diff options
Diffstat (limited to 'kernel/drivers/block/zram/zram_drv.c')
-rw-r--r-- | kernel/drivers/block/zram/zram_drv.c | 1024 |
1 files changed, 578 insertions, 446 deletions
diff --git a/kernel/drivers/block/zram/zram_drv.c b/kernel/drivers/block/zram/zram_drv.c index 6e134f475..65e0b375a 100644 --- a/kernel/drivers/block/zram/zram_drv.c +++ b/kernel/drivers/block/zram/zram_drv.c @@ -15,10 +15,6 @@ #define KMSG_COMPONENT "zram" #define pr_fmt(fmt) KMSG_COMPONENT ": " fmt -#ifdef CONFIG_ZRAM_DEBUG -#define DEBUG -#endif - #include <linux/module.h> #include <linux/kernel.h> #include <linux/bio.h> @@ -32,12 +28,16 @@ #include <linux/string.h> #include <linux/vmalloc.h> #include <linux/err.h> +#include <linux/idr.h> +#include <linux/sysfs.h> #include "zram_drv.h" -/* Globals */ +static DEFINE_IDR(zram_index_idr); +/* idr index must be protected */ +static DEFINE_MUTEX(zram_index_mutex); + static int zram_major; -static struct zram *zram_devices; static const char *default_compressor = "lzo"; /* Module params (documentation at end) */ @@ -53,7 +53,7 @@ static inline void deprecated_attr_warn(const char *name) } #define ZRAM_ATTR_RO(name) \ -static ssize_t name##_show(struct device *d, \ +static ssize_t name##_show(struct device *d, \ struct device_attribute *attr, char *b) \ { \ struct zram *zram = dev_to_zram(d); \ @@ -74,33 +74,117 @@ static inline struct zram *dev_to_zram(struct device *dev) return (struct zram *)dev_to_disk(dev)->private_data; } -static ssize_t compact_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t len) +/* flag operations require table entry bit_spin_lock() being held */ +static int zram_test_flag(struct zram_meta *meta, u32 index, + enum zram_pageflags flag) { - unsigned long nr_migrated; - struct zram *zram = dev_to_zram(dev); - struct zram_meta *meta; + return meta->table[index].value & BIT(flag); +} - down_read(&zram->init_lock); - if (!init_done(zram)) { - up_read(&zram->init_lock); - return -EINVAL; - } +static void zram_set_flag(struct zram_meta *meta, u32 index, + enum zram_pageflags flag) +{ + meta->table[index].value |= BIT(flag); +} - meta = zram->meta; - nr_migrated = zs_compact(meta->mem_pool); - atomic64_add(nr_migrated, &zram->stats.num_migrated); - up_read(&zram->init_lock); +static void zram_clear_flag(struct zram_meta *meta, u32 index, + enum zram_pageflags flag) +{ + meta->table[index].value &= ~BIT(flag); +} - return len; +static size_t zram_get_obj_size(struct zram_meta *meta, u32 index) +{ + return meta->table[index].value & (BIT(ZRAM_FLAG_SHIFT) - 1); } -static ssize_t disksize_show(struct device *dev, - struct device_attribute *attr, char *buf) +static void zram_set_obj_size(struct zram_meta *meta, + u32 index, size_t size) { - struct zram *zram = dev_to_zram(dev); + unsigned long flags = meta->table[index].value >> ZRAM_FLAG_SHIFT; - return scnprintf(buf, PAGE_SIZE, "%llu\n", zram->disksize); + meta->table[index].value = (flags << ZRAM_FLAG_SHIFT) | size; +} + +static inline bool is_partial_io(struct bio_vec *bvec) +{ + return bvec->bv_len != PAGE_SIZE; +} + +/* + * Check if request is within bounds and aligned on zram logical blocks. + */ +static inline bool valid_io_request(struct zram *zram, + sector_t start, unsigned int size) +{ + u64 end, bound; + + /* unaligned request */ + if (unlikely(start & (ZRAM_SECTOR_PER_LOGICAL_BLOCK - 1))) + return false; + if (unlikely(size & (ZRAM_LOGICAL_BLOCK_SIZE - 1))) + return false; + + end = start + (size >> SECTOR_SHIFT); + bound = zram->disksize >> SECTOR_SHIFT; + /* out of range range */ + if (unlikely(start >= bound || end > bound || start > end)) + return false; + + /* I/O request is valid */ + return true; +} + +static void update_position(u32 *index, int *offset, struct bio_vec *bvec) +{ + if (*offset + bvec->bv_len >= PAGE_SIZE) + (*index)++; + *offset = (*offset + bvec->bv_len) % PAGE_SIZE; +} + +static inline void update_used_max(struct zram *zram, + const unsigned long pages) +{ + unsigned long old_max, cur_max; + + old_max = atomic_long_read(&zram->stats.max_used_pages); + + do { + cur_max = old_max; + if (pages > cur_max) + old_max = atomic_long_cmpxchg( + &zram->stats.max_used_pages, cur_max, pages); + } while (old_max != cur_max); +} + +static bool page_zero_filled(void *ptr) +{ + unsigned int pos; + unsigned long *page; + + page = (unsigned long *)ptr; + + for (pos = 0; pos != PAGE_SIZE / sizeof(*page); pos++) { + if (page[pos]) + return false; + } + + return true; +} + +static void handle_zero_page(struct bio_vec *bvec) +{ + struct page *page = bvec->bv_page; + void *user_mem; + + user_mem = kmap_atomic(page); + if (is_partial_io(bvec)) + memset(user_mem + bvec->bv_offset, 0, bvec->bv_len); + else + clear_page(user_mem); + kunmap_atomic(user_mem); + + flush_dcache_page(page); } static ssize_t initstate_show(struct device *dev, @@ -116,6 +200,14 @@ static ssize_t initstate_show(struct device *dev, return scnprintf(buf, PAGE_SIZE, "%u\n", val); } +static ssize_t disksize_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct zram *zram = dev_to_zram(dev); + + return scnprintf(buf, PAGE_SIZE, "%llu\n", zram->disksize); +} + static ssize_t orig_data_size_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -143,19 +235,6 @@ static ssize_t mem_used_total_show(struct device *dev, return scnprintf(buf, PAGE_SIZE, "%llu\n", val << PAGE_SHIFT); } -static ssize_t max_comp_streams_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - int val; - struct zram *zram = dev_to_zram(dev); - - down_read(&zram->init_lock); - val = zram->max_comp_streams; - up_read(&zram->init_lock); - - return scnprintf(buf, PAGE_SIZE, "%d\n", val); -} - static ssize_t mem_limit_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -225,6 +304,19 @@ static ssize_t mem_used_max_store(struct device *dev, return len; } +static ssize_t max_comp_streams_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int val; + struct zram *zram = dev_to_zram(dev); + + down_read(&zram->init_lock); + val = zram->max_comp_streams; + up_read(&zram->init_lock); + + return scnprintf(buf, PAGE_SIZE, "%d\n", val); +} + static ssize_t max_comp_streams_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t len) { @@ -271,6 +363,11 @@ static ssize_t comp_algorithm_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t len) { struct zram *zram = dev_to_zram(dev); + size_t sz; + + if (!zcomp_available_algorithm(buf)) + return -EINVAL; + down_write(&zram->init_lock); if (init_done(zram)) { up_write(&zram->init_lock); @@ -278,69 +375,108 @@ static ssize_t comp_algorithm_store(struct device *dev, return -EBUSY; } strlcpy(zram->compressor, buf, sizeof(zram->compressor)); + + /* ignore trailing newline */ + sz = strlen(zram->compressor); + if (sz > 0 && zram->compressor[sz - 1] == '\n') + zram->compressor[sz - 1] = 0x00; + up_write(&zram->init_lock); return len; } -/* flag operations needs meta->tb_lock */ -static int zram_test_flag(struct zram_meta *meta, u32 index, - enum zram_pageflags flag) +static ssize_t compact_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) { - return meta->table[index].value & BIT(flag); -} + struct zram *zram = dev_to_zram(dev); + struct zram_meta *meta; -static void zram_set_flag(struct zram_meta *meta, u32 index, - enum zram_pageflags flag) -{ - meta->table[index].value |= BIT(flag); -} + down_read(&zram->init_lock); + if (!init_done(zram)) { + up_read(&zram->init_lock); + return -EINVAL; + } -static void zram_clear_flag(struct zram_meta *meta, u32 index, - enum zram_pageflags flag) -{ - meta->table[index].value &= ~BIT(flag); + meta = zram->meta; + zs_compact(meta->mem_pool); + up_read(&zram->init_lock); + + return len; } -static size_t zram_get_obj_size(struct zram_meta *meta, u32 index) +static ssize_t io_stat_show(struct device *dev, + struct device_attribute *attr, char *buf) { - return meta->table[index].value & (BIT(ZRAM_FLAG_SHIFT) - 1); + struct zram *zram = dev_to_zram(dev); + ssize_t ret; + + down_read(&zram->init_lock); + ret = scnprintf(buf, PAGE_SIZE, + "%8llu %8llu %8llu %8llu\n", + (u64)atomic64_read(&zram->stats.failed_reads), + (u64)atomic64_read(&zram->stats.failed_writes), + (u64)atomic64_read(&zram->stats.invalid_io), + (u64)atomic64_read(&zram->stats.notify_free)); + up_read(&zram->init_lock); + + return ret; } -static void zram_set_obj_size(struct zram_meta *meta, - u32 index, size_t size) +static ssize_t mm_stat_show(struct device *dev, + struct device_attribute *attr, char *buf) { - unsigned long flags = meta->table[index].value >> ZRAM_FLAG_SHIFT; + struct zram *zram = dev_to_zram(dev); + struct zs_pool_stats pool_stats; + u64 orig_size, mem_used = 0; + long max_used; + ssize_t ret; - meta->table[index].value = (flags << ZRAM_FLAG_SHIFT) | size; + memset(&pool_stats, 0x00, sizeof(struct zs_pool_stats)); + + down_read(&zram->init_lock); + if (init_done(zram)) { + mem_used = zs_get_total_pages(zram->meta->mem_pool); + zs_pool_stats(zram->meta->mem_pool, &pool_stats); + } + + orig_size = atomic64_read(&zram->stats.pages_stored); + max_used = atomic_long_read(&zram->stats.max_used_pages); + + ret = scnprintf(buf, PAGE_SIZE, + "%8llu %8llu %8llu %8lu %8ld %8llu %8lu\n", + orig_size << PAGE_SHIFT, + (u64)atomic64_read(&zram->stats.compr_data_size), + mem_used << PAGE_SHIFT, + zram->limit_pages << PAGE_SHIFT, + max_used << PAGE_SHIFT, + (u64)atomic64_read(&zram->stats.zero_pages), + pool_stats.pages_compacted); + up_read(&zram->init_lock); + + return ret; } -static inline int is_partial_io(struct bio_vec *bvec) +static DEVICE_ATTR_RO(io_stat); +static DEVICE_ATTR_RO(mm_stat); +ZRAM_ATTR_RO(num_reads); +ZRAM_ATTR_RO(num_writes); +ZRAM_ATTR_RO(failed_reads); +ZRAM_ATTR_RO(failed_writes); +ZRAM_ATTR_RO(invalid_io); +ZRAM_ATTR_RO(notify_free); +ZRAM_ATTR_RO(zero_pages); +ZRAM_ATTR_RO(compr_data_size); + +static inline bool zram_meta_get(struct zram *zram) { - return bvec->bv_len != PAGE_SIZE; + if (atomic_inc_not_zero(&zram->refcount)) + return true; + return false; } -/* - * Check if request is within bounds and aligned on zram logical blocks. - */ -static inline int valid_io_request(struct zram *zram, - sector_t start, unsigned int size) +static inline void zram_meta_put(struct zram *zram) { - u64 end, bound; - - /* unaligned request */ - if (unlikely(start & (ZRAM_SECTOR_PER_LOGICAL_BLOCK - 1))) - return 0; - if (unlikely(size & (ZRAM_LOGICAL_BLOCK_SIZE - 1))) - return 0; - - end = start + (size >> SECTOR_SHIFT); - bound = zram->disksize >> SECTOR_SHIFT; - /* out of range range */ - if (unlikely(start >= bound || end > bound || start > end)) - return 0; - - /* I/O request is valid */ - return 1; + atomic_dec(&zram->refcount); } static void zram_meta_free(struct zram_meta *meta, u64 disksize) @@ -363,10 +499,9 @@ static void zram_meta_free(struct zram_meta *meta, u64 disksize) kfree(meta); } -static struct zram_meta *zram_meta_alloc(int device_id, u64 disksize) +static struct zram_meta *zram_meta_alloc(char *pool_name, u64 disksize) { size_t num_pages; - char pool_name[8]; struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL); if (!meta) @@ -379,13 +514,14 @@ static struct zram_meta *zram_meta_alloc(int device_id, u64 disksize) goto out_error; } - snprintf(pool_name, sizeof(pool_name), "zram%d", device_id); meta->mem_pool = zs_create_pool(pool_name, GFP_NOIO | __GFP_HIGHMEM); if (!meta->mem_pool) { pr_err("Error creating memory pool\n"); goto out_error; } + zram_meta_init_table_locks(meta, disksize); + return meta; out_error: @@ -394,56 +530,6 @@ out_error: return NULL; } -static inline bool zram_meta_get(struct zram *zram) -{ - if (atomic_inc_not_zero(&zram->refcount)) - return true; - return false; -} - -static inline void zram_meta_put(struct zram *zram) -{ - atomic_dec(&zram->refcount); -} - -static void update_position(u32 *index, int *offset, struct bio_vec *bvec) -{ - if (*offset + bvec->bv_len >= PAGE_SIZE) - (*index)++; - *offset = (*offset + bvec->bv_len) % PAGE_SIZE; -} - -static int page_zero_filled(void *ptr) -{ - unsigned int pos; - unsigned long *page; - - page = (unsigned long *)ptr; - - for (pos = 0; pos != PAGE_SIZE / sizeof(*page); pos++) { - if (page[pos]) - return 0; - } - - return 1; -} - -static void handle_zero_page(struct bio_vec *bvec) -{ - struct page *page = bvec->bv_page; - void *user_mem; - - user_mem = kmap_atomic(page); - if (is_partial_io(bvec)) - memset(user_mem + bvec->bv_offset, 0, bvec->bv_len); - else - clear_page(user_mem); - kunmap_atomic(user_mem); - - flush_dcache_page(page); -} - - /* * To protect concurrent access to the same index entry, * caller should hold this table index entry's bit_spinlock to @@ -484,12 +570,12 @@ static int zram_decompress_page(struct zram *zram, char *mem, u32 index) unsigned long handle; size_t size; - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value); + zram_lock_table(&meta->table[index]); handle = meta->table[index].handle; size = zram_get_obj_size(meta, index); if (!handle || zram_test_flag(meta, index, ZRAM_ZERO)) { - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); clear_page(mem); return 0; } @@ -500,7 +586,7 @@ static int zram_decompress_page(struct zram *zram, char *mem, u32 index) else ret = zcomp_decompress(zram->comp, cmem, size, mem); zs_unmap_object(meta->mem_pool, handle); - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); /* Should NEVER happen. Return bio error if it does. */ if (unlikely(ret)) { @@ -520,14 +606,14 @@ static int zram_bvec_read(struct zram *zram, struct bio_vec *bvec, struct zram_meta *meta = zram->meta; page = bvec->bv_page; - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value); + zram_lock_table(&meta->table[index]); if (unlikely(!meta->table[index].handle) || zram_test_flag(meta, index, ZRAM_ZERO)) { - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); handle_zero_page(bvec); return 0; } - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); if (is_partial_io(bvec)) /* Use a temporary buffer to decompress the page */ @@ -538,7 +624,7 @@ static int zram_bvec_read(struct zram *zram, struct bio_vec *bvec, uncmem = user_mem; if (!uncmem) { - pr_info("Unable to allocate temp memory\n"); + pr_err("Unable to allocate temp memory\n"); ret = -ENOMEM; goto out_cleanup; } @@ -561,21 +647,6 @@ out_cleanup: return ret; } -static inline void update_used_max(struct zram *zram, - const unsigned long pages) -{ - unsigned long old_max, cur_max; - - old_max = atomic_long_read(&zram->stats.max_used_pages); - - do { - cur_max = old_max; - if (pages > cur_max) - old_max = atomic_long_cmpxchg( - &zram->stats.max_used_pages, cur_max, pages); - } while (old_max != cur_max); -} - static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index, int offset) { @@ -585,8 +656,7 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index, struct page *page; unsigned char *user_mem, *cmem, *src, *uncmem = NULL; struct zram_meta *meta = zram->meta; - struct zcomp_strm *zstrm; - bool locked = false; + struct zcomp_strm *zstrm = NULL; unsigned long alloced_pages; page = bvec->bv_page; @@ -606,7 +676,6 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index, } zstrm = zcomp_strm_find(zram->comp); - locked = true; user_mem = kmap_atomic(page); if (is_partial_io(bvec)) { @@ -622,10 +691,10 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index, if (user_mem) kunmap_atomic(user_mem); /* Free memory associated with this sector now. */ - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value); + zram_lock_table(&meta->table[index]); zram_free_page(zram, index); zram_set_flag(meta, index, ZRAM_ZERO); - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); atomic64_inc(&zram->stats.zero_pages); ret = 0; @@ -652,21 +721,21 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index, handle = zs_malloc(meta->mem_pool, clen); if (!handle) { - pr_info("Error allocating memory for compressed page: %u, size=%zu\n", + pr_err("Error allocating memory for compressed page: %u, size=%zu\n", index, clen); ret = -ENOMEM; goto out; } alloced_pages = zs_get_total_pages(meta->mem_pool); + update_used_max(zram, alloced_pages); + if (zram->limit_pages && alloced_pages > zram->limit_pages) { zs_free(meta->mem_pool, handle); ret = -ENOMEM; goto out; } - update_used_max(zram, alloced_pages); - cmem = zs_map_object(meta->mem_pool, handle, ZS_MM_WO); if ((clen == PAGE_SIZE) && !is_partial_io(bvec)) { @@ -678,60 +747,31 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index, } zcomp_strm_release(zram->comp, zstrm); - locked = false; + zstrm = NULL; zs_unmap_object(meta->mem_pool, handle); /* * Free memory associated with this sector * before overwriting unused sectors. */ - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value); + zram_lock_table(&meta->table[index]); zram_free_page(zram, index); meta->table[index].handle = handle; zram_set_obj_size(meta, index, clen); - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); /* Update stats */ atomic64_add(clen, &zram->stats.compr_data_size); atomic64_inc(&zram->stats.pages_stored); out: - if (locked) + if (zstrm) zcomp_strm_release(zram->comp, zstrm); if (is_partial_io(bvec)) kfree(uncmem); return ret; } -static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index, - int offset, int rw) -{ - unsigned long start_time = jiffies; - int ret; - - generic_start_io_acct(rw, bvec->bv_len >> SECTOR_SHIFT, - &zram->disk->part0); - - if (rw == READ) { - atomic64_inc(&zram->stats.num_reads); - ret = zram_bvec_read(zram, bvec, index, offset); - } else { - atomic64_inc(&zram->stats.num_writes); - ret = zram_bvec_write(zram, bvec, index, offset); - } - - generic_end_io_acct(rw, &zram->disk->part0, start_time); - - if (unlikely(ret)) { - if (rw == READ) - atomic64_inc(&zram->stats.failed_reads); - else - atomic64_inc(&zram->stats.failed_writes); - } - - return ret; -} - /* * zram_bio_discard - handler on discard request * @index: physical block index in PAGE_SIZE units @@ -762,160 +802,41 @@ static void zram_bio_discard(struct zram *zram, u32 index, } while (n >= PAGE_SIZE) { - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value); + zram_lock_table(&meta->table[index]); zram_free_page(zram, index); - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); atomic64_inc(&zram->stats.notify_free); index++; n -= PAGE_SIZE; } } -static void zram_reset_device(struct zram *zram) -{ - struct zram_meta *meta; - struct zcomp *comp; - u64 disksize; - - down_write(&zram->init_lock); - - zram->limit_pages = 0; - - if (!init_done(zram)) { - up_write(&zram->init_lock); - return; - } - - meta = zram->meta; - comp = zram->comp; - disksize = zram->disksize; - /* - * Refcount will go down to 0 eventually and r/w handler - * cannot handle further I/O so it will bail out by - * check zram_meta_get. - */ - zram_meta_put(zram); - /* - * We want to free zram_meta in process context to avoid - * deadlock between reclaim path and any other locks. - */ - wait_event(zram->io_done, atomic_read(&zram->refcount) == 0); - - /* Reset stats */ - memset(&zram->stats, 0, sizeof(zram->stats)); - zram->disksize = 0; - zram->max_comp_streams = 1; - - set_capacity(zram->disk, 0); - part_stat_set_all(&zram->disk->part0, 0); - - up_write(&zram->init_lock); - /* I/O operation under all of CPU are done so let's free */ - zram_meta_free(meta, disksize); - zcomp_destroy(comp); -} - -static ssize_t disksize_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t len) -{ - u64 disksize; - struct zcomp *comp; - struct zram_meta *meta; - struct zram *zram = dev_to_zram(dev); - int err; - - disksize = memparse(buf, NULL); - if (!disksize) - return -EINVAL; - - disksize = PAGE_ALIGN(disksize); - meta = zram_meta_alloc(zram->disk->first_minor, disksize); - if (!meta) - return -ENOMEM; - - comp = zcomp_create(zram->compressor, zram->max_comp_streams); - if (IS_ERR(comp)) { - pr_info("Cannot initialise %s compressing backend\n", - zram->compressor); - err = PTR_ERR(comp); - goto out_free_meta; - } - - down_write(&zram->init_lock); - if (init_done(zram)) { - pr_info("Cannot change disksize for initialized device\n"); - err = -EBUSY; - goto out_destroy_comp; - } - - init_waitqueue_head(&zram->io_done); - atomic_set(&zram->refcount, 1); - zram->meta = meta; - zram->comp = comp; - zram->disksize = disksize; - set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT); - up_write(&zram->init_lock); - - /* - * Revalidate disk out of the init_lock to avoid lockdep splat. - * It's okay because disk's capacity is protected by init_lock - * so that revalidate_disk always sees up-to-date capacity. - */ - revalidate_disk(zram->disk); - - return len; - -out_destroy_comp: - up_write(&zram->init_lock); - zcomp_destroy(comp); -out_free_meta: - zram_meta_free(meta, disksize); - return err; -} - -static ssize_t reset_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t len) +static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index, + int offset, int rw) { + unsigned long start_time = jiffies; int ret; - unsigned short do_reset; - struct zram *zram; - struct block_device *bdev; - zram = dev_to_zram(dev); - bdev = bdget_disk(zram->disk, 0); - - if (!bdev) - return -ENOMEM; + generic_start_io_acct(rw, bvec->bv_len >> SECTOR_SHIFT, + &zram->disk->part0); - mutex_lock(&bdev->bd_mutex); - /* Do not reset an active device! */ - if (bdev->bd_openers) { - ret = -EBUSY; - goto out; + if (rw == READ) { + atomic64_inc(&zram->stats.num_reads); + ret = zram_bvec_read(zram, bvec, index, offset); + } else { + atomic64_inc(&zram->stats.num_writes); + ret = zram_bvec_write(zram, bvec, index, offset); } - ret = kstrtou16(buf, 10, &do_reset); - if (ret) - goto out; + generic_end_io_acct(rw, &zram->disk->part0, start_time); - if (!do_reset) { - ret = -EINVAL; - goto out; + if (unlikely(ret)) { + if (rw == READ) + atomic64_inc(&zram->stats.failed_reads); + else + atomic64_inc(&zram->stats.failed_writes); } - /* Make sure all pending I/O is finished */ - fsync_bdev(bdev); - zram_reset_device(zram); - - mutex_unlock(&bdev->bd_mutex); - revalidate_disk(zram->disk); - bdput(bdev); - - return len; - -out: - mutex_unlock(&bdev->bd_mutex); - bdput(bdev); return ret; } @@ -932,7 +853,7 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) if (unlikely(bio->bi_rw & REQ_DISCARD)) { zram_bio_discard(zram, index, offset, bio); - bio_endio(bio, 0); + bio_endio(bio); return; } @@ -965,8 +886,7 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) update_position(&index, &offset, &bvec); } - set_bit(BIO_UPTODATE, &bio->bi_flags); - bio_endio(bio, 0); + bio_endio(bio); return; out: @@ -976,13 +896,15 @@ out: /* * Handler function for all zram I/O requests. */ -static void zram_make_request(struct request_queue *queue, struct bio *bio) +static blk_qc_t zram_make_request(struct request_queue *queue, struct bio *bio) { struct zram *zram = queue->queuedata; if (unlikely(!zram_meta_get(zram))) goto error; + blk_queue_split(queue, &bio, queue->bio_split); + if (!valid_io_request(zram, bio->bi_iter.bi_sector, bio->bi_iter.bi_size)) { atomic64_inc(&zram->stats.invalid_io); @@ -991,11 +913,12 @@ static void zram_make_request(struct request_queue *queue, struct bio *bio) __zram_make_request(zram, bio); zram_meta_put(zram); - return; + return BLK_QC_T_NONE; put_zram: zram_meta_put(zram); error: bio_io_error(bio); + return BLK_QC_T_NONE; } static void zram_slot_free_notify(struct block_device *bdev, @@ -1007,9 +930,9 @@ static void zram_slot_free_notify(struct block_device *bdev, zram = bdev->bd_disk->private_data; meta = zram->meta; - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value); + zram_lock_table(&meta->table[index]); zram_free_page(zram, index); - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value); + zram_unlock_table(&meta->table[index]); atomic64_inc(&zram->stats.notify_free); } @@ -1055,80 +978,185 @@ out: return err; } -static const struct block_device_operations zram_devops = { - .swap_slot_free_notify = zram_slot_free_notify, - .rw_page = zram_rw_page, - .owner = THIS_MODULE -}; +static void zram_reset_device(struct zram *zram) +{ + struct zram_meta *meta; + struct zcomp *comp; + u64 disksize; -static DEVICE_ATTR_WO(compact); -static DEVICE_ATTR_RW(disksize); -static DEVICE_ATTR_RO(initstate); -static DEVICE_ATTR_WO(reset); -static DEVICE_ATTR_RO(orig_data_size); -static DEVICE_ATTR_RO(mem_used_total); -static DEVICE_ATTR_RW(mem_limit); -static DEVICE_ATTR_RW(mem_used_max); -static DEVICE_ATTR_RW(max_comp_streams); -static DEVICE_ATTR_RW(comp_algorithm); + down_write(&zram->init_lock); -static ssize_t io_stat_show(struct device *dev, - struct device_attribute *attr, char *buf) + zram->limit_pages = 0; + + if (!init_done(zram)) { + up_write(&zram->init_lock); + return; + } + + meta = zram->meta; + comp = zram->comp; + disksize = zram->disksize; + /* + * Refcount will go down to 0 eventually and r/w handler + * cannot handle further I/O so it will bail out by + * check zram_meta_get. + */ + zram_meta_put(zram); + /* + * We want to free zram_meta in process context to avoid + * deadlock between reclaim path and any other locks. + */ + wait_event(zram->io_done, atomic_read(&zram->refcount) == 0); + + /* Reset stats */ + memset(&zram->stats, 0, sizeof(zram->stats)); + zram->disksize = 0; + zram->max_comp_streams = 1; + + set_capacity(zram->disk, 0); + part_stat_set_all(&zram->disk->part0, 0); + + up_write(&zram->init_lock); + /* I/O operation under all of CPU are done so let's free */ + zram_meta_free(meta, disksize); + zcomp_destroy(comp); +} + +static ssize_t disksize_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) { + u64 disksize; + struct zcomp *comp; + struct zram_meta *meta; struct zram *zram = dev_to_zram(dev); - ssize_t ret; + int err; - down_read(&zram->init_lock); - ret = scnprintf(buf, PAGE_SIZE, - "%8llu %8llu %8llu %8llu\n", - (u64)atomic64_read(&zram->stats.failed_reads), - (u64)atomic64_read(&zram->stats.failed_writes), - (u64)atomic64_read(&zram->stats.invalid_io), - (u64)atomic64_read(&zram->stats.notify_free)); - up_read(&zram->init_lock); + disksize = memparse(buf, NULL); + if (!disksize) + return -EINVAL; - return ret; + disksize = PAGE_ALIGN(disksize); + meta = zram_meta_alloc(zram->disk->disk_name, disksize); + if (!meta) + return -ENOMEM; + + comp = zcomp_create(zram->compressor, zram->max_comp_streams); + if (IS_ERR(comp)) { + pr_err("Cannot initialise %s compressing backend\n", + zram->compressor); + err = PTR_ERR(comp); + goto out_free_meta; + } + + down_write(&zram->init_lock); + if (init_done(zram)) { + pr_info("Cannot change disksize for initialized device\n"); + err = -EBUSY; + goto out_destroy_comp; + } + + init_waitqueue_head(&zram->io_done); + atomic_set(&zram->refcount, 1); + zram->meta = meta; + zram->comp = comp; + zram->disksize = disksize; + set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT); + up_write(&zram->init_lock); + + /* + * Revalidate disk out of the init_lock to avoid lockdep splat. + * It's okay because disk's capacity is protected by init_lock + * so that revalidate_disk always sees up-to-date capacity. + */ + revalidate_disk(zram->disk); + + return len; + +out_destroy_comp: + up_write(&zram->init_lock); + zcomp_destroy(comp); +out_free_meta: + zram_meta_free(meta, disksize); + return err; } -static ssize_t mm_stat_show(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t reset_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) { - struct zram *zram = dev_to_zram(dev); - u64 orig_size, mem_used = 0; - long max_used; - ssize_t ret; + int ret; + unsigned short do_reset; + struct zram *zram; + struct block_device *bdev; - down_read(&zram->init_lock); - if (init_done(zram)) - mem_used = zs_get_total_pages(zram->meta->mem_pool); + ret = kstrtou16(buf, 10, &do_reset); + if (ret) + return ret; - orig_size = atomic64_read(&zram->stats.pages_stored); - max_used = atomic_long_read(&zram->stats.max_used_pages); + if (!do_reset) + return -EINVAL; - ret = scnprintf(buf, PAGE_SIZE, - "%8llu %8llu %8llu %8lu %8ld %8llu %8llu\n", - orig_size << PAGE_SHIFT, - (u64)atomic64_read(&zram->stats.compr_data_size), - mem_used << PAGE_SHIFT, - zram->limit_pages << PAGE_SHIFT, - max_used << PAGE_SHIFT, - (u64)atomic64_read(&zram->stats.zero_pages), - (u64)atomic64_read(&zram->stats.num_migrated)); - up_read(&zram->init_lock); + zram = dev_to_zram(dev); + bdev = bdget_disk(zram->disk, 0); + if (!bdev) + return -ENOMEM; + + mutex_lock(&bdev->bd_mutex); + /* Do not reset an active device or claimed device */ + if (bdev->bd_openers || zram->claim) { + mutex_unlock(&bdev->bd_mutex); + bdput(bdev); + return -EBUSY; + } + + /* From now on, anyone can't open /dev/zram[0-9] */ + zram->claim = true; + mutex_unlock(&bdev->bd_mutex); + + /* Make sure all the pending I/O are finished */ + fsync_bdev(bdev); + zram_reset_device(zram); + revalidate_disk(zram->disk); + bdput(bdev); + + mutex_lock(&bdev->bd_mutex); + zram->claim = false; + mutex_unlock(&bdev->bd_mutex); + + return len; +} + +static int zram_open(struct block_device *bdev, fmode_t mode) +{ + int ret = 0; + struct zram *zram; + + WARN_ON(!mutex_is_locked(&bdev->bd_mutex)); + + zram = bdev->bd_disk->private_data; + /* zram was claimed to reset so open request fails */ + if (zram->claim) + ret = -EBUSY; return ret; } -static DEVICE_ATTR_RO(io_stat); -static DEVICE_ATTR_RO(mm_stat); -ZRAM_ATTR_RO(num_reads); -ZRAM_ATTR_RO(num_writes); -ZRAM_ATTR_RO(failed_reads); -ZRAM_ATTR_RO(failed_writes); -ZRAM_ATTR_RO(invalid_io); -ZRAM_ATTR_RO(notify_free); -ZRAM_ATTR_RO(zero_pages); -ZRAM_ATTR_RO(compr_data_size); +static const struct block_device_operations zram_devops = { + .open = zram_open, + .swap_slot_free_notify = zram_slot_free_notify, + .rw_page = zram_rw_page, + .owner = THIS_MODULE +}; + +static DEVICE_ATTR_WO(compact); +static DEVICE_ATTR_RW(disksize); +static DEVICE_ATTR_RO(initstate); +static DEVICE_ATTR_WO(reset); +static DEVICE_ATTR_RO(orig_data_size); +static DEVICE_ATTR_RO(mem_used_total); +static DEVICE_ATTR_RW(mem_limit); +static DEVICE_ATTR_RW(mem_used_max); +static DEVICE_ATTR_RW(max_comp_streams); +static DEVICE_ATTR_RW(comp_algorithm); static struct attribute *zram_disk_attrs[] = { &dev_attr_disksize.attr, @@ -1158,10 +1186,24 @@ static struct attribute_group zram_disk_attr_group = { .attrs = zram_disk_attrs, }; -static int create_device(struct zram *zram, int device_id) +/* + * Allocate and initialize new zram device. the function returns + * '>= 0' device_id upon success, and negative value otherwise. + */ +static int zram_add(void) { + struct zram *zram; struct request_queue *queue; - int ret = -ENOMEM; + int ret, device_id; + + zram = kzalloc(sizeof(struct zram), GFP_KERNEL); + if (!zram) + return -ENOMEM; + + ret = idr_alloc(&zram_index_idr, zram, 0, 0, GFP_KERNEL); + if (ret < 0) + goto out_free_dev; + device_id = ret; init_rwsem(&zram->init_lock); @@ -1169,15 +1211,16 @@ static int create_device(struct zram *zram, int device_id) if (!queue) { pr_err("Error allocating disk queue for device %d\n", device_id); - goto out; + ret = -ENOMEM; + goto out_free_idr; } blk_queue_make_request(queue, zram_make_request); - /* gendisk structure */ + /* gendisk structure */ zram->disk = alloc_disk(1); if (!zram->disk) { - pr_warn("Error allocating disk structure for device %d\n", + pr_err("Error allocating disk structure for device %d\n", device_id); ret = -ENOMEM; goto out_free_queue; @@ -1206,7 +1249,7 @@ static int create_device(struct zram *zram, int device_id) blk_queue_io_min(zram->disk->queue, PAGE_SIZE); blk_queue_io_opt(zram->disk->queue, PAGE_SIZE); zram->disk->queue->limits.discard_granularity = PAGE_SIZE; - zram->disk->queue->limits.max_discard_sectors = UINT_MAX; + blk_queue_max_discard_sectors(zram->disk->queue, UINT_MAX); /* * zram_bio_discard() will clear all logical blocks if logical block * size is identical with physical block size(PAGE_SIZE). But if it is @@ -1226,96 +1269,185 @@ static int create_device(struct zram *zram, int device_id) ret = sysfs_create_group(&disk_to_dev(zram->disk)->kobj, &zram_disk_attr_group); if (ret < 0) { - pr_warn("Error creating sysfs group"); + pr_err("Error creating sysfs group for device %d\n", + device_id); goto out_free_disk; } strlcpy(zram->compressor, default_compressor, sizeof(zram->compressor)); zram->meta = NULL; zram->max_comp_streams = 1; - return 0; + + pr_info("Added device: %s\n", zram->disk->disk_name); + return device_id; out_free_disk: del_gendisk(zram->disk); put_disk(zram->disk); out_free_queue: blk_cleanup_queue(queue); -out: +out_free_idr: + idr_remove(&zram_index_idr, device_id); +out_free_dev: + kfree(zram); return ret; } -static void destroy_devices(unsigned int nr) +static int zram_remove(struct zram *zram) +{ + struct block_device *bdev; + + bdev = bdget_disk(zram->disk, 0); + if (!bdev) + return -ENOMEM; + + mutex_lock(&bdev->bd_mutex); + if (bdev->bd_openers || zram->claim) { + mutex_unlock(&bdev->bd_mutex); + bdput(bdev); + return -EBUSY; + } + + zram->claim = true; + mutex_unlock(&bdev->bd_mutex); + + /* + * Remove sysfs first, so no one will perform a disksize + * store while we destroy the devices. This also helps during + * hot_remove -- zram_reset_device() is the last holder of + * ->init_lock, no later/concurrent disksize_store() or any + * other sysfs handlers are possible. + */ + sysfs_remove_group(&disk_to_dev(zram->disk)->kobj, + &zram_disk_attr_group); + + /* Make sure all the pending I/O are finished */ + fsync_bdev(bdev); + zram_reset_device(zram); + bdput(bdev); + + pr_info("Removed device: %s\n", zram->disk->disk_name); + + blk_cleanup_queue(zram->disk->queue); + del_gendisk(zram->disk); + put_disk(zram->disk); + kfree(zram); + return 0; +} + +/* zram-control sysfs attributes */ +static ssize_t hot_add_show(struct class *class, + struct class_attribute *attr, + char *buf) +{ + int ret; + + mutex_lock(&zram_index_mutex); + ret = zram_add(); + mutex_unlock(&zram_index_mutex); + + if (ret < 0) + return ret; + return scnprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t hot_remove_store(struct class *class, + struct class_attribute *attr, + const char *buf, + size_t count) { struct zram *zram; - unsigned int i; + int ret, dev_id; - for (i = 0; i < nr; i++) { - zram = &zram_devices[i]; - /* - * Remove sysfs first, so no one will perform a disksize - * store while we destroy the devices - */ - sysfs_remove_group(&disk_to_dev(zram->disk)->kobj, - &zram_disk_attr_group); + /* dev_id is gendisk->first_minor, which is `int' */ + ret = kstrtoint(buf, 10, &dev_id); + if (ret) + return ret; + if (dev_id < 0) + return -EINVAL; - zram_reset_device(zram); + mutex_lock(&zram_index_mutex); - blk_cleanup_queue(zram->disk->queue); - del_gendisk(zram->disk); - put_disk(zram->disk); + zram = idr_find(&zram_index_idr, dev_id); + if (zram) { + ret = zram_remove(zram); + idr_remove(&zram_index_idr, dev_id); + } else { + ret = -ENODEV; } - kfree(zram_devices); + mutex_unlock(&zram_index_mutex); + return ret ? ret : count; +} + +static struct class_attribute zram_control_class_attrs[] = { + __ATTR_RO(hot_add), + __ATTR_WO(hot_remove), + __ATTR_NULL, +}; + +static struct class zram_control_class = { + .name = "zram-control", + .owner = THIS_MODULE, + .class_attrs = zram_control_class_attrs, +}; + +static int zram_remove_cb(int id, void *ptr, void *data) +{ + zram_remove(ptr); + return 0; +} + +static void destroy_devices(void) +{ + class_unregister(&zram_control_class); + idr_for_each(&zram_index_idr, &zram_remove_cb, NULL); + idr_destroy(&zram_index_idr); unregister_blkdev(zram_major, "zram"); - pr_info("Destroyed %u device(s)\n", nr); } static int __init zram_init(void) { - int ret, dev_id; + int ret; - if (num_devices > max_num_devices) { - pr_warn("Invalid value for num_devices: %u\n", - num_devices); - return -EINVAL; + ret = class_register(&zram_control_class); + if (ret) { + pr_err("Unable to register zram-control class\n"); + return ret; } zram_major = register_blkdev(0, "zram"); if (zram_major <= 0) { - pr_warn("Unable to get major number\n"); + pr_err("Unable to get major number\n"); + class_unregister(&zram_control_class); return -EBUSY; } - /* Allocate the device array and initialize each one */ - zram_devices = kzalloc(num_devices * sizeof(struct zram), GFP_KERNEL); - if (!zram_devices) { - unregister_blkdev(zram_major, "zram"); - return -ENOMEM; - } - - for (dev_id = 0; dev_id < num_devices; dev_id++) { - ret = create_device(&zram_devices[dev_id], dev_id); - if (ret) + while (num_devices != 0) { + mutex_lock(&zram_index_mutex); + ret = zram_add(); + mutex_unlock(&zram_index_mutex); + if (ret < 0) goto out_error; + num_devices--; } - pr_info("Created %u device(s)\n", num_devices); return 0; out_error: - destroy_devices(dev_id); + destroy_devices(); return ret; } static void __exit zram_exit(void) { - destroy_devices(num_devices); + destroy_devices(); } module_init(zram_init); module_exit(zram_exit); module_param(num_devices, uint, 0); -MODULE_PARM_DESC(num_devices, "Number of zram devices"); +MODULE_PARM_DESC(num_devices, "Number of pre-created zram devices"); MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Nitin Gupta <ngupta@vflare.org>"); |