Thanks for your reply.
as far as I can tell bcache is caching BtrFS metadata for me, with or
without my patch. Perhaps BtrFS correctly flags it as REQ_PRIO. In my
testing bcache was consistently (and repeatedly) bypassing file contents
just because they were read as a result of a readahead operation, hence
the need for my patch. I also found that, before my patch,
sequential_cutoff and congestion thresholds had close to no effect on
how much would be cached or bypassed.
The problem you linked to seems to be a separate issue where different
filesystems flag their internal IO inconsistently leading to similar yet
different problems.
I did not know what REQ_BACKGROUND did or how it's handled so I made a
seperate switch for it just to be on the safer side.
I did not intend my patch to be a thorough solution to all problems
currently ongoing with the hardcoded IO flag based bypass, just a quick
fix to restore previous performance. Neither do I intend to develop such
a solution as I feel like my knowledge of bcache and the Linux kernel in
general is far too limited to produce a patch of sufficient quality.
Am 2019-02-16 um 0003 schrieb Nix:
> On 15 Feb 2019, Andreas said:
>
>> I created a patch to make the bypasses for readahead and background IO
>> that were added in late 2017 configurable via SysFS switches. Since
>> receiving that original patch in my distro's kernel I noticed
>> performance degradation and found a few people asking about similar
>> symptoms as I noticed online who weren't able to identify the problem.
> Hm.
>
>> diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
>> index 15070412a32e..8028638b348e 100644
>> --- a/drivers/md/bcache/request.c
>> +++ b/drivers/md/bcache/request.c
>> @@ -394,9 +394,13 @@ static bool check_should_bypass(struct cached_dev *dc, struct bio *bio)
>> * Flag for bypass if the IO is for read-ahead or background,
>> * unless the read-ahead request is for metadata (eg, for gfs2).
>> */
>> - if (bio->bi_opf & (REQ_RAHEAD|REQ_BACKGROUND) &&
>> - !(bio->bi_opf & REQ_PRIO))
>> - goto skip;
>> + if (!(bio->bi_opf & REQ_PRIO))
>> + {
>> + if (dc->bypass_readahead_io && (bio->bi_opf & REQ_RAHEAD))
>> + goto skip;
>> + if (dc->bypass_background_io && (bio->bi_opf & REQ_BACKGROUND))
>> + goto skip;
>> + }
> The thing you based this on is buggy: so, as a result, your patch is
> buggy too. You want to apply this atop the patch in
> <https://lkml.org/lkml/2019/2/7/77>, I think: without it, metadata I/O
> will often not be cached at all. I also suspect this is the cause of
> your performance degradation.
>
> FYI, REQ_BACKGROUND is only used for writeback that I can see, so I'm
> not sure that bit of the patch does anything at all, given that
> as I understand it bypassing relates solely to reads.
>
