Re: [RFC PATCH 0/1] Large folios in block buffered IO path

From: Mateusz Guzik
Date: Wed Nov 27 2024 - 07:28:32 EST


On Wed, Nov 27, 2024 at 1:18 PM Bharata B Rao <bharata@xxxxxxx> wrote:
>
> On 27-Nov-24 11:49 AM, Mateusz Guzik wrote:
> > That is to say bare minimum this needs to be benchmarked before/after
> > with the lock removed from the picture, like so:
> >
> > diff --git a/block/fops.c b/block/fops.c
> > index 2d01c9007681..7f9e9e2f9081 100644
> > --- a/block/fops.c
> > +++ b/block/fops.c
> > @@ -534,12 +534,8 @@ const struct address_space_operations def_blk_aops = {
> > static loff_t blkdev_llseek(struct file *file, loff_t offset, int whence)
> > {
> > struct inode *bd_inode = bdev_file_inode(file);
> > - loff_t retval;
> >
> > - inode_lock(bd_inode);
> > - retval = fixed_size_llseek(file, offset, whence, i_size_read(bd_inode));
> > - inode_unlock(bd_inode);
> > - return retval;
> > + return fixed_size_llseek(file, offset, whence, i_size_read(bd_inode));
> > }
> >
> > static int blkdev_fsync(struct file *filp, loff_t start, loff_t end,
> >
> > To be aborted if it blows up (but I don't see why it would).
>
> Thanks for this fix, will try and get back with results.
>

Please make sure to have results just with this change, no messing
with folio sizes so that I have something for the patch submission.

--
Mateusz Guzik <mjguzik gmail.com>