linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Fabio M. De Francesco" <fmdefrancesco@gmail.com>
To: Matthew Wilcox <willy@infradead.org>
Cc: linux-fscrypt@vger.kernel.org, Eric Biggers <ebiggers@kernel.org>,
	linux-fsdevel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH] fs-verity: use kmap_local_page() instead of kmap()
Date: Sat, 20 Aug 2022 00:31:31 +0200	[thread overview]
Message-ID: <2255194.ElGaqSPkdT@opensuse> (raw)
In-Reply-To: <Yv/Wi/2IH/bY05zG@casper.infradead.org>

On venerdì 19 agosto 2022 20:29:31 CEST Matthew Wilcox wrote:
> On Fri, Aug 19, 2022 at 09:50:37AM +0200, Fabio M. De Francesco wrote:
> > On venerdì 19 agosto 2022 00:40:10 CEST Eric Biggers wrote:
> > > From: Eric Biggers <ebiggers@google.com>
> > > 
> > > Convert the use of kmap() to its recommended replacement
> > > kmap_local_page().  This avoids the overhead of doing a non-local
> > > mapping, which is unnecessary in this case.
> > > 
> > > Signed-off-by: Eric Biggers <ebiggers@google.com>
> > > ---
> > > 
> > >  fs/verity/read_metadata.c | 6 +++---
> > >  1 file changed, 3 insertions(+), 3 deletions(-)
> > 
> > It looks good to me...
> > 
> > > -		virt = kmap(page);
> > > +		virt = kmap_local_page(page);
> > > 
> > >  		if (copy_to_user(buf, virt + offs_in_page, bytes_to_copy))
> > 
> > {
> > 
> > > -			kunmap(page);
> > > +			kunmap_local(virt);
> > > 
> > >  			put_page(page);
> > >  			err = -EFAULT;
> > >  			break;
> > >  		
> > >  		}
> > > 
> > > -		kunmap(page);
> > > +		kunmap_local(virt);
> 
> Is this a common pattern?  eg do we want something like:
> 
> static inline int copy_user_page(void __user *dst, struct page *page,
> 		size_t offset, size_t len)
> {
> 	char *src = kmap_local_page(page) + offset;
> 	int err = 0;
> 
> 	VM_BUG_ON(offset + len > PAGE_SIZE);
> 	if (copy_to_user(dst, src, len))
> 		err = -EFAULT;
> 
> 	kunmap_local(src);
> 	return err;
> }
> 
> in highmem.h?

Not sure that it is much common...

Can the following command provide any insight?

opensuse:/usr/src/git/kernels/linux> grep -rn copy_to_user -B7 . --exclude-
dir\=Documentation | grep kmap

./drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c-2306-         ptr = 
kmap_local_page(p);
./drivers/gpu/drm/i915/i915_gem.c-215-  vaddr = kmap_local_page(page);
./drivers/gpu/drm/radeon/radeon_ttm.c-872-                      ptr = 
kmap(page);
./drivers/vfio/pci/mlx5/main.c-182-             from_buff = 
kmap_local_page(page);
./arch/parisc/kernel/cache.c-580-       kto = kmap_local_page(to);
./mm/memory.c-5474-                     maddr = kmap(page);
./fs/verity/read_metadata.c-56-         virt = kmap(page);
./fs/aio.c-1252-                ev = kmap_local_page(page);
./fs/exec.c-883-                char *src = kmap_local_page(bprm->page[index]) 
+ offset;

If this command is good to provide any hint, it suggests that having 
copy_to_user() in highmem.h is not probably worth.

Thanks,

Fabio 




  reply	other threads:[~2022-08-19 22:31 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-18 22:40 [PATCH] fs-verity: use kmap_local_page() instead of kmap() Eric Biggers
2022-08-19  7:50 ` Fabio M. De Francesco
2022-08-19 18:29   ` Matthew Wilcox
2022-08-19 22:31     ` Fabio M. De Francesco [this message]
2022-08-19 11:14 ` Fabio M. De Francesco

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=2255194.ElGaqSPkdT@opensuse \
    --to=fmdefrancesco@gmail.com \
    --cc=ebiggers@kernel.org \
    --cc=linux-fscrypt@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).