[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240329094050.2815699-1-senozhatsky@chromium.org>
Date: Fri, 29 Mar 2024 18:39:41 +0900
From: Sergey Senozhatsky <senozhatsky@...omium.org>
To: Andrew Morton <akpm@...ux-foundation.org>,
Minchan Kim <minchan@...nel.org>
Cc: Brian Geffon <bgeffon@...gle.com>,
linux-kernel@...r.kernel.org,
linux-block@...r.kernel.org,
Sergey Senozhatsky <senozhatsky@...omium.org>
Subject: [PATCHv2] zram: add max_pages param to recompression
Introduce "max_pages" param to recompress device attribute
which sets an upper limit on the number of entries (pages)
zram attempts to recompress (in this particular recompression
call). S/W recompression can be quite expensive so limiting
the number of pages recompress touches can be quite helpful.
Signed-off-by: Sergey Senozhatsky <senozhatsky@...omium.org>
---
Documentation/admin-guide/blockdev/zram.rst | 5 ++++
drivers/block/zram/zram_drv.c | 31 +++++++++++++++++++--
2 files changed, 33 insertions(+), 3 deletions(-)
diff --git a/Documentation/admin-guide/blockdev/zram.rst b/Documentation/admin-guide/blockdev/zram.rst
index ee2b0030d416..091e8bb38887 100644
--- a/Documentation/admin-guide/blockdev/zram.rst
+++ b/Documentation/admin-guide/blockdev/zram.rst
@@ -466,6 +466,11 @@ of equal or greater size:::
#recompress idle pages larger than 2000 bytes
echo "type=idle threshold=2000" > /sys/block/zramX/recompress
+It is also possible to limit the number of pages zram re-compression will
+attempt to recompress:::
+
+ echo "type=huge_idle max_pages=42" > /sys/block/zramX/recompress
+
Recompression of idle pages requires memory tracking.
During re-compression for every page, that matches re-compression criteria,
diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
index f0639df6cd18..4cf38f7d3e0a 100644
--- a/drivers/block/zram/zram_drv.c
+++ b/drivers/block/zram/zram_drv.c
@@ -1568,7 +1568,8 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec,
* Corresponding ZRAM slot should be locked.
*/
static int zram_recompress(struct zram *zram, u32 index, struct page *page,
- u32 threshold, u32 prio, u32 prio_max)
+ u64 *num_recomp_pages, u32 threshold, u32 prio,
+ u32 prio_max)
{
struct zcomp_strm *zstrm = NULL;
unsigned long handle_old;
@@ -1645,6 +1646,15 @@ static int zram_recompress(struct zram *zram, u32 index, struct page *page,
if (!zstrm)
return 0;
+ /*
+ * Decrement the limit (if set) on pages we can recompress, even
+ * when current recompression was unsuccessful or did not compress
+ * the page below the threshold, because we still spent resources
+ * on it.
+ */
+ if (*num_recomp_pages)
+ *num_recomp_pages -= 1;
+
if (class_index_new >= class_index_old) {
/*
* Secondary algorithms failed to re-compress the page
@@ -1710,6 +1720,7 @@ static ssize_t recompress_store(struct device *dev,
struct zram *zram = dev_to_zram(dev);
unsigned long nr_pages = zram->disksize >> PAGE_SHIFT;
char *args, *param, *val, *algo = NULL;
+ u64 num_recomp_pages = ULLONG_MAX;
u32 mode = 0, threshold = 0;
unsigned long index;
struct page *page;
@@ -1732,6 +1743,17 @@ static ssize_t recompress_store(struct device *dev,
continue;
}
+ if (!strcmp(param, "max_pages")) {
+ /*
+ * Limit the number of entries (pages) we attempt to
+ * recompress.
+ */
+ ret = kstrtoull(val, 10, &num_recomp_pages);
+ if (ret)
+ return ret;
+ continue;
+ }
+
if (!strcmp(param, "threshold")) {
/*
* We will re-compress only idle objects equal or
@@ -1788,6 +1810,9 @@ static ssize_t recompress_store(struct device *dev,
for (index = 0; index < nr_pages; index++) {
int err = 0;
+ if (!num_recomp_pages)
+ break;
+
zram_slot_lock(zram, index);
if (!zram_allocated(zram, index))
@@ -1807,8 +1832,8 @@ static ssize_t recompress_store(struct device *dev,
zram_test_flag(zram, index, ZRAM_INCOMPRESSIBLE))
goto next;
- err = zram_recompress(zram, index, page, threshold,
- prio, prio_max);
+ err = zram_recompress(zram, index, page, &num_recomp_pages,
+ threshold, prio, prio_max);
next:
zram_slot_unlock(zram, index);
if (err) {
--
2.44.0.478.gd926399ef9-goog
Powered by blists - more mailing lists