iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

zram: always expose rw_page

Browse Source 
Currently zram will adjust its fops to a version which does not contain
rw_page when a backing device has been assigned.  This is done to prevent
upper layers from assuming a synchronous operation when a page may have
been written back.  This forces every operation through bio which has
overhead associated with bio_alloc/frees.

The code can be simplified to always expose an rw_page method and only in
the rare event that a page is written back we instead will return
-EOPNOTSUPP forcing the upper layer to fallback to bio.

Link: https://lkml.kernel.org/r/20221003144832.2906610-1-bgeffon@google.com
Signed-off-by: Brian Geffon <bgeffon@google.com>
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Rom Lemarchand <romlem@google.com>
Cc: Suleiman Souhlal <suleiman@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
Brian Geffon 2022-10-03 10:48:32 -04:00 committed by Andrew Morton
parent 14c2ac3681
commit 94541bc3fb
1 changed files with 3 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
drivers/block/zram/zram_drv.c
View File

@ -52,9 +52,6 @@ static unsigned int num_devices = 1;
static size_t huge_class_size;
static const struct block_device_operations zram_devops;
#ifdef CONFIG_ZRAM_WRITEBACK
static const struct block_device_operations zram_wb_devops;
#endif
static void zram_free_page(struct zram *zram, size_t index);
static int zram_bvec_read(struct zram *zram, struct bio_vec *bvec,
@ -546,17 +543,6 @@ static ssize_t backing_dev_store(struct device *dev,
zram->backing_dev = backing_dev;
zram->bitmap = bitmap;
zram->nr_pages = nr_pages;
/*
* With writeback feature, zram does asynchronous IO so it's no longer
* synchronous device so let's remove synchronous io flag. Othewise,
* upper layer(e.g., swap) could wait IO completion rather than
* (submit and return), which will cause system sluggish.
* Furthermore, when the IO function returns(e.g., swap_readpage),
* upper layer expects IO was done so it could deallocate the page
* freely but in fact, IO is going on so finally could cause
* use-after-free when the IO is really done.
*/
zram->disk->fops = &zram_wb_devops;
up_write(&zram->init_lock);
pr_info("setup backing device %s\n", file_name);
@ -1270,6 +1256,9 @@ static int __zram_bvec_read(struct zram *zram, struct page *page, u32 index,
struct bio_vec bvec;
zram_slot_unlock(zram, index);
/* A null bio means rw_page was used, we must fallback to bio */
if (!bio)
return -EOPNOTSUPP;
bvec.bv_page = page;
bvec.bv_len = PAGE_SIZE;
@ -1856,15 +1845,6 @@ static const struct block_device_operations zram_devops = {
.owner = THIS_MODULE
};
#ifdef CONFIG_ZRAM_WRITEBACK
static const struct block_device_operations zram_wb_devops = {
.open = zram_open,
.submit_bio = zram_submit_bio,
.swap_slot_free_notify = zram_slot_free_notify,
.owner = THIS_MODULE
};
#endif
static DEVICE_ATTR_WO(compact);
static DEVICE_ATTR_RW(disksize);
static DEVICE_ATTR_RO(initstate);