Re: [PATCH] mm: honor FGP_NOWAIT for page cache page allocation

[Jens Axboe <axboe@kernel.dk>]

> On Jul 1, 2022, at 6:56 PM, Shakeel Butt <shakeelb@google.com> wrote:
> 
> On Fri, Jul 01, 2022 at 02:08:26PM -0600, Jens Axboe wrote:
>> If we're creating a page cache page with FGP_CREAT but FGP_NOWAIT is
>> set, we should dial back the gfp flags to avoid frivolous blocking
>> which is trivial to hit in low memory conditions:
>> 
>> [   10.117661]  __schedule+0x8c/0x550
>> [   10.118305]  schedule+0x58/0xa0
>> [   10.118897]  schedule_timeout+0x30/0xdc
>> [   10.119610]  __wait_for_common+0x88/0x114
>> [   10.120348]  wait_for_completion+0x1c/0x24
>> [   10.121103]  __flush_work.isra.0+0x16c/0x19c
>> [   10.121896]  flush_work+0xc/0x14
>> [   10.122496]  __drain_all_pages+0x144/0x218
>> [   10.123267]  drain_all_pages+0x10/0x18
>> [   10.123941]  __alloc_pages+0x464/0x9e4
>> [   10.124633]  __folio_alloc+0x18/0x3c
>> [   10.125294]  __filemap_get_folio+0x17c/0x204
>> [   10.126084]  iomap_write_begin+0xf8/0x428
>> [   10.126829]  iomap_file_buffered_write+0x144/0x24c
>> [   10.127710]  xfs_file_buffered_write+0xe8/0x248
>> [   10.128553]  xfs_file_write_iter+0xa8/0x120
>> [   10.129324]  io_write+0x16c/0x38c
>> [   10.129940]  io_issue_sqe+0x70/0x1cc
>> [   10.130617]  io_queue_sqe+0x18/0xfc
>> [   10.131277]  io_submit_sqes+0x5d4/0x600
>> [   10.131946]  __arm64_sys_io_uring_enter+0x224/0x600
>> [   10.132752]  invoke_syscall.constprop.0+0x70/0xc0
>> [   10.133616]  do_el0_svc+0xd0/0x118
>> [   10.134238]  el0_svc+0x78/0xa0
>> 
>> Clear IO, FS, and reclaim flags and mark the allocation as nowarn,
>> as FGP_NOWAIT are expected to be able to handle -EAGAIN failures and
>> retry from a suitable context without NOWAIT set.
>> 
>> Signed-off-by: Jens Axboe <axboe@kernel.dk>
>> ---
>> mm/filemap.c | 4 ++++
>> 1 file changed, 4 insertions(+)
>> 
>> diff --git a/mm/filemap.c b/mm/filemap.c
>> index ffdfbc8b0e3c..9b747d3247c6 100644
>> --- a/mm/filemap.c
>> +++ b/mm/filemap.c
>> @@ -1988,6 +1988,10 @@ struct folio *__filemap_get_folio(struct address_space *mapping, pgoff_t index,
>>            gfp |= __GFP_WRITE;
>>        if (fgp_flags & FGP_NOFS)
>>            gfp &= ~__GFP_FS;
>> +        if (fgp_flags & FGP_NOWAIT) {
>> +            gfp &= ~GFP_KERNEL;
> 
> Do you think the semantic of FGP_NOWAIT should be similar to GFP_NOWAIT?
> That is allow kswapd wakeup.

Yep I think that’d be fine, and probably appropriate.