search for: vmscan

...lloon_page(newpage)) || PageMovable(newpage) by patch 02/16? And this patch should be just removing the balloon-specific check? Otherwise it seems like between patches 02 and 04, other kinds of PageMovable pages were unnecessarily/wrongly routed through putback_lru_page()? > diff --git a/mm/vmscan.c b/mm/vmscan.c > index d82196244340..c7696a2e11c7 100644 > --- a/mm/vmscan.c > +++ b/mm/vmscan.c > @@ -1254,7 +1254,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone, > > list_for_each_entry_safe(page, next, page_list, lru) { > if (page_is_file_cache(pa...

...lloon_page(newpage)) || PageMovable(newpage) by patch 02/16? And this patch should be just removing the balloon-specific check? Otherwise it seems like between patches 02 and 04, other kinds of PageMovable pages were unnecessarily/wrongly routed through putback_lru_page()? > diff --git a/mm/vmscan.c b/mm/vmscan.c > index d82196244340..c7696a2e11c7 100644 > --- a/mm/vmscan.c > +++ b/mm/vmscan.c > @@ -1254,7 +1254,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone, > > list_for_each_entry_safe(page, next, page_list, lru) { > if (page_is_file_cache(pa...

...shrink_control data structure already and priority could be added there. > There was a report that this results in undesired side effects when > inflating the balloon to shrink the page cache. [1] > "When inflating the balloon against page cache (i.e. no free memory > remains) vmscan.c will both shrink page cache, but also invoke the > shrinkers -- including the balloon's shrinker. So the balloon > driver allocates memory which requires reclaim, vmscan gets this > memory by shrinking the balloon, and then the driver adds the > memory back to the balloon....

...shrink_control data structure already and priority could be added there. > There was a report that this results in undesired side effects when > inflating the balloon to shrink the page cache. [1] > "When inflating the balloon against page cache (i.e. no free memory > remains) vmscan.c will both shrink page cache, but also invoke the > shrinkers -- including the balloon's shrinker. So the balloon > driver allocates memory which requires reclaim, vmscan gets this > memory by shrinking the balloon, and then the driver adds the > memory back to the balloon....

...ave not been freed yet. A call to get_page_unless_zero can then be used to get a proper page reference. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> --- include/linux/page-states.h | 4 +++ mm/page_alloc.c | 56 ++++++++++++++++++++++++++++++++++++++++++++ mm/vmscan.c | 3 ++ 3 files changed, 63 insertions(+) diff -urpN linux-2.6/include/linux/page-states.h linux-2.6-patched/include/linux/page-states.h --- linux-2.6/include/linux/page-states.h 2006-09-01 12:50:25.000000000 +0200 +++ linux-2.6-patched/include/linux/page-states.h 2006-09-01 1...

...ave not been freed yet. A call to get_page_unless_zero can then be used to get a proper page reference. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> --- include/linux/page-states.h | 4 +++ mm/page_alloc.c | 56 ++++++++++++++++++++++++++++++++++++++++++++ mm/vmscan.c | 3 ++ 3 files changed, 63 insertions(+) diff -urpN linux-2.6/include/linux/page-states.h linux-2.6-patched/include/linux/page-states.h --- linux-2.6/include/linux/page-states.h 2006-09-01 12:50:25.000000000 +0200 +++ linux-2.6-patched/include/linux/page-states.h 2006-09-01 1...

...tch ext3-2.4-0.9.14-2414p5 fails: patching file include/linux/sched.h Hunk #2 succeeded at 489 (offset 1 line). patching file include/linux/fs.h patching file include/linux/capability.h patching file fs/buffer.c patching file mm/filemap.c Hunk #1 succeeded at 209 (offset 4 lines). patching file mm/vmscan.c Hunk #2 FAILED at 114. Hunk #3 succeeded at 149 (offset -3 lines). Hunk #4 succeeded at 429 (offset -3 lines). Hunk #5 succeeded at 450 (offset -3 lines). 1 out of 5 hunks FAILED -- saving rejects to file mm/vmscan.c.rej patching file drivers/block/loop.c patching file fs/Makefile patching file f...

...hat the >>>>>> intended >>>>>>>> device implementation is. >>>>>>>> >>>>>>>> When inflating the balloon against page cache (i.e. no free >>>>> memory >>>>>>>> remains) vmscan.c will both shrink page cache, but also invoke >>>>> the >>>>>>>> shrinkers -- including the balloon's shrinker. So the balloon >>>>> driver >>>>>>>> allocates memory which requires reclaim, vmscan gets this memor...

...hat the >>>>>> intended >>>>>>>> device implementation is. >>>>>>>> >>>>>>>> When inflating the balloon against page cache (i.e. no free >>>>> memory >>>>>>>> remains) vmscan.c will both shrink page cache, but also invoke >>>>> the >>>>>>>> shrinkers -- including the balloon's shrinker. So the balloon >>>>> driver >>>>>>>> allocates memory which requires reclaim, vmscan gets this memor...

...t the >>> > intended >>> > > > device implementation is. >>> > > > >>> > > > When inflating the balloon against page cache (i.e. no free >>> memory >>> > > > remains) vmscan.c will both shrink page cache, but also invoke >>> the >>> > > > shrinkers -- including the balloon's shrinker. So the balloon >>> driver >>> > > > allocates memory which requires reclaim, vmscan gets this memory >...

...t the >>> > intended >>> > > > device implementation is. >>> > > > >>> > > > When inflating the balloon against page cache (i.e. no free >>> memory >>> > > > remains) vmscan.c will both shrink page cache, but also invoke >>> the >>> > > > shrinkers -- including the balloon's shrinker. So the balloon >>> driver >>> > > > allocates memory which requires reclaim, vmscan gets this memory >...

...a concept of > priorities, so this behavior cannot be configured. > > There was a report that this results in undesired side effects when > inflating the balloon to shrink the page cache. [1] > "When inflating the balloon against page cache (i.e. no free memory > remains) vmscan.c will both shrink page cache, but also invoke the > shrinkers -- including the balloon's shrinker. So the balloon > driver allocates memory which requires reclaim, vmscan gets this > memory by shrinking the balloon, and then the driver adds the > memory back to the balloon....

...a concept of > priorities, so this behavior cannot be configured. > > There was a report that this results in undesired side effects when > inflating the balloon to shrink the page cache. [1] > "When inflating the balloon against page cache (i.e. no free memory > remains) vmscan.c will both shrink page cache, but also invoke the > shrinkers -- including the balloon's shrinker. So the balloon > driver allocates memory which requires reclaim, vmscan gets this > memory by shrinking the balloon, and then the driver adds the > memory back to the balloon....

...avid at redhat.com> wrote: > >> There was a report that this results in undesired side effects when > >> inflating the balloon to shrink the page cache. [1] > >> "When inflating the balloon against page cache (i.e. no free memory > >> remains) vmscan.c will both shrink page cache, but also invoke the > >> shrinkers -- including the balloon's shrinker. So the balloon > >> driver allocates memory which requires reclaim, vmscan gets this > >> memory by shrinking the balloon, and then the driver adds...

...++++++++++++-- include/linux/balloon_compaction.h | 51 ++++++--------------- include/uapi/linux/magic.h | 1 + mm/balloon_compaction.c | 94 +++++++------------------------------- mm/compaction.c | 7 --- mm/migrate.c | 19 +------- mm/vmscan.c | 2 +- 7 files changed, 83 insertions(+), 143 deletions(-) diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c index 7b6d74f0c72f..04fc63b4a735 100644 --- a/drivers/virtio/virtio_balloon.c +++ b/drivers/virtio/virtio_balloon.c @@ -30,6 +30,7 @@...

...switch back after doing proper >> testing. >> >> There was a report that this results in undesired side effects when >> inflating the balloon to shrink the page cache. [1] >> "When inflating the balloon against page cache (i.e. no free memory >> remains) vmscan.c will both shrink page cache, but also invoke the >> shrinkers -- including the balloon's shrinker. So the balloon >> driver allocates memory which requires reclaim, vmscan gets this >> memory by shrinking the balloon, and then the driver adds the >> memory back...

...switch back after doing proper >> testing. >> >> There was a report that this results in undesired side effects when >> inflating the balloon to shrink the page cache. [1] >> "When inflating the balloon against page cache (i.e. no free memory >> remains) vmscan.c will both shrink page cache, but also invoke the >> shrinkers -- including the balloon's shrinker. So the balloon >> driver allocates memory which requires reclaim, vmscan gets this >> memory by shrinking the balloon, and then the driver adds the >> memory back...

...+++++++++++--- include/linux/balloon_compaction.h | 53 +++++++-------------- include/uapi/linux/magic.h | 1 + mm/balloon_compaction.c | 94 +++++++------------------------------- mm/compaction.c | 7 --- mm/migrate.c | 19 +------- mm/vmscan.c | 2 +- 7 files changed, 85 insertions(+), 145 deletions(-) diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c index 476c0e3a7150..88d5609375de 100644 --- a/drivers/virtio/virtio_balloon.c +++ b/drivers/virtio/virtio_balloon.c @@ -30,6 +30,7 @@...

...t; >>>>>>>>> device implementation is. > >>>>>>>>> > >>>>>>>>> When inflating the balloon against page cache (i.e. no free > >>>>>> memory > >>>>>>>>> remains) vmscan.c will both shrink page cache, but also invoke > >>>>>> the > >>>>>>>>> shrinkers -- including the balloon's shrinker. So the balloon > >>>>>> driver > >>>>>>>>> allocates memory which requ...

Two fixes for issues I stumbled over while working on patch #3. Switch back to the good ol' OOM handler for VIRTIO_BALLOON_F_DEFLATE_ON_OOM as the switch to the shrinker introduce some undesired side effects. Keep the shrinker in place to handle VIRTIO_BALLOON_F_FREE_PAGE_HINT. Lengthy discussion under [1]. I tested with QEMU and "deflate-on-oom=on". Works as expected. Did not test