Re: [PATCH] fuse: enable larger read buffers for readdir.

[Bernd Schubert <bernd.schubert@fastmail.fm>]

On 7/26/23 17:26, Jaco Kroon wrote:
> Hi,
> 
> On 2023/07/26 15:53, Bernd Schubert wrote:
>>
>>
>> On 7/26/23 12:59, Jaco Kroon wrote:
>>> Signed-off-by: Jaco Kroon <jaco@uls.co.za>
>>> ---
>>>   fs/fuse/Kconfig   | 16 ++++++++++++++++
>>>   fs/fuse/readdir.c | 42 ++++++++++++++++++++++++------------------
>>>   2 files changed, 40 insertions(+), 18 deletions(-)
>>>
>>> diff --git a/fs/fuse/Kconfig b/fs/fuse/Kconfig
>>> index 038ed0b9aaa5..0783f9ee5cd3 100644
>>> --- a/fs/fuse/Kconfig
>>> +++ b/fs/fuse/Kconfig
>>> @@ -18,6 +18,22 @@ config FUSE_FS
>>>         If you want to develop a userspace FS, or if you want to use
>>>         a filesystem based on FUSE, answer Y or M.
>>>   +config FUSE_READDIR_ORDER
>>> +    int
>>> +    range 0 5
>>> +    default 5
>>> +    help
>>> +        readdir performance varies greatly depending on the size of 
>>> the read.
>>> +        Larger buffers results in larger reads, thus fewer reads and 
>>> higher
>>> +        performance in return.
>>> +
>>> +        You may want to reduce this value on seriously constrained 
>>> memory
>>> +        systems where 128KiB (assuming 4KiB pages) cache pages is 
>>> not ideal.
>>> +
>>> +        This value reprents the order of the number of pages to 
>>> allocate (ie,
>>> +        the shift value).  A value of 0 is thus 1 page (4KiB) where 
>>> 5 is 32
>>> +        pages (128KiB).
>>> +
>>
>> I like the idea of a larger readdir size, but shouldn't that be a 
>> server/daemon/library decision which size to use, instead of kernel 
>> compile time? So should be part of FUSE_INIT negotiation?
> 
> Yes sure, but there still needs to be a default.  And one page at a time 
> doesn't cut it.
> 
> -- snip --
> 
>>>   -    page = alloc_page(GFP_KERNEL);
>>> +    page = alloc_pages(GFP_KERNEL, READDIR_PAGES_ORDER);
>>
>> I guess that should become folio alloc(), one way or the other. Now I 
>> think order 0 was chosen before to avoid risk of allocation failure. I 
>> guess it might work to try a large size and to fall back to 0 when 
>> that failed. Or fail back to the slower vmalloc.
> 
> If this varies then a bunch of other code will become somewhat more 
> complex, especially if one alloc succeeds, and then a follow-up succeeds.

Yeah, the better choice is kvmalloc/kvfree which handles it internally.

> 
> I'm not familiar with the differences between the different mechanisms 
> available for allocation.
> 
> -- snip --
> 
>> Thanks,
> My pleasure,
> Jaco