send. + */ + max_blindex = build_bl_tree(s); + + /* Determine the best encoding. Compute first the block length in bytes*/ + opt_lenb = (s->opt_len+3+7)>>3; + static_lenb = (s->static_len+3+7)>>3; + + Tracev((stderr, "\nopt %lu(%lu) stat %lu(%lu) stored %lu lit %u ", + opt_lenb, s->opt_len, static_lenb, s->static_len, stored_len, + s->last_lit)); + + if (static_lenb <= opt_lenb) opt_lenb = static_lenb; + + } else { + Assert(buf != (char*)0, "lost buf"); + opt_lenb = static_lenb = stored_len + 5; /* force a stored block */ + } + + /* If compression failed and this is the first and last block, + * and if the .zip file can be seeked (to rewrite the local header), + * the whole file is transformed into a stored file: + */ +#ifdef STORED_FILE_OK +# ifdef FORCE_STORED_FILE + if (eof && s->compressed_len == 0L) { /* force stored file */ +# else + if (stored_len <= opt_lenb && eof && s->compressed_len==0L && seekable()) { +# endif + /* Since LIT_BUFSIZE <= 2*WSIZE, the input data must be there: */ + if (buf == (char*)0) error ("block vanished"); + + copy_block(s, buf, (unsigned)stored_len, 0); /* without header */ + s->compressed_len = stored_len << 3; + s->method = STORED; + } else +#endif /* STORED_FILE_OK */ + +#ifdef FORCE_STORED + if (buf != (char*)0) { /* force stored block */ +#else + if (stored_len+4 <= opt_lenb && buf != (char*)0) { + /* 4: two words for the lengths */ +#endif + /* The test buf != NULL is only necessary if LIT_BUFSIZE > WSIZE. + * Otherwise we can't have processed more than WSIZE input bytes since + * the last block flush, because compression would have been + * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to + * transform a block into a stored block. + */ + zlib_tr_stored_block(s, buf, stored_len, eof); + +#ifdef FORCE_STATIC + } else if (static_lenb >= 0) { /* force static trees */ +#else + } else if (static_lenb == opt_lenb) { +#endif + send_bits(s, (STATIC_TREES<<1)+eof, 3); + compress_block(s, (ct_data *)static_ltree, (ct_data *)static_dtree); + s->compressed_len += 3 + s->static_len; + } else { + send_bits(s, (DYN_TREES<<1)+eof, 3); + send_all_trees(s, s->l_desc.max_code+1, s->d_desc.max_code+1, + max_blindex+1); + compress_block(s, (ct_data *)s->dyn_ltree, (ct_data *)s->dyn_dtree); + s->compressed_len += 3 + s->opt_len; + } + Assert (s->compressed_len == s->bits_sent, "bad compressed size"); + init_block(s); + + if (eof) { + bi_windup(s); + s->compressed_len += 7; /* align on byte boundary */ + } + Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3, + s->compressed_len-7*eof)); + + return s->compressed_len >> 3; +} + +/* =========================================================================== + * Save the match info and tally the frequency counts. Return true if + * the current block must be flushed. + */ +int zlib_tr_tally( + deflate_state *s, + unsigned dist, /* distance of matched string */ + unsigned lc /* match length-MIN_MATCH or unmatched char (if dist==0) */ +) +{ + s->d_buf[s->last_lit] = (ush)dist; + s->l_buf[s->last_lit++] = (uch)lc; + if (dist == 0) { + /* lc is the unmatched char */ + s->dyn_ltree[lc].Freq++; + } else { + s->matches++; + /* Here, lc is the match length - MIN_MATCH */ + dist--; /* dist = match distance - 1 */ + Assert((ush)dist < (ush)MAX_DIST(s) && + (ush)lc <= (ush)(MAX_MATCH-MIN_MATCH) && + (ush)d_code(dist) < (ush)D_CODES, "zlib_tr_tally: bad match"); + + s->dyn_ltree[length_code[lc]+LITERALS+1].Freq++; + s->dyn_dtree[d_code(dist)].Freq++; + } + + /* Try to guess if it is profitable to stop the current block here */ + if ((s->last_lit & 0xfff) == 0 && s->level > 2) { + /* Compute an upper bound for the compressed length */ + ulg out_length = (ulg)s->last_lit*8L; + ulg in_length = (ulg)((long)s->strstart - s->block_start); + int dcode; + for (dcode = 0; dcode < D_CODES; dcode++) { + out_length += (ulg)s->dyn_dtree[dcode].Freq * + (5L+extra_dbits[dcode]); + } + out_length >>= 3; + Tracev((stderr,"\nlast_lit %u, in %ld, out ~%ld(%ld%%) ", + s->last_lit, in_length, out_length, + 100L - out_length*100L/in_length)); + if (s->matches < s->last_lit/2 && out_length < in_length/2) return 1; + } + return (s->last_lit == s->lit_bufsize-1); + /* We avoid equality with lit_bufsize because of wraparound at 64K + * on 16 bit machines and because stored blocks are restricted to + * 64K-1 bytes. + */ +} + +/* =========================================================================== + * Send the block data compressed using the given Huffman trees + */ +static void compress_block( + deflate_state *s, + ct_data *ltree, /* literal tree */ + ct_data *dtree /* distance tree */ +) +{ + unsigned dist; /* distance of matched string */ + int lc; /* match length or unmatched char (if dist == 0) */ + unsigned lx = 0; /* running index in l_buf */ + unsigned code; /* the code to send */ + int extra; /* number of extra bits to send */ + + if (s->last_lit != 0) do { + dist = s->d_buf[lx]; + lc = s->l_buf[lx++]; + if (dist == 0) { + send_code(s, lc, ltree); /* send a literal byte */ + Tracecv(isgraph(lc), (stderr," '%c' ", lc)); + } else { + /* Here, lc is the match length - MIN_MATCH */ + code = length_code[lc]; + send_code(s, code+LITERALS+1, ltree); /* send the length code */ + extra = extra_lbits[code]; + if (extra != 0) { + lc -= base_length[code]; + send_bits(s, lc, extra); /* send the extra length bits */ + } + dist--; /* dist is now the match distance - 1 */ + code = d_code(dist); + Assert (code < D_CODES, "bad d_code"); + + send_code(s, code, dtree); /* send the distance code */ + extra = extra_dbits[code]; + if (extra != 0) { + dist -= base_dist[code]; + send_bits(s, dist, extra); /* send the extra distance bits */ + } + } /* literal or match pair ? */ + + /* Check that the overlay between pending_buf and d_buf+l_buf is ok: */ + Assert(s->pending < s->lit_bufsize + 2*lx, "pendingBuf overflow"); + + } while (lx < s->last_lit); + + send_code(s, END_BLOCK, ltree); + s->last_eob_len = ltree[END_BLOCK].Len; +} + +/* =========================================================================== + * Set the data type to ASCII or BINARY, using a crude approximation: + * binary if more than 20% of the bytes are <= 6 or >= 128, ascii otherwise. + * IN assertion: the fields freq of dyn_ltree are set and the total of all + * frequencies does not exceed 64K (to fit in an int on 16 bit machines). + */ +static void set_data_type( + deflate_state *s +) +{ + int n = 0; + unsigned ascii_freq = 0; + unsigned bin_freq = 0; + while (n < 7) bin_freq += s->dyn_ltree[n++].Freq; + while (n < 128) ascii_freq += s->dyn_ltree[n++].Freq; + while (n < LITERALS) bin_freq += s->dyn_ltree[n++].Freq; + s->data_type = (Byte)(bin_freq > (ascii_freq >> 2) ? Z_BINARY : Z_ASCII); +} + +/* =========================================================================== + * Copy a stored block, storing first the length and its + * one's complement if requested. + */ +static void copy_block( + deflate_state *s, + char *buf, /* the input data */ + unsigned len, /* its length */ + int header /* true if block header must be written */ +) +{ + bi_windup(s); /* align on byte boundary */ + s->last_eob_len = 8; /* enough lookahead for inflate */ + + if (header) { + put_short(s, (ush)len); + put_short(s, (ush)~len); +#ifdef DEBUG_ZLIB + s->bits_sent += 2*16; +#endif + } +#ifdef DEBUG_ZLIB + s->bits_sent += (ulg)len<<3; +#endif + /* bundle up the put_byte(s, *buf++) calls */ + memcpy(&s->pending_buf[s->pending], buf, len); + s->pending += len; +} + diff --git a/linux/zlib_deflate/defutil.h b/linux/zlib_deflate/defutil.h new file mode 100644 index 0000000..a8c3708 --- /dev/null +++ b/linux/zlib_deflate/defutil.h @@ -0,0 +1,327 @@ + + + +#define Assert(err, str) +#define Trace(dummy) +#define Tracev(dummy) +#define Tracecv(err, dummy) +#define Tracevv(dummy) + + + +#define LENGTH_CODES 29 +/* number of length codes, not counting the special END_BLOCK code */ + +#define LITERALS 256 +/* number of literal bytes 0..255 */ + +#define L_CODES (LITERALS+1+LENGTH_CODES) +/* number of Literal or Length codes, including the END_BLOCK code */ + +#define D_CODES 30 +/* number of distance codes */ + +#define BL_CODES 19 +/* number of codes used to transfer the bit lengths */ + +#define HEAP_SIZE (2*L_CODES+1) +/* maximum heap size */ + +#define MAX_BITS 15 +/* All codes must not exceed MAX_BITS bits */ + +#define INIT_STATE 42 +#define BUSY_STATE 113 +#define FINISH_STATE 666 +/* Stream status */ + + +/* Data structure describing a single value and its code string. */ +typedef struct ct_data_s { + union { + ush freq; /* frequency count */ + ush code; /* bit string */ + } fc; + union { + ush dad; /* father node in Huffman tree */ + ush len; /* length of bit string */ + } dl; +} ct_data; + +#define Freq fc.freq +#define Code fc.code +#define Dad dl.dad +#define Len dl.len + +typedef struct static_tree_desc_s static_tree_desc; + +typedef struct tree_desc_s { + ct_data *dyn_tree; /* the dynamic tree */ + int max_code; /* largest code with non zero frequency */ + static_tree_desc *stat_desc; /* the corresponding static tree */ +} tree_desc; + +typedef ush Pos; +typedef unsigned IPos; + +/* A Pos is an index in the character window. We use short instead of int to + * save space in the various tables. IPos is used only for parameter passing. + */ + +typedef struct deflate_state { + z_streamp strm; /* pointer back to this zlib stream */ + int status; /* as the name implies */ + Byte *pending_buf; /* output still pending */ + ulg pending_buf_size; /* size of pending_buf */ + Byte *pending_out; /* next pending byte to output to the stream */ + int pending; /* nb of bytes in the pending buffer */ + int noheader; /* suppress zlib header and adler32 */ + Byte data_type; /* UNKNOWN, BINARY or ASCII */ + Byte method; /* STORED (for zip only) or DEFLATED */ + int last_flush; /* value of flush param for previous deflate call */ + + /* used by deflate.c: */ + + uInt w_size; /* LZ77 window size (32K by default) */ + uInt w_bits; /* log2(w_size) (8..16) */ + uInt w_mask; /* w_size - 1 */ + + Byte *window; + /* Sliding window. Input bytes are read into the second half of the window, + * and move to the first half later to keep a dictionary of at least wSize + * bytes. With this organization, matches are limited to a distance of + * wSize-MAX_MATCH bytes, but this ensures that IO is always + * performed with a length multiple of the block size. Also, it limits + * the window size to 64K, which is quite useful on MSDOS. + * To do: use the user input buffer as sliding window. + */ + + ulg window_size; + /* Actual size of window: 2*wSize, except when the user input buffer + * is directly used as sliding window. + */ + + Pos *prev; + /* Link to older string with same hash index. To limit the size of this + * array to 64K, this link is maintained only for the last 32K strings. + * An index in this array is thus a window index modulo 32K. + */ + + Pos *head; /* Heads of the hash chains or NIL. */ + + uInt ins_h; /* hash index of string to be inserted */ + uInt hash_size; /* number of elements in hash table */ + uInt hash_bits; /* log2(hash_size) */ + uInt hash_mask; /* hash_size-1 */ + + uInt hash_shift; + /* Number of bits by which ins_h must be shifted at each input + * step. It must be such that after MIN_MATCH steps, the oldest + * byte no longer takes part in the hash key, that is: + * hash_shift * MIN_MATCH >= hash_bits + */ + + long block_start; + /* Window position at the beginning of the current output block. Gets + * negative when the window is moved backwards. + */ + + uInt match_length; /* length of best match */ + IPos prev_match; /* previous match */ + int match_available; /* set if previous match exists */ + uInt strstart; /* start of string to insert */ + uInt match_start; /* start of matching string */ + uInt lookahead; /* number of valid bytes ahead in window */ + + uInt prev_length; + /* Length of the best match at previous step. Matches not greater than this + * are discarded. This is used in the lazy match evaluation. + */ + + uInt max_chain_length; + /* To speed up deflation, hash chains are never searched beyond this + * length. A higher limit improves compression ratio but degrades the + * speed. + */ + + uInt max_lazy_match; + /* Attempt to find a better match only when the current match is strictly + * smaller than this value. This mechanism is used only for compression + * levels >= 4. + */ +# define max_insert_length max_lazy_match + /* Insert new strings in the hash table only if the match length is not + * greater than this length. This saves time but degrades compression. + * max_insert_length is used only for compression levels <= 3. + */ + + int level; /* compression level (1..9) */ + int strategy; /* favor or force Huffman coding*/ + + uInt good_match; + /* Use a faster search when the previous match is longer than this */ + + int nice_match; /* Stop searching when current match exceeds this */ + + /* used by trees.c: */ + /* Didn't use ct_data typedef below to suppress compiler warning */ + struct ct_data_s dyn_ltree[HEAP_SIZE]; /* literal and length tree */ + struct ct_data_s dyn_dtree[2*D_CODES+1]; /* distance tree */ + struct ct_data_s bl_tree[2*BL_CODES+1]; /* Huffman tree for bit lengths */ + + struct tree_desc_s l_desc; /* desc. for literal tree */ + struct tree_desc_s d_desc; /* desc. for distance tree */ + struct tree_desc_s bl_desc; /* desc. for bit length tree */ + + ush bl_count[MAX_BITS+1]; + /* number of codes at each bit length for an optimal tree */ + + int heap[2*L_CODES+1]; /* heap used to build the Huffman trees */ + int heap_len; /* number of elements in the heap */ + int heap_max; /* element of largest frequency */ + /* The sons of heap[n] are heap[2*n] and heap[2*n+1]. heap[0] is not used. + * The same heap array is used to build all trees. + */ + + uch depth[2*L_CODES+1]; + /* Depth of each subtree used as tie breaker for trees of equal frequency + */ + + uch *l_buf; /* buffer for literals or lengths */ + + uInt lit_bufsize; + /* Size of match buffer for literals/lengths. There are 4 reasons for + * limiting lit_bufsize to 64K: + * - frequencies can be kept in 16 bit counters + * - if compression is not successful for the first block, all input + * data is still in the window so we can still emit a stored block even + * when input comes from standard input. (This can also be done for + * all blocks if lit_bufsize is not greater than 32K.) + * - if compression is not successful for a file smaller than 64K, we can + * even emit a stored file instead of a stored block (saving 5 bytes). + * This is applicable only for zip (not gzip or zlib). + * - creating new Huffman trees less frequently may not provide fast + * adaptation to changes in the input data statistics. (Take for + * example a binary file with poorly compressible code followed by + * a highly compressible string table.) Smaller buffer sizes give + * fast adaptation but have of course the overhead of transmitting + * trees more frequently. + * - I can't count above 4 + */ + + uInt last_lit; /* running index in l_buf */ + + ush *d_buf; + /* Buffer for distances. To simplify the code, d_buf and l_buf have + * the same number of elements. To use different lengths, an extra flag + * array would be necessary. + */ + + ulg opt_len; /* bit length of current block with optimal trees */ + ulg static_len; /* bit length of current block with static trees */ + ulg compressed_len; /* total bit length of compressed file */ + uInt matches; /* number of string matches in current block */ + int last_eob_len; /* bit length of EOB code for last block */ + +#ifdef DEBUG_ZLIB + ulg bits_sent; /* bit length of the compressed data */ +#endif + + ush bi_buf; + /* Output buffer. bits are inserted starting at the bottom (least + * significant bits). + */ + int bi_valid; + /* Number of valid bits in bi_buf. All bits above the last valid bit + * are always zero. + */ + +} deflate_state; + +typedef struct deflate_workspace { + /* State memory for the deflator */ + deflate_state deflate_memory; + Byte *window_memory; + Pos *prev_memory; + Pos *head_memory; + char *overlay_memory; +} deflate_workspace; + +#define zlib_deflate_window_memsize(windowBits) \ + (2 * (1 << (windowBits)) * sizeof(Byte)) +#define zlib_deflate_prev_memsize(windowBits) \ + ((1 << (windowBits)) * sizeof(Pos)) +#define zlib_deflate_head_memsize(memLevel) \ + ((1 << ((memLevel)+7)) * sizeof(Pos)) +#define zlib_deflate_overlay_memsize(memLevel) \ + ((1 << ((memLevel)+6)) * (sizeof(ush)+2)) + +/* Output a byte on the stream. + * IN assertion: there is enough room in pending_buf. + */ +#define put_byte(s, c) {s->pending_buf[s->pending++] = (c);} + + +#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1) +/* Minimum amount of lookahead, except at the end of the input file. + * See deflate.c for comments about the MIN_MATCH+1. + */ + +#define MAX_DIST(s) ((s)->w_size-MIN_LOOKAHEAD) +/* In order to simplify the code, particularly on 16 bit machines, match + * distances are limited to MAX_DIST instead of WSIZE. + */ + + /* in trees.c */ +void zlib_tr_init (deflate_state *s); +int zlib_tr_tally (deflate_state *s, unsigned dist, unsigned lc); +ulg zlib_tr_flush_block (deflate_state *s, char *buf, ulg stored_len, + int eof); +void zlib_tr_align (deflate_state *s); +void zlib_tr_stored_block (deflate_state *s, char *buf, ulg stored_len, + int eof); +void zlib_tr_stored_type_only (deflate_state *); + + +/* =========================================================================== + * Output a short LSB first on the stream. + * IN assertion: there is enough room in pendingBuf. + */ +#define put_short(s, w) { \ + put_byte(s, (uch)((w) & 0xff)); \ + put_byte(s, (uch)((ush)(w) >> 8)); \ +} + +/* =========================================================================== + * Flush the bit buffer, keeping at most 7 bits in it. + */ +static inline void bi_flush(deflate_state *s) +{ + if (s->bi_valid == 16) { + put_short(s, s->bi_buf); + s->bi_buf = 0; + s->bi_valid = 0; + } else if (s->bi_valid >= 8) { + put_byte(s, (Byte)s->bi_buf); + s->bi_buf >>= 8; + s->bi_valid -= 8; + } +} + +/* =========================================================================== + * Flush the bit buffer and align the output on a byte boundary + */ +static inline void bi_windup(deflate_state *s) +{ + if (s->bi_valid > 8) { + put_short(s, s->bi_buf); + } else if (s->bi_valid > 0) { + put_byte(s, (Byte)s->bi_buf); + } + s->bi_buf = 0; + s->bi_valid = 0; +#ifdef DEBUG_ZLIB + s->bits_sent = (s->bits_sent+7) & ~7; +#endif +} + diff --git a/linux/zlib_inflate/inffast.c b/linux/zlib_inflate/inffast.c new file mode 100644 index 0000000..2c13ecc --- /dev/null +++ b/linux/zlib_inflate/inffast.c @@ -0,0 +1,363 @@ +/* inffast.c -- fast decoding + * Copyright (C) 1995-2004 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include <linux/zutil.h> +#include "inftrees.h" +#include "inflate.h" +#include "inffast.h" + +#ifndef ASMINF + +/* Allow machine dependent optimization for post-increment or pre-increment. + Based on testing to date, + Pre-increment preferred for: + - PowerPC G3 (Adler) + - MIPS R5000 (Randers-Pehrson) + Post-increment preferred for: + - none + No measurable difference: + - Pentium III (Anderson) + - M68060 (Nikl) + */ +union uu { + unsigned short us; + unsigned char b[2]; +}; + +/* Endian independed version */ +static inline unsigned short +get_unaligned16(const unsigned short *p) +{ + union uu mm; + unsigned char *b = (unsigned char *)p; + + mm.b[0] = b[0]; + mm.b[1] = b[1]; + return mm.us; +} + +#ifdef POSTINC +# define OFF 0 +# define PUP(a) *(a)++ +# define UP_UNALIGNED(a) get_unaligned16((a)++) +#else +# define OFF 1 +# define PUP(a) *++(a) +# define UP_UNALIGNED(a) get_unaligned16(++(a)) +#endif + +/* + Decode literal, length, and distance codes and write out the resulting + literal and match bytes until either not enough input or output is + available, an end-of-block is encountered, or a data error is encountered. + When large enough input and output buffers are supplied to inflate(), for + example, a 16K input buffer and a 64K output buffer, more than 95% of the + inflate execution time is spent in this routine. + + Entry assumptions: + + state->mode == LEN + strm->avail_in >= 6 + strm->avail_out >= 258 + start >= strm->avail_out + state->bits < 8 + + On return, state->mode is one of: + + LEN -- ran out of enough output space or enough available input + TYPE -- reached end of block code, inflate() to interpret next block + BAD -- error in block data + + Notes: + + - The maximum input bits used by a length/distance pair is 15 bits for the + length code, 5 bits for the length extra, 15 bits for the distance code, + and 13 bits for the distance extra. This totals 48 bits, or six bytes. + Therefore if strm->avail_in >= 6, then there is enough input to avoid + checking for available input while decoding. + + - The maximum bytes that a single length/distance pair can output is 258 + bytes, which is the maximum length that can be coded. inflate_fast() + requires strm->avail_out >= 258 for each loop to avoid checking for + output space. + + - @start: inflate()'s starting value for strm->avail_out + */ +void inflate_fast(z_streamp strm, unsigned start) +{ + struct inflate_state *state; + const unsigned char *in; /* local strm->next_in */ + const unsigned char *last; /* while in < last, enough input available */ + unsigned char *out; /* local strm->next_out */ + unsigned char *beg; /* inflate()'s initial strm->next_out */ + unsigned char *end; /* while out < end, enough space available */ +#ifdef INFLATE_STRICT + unsigned dmax; /* maximum distance from zlib header */ +#endif + unsigned wsize; /* window size or zero if not using window */ + unsigned whave; /* valid bytes in the window */ + unsigned write; /* window write index */ + unsigned char *window; /* allocated sliding window, if wsize != 0 */ + unsigned long hold; /* local strm->hold */ + unsigned bits; /* local strm->bits */ + code const *lcode; /* local strm->lencode */ + code const *dcode; /* local strm->distcode */ + unsigned lmask; /* mask for first level of length codes */ + unsigned dmask; /* mask for first level of distance codes */ + code this; /* retrieved table entry */ + unsigned op; /* code bits, operation, extra bits, or */ + /* window position, window bytes to copy */ + unsigned len; /* match length, unused bytes */ + unsigned dist; /* match distance */ + unsigned char *from; /* where to copy match from */ + + /* copy state to local variables */ + state = (struct inflate_state *)strm->state; + in = strm->next_in - OFF; + last = in + (strm->avail_in - 5); + out = strm->next_out - OFF; + beg = out - (start - strm->avail_out); + end = out + (strm->avail_out - 257); +#ifdef INFLATE_STRICT + dmax = state->dmax; +#endif + wsize = state->wsize; + whave = state->whave; + write = state->write; + window = state->window; + hold = state->hold; + bits = state->bits; + lcode = state->lencode; + dcode = state->distcode; + lmask = (1U << state->lenbits) - 1; + dmask = (1U << state->distbits) - 1; + + /* decode literals and length/distances until end-of-block or not enough + input data or output space */ + do { + if (bits < 15) { + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + } + this = lcode[hold & lmask]; + dolen: + op = (unsigned)(this.bits); + hold >>= op; + bits -= op; + op = (unsigned)(this.op); + if (op == 0) { /* literal */ + PUP(out) = (unsigned char)(this.val); + } + else if (op & 16) { /* length base */ + len = (unsigned)(this.val); + op &= 15; /* number of extra bits */ + if (op) { + if (bits < op) { + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + } + len += (unsigned)hold & ((1U << op) - 1); + hold >>= op; + bits -= op; + } + if (bits < 15) { + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + } + this = dcode[hold & dmask]; + dodist: + op = (unsigned)(this.bits); + hold >>= op; + bits -= op; + op = (unsigned)(this.op); + if (op & 16) { /* distance base */ + dist = (unsigned)(this.val); + op &= 15; /* number of extra bits */ + if (bits < op) { + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + if (bits < op) { + hold += (unsigned long)(PUP(in)) << bits; + bits += 8; + } + } + dist += (unsigned)hold & ((1U << op) - 1); +#ifdef INFLATE_STRICT + if (dist > dmax) { + strm->msg = (char *)"invalid distance too far back"; + state->mode = BAD; + break; + } +#endif + hold >>= op; + bits -= op; + op = (unsigned)(out - beg); /* max distance in output */ + if (dist > op) { /* see if copy from window */ + op = dist - op; /* distance back in window */ + if (op > whave) { + strm->msg = (char *)"invalid distance too far back"; + state->mode = BAD; + break; + } + from = window - OFF; + if (write == 0) { /* very common case */ + from += wsize - op; + if (op < len) { /* some from window */ + len -= op; + do { + PUP(out) = PUP(from); + } while (--op); + from = out - dist; /* rest from output */ + } + } + else if (write < op) { /* wrap around window */ + from += wsize + write - op; + op -= write; + if (op < len) { /* some from end of window */ + len -= op; + do { + PUP(out) = PUP(from); + } while (--op); + from = window - OFF; + if (write < len) { /* some from start of window */ + op = write; + len -= op; + do { + PUP(out) = PUP(from); + } while (--op); + from = out - dist; /* rest from output */ + } + } + } + else { /* contiguous in window */ + from += write - op; + if (op < len) { /* some from window */ + len -= op; + do { + PUP(out) = PUP(from); + } while (--op); + from = out - dist; /* rest from output */ + } + } + while (len > 2) { + PUP(out) = PUP(from); + PUP(out) = PUP(from); + PUP(out) = PUP(from); + len -= 3; + } + if (len) { + PUP(out) = PUP(from); + if (len > 1) + PUP(out) = PUP(from); + } + } + else { + unsigned short *sout; + unsigned long loops; + + from = out - dist; /* copy direct from output */ + /* minimum length is three */ + /* Align out addr */ + if (!((long)(out - 1 + OFF) & 1)) { + PUP(out) = PUP(from); + len--; + } + sout = (unsigned short *)(out - OFF); + if (dist > 2) { + unsigned short *sfrom; + + sfrom = (unsigned short *)(from - OFF); + loops = len >> 1; + do +#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS + PUP(sout) = PUP(sfrom); +#else + PUP(sout) = UP_UNALIGNED(sfrom); +#endif + while (--loops); + out = (unsigned char *)sout + OFF; + from = (unsigned char *)sfrom + OFF; + } else { /* dist == 1 or dist == 2 */ + unsigned short pat16; + + pat16 = *(sout-1+OFF); + if (dist == 1) { + union uu mm; + /* copy one char pattern to both bytes */ + mm.us = pat16; + mm.b[0] = mm.b[1]; + pat16 = mm.us; + } + loops = len >> 1; + do + PUP(sout) = pat16; + while (--loops); + out = (unsigned char *)sout + OFF; + } + if (len & 1) + PUP(out) = PUP(from); + } + } + else if ((op & 64) == 0) { /* 2nd level distance code */ + this = dcode[this.val + (hold & ((1U << op) - 1))]; + goto dodist; + } + else { + strm->msg = (char *)"invalid distance code"; + state->mode = BAD; + break; + } + } + else if ((op & 64) == 0) { /* 2nd level length code */ + this = lcode[this.val + (hold & ((1U << op) - 1))]; + goto dolen; + } + else if (op & 32) { /* end-of-block */ + state->mode = TYPE; + break; + } + else { + strm->msg = (char *)"invalid literal/length code"; + state->mode = BAD; + break; + } + } while (in < last && out < end); + + /* return unused bytes (on entry, bits < 8, so in won't go too far back) */ + len = bits >> 3; + in -= len; + bits -= len << 3; + hold &= (1U << bits) - 1; + + /* update state and return */ + strm->next_in = in + OFF; + strm->next_out = out + OFF; + strm->avail_in = (unsigned)(in < last ? 5 + (last - in) : 5 - (in - last)); + strm->avail_out = (unsigned)(out < end ? + 257 + (end - out) : 257 - (out - end)); + state->hold = hold; + state->bits = bits; + return; +} + +/* + inflate_fast() speedups that turned out slower (on a PowerPC G3 750CXe): + - Using bit fields for code structure + - Different op definition to avoid & for extra bits (do & for table bits) + - Three separate decoding do-loops for direct, window, and write == 0 + - Special case for distance > 1 copies to do overlapped load and store copy + - Explicit branch predictions (based on measured branch probabilities) + - Deferring match copy and interspersed it with decoding subsequent codes + - Swapping literal/length else + - Swapping window/direct else + - Larger unrolled copy loops (three is about right) + - Moving len -= 3 statement into middle of loop + */ + +#endif /* !ASMINF */ diff --git a/linux/zlib_inflate/inffast.h b/linux/zlib_inflate/inffast.h new file mode 100644 index 0000000..40315d9 --- /dev/null +++ b/linux/zlib_inflate/inffast.h @@ -0,0 +1,11 @@ +/* inffast.h -- header to use inffast.c + * Copyright (C) 1995-2003 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +void inflate_fast (z_streamp strm, unsigned start); diff --git a/linux/zlib_inflate/inffixed.h b/linux/zlib_inflate/inffixed.h new file mode 100644 index 0000000..75ed4b5 --- /dev/null +++ b/linux/zlib_inflate/inffixed.h @@ -0,0 +1,94 @@ + /* inffixed.h -- table for decoding fixed codes + * Generated automatically by makefixed(). + */ + + /* WARNING: this file should *not* be used by applications. It + is part of the implementation of the compression library and + is subject to change. Applications should only use zlib.h. + */ + + static const code lenfix[512] = { + {96,7,0},{0,8,80},{0,8,16},{20,8,115},{18,7,31},{0,8,112},{0,8,48}, + {0,9,192},{16,7,10},{0,8,96},{0,8,32},{0,9,160},{0,8,0},{0,8,128}, + {0,8,64},{0,9,224},{16,7,6},{0,8,88},{0,8,24},{0,9,144},{19,7,59}, + {0,8,120},{0,8,56},{0,9,208},{17,7,17},{0,8,104},{0,8,40},{0,9,176}, + {0,8,8},{0,8,136},{0,8,72},{0,9,240},{16,7,4},{0,8,84},{0,8,20}, + {21,8,227},{19,7,43},{0,8,116},{0,8,52},{0,9,200},{17,7,13},{0,8,100}, + {0,8,36},{0,9,168},{0,8,4},{0,8,132},{0,8,68},{0,9,232},{16,7,8}, + {0,8,92},{0,8,28},{0,9,152},{20,7,83},{0,8,124},{0,8,60},{0,9,216}, + {18,7,23},{0,8,108},{0,8,44},{0,9,184},{0,8,12},{0,8,140},{0,8,76}, + {0,9,248},{16,7,3},{0,8,82},{0,8,18},{21,8,163},{19,7,35},{0,8,114}, + {0,8,50},{0,9,196},{17,7,11},{0,8,98},{0,8,34},{0,9,164},{0,8,2}, + {0,8,130},{0,8,66},{0,9,228},{16,7,7},{0,8,90},{0,8,26},{0,9,148}, + {20,7,67},{0,8,122},{0,8,58},{0,9,212},{18,7,19},{0,8,106},{0,8,42}, + {0,9,180},{0,8,10},{0,8,138},{0,8,74},{0,9,244},{16,7,5},{0,8,86}, + {0,8,22},{64,8,0},{19,7,51},{0,8,118},{0,8,54},{0,9,204},{17,7,15}, + {0,8,102},{0,8,38},{0,9,172},{0,8,6},{0,8,134},{0,8,70},{0,9,236}, + {16,7,9},{0,8,94},{0,8,30},{0,9,156},{20,7,99},{0,8,126},{0,8,62}, + {0,9,220},{18,7,27},{0,8,110},{0,8,46},{0,9,188},{0,8,14},{0,8,142}, + {0,8,78},{0,9,252},{96,7,0},{0,8,81},{0,8,17},{21,8,131},{18,7,31}, + {0,8,113},{0,8,49},{0,9,194},{16,7,10},{0,8,97},{0,8,33},{0,9,162}, + {0,8,1},{0,8,129},{0,8,65},{0,9,226},{16,7,6},{0,8,89},{0,8,25}, + {0,9,146},{19,7,59},{0,8,121},{0,8,57},{0,9,210},{17,7,17},{0,8,105}, + {0,8,41},{0,9,178},{0,8,9},{0,8,137},{0,8,73},{0,9,242},{16,7,4}, + {0,8,85},{0,8,21},{16,8,258},{19,7,43},{0,8,117},{0,8,53},{0,9,202}, + {17,7,13},{0,8,101},{0,8,37},{0,9,170},{0,8,5},{0,8,133},{0,8,69}, + {0,9,234},{16,7,8},{0,8,93},{0,8,29},{0,9,154},{20,7,83},{0,8,125}, + {0,8,61},{0,9,218},{18,7,23},{0,8,109},{0,8,45},{0,9,186},{0,8,13}, + {0,8,141},{0,8,77},{0,9,250},{16,7,3},{0,8,83},{0,8,19},{21,8,195}, + {19,7,35},{0,8,115},{0,8,51},{0,9,198},{17,7,11},{0,8,99},{0,8,35}, + {0,9,166},{0,8,3},{0,8,131},{0,8,67},{0,9,230},{16,7,7},{0,8,91}, + {0,8,27},{0,9,150},{20,7,67},{0,8,123},{0,8,59},{0,9,214},{18,7,19}, + {0,8,107},{0,8,43},{0,9,182},{0,8,11},{0,8,139},{0,8,75},{0,9,246}, + {16,7,5},{0,8,87},{0,8,23},{64,8,0},{19,7,51},{0,8,119},{0,8,55}, + {0,9,206},{17,7,15},{0,8,103},{0,8,39},{0,9,174},{0,8,7},{0,8,135}, + {0,8,71},{0,9,238},{16,7,9},{0,8,95},{0,8,31},{0,9,158},{20,7,99}, + {0,8,127},{0,8,63},{0,9,222},{18,7,27},{0,8,111},{0,8,47},{0,9,190}, + {0,8,15},{0,8,143},{0,8,79},{0,9,254},{96,7,0},{0,8,80},{0,8,16}, + {20,8,115},{18,7,31},{0,8,112},{0,8,48},{0,9,193},{16,7,10},{0,8,96}, + {0,8,32},{0,9,161},{0,8,0},{0,8,128},{0,8,64},{0,9,225},{16,7,6}, + {0,8,88},{0,8,24},{0,9,145},{19,7,59},{0,8,120},{0,8,56},{0,9,209}, + {17,7,17},{0,8,104},{0,8,40},{0,9,177},{0,8,8},{0,8,136},{0,8,72}, + {0,9,241},{16,7,4},{0,8,84},{0,8,20},{21,8,227},{19,7,43},{0,8,116}, + {0,8,52},{0,9,201},{17,7,13},{0,8,100},{0,8,36},{0,9,169},{0,8,4}, + {0,8,132},{0,8,68},{0,9,233},{16,7,8},{0,8,92},{0,8,28},{0,9,153}, + {20,7,83},{0,8,124},{0,8,60},{0,9,217},{18,7,23},{0,8,108},{0,8,44}, + {0,9,185},{0,8,12},{0,8,140},{0,8,76},{0,9,249},{16,7,3},{0,8,82}, + {0,8,18},{21,8,163},{19,7,35},{0,8,114},{0,8,50},{0,9,197},{17,7,11}, + {0,8,98},{0,8,34},{0,9,165},{0,8,2},{0,8,130},{0,8,66},{0,9,229}, + {16,7,7},{0,8,90},{0,8,26},{0,9,149},{20,7,67},{0,8,122},{0,8,58}, + {0,9,213},{18,7,19},{0,8,106},{0,8,42},{0,9,181},{0,8,10},{0,8,138}, + {0,8,74},{0,9,245},{16,7,5},{0,8,86},{0,8,22},{64,8,0},{19,7,51}, + {0,8,118},{0,8,54},{0,9,205},{17,7,15},{0,8,102},{0,8,38},{0,9,173}, + {0,8,6},{0,8,134},{0,8,70},{0,9,237},{16,7,9},{0,8,94},{0,8,30}, + {0,9,157},{20,7,99},{0,8,126},{0,8,62},{0,9,221},{18,7,27},{0,8,110}, + {0,8,46},{0,9,189},{0,8,14},{0,8,142},{0,8,78},{0,9,253},{96,7,0}, + {0,8,81},{0,8,17},{21,8,131},{18,7,31},{0,8,113},{0,8,49},{0,9,195}, + {16,7,10},{0,8,97},{0,8,33},{0,9,163},{0,8,1},{0,8,129},{0,8,65}, + {0,9,227},{16,7,6},{0,8,89},{0,8,25},{0,9,147},{19,7,59},{0,8,121}, + {0,8,57},{0,9,211},{17,7,17},{0,8,105},{0,8,41},{0,9,179},{0,8,9}, + {0,8,137},{0,8,73},{0,9,243},{16,7,4},{0,8,85},{0,8,21},{16,8,258}, + {19,7,43},{0,8,117},{0,8,53},{0,9,203},{17,7,13},{0,8,101},{0,8,37}, + {0,9,171},{0,8,5},{0,8,133},{0,8,69},{0,9,235},{16,7,8},{0,8,93}, + {0,8,29},{0,9,155},{20,7,83},{0,8,125},{0,8,61},{0,9,219},{18,7,23}, + {0,8,109},{0,8,45},{0,9,187},{0,8,13},{0,8,141},{0,8,77},{0,9,251}, + {16,7,3},{0,8,83},{0,8,19},{21,8,195},{19,7,35},{0,8,115},{0,8,51}, + {0,9,199},{17,7,11},{0,8,99},{0,8,35},{0,9,167},{0,8,3},{0,8,131}, + {0,8,67},{0,9,231},{16,7,7},{0,8,91},{0,8,27},{0,9,151},{20,7,67}, + {0,8,123},{0,8,59},{0,9,215},{18,7,19},{0,8,107},{0,8,43},{0,9,183}, + {0,8,11},{0,8,139},{0,8,75},{0,9,247},{16,7,5},{0,8,87},{0,8,23}, + {64,8,0},{19,7,51},{0,8,119},{0,8,55},{0,9,207},{17,7,15},{0,8,103}, + {0,8,39},{0,9,175},{0,8,7},{0,8,135},{0,8,71},{0,9,239},{16,7,9}, + {0,8,95},{0,8,31},{0,9,159},{20,7,99},{0,8,127},{0,8,63},{0,9,223}, + {18,7,27},{0,8,111},{0,8,47},{0,9,191},{0,8,15},{0,8,143},{0,8,79}, + {0,9,255} + }; + + static const code distfix[32] = { + {16,5,1},{23,5,257},{19,5,17},{27,5,4097},{17,5,5},{25,5,1025}, + {21,5,65},{29,5,16385},{16,5,3},{24,5,513},{20,5,33},{28,5,8193}, + {18,5,9},{26,5,2049},{22,5,129},{64,5,0},{16,5,2},{23,5,385}, + {19,5,25},{27,5,6145},{17,5,7},{25,5,1537},{21,5,97},{29,5,24577}, + {16,5,4},{24,5,769},{20,5,49},{28,5,12289},{18,5,13},{26,5,3073}, + {22,5,193},{64,5,0} + }; diff --git a/linux/zlib_inflate/inflate.c b/linux/zlib_inflate/inflate.c new file mode 100644 index 0000000..58a733b --- /dev/null +++ b/linux/zlib_inflate/inflate.c @@ -0,0 +1,786 @@ +/* inflate.c -- zlib decompression + * Copyright (C) 1995-2005 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + * + * Based on zlib 1.2.3 but modified for the Linux Kernel by + * Richard Purdie <richard@openedhand.com> + * + * Changes mainly for static instead of dynamic memory allocation + * + */ + +#include <linux/zutil.h> +#include "inftrees.h" +#include "inflate.h" +#include "inffast.h" +#include "infutil.h" + +int zlib_inflate_workspacesize(void) +{ + return sizeof(struct inflate_workspace); +} + +int zlib_inflateReset(z_streamp strm) +{ + struct inflate_state *state; + + if (strm == NULL || strm->state == NULL) return Z_STREAM_ERROR; + state = (struct inflate_state *)strm->state; + strm->total_in = strm->total_out = state->total = 0; + strm->msg = NULL; + strm->adler = 1; /* to support ill-conceived Java test suite */ + state->mode = HEAD; + state->last = 0; + state->havedict = 0; + state->dmax = 32768U; + state->hold = 0; + state->bits = 0; + state->lencode = state->distcode = state->next = state->codes; + + /* Initialise Window */ + state->wsize = 1U << state->wbits; + state->write = 0; + state->whave = 0; + + return Z_OK; +} + +int zlib_inflateInit2(z_streamp strm, int windowBits) +{ + struct inflate_state *state; + + if (strm == NULL) return Z_STREAM_ERROR; + strm->msg = NULL; /* in case we return an error */ + + state = &WS(strm)->inflate_state; + strm->state = (struct internal_state *)state; + + if (windowBits < 0) { + state->wrap = 0; + windowBits = -windowBits; + } + else { + state->wrap = (windowBits >> 4) + 1; + } + if (windowBits < 8 || windowBits > 15) { + return Z_STREAM_ERROR; + } + state->wbits = (unsigned)windowBits; + state->window = &WS(strm)->working_window[0]; + + return zlib_inflateReset(strm); +} + +/* + Return state with length and distance decoding tables and index sizes set to + fixed code decoding. This returns fixed tables from inffixed.h. + */ +static void zlib_fixedtables(struct inflate_state *state) +{ +# include "inffixed.h" + state->lencode = lenfix; + state->lenbits = 9; + state->distcode = distfix; + state->distbits = 5; +} + + +/* + Update the window with the last wsize (normally 32K) bytes written before + returning. This is only called when a window is already in use, or when + output has been written during this inflate call, but the end of the deflate + stream has not been reached yet. It is also called to window dictionary data + when a dictionary is loaded. + + Providing output buffers larger than 32K to inflate() should provide a speed + advantage, since only the last 32K of output is copied to the sliding window + upon return from inflate(), and since all distances after the first 32K of + output will fall in the output data, making match copies simpler and faster. + The advantage may be dependent on the size of the processor's data caches. + */ +static void zlib_updatewindow(z_streamp strm, unsigned out) +{ + struct inflate_state *state; + unsigned copy, dist; + + state = (struct inflate_state *)strm->state; + + /* copy state->wsize or less output bytes into the circular window */ + copy = out - strm->avail_out; + if (copy >= state->wsize) { + memcpy(state->window, strm->next_out - state->wsize, state->wsize); + state->write = 0; + state->whave = state->wsize; + } + else { + dist = state->wsize - state->write; + if (dist > copy) dist = copy; + memcpy(state->window + state->write, strm->next_out - copy, dist); + copy -= dist; + if (copy) { + memcpy(state->window, strm->next_out - copy, copy); + state->write = copy; + state->whave = state->wsize; + } + else { + state->write += dist; + if (state->write == state->wsize) state->write = 0; + if (state->whave < state->wsize) state->whave += dist; + } + } +} + + +/* + * At the end of a Deflate-compressed PPP packet, we expect to have seen + * a `stored' block type value but not the (zero) length bytes. + */ +/* + Returns true if inflate is currently at the end of a block generated by + Z_SYNC_FLUSH or Z_FULL_FLUSH. This function is used by one PPP + implementation to provide an additional safety check. PPP uses + Z_SYNC_FLUSH but removes the length bytes of the resulting empty stored + block. When decompressing, PPP checks that at the end of input packet, + inflate is waiting for these length bytes. + */ +static int zlib_inflateSyncPacket(z_streamp strm) +{ + struct inflate_state *state; + + if (strm == NULL || strm->state == NULL) return Z_STREAM_ERROR; + state = (struct inflate_state *)strm->state; + + if (state->mode == STORED && state->bits == 0) { + state->mode = TYPE; + return Z_OK; + } + return Z_DATA_ERROR; +} + +/* Macros for inflate(): */ + +/* check function to use adler32() for zlib or crc32() for gzip */ +#define UPDATE(check, buf, len) zlib_adler32(check, buf, len) + +/* Load registers with state in inflate() for speed */ +#define LOAD() \ + do { \ + put = strm->next_out; \ + left = strm->avail_out; \ + next = strm->next_in; \ + have = strm->avail_in; \ + hold = state->hold; \ + bits = state->bits; \ + } while (0) + +/* Restore state from registers in inflate() */ +#define RESTORE() \ + do { \ + strm->next_out = put; \ + strm->avail_out = left; \ + strm->next_in = next; \ + strm->avail_in = have; \ + state->hold = hold; \ + state->bits = bits; \ + } while (0) + +/* Clear the input bit accumulator */ +#define INITBITS() \ + do { \ + hold = 0; \ + bits = 0; \ + } while (0) + +/* Get a byte of input into the bit accumulator, or return from inflate() + if there is no input available. */ +#define PULLBYTE() \ + do { \ + if (have == 0) goto inf_leave; \ + have--; \ + hold += (unsigned long)(*next++) << bits; \ + bits += 8; \ + } while (0) + +/* Assure that there are at least n bits in the bit accumulator. If there is + not enough available input to do that, then return from inflate(). */ +#define NEEDBITS(n) \ + do { \ + while (bits < (unsigned)(n)) \ + PULLBYTE(); \ + } while (0) + +/* Return the low n bits of the bit accumulator (n < 16) */ +#define BITS(n) \ + ((unsigned)hold & ((1U << (n)) - 1)) + +/* Remove n bits from the bit accumulator */ +#define DROPBITS(n) \ + do { \ + hold >>= (n); \ + bits -= (unsigned)(n); \ + } while (0) + +/* Remove zero to seven bits as needed to go to a byte boundary */ +#define BYTEBITS() \ + do { \ + hold >>= bits & 7; \ + bits -= bits & 7; \ + } while (0) + +/* Reverse the bytes in a 32-bit value */ +#define REVERSE(q) \ + ((((q) >> 24) & 0xff) + (((q) >> 8) & 0xff00) + \ + (((q) & 0xff00) << 8) + (((q) & 0xff) << 24)) + +/* + inflate() uses a state machine to process as much input data and generate as + much output data as possible before returning. The state machine is + structured roughly as follows: + + for (;;) switch (state) { + ... + case STATEn: + if (not enough input data or output space to make progress) + return; + ... make progress ... + state = STATEm; + break; + ... + } + + so when inflate() is called again, the same case is attempted again, and + if the appropriate resources are provided, the machine proceeds to the + next state. The NEEDBITS() macro is usually the way the state evaluates + whether it can proceed or should return. NEEDBITS() does the return if + the requested bits are not available. The typical use of the BITS macros + is: + + NEEDBITS(n); + ... do something with BITS(n) ... + DROPBITS(n); + + where NEEDBITS(n) either returns from inflate() if there isn't enough + input left to load n bits into the accumulator, or it continues. BITS(n) + gives the low n bits in the accumulator. When done, DROPBITS(n) drops + the low n bits off the accumulator. INITBITS() clears the accumulator + and sets the number of available bits to zero. BYTEBITS() discards just + enough bits to put the accumulator on a byte boundary. After BYTEBITS() + and a NEEDBITS(8), then BITS(8) would return the next byte in the stream. + + NEEDBITS(n) uses PULLBYTE() to get an available byte of input, or to return + if there is no input available. The decoding of variable length codes uses + PULLBYTE() directly in order to pull just enough bytes to decode the next + code, and no more. + + Some states loop until they get enough input, making sure that enough + state information is maintained to continue the loop where it left off + if NEEDBITS() returns in the loop. For example, want, need, and keep + would all have to actually be part of the saved state in case NEEDBITS() + returns: + + case STATEw: + while (want < need) { + NEEDBITS(n); + keep[want++] = BITS(n); + DROPBITS(n); + } + state = STATEx; + case STATEx: + + As shown above, if the next state is also the next case, then the break + is omitted. + + A state may also return if there is not enough output space available to + complete that state. Those states are copying stored data, writing a + literal byte, and copying a matching string. + + When returning, a "goto inf_leave" is used to update the total counters, + update the check value, and determine whether any progress has been made + during that inflate() call in order to return the proper return code. + Progress is defined as a change in either strm->avail_in or strm->avail_out. + When there is a window, goto inf_leave will update the window with the last + output written. If a goto inf_leave occurs in the middle of decompression + and there is no window currently, goto inf_leave will create one and copy + output to the window for the next call of inflate(). + + In this implementation, the flush parameter of inflate() only affects the + return code (per zlib.h). inflate() always writes as much as possible to + strm->next_out, given the space available and the provided input--the effect + documented in zlib.h of Z_SYNC_FLUSH. Furthermore, inflate() always defers + the allocation of and copying into a sliding window until necessary, which + provides the effect documented in zlib.h for Z_FINISH when the entire input + stream available. So the only thing the flush parameter actually does is: + when flush is set to Z_FINISH, inflate() cannot return Z_OK. Instead it + will return Z_BUF_ERROR if it has not reached the end of the stream. + */ + +int zlib_inflate(z_streamp strm, int flush) +{ + struct inflate_state *state; + const unsigned char *next; /* next input */ + unsigned char *put; /* next output */ + unsigned have, left; /* available input and output */ + unsigned long hold; /* bit buffer */ + unsigned bits; /* bits in bit buffer */ + unsigned in, out; /* save starting available input and output */ + unsigned copy; /* number of stored or match bytes to copy */ + unsigned char *from; /* where to copy match bytes from */ + code this; /* current decoding table entry */ + code last; /* parent table entry */ + unsigned len; /* length to copy for repeats, bits to drop */ + int ret; /* return code */ + static const unsigned short order[19] = /* permutation of code lengths */ + {16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; + + /* Do not check for strm->next_out == NULL here as ppc zImage + inflates to strm->next_out = 0 */ + + if (strm == NULL || strm->state == NULL || + (strm->next_in == NULL && strm->avail_in != 0)) + return Z_STREAM_ERROR; + + state = (struct inflate_state *)strm->state; + + if (state->mode == TYPE) state->mode = TYPEDO; /* skip check */ + LOAD(); + in = have; + out = left; + ret = Z_OK; + for (;;) + switch (state->mode) { + case HEAD: + if (state->wrap == 0) { + state->mode = TYPEDO; + break; + } + NEEDBITS(16); + if ( + ((BITS(8) << 8) + (hold >> 8)) % 31) { + strm->msg = (char *)"incorrect header check"; + state->mode = BAD; + break; + } + if (BITS(4) != Z_DEFLATED) { + strm->msg = (char *)"unknown compression method"; + state->mode = BAD; + break; + } + DROPBITS(4); + len = BITS(4) + 8; + if (len > state->wbits) { + strm->msg = (char *)"invalid window size"; + state->mode = BAD; + break; + } + state->dmax = 1U << len; + strm->adler = state->check = zlib_adler32(0L, NULL, 0); + state->mode = hold & 0x200 ? DICTID : TYPE; + INITBITS(); + break; + case DICTID: + NEEDBITS(32); + strm->adler = state->check = REVERSE(hold); + INITBITS(); + state->mode = DICT; + case DICT: + if (state->havedict == 0) { + RESTORE(); + return Z_NEED_DICT; + } + strm->adler = state->check = zlib_adler32(0L, NULL, 0); + state->mode = TYPE; + case TYPE: + if (flush == Z_BLOCK) goto inf_leave; + case TYPEDO: + if (state->last) { + BYTEBITS(); + state->mode = CHECK; + break; + } + NEEDBITS(3); + state->last = BITS(1); + DROPBITS(1); + switch (BITS(2)) { + case 0: /* stored block */ + state->mode = STORED; + break; + case 1: /* fixed block */ + zlib_fixedtables(state); + state->mode = LEN; /* decode codes */ + break; + case 2: /* dynamic block */ + state->mode = TABLE; + break; + case 3: + strm->msg = (char *)"invalid block type"; + state->mode = BAD; + } + DROPBITS(2); + break; + case STORED: + BYTEBITS(); /* go to byte boundary */ + NEEDBITS(32); + if ((hold & 0xffff) != ((hold >> 16) ^ 0xffff)) { + strm->msg = (char *)"invalid stored block lengths"; + state->mode = BAD; + break; + } + state->length = (unsigned)hold & 0xffff; + INITBITS(); + state->mode = COPY; + case COPY: + copy = state->length; + if (copy) { + if (copy > have) copy = have; + if (copy > left) copy = left; + if (copy == 0) goto inf_leave; + memcpy(put, next, copy); + have -= copy; + next += copy; + left -= copy; + put += copy; + state->length -= copy; + break; + } + state->mode = TYPE; + break; + case TABLE: + NEEDBITS(14); + state->nlen = BITS(5) + 257; + DROPBITS(5); + state->ndist = BITS(5) + 1; + DROPBITS(5); + state->ncode = BITS(4) + 4; + DROPBITS(4); +#ifndef PKZIP_BUG_WORKAROUND + if (state->nlen > 286 || state->ndist > 30) { + strm->msg = (char *)"too many length or distance symbols"; + state->mode = BAD; + break; + } +#endif + state->have = 0; + state->mode = LENLENS; + case LENLENS: + while (state->have < state->ncode) { + NEEDBITS(3); + state->lens[order[state->have++]] = (unsigned short)BITS(3); + DROPBITS(3); + } + while (state->have < 19) + state->lens[order[state->have++]] = 0; + state->next = state->codes; + state->lencode = (code const *)(state->next); + state->lenbits = 7; + ret = zlib_inflate_table(CODES, state->lens, 19, &(state->next), + &(state->lenbits), state->work); + if (ret) { + strm->msg = (char *)"invalid code lengths set"; + state->mode = BAD; + break; + } + state->have = 0; + state->mode = CODELENS; + case CODELENS: + while (state->have < state->nlen + state->ndist) { + for (;;) { + this = state->lencode[BITS(state->lenbits)]; + if ((unsigned)(this.bits) <= bits) break; + PULLBYTE(); + } + if (this.val < 16) { + NEEDBITS(this.bits); + DROPBITS(this.bits); + state->lens[state->have++] = this.val; + } + else { + if (this.val == 16) { + NEEDBITS(this.bits + 2); + DROPBITS(this.bits); + if (state->have == 0) { + strm->msg = (char *)"invalid bit length repeat"; + state->mode = BAD; + break; + } + len = state->lens[state->have - 1]; + copy = 3 + BITS(2); + DROPBITS(2); + } + else if (this.val == 17) { + NEEDBITS(this.bits + 3); + DROPBITS(this.bits); + len = 0; + copy = 3 + BITS(3); + DROPBITS(3); + } + else { + NEEDBITS(this.bits + 7); + DROPBITS(this.bits); + len = 0; + copy = 11 + BITS(7); + DROPBITS(7); + } + if (state->have + copy > state->nlen + state->ndist) { + strm->msg = (char *)"invalid bit length repeat"; + state->mode = BAD; + break; + } + while (copy--) + state->lens[state->have++] = (unsigned short)len; + } + } + + /* handle error breaks in while */ + if (state->mode == BAD) break; + + /* build code tables */ + state->next = state->codes; + state->lencode = (code const *)(state->next); + state->lenbits = 9; + ret = zlib_inflate_table(LENS, state->lens, state->nlen, &(state->next), + &(state->lenbits), state->work); + if (ret) { + strm->msg = (char *)"invalid literal/lengths set"; + state->mode = BAD; + break; + } + state->distcode = (code const *)(state->next); + state->distbits = 6; + ret = zlib_inflate_table(DISTS, state->lens + state->nlen, state->ndist, + &(state->next), &(state->distbits), state->work); + if (ret) { + strm->msg = (char *)"invalid distances set"; + state->mode = BAD; + break; + } + state->mode = LEN; + case LEN: + if (have >= 6 && left >= 258) { + RESTORE(); + inflate_fast(strm, out); + LOAD(); + break; + } + for (;;) { + this = state->lencode[BITS(state->lenbits)]; + if ((unsigned)(this.bits) <= bits) break; + PULLBYTE(); + } + if (this.op && (this.op & 0xf0) == 0) { + last = this; + for (;;) { + this = state->lencode[last.val + + (BITS(last.bits + last.op) >> last.bits)]; + if ((unsigned)(last.bits + this.bits) <= bits) break; + PULLBYTE(); + } + DROPBITS(last.bits); + } + DROPBITS(this.bits); + state->length = (unsigned)this.val; + if ((int)(this.op) == 0) { + state->mode = LIT; + break; + } + if (this.op & 32) { + state->mode = TYPE; + break; + } + if (this.op & 64) { + strm->msg = (char *)"invalid literal/length code"; + state->mode = BAD; + break; + } + state->extra = (unsigned)(this.op) & 15; + state->mode = LENEXT; + case LENEXT: + if (state->extra) { + NEEDBITS(state->extra); + state->length += BITS(state->extra); + DROPBITS(state->extra); + } + state->mode = DIST; + case DIST: + for (;;) { + this = state->distcode[BITS(state->distbits)]; + if ((unsigned)(this.bits) <= bits) break; + PULLBYTE(); + } + if ((this.op & 0xf0) == 0) { + last = this; + for (;;) { + this = state->distcode[last.val + + (BITS(last.bits + last.op) >> last.bits)]; + if ((unsigned)(last.bits + this.bits) <= bits) break; + PULLBYTE(); + } + DROPBITS(last.bits); + } + DROPBITS(this.bits); + if (this.op & 64) { + strm->msg = (char *)"invalid distance code"; + state->mode = BAD; + break; + } + state->offset = (unsigned)this.val; + state->extra = (unsigned)(this.op) & 15; + state->mode = DISTEXT; + case DISTEXT: + if (state->extra) { + NEEDBITS(state->extra); + state->offset += BITS(state->extra); + DROPBITS(state->extra); + } +#ifdef INFLATE_STRICT + if (state->offset > state->dmax) { + strm->msg = (char *)"invalid distance too far back"; + state->mode = BAD; + break; + } +#endif + if (state->offset > state->whave + out - left) { + strm->msg = (char *)"invalid distance too far back"; + state->mode = BAD; + break; + } + state->mode = MATCH; + case MATCH: + if (left == 0) goto inf_leave; + copy = out - left; + if (state->offset > copy) { /* copy from window */ + copy = state->offset - copy; + if (copy > state->write) { + copy -= state->write; + from = state->window + (state->wsize - copy); + } + else + from = state->window + (state->write - copy); + if (copy > state->length) copy = state->length; + } + else { /* copy from output */ + from = put - state->offset; + copy = state->length; + } + if (copy > left) copy = left; + left -= copy; + state->length -= copy; + do { + *put++ = *from++; + } while (--copy); + if (state->length == 0) state->mode = LEN; + break; + case LIT: + if (left == 0) goto inf_leave; + *put++ = (unsigned char)(state->length); + left--; + state->mode = LEN; + break; + case CHECK: + if (state->wrap) { + NEEDBITS(32); + out -= left; + strm->total_out += out; + state->total += out; + if (out) + strm->adler = state->check = + UPDATE(state->check, put - out, out); + out = left; + if (( + REVERSE(hold)) != state->check) { + strm->msg = (char *)"incorrect data check"; + state->mode = BAD; + break; + } + INITBITS(); + } + state->mode = DONE; + case DONE: + ret = Z_STREAM_END; + goto inf_leave; + case BAD: + ret = Z_DATA_ERROR; + goto inf_leave; + case MEM: + return Z_MEM_ERROR; + case SYNC: + default: + return Z_STREAM_ERROR; + } + + /* + Return from inflate(), updating the total counts and the check value. + If there was no progress during the inflate() call, return a buffer + error. Call zlib_updatewindow() to create and/or update the window state. + */ + inf_leave: + RESTORE(); + if (state->wsize || (state->mode < CHECK && out != strm->avail_out)) + zlib_updatewindow(strm, out); + + in -= strm->avail_in; + out -= strm->avail_out; + strm->total_in += in; + strm->total_out += out; + state->total += out; + if (state->wrap && out) + strm->adler = state->check = + UPDATE(state->check, strm->next_out - out, out); + + strm->data_type = state->bits + (state->last ? 64 : 0) + + (state->mode == TYPE ? 128 : 0); + + if (flush == Z_PACKET_FLUSH && ret == Z_OK && + strm->avail_out != 0 && strm->avail_in == 0) + return zlib_inflateSyncPacket(strm); + + if (((in == 0 && out == 0) || flush == Z_FINISH) && ret == Z_OK) + ret = Z_BUF_ERROR; + + return ret; +} + +int zlib_inflateEnd(z_streamp strm) +{ + if (strm == NULL || strm->state == NULL) + return Z_STREAM_ERROR; + return Z_OK; +} + +/* + * This subroutine adds the data at next_in/avail_in to the output history + * without performing any output. The output buffer must be "caught up"; + * i.e. no pending output but this should always be the case. The state must + * be waiting on the start of a block (i.e. mode == TYPE or HEAD). On exit, + * the output will also be caught up, and the checksum will have been updated + * if need be. + */ +int zlib_inflateIncomp(z_stream *z) +{ + struct inflate_state *state = (struct inflate_state *)z->state; + Byte *saved_no = z->next_out; + uInt saved_ao = z->avail_out; + + if (state->mode != TYPE && state->mode != HEAD) + return Z_DATA_ERROR; + + /* Setup some variables to allow misuse of updateWindow */ + z->avail_out = 0; + z->next_out = (unsigned char*)z->next_in + z->avail_in; + + zlib_updatewindow(z, z->avail_in); + + /* Restore saved variables */ + z->avail_out = saved_ao; + z->next_out = saved_no; + + z->adler = state->check = + UPDATE(state->check, z->next_in, z->avail_in); + + z->total_out += z->avail_in; + z->total_in += z->avail_in; + z->next_in += z->avail_in; + state->total += z->avail_in; + z->avail_in = 0; + + return Z_OK; +} diff --git a/linux/zlib_inflate/inflate.h b/linux/zlib_inflate/inflate.h new file mode 100644 index 0000000..3d17b3d --- /dev/null +++ b/linux/zlib_inflate/inflate.h @@ -0,0 +1,111 @@ +#ifndef INFLATE_H +#define INFLATE_H + +/* inflate.h -- internal inflate state definition + * Copyright (C) 1995-2004 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +/* Possible inflate modes between inflate() calls */ +typedef enum { + HEAD, /* i: waiting for magic header */ + FLAGS, /* i: waiting for method and flags (gzip) */ + TIME, /* i: waiting for modification time (gzip) */ + OS, /* i: waiting for extra flags and operating system (gzip) */ + EXLEN, /* i: waiting for extra length (gzip) */ + EXTRA, /* i: waiting for extra bytes (gzip) */ + NAME, /* i: waiting for end of file name (gzip) */ + COMMENT, /* i: waiting for end of comment (gzip) */ + HCRC, /* i: waiting for header crc (gzip) */ + DICTID, /* i: waiting for dictionary check value */ + DICT, /* waiting for inflateSetDictionary() call */ + TYPE, /* i: waiting for type bits, including last-flag bit */ + TYPEDO, /* i: same, but skip check to exit inflate on new block */ + STORED, /* i: waiting for stored size (length and complement) */ + COPY, /* i/o: waiting for input or output to copy stored block */ + TABLE, /* i: waiting for dynamic block table lengths */ + LENLENS, /* i: waiting for code length code lengths */ + CODELENS, /* i: waiting for length/lit and distance code lengths */ + LEN, /* i: waiting for length/lit code */ + LENEXT, /* i: waiting for length extra bits */ + DIST, /* i: waiting for distance code */ + DISTEXT, /* i: waiting for distance extra bits */ + MATCH, /* o: waiting for output space to copy string */ + LIT, /* o: waiting for output space to write literal */ + CHECK, /* i: waiting for 32-bit check value */ + LENGTH, /* i: waiting for 32-bit length (gzip) */ + DONE, /* finished check, done -- remain here until reset */ + BAD, /* got a data error -- remain here until reset */ + MEM, /* got an inflate() memory error -- remain here until reset */ + SYNC /* looking for synchronization bytes to restart inflate() */ +} inflate_mode; + +/* + State transitions between above modes - + + (most modes can go to the BAD or MEM mode -- not shown for clarity) + + Process header: + HEAD -> (gzip) or (zlib) + (gzip) -> FLAGS -> TIME -> OS -> EXLEN -> EXTRA -> NAME + NAME -> COMMENT -> HCRC -> TYPE + (zlib) -> DICTID or TYPE + DICTID -> DICT -> TYPE + Read deflate blocks: + TYPE -> STORED or TABLE or LEN or CHECK + STORED -> COPY -> TYPE + TABLE -> LENLENS -> CODELENS -> LEN + Read deflate codes: + LEN -> LENEXT or LIT or TYPE + LENEXT -> DIST -> DISTEXT -> MATCH -> LEN + LIT -> LEN + Process trailer: + CHECK -> LENGTH -> DONE + */ + +/* state maintained between inflate() calls. Approximately 7K bytes. */ +struct inflate_state { + inflate_mode mode; /* current inflate mode */ + int last; /* true if processing last block */ + int wrap; /* bit 0 true for zlib, bit 1 true for gzip */ + int havedict; /* true if dictionary provided */ + int flags; /* gzip header method and flags (0 if zlib) */ + unsigned dmax; /* zlib header max distance (INFLATE_STRICT) */ + unsigned long check; /* protected copy of check value */ + unsigned long total; /* protected copy of output count */ + /* gz_headerp head; */ /* where to save gzip header information */ + /* sliding window */ + unsigned wbits; /* log base 2 of requested window size */ + unsigned wsize; /* window size or zero if not using window */ + unsigned whave; /* valid bytes in the window */ + unsigned write; /* window write index */ + unsigned char *window; /* allocated sliding window, if needed */ + /* bit accumulator */ + unsigned long hold; /* input bit accumulator */ + unsigned bits; /* number of bits in "in" */ + /* for string and stored block copying */ + unsigned length; /* literal or length of data to copy */ + unsigned offset; /* distance back to copy string from */ + /* for table and code decoding */ + unsigned extra; /* extra bits needed */ + /* fixed and dynamic code tables */ + code const *lencode; /* starting table for length/literal codes */ + code const *distcode; /* starting table for distance codes */ + unsigned lenbits; /* index bits for lencode */ + unsigned distbits; /* index bits for distcode */ + /* dynamic table building */ + unsigned ncode; /* number of code length code lengths */ + unsigned nlen; /* number of length code lengths */ + unsigned ndist; /* number of distance code lengths */ + unsigned have; /* number of code lengths in lens[] */ + code *next; /* next available space in codes[] */ + unsigned short lens[320]; /* temporary storage for code lengths */ + unsigned short work[288]; /* work area for code table building */ + code codes[ENOUGH]; /* space for code tables */ +}; +#endif diff --git a/linux/zlib_inflate/inftrees.c b/linux/zlib_inflate/inftrees.c new file mode 100644 index 0000000..3fe6ce5 --- /dev/null +++ b/linux/zlib_inflate/inftrees.c @@ -0,0 +1,315 @@ +/* inftrees.c -- generate Huffman trees for efficient decoding + * Copyright (C) 1995-2005 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include <linux/zutil.h> +#include "inftrees.h" + +#define MAXBITS 15 + +/* + Build a set of tables to decode the provided canonical Huffman code. + The code lengths are lens[0..codes-1]. The result starts at *table, + whose indices are 0..2^bits-1. work is a writable array of at least + lens shorts, which is used as a work area. type is the type of code + to be generated, CODES, LENS, or DISTS. On return, zero is success, + -1 is an invalid code, and +1 means that ENOUGH isn't enough. table + on return points to the next available entry's address. bits is the + requested root table index bits, and on return it is the actual root + table index bits. It will differ if the request is greater than the + longest code or if it is less than the shortest code. + */ +int zlib_inflate_table(codetype type, unsigned short *lens, unsigned codes, + code **table, unsigned *bits, unsigned short *work) +{ + unsigned len; /* a code's length in bits */ + unsigned sym; /* index of code symbols */ + unsigned min, max; /* minimum and maximum code lengths */ + unsigned root; /* number of index bits for root table */ + unsigned curr; /* number of index bits for current table */ + unsigned drop; /* code bits to drop for sub-table */ + int left; /* number of prefix codes available */ + unsigned used; /* code entries in table used */ + unsigned huff; /* Huffman code */ + unsigned incr; /* for incrementing code, index */ + unsigned fill; /* index for replicating entries */ + unsigned low; /* low bits for current root entry */ + unsigned mask; /* mask for low root bits */ + code this; /* table entry for duplication */ + code *next; /* next available space in table */ + const unsigned short *base; /* base value table to use */ + const unsigned short *extra; /* extra bits table to use */ + int end; /* use base and extra for symbol > end */ + unsigned short count[MAXBITS+1]; /* number of codes of each length */ + unsigned short offs[MAXBITS+1]; /* offsets in table for each length */ + static const unsigned short lbase[31] = { /* Length codes 257..285 base */ + 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, + 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; + static const unsigned short lext[31] = { /* Length codes 257..285 extra */ + 16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 18, 18, 18, 18, + 19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 201, 196}; + static const unsigned short dbase[32] = { /* Distance codes 0..29 base */ + 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193, + 257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145, + 8193, 12289, 16385, 24577, 0, 0}; + static const unsigned short dext[32] = { /* Distance codes 0..29 extra */ + 16, 16, 16, 16, 17, 17, 18, 18, 19, 19, 20, 20, 21, 21, 22, 22, + 23, 23, 24, 24, 25, 25, 26, 26, 27, 27, + 28, 28, 29, 29, 64, 64}; + + /* + Process a set of code lengths to create a canonical Huffman code. The + code lengths are lens[0..codes-1]. Each length corresponds to the + symbols 0..codes-1. The Huffman code is generated by first sorting the + symbols by length from short to long, and retaining the symbol order + for codes with equal lengths. Then the code starts with all zero bits + for the first code of the shortest length, and the codes are integer + increments for the same length, and zeros are appended as the length + increases. For the deflate format, these bits are stored backwards + from their more natural integer increment ordering, and so when the + decoding tables are built in the large loop below, the integer codes + are incremented backwards. + + This routine assumes, but does not check, that all of the entries in + lens[] are in the range 0..MAXBITS. The caller must assure this. + 1..MAXBITS is interpreted as that code length. zero means that that + symbol does not occur in this code. + + The codes are sorted by computing a count of codes for each length, + creating from that a table of starting indices for each length in the + sorted table, and then entering the symbols in order in the sorted + table. The sorted table is work[], with that space being provided by + the caller. + + The length counts are used for other purposes as well, i.e. finding + the minimum and maximum length codes, determining if there are any + codes at all, checking for a valid set of lengths, and looking ahead + at length counts to determine sub-table sizes when building the + decoding tables. + */ + + /* accumulate lengths for codes (assumes lens[] all in 0..MAXBITS) */ + for (len = 0; len <= MAXBITS; len++) + count[len] = 0; + for (sym = 0; sym < codes; sym++) + count[lens[sym]]++; + + /* bound code lengths, force root to be within code lengths */ + root = *bits; + for (max = MAXBITS; max >= 1; max--) + if (count[max] != 0) break; + if (root > max) root = max; + if (max == 0) { /* no symbols to code at all */ + this.op = (unsigned char)64; /* invalid code marker */ + this.bits = (unsigned char)1; + this.val = (unsigned short)0; + *(*table)++ = this; /* make a table to force an error */ + *(*table)++ = this; + *bits = 1; + return 0; /* no symbols, but wait for decoding to report error */ + } + for (min = 1; min <= MAXBITS; min++) + if (count[min] != 0) break; + if (root < min) root = min; + + /* check for an over-subscribed or incomplete set of lengths */ + left = 1; + for (len = 1; len <= MAXBITS; len++) { + left <<= 1; + left -= count[len]; + if (left < 0) return -1; /* over-subscribed */ + } + if (left > 0 && (type == CODES || max != 1)) + return -1; /* incomplete set */ + + /* generate offsets into symbol table for each length for sorting */ + offs[1] = 0; + for (len = 1; len < MAXBITS; len++) + offs[len + 1] = offs[len] + count[len]; + + /* sort symbols by length, by symbol order within each length */ + for (sym = 0; sym < codes; sym++) + if (lens[sym] != 0) work[offs[lens[sym]]++] = (unsigned short)sym; + + /* + Create and fill in decoding tables. In this loop, the table being + filled is at next and has curr index bits. The code being used is huff + with length len. That code is converted to an index by dropping drop + bits off of the bottom. For codes where len is less than drop + curr, + those top drop + curr - len bits are incremented through all values to + fill the table with replicated entries. + + root is the number of index bits for the root table. When len exceeds + root, sub-tables are created pointed to by the root entry with an index + of the low root bits of huff. This is saved in low to check for when a + new sub-table should be started. drop is zero when the root table is + being filled, and drop is root when sub-tables are being filled. + + When a new sub-table is needed, it is necessary to look ahead in the + code lengths to determine what size sub-table is needed. The length + counts are used for this, and so count[] is decremented as codes are + entered in the tables. + + used keeps track of how many table entries have been allocated from the + provided *table space. It is checked when a LENS table is being made + against the space in *table, ENOUGH, minus the maximum space needed by + the worst case distance code, MAXD. This should never happen, but the + sufficiency of ENOUGH has not been proven exhaustively, hence the check. + This assumes that when type == LENS, bits == 9. + + sym increments through all symbols, and the loop terminates when + all codes of length max, i.e. all codes, have been processed. This + routine permits incomplete codes, so another loop after this one fills + in the rest of the decoding tables with invalid code markers. + */ + + /* set up for code type */ + switch (type) { + case CODES: + base = extra = work; /* dummy value--not used */ + end = 19; + break; + case LENS: + base = lbase; + base -= 257; + extra = lext; + extra -= 257; + end = 256; + break; + default: /* DISTS */ + base = dbase; + extra = dext; + end = -1; + } + + /* initialize state for loop */ + huff = 0; /* starting code */ + sym = 0; /* starting code symbol */ + len = min; /* starting code length */ + next = *table; /* current table to fill in */ + curr = root; /* current table index bits */ + drop = 0; /* current bits to drop from code for index */ + low = (unsigned)(-1); /* trigger new sub-table when len > root */ + used = 1U << root; /* use root table entries */ + mask = used - 1; /* mask for comparing low */ + + /* check available table space */ + if (type == LENS && used >= ENOUGH - MAXD) + return 1; + + /* process all codes and make table entries */ + for (;;) { + /* create table entry */ + this.bits = (unsigned char)(len - drop); + if ((int)(work[sym]) < end) { + this.op = (unsigned char)0; + this.val = work[sym]; + } + else if ((int)(work[sym]) > end) { + this.op = (unsigned char)(extra[work[sym]]); + this.val = base[work[sym]]; + } + else { + this.op = (unsigned char)(32 + 64); /* end of block */ + this.val = 0; + } + + /* replicate for those indices with low len bits equal to huff */ + incr = 1U << (len - drop); + fill = 1U << curr; + min = fill; /* save offset to next table */ + do { + fill -= incr; + next[(huff >> drop) + fill] = this; + } while (fill != 0); + + /* backwards increment the len-bit code huff */ + incr = 1U << (len - 1); + while (huff & incr) + incr >>= 1; + if (incr != 0) { + huff &= incr - 1; + huff += incr; + } + else + huff = 0; + + /* go to next symbol, update count, len */ + sym++; + if (--(count[len]) == 0) { + if (len == max) break; + len = lens[work[sym]]; + } + + /* create new sub-table if needed */ + if (len > root && (huff & mask) != low) { + /* if first time, transition to sub-tables */ + if (drop == 0) + drop = root; + + /* increment past last table */ + next += min; /* here min is 1 << curr */ + + /* determine length of next table */ + curr = len - drop; + left = (int)(1 << curr); + while (curr + drop < max) { + left -= count[curr + drop]; + if (left <= 0) break; + curr++; + left <<= 1; + } + + /* check for enough space */ + used += 1U << curr; + if (type == LENS && used >= ENOUGH - MAXD) + return 1; + + /* point entry in root table to sub-table */ + low = huff & mask; + (*table)[low].op = (unsigned char)curr; + (*table)[low].bits = (unsigned char)root; + (*table)[low].val = (unsigned short)(next - *table); + } + } + + /* + Fill in rest of table for incomplete codes. This loop is similar to the + loop above in incrementing huff for table indices. It is assumed that + len is equal to curr + drop, so there is no loop needed to increment + through high index bits. When the current sub-table is filled, the loop + drops back to the root table to fill in any remaining entries there. + */ + this.op = (unsigned char)64; /* invalid code marker */ + this.bits = (unsigned char)(len - drop); + this.val = (unsigned short)0; + while (huff != 0) { + /* when done with sub-table, drop back to root table */ + if (drop != 0 && (huff & mask) != low) { + drop = 0; + len = root; + next = *table; + this.bits = (unsigned char)len; + } + + /* put invalid code marker in table */ + next[huff >> drop] = this; + + /* backwards increment the len-bit code huff */ + incr = 1U << (len - 1); + while (huff & incr) + incr >>= 1; + if (incr != 0) { + huff &= incr - 1; + huff += incr; + } + else + huff = 0; + } + + /* set return parameters */ + *table += used; + *bits = root; + return 0; +} diff --git a/linux/zlib_inflate/inftrees.h b/linux/zlib_inflate/inftrees.h new file mode 100644 index 0000000..b70b473 --- /dev/null +++ b/linux/zlib_inflate/inftrees.h @@ -0,0 +1,59 @@ +#ifndef INFTREES_H +#define INFTREES_H + +/* inftrees.h -- header to use inftrees.c + * Copyright (C) 1995-2005 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +/* Structure for decoding tables. Each entry provides either the + information needed to do the operation requested by the code that + indexed that table entry, or it provides a pointer to another + table that indexes more bits of the code. op indicates whether + the entry is a pointer to another table, a literal, a length or + distance, an end-of-block, or an invalid code. For a table + pointer, the low four bits of op is the number of index bits of + that table. For a length or distance, the low four bits of op + is the number of extra bits to get after the code. bits is + the number of bits in this code or part of the code to drop off + of the bit buffer. val is the actual byte to output in the case + of a literal, the base length or distance, or the offset from + the current table to the next table. Each entry is four bytes. */ +typedef struct { + unsigned char op; /* operation, extra bits, table bits */ + unsigned char bits; /* bits in this part of the code */ + unsigned short val; /* offset in table or code value */ +} code; + +/* op values as set by inflate_table(): + 00000000 - literal + 0000tttt - table link, tttt != 0 is the number of table index bits + 0001eeee - length or distance, eeee is the number of extra bits + 01100000 - end of block + 01000000 - invalid code + */ + +/* Maximum size of dynamic tree. The maximum found in a long but non- + exhaustive search was 1444 code structures (852 for length/literals + and 592 for distances, the latter actually the result of an + exhaustive search). The true maximum is not known, but the value + below is more than safe. */ +#define ENOUGH 2048 +#define MAXD 592 + +/* Type of code to build for inftable() */ +typedef enum { + CODES, + LENS, + DISTS +} codetype; + +extern int zlib_inflate_table (codetype type, unsigned short *lens, + unsigned codes, code **table, + unsigned *bits, unsigned short *work); +#endif diff --git a/linux/zlib_inflate/infutil.c b/linux/zlib_inflate/infutil.c new file mode 100644 index 0000000..4824c2c --- /dev/null +++ b/linux/zlib_inflate/infutil.c @@ -0,0 +1,49 @@ +#include <linux/zutil.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> + +/* Utility function: initialize zlib, unpack binary blob, clean up zlib, + * return len or negative error code. + */ +int zlib_inflate_blob(void *gunzip_buf, unsigned int sz, + const void *buf, unsigned int len) +{ + const u8 *zbuf = buf; + struct z_stream_s *strm; + int rc; + + rc = -ENOMEM; + strm = kmalloc(sizeof(*strm), GFP_KERNEL); + if (strm == NULL) + goto gunzip_nomem1; + strm->workspace = kmalloc(zlib_inflate_workspacesize(), GFP_KERNEL); + if (strm->workspace == NULL) + goto gunzip_nomem2; + + /* gzip header (1f,8b,08... 10 bytes total + possible asciz filename) + * expected to be stripped from input + */ + strm->next_in = zbuf; + strm->avail_in = len; + strm->next_out = gunzip_buf; + strm->avail_out = sz; + + rc = zlib_inflateInit2(strm, -MAX_WBITS); + if (rc == Z_OK) { + rc = zlib_inflate(strm, Z_FINISH); + /* after Z_FINISH, only Z_STREAM_END is "we unpacked it all" */ + if (rc == Z_STREAM_END) + rc = sz - strm->avail_out; + else + rc = -EINVAL; + zlib_inflateEnd(strm); + } else + rc = -EINVAL; + + kfree(strm->workspace); +gunzip_nomem2: + kfree(strm); +gunzip_nomem1: + return rc; /* returns Z_OK (0) if successful */ +} diff --git a/linux/zlib_inflate/infutil.h b/linux/zlib_inflate/infutil.h new file mode 100644 index 0000000..eb1a900 --- /dev/null +++ b/linux/zlib_inflate/infutil.h @@ -0,0 +1,25 @@ +/* infutil.h -- types and macros common to blocks and codes + * Copyright (C) 1995-1998 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +#ifndef _INFUTIL_H +#define _INFUTIL_H + +#include <linux/zlib.h> + +/* memory allocation for inflation */ + +struct inflate_workspace { + struct inflate_state inflate_state; + unsigned char working_window[1 << MAX_WBITS]; +}; + +#define WS(z) ((struct inflate_workspace *)(z->workspace)) + +#endif |