[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <cdnpycsyf37gbcp6yxx36pxgilothhdpajmtwle5pphjxshn6o@j5enpjtww3xx>
Date: Sun, 29 Dec 2024 13:58:19 +0100
From: Andrey Albershteyn <aalbersh@...hat.com>
To: "Darrick J. Wong" <djwong@...nel.org>
Cc: Mirsad Todorovac <mtodorovac69@...il.com>,
Alex Deucher <alexander.deucher@....com>, Victor Skvortsov <victor.skvortsov@....com>,
amd-gfx@...ts.freedesktop.org, dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org,
Christian König <christian.koenig@....com>, Xinhui Pan <Xinhui.Pan@....com>,
David Airlie <airlied@...il.com>, Simona Vetter <simona@...ll.ch>,
Carlos Maiolino <cem@...nel.org>, Chandan Babu R <chandanbabu@...nel.org>,
Dave Chinner <dchinner@...hat.com>, linux-xfs@...r.kernel.org
Subject: Re: [PATCH v1 2/3] xfs/libxfs: replace kmalloc() and memcpy() with
kmemdup()
On 2024-12-18 16:35:21, Darrick J. Wong wrote:
> On Tue, Dec 17, 2024 at 11:58:12PM +0100, Mirsad Todorovac wrote:
> > The source static analysis tool gave the following advice:
> >
> > ./fs/xfs/libxfs/xfs_dir2.c:382:15-22: WARNING opportunity for kmemdup
> >
> > → 382 args->value = kmalloc(len,
> > 383 GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_RETRY_MAYFAIL);
> > 384 if (!args->value)
> > 385 return -ENOMEM;
> > 386
> > → 387 memcpy(args->value, name, len);
> > 388 args->valuelen = len;
> > 389 return -EEXIST;
> >
> > Replacing kmalloc() + memcpy() with kmemdump() doesn't change semantics.
> > Original code works without fault, so this is not a bug fix but proposed improvement.
>
> I guess this is all right, but seeing as this code is shared with
> userspace ("libxfs"), making this change will just add to the wrappers
> that we have to have:
>
> void *kmemdup_noprof(const void *src, size_t len, gfp_t gfp)
> {
> void *p;
>
> p = kmalloc_node_track_caller_noprof(len, gfp, NUMA_NO_NODE, _RET_IP_);
> if (p)
> memcpy(p, src, len);
> return p;
> }
>
> Is this sufficiently better? That's a question for the kernel
> maintainer (cem) and the userspace maintainer (andrey, now cc'd).
>
> --D
There's still possibility to set wrong length in args->valuelen,
which I suppose what this change tries to prevent.
But otherwise wrapper looks good to me
>
> > Link: https://lwn.net/Articles/198928/
> > Fixes: 94a69db2367ef ("xfs: use __GFP_NOLOCKDEP instead of GFP_NOFS")
> > Fixes: 384f3ced07efd ("[XFS] Return case-insensitive match for dentry cache")
> > Fixes: 2451337dd0439 ("xfs: global error sign conversion")
> > Cc: Carlos Maiolino <cem@...nel.org>
> > Cc: "Darrick J. Wong" <djwong@...nel.org>
> > Cc: Chandan Babu R <chandanbabu@...nel.org>
> > Cc: Dave Chinner <dchinner@...hat.com>
> > Cc: linux-xfs@...r.kernel.org
> > Cc: linux-kernel@...r.kernel.org
> > Signed-off-by: Mirsad Todorovac <mtodorovac69@...il.com>
> > ---
> > v1:
> > initial version.
> >
> > fs/xfs/libxfs/xfs_dir2.c | 3 +--
> > 1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
> > index 202468223bf9..24251e42bdeb 100644
> > --- a/fs/xfs/libxfs/xfs_dir2.c
> > +++ b/fs/xfs/libxfs/xfs_dir2.c
> > @@ -379,12 +379,11 @@ xfs_dir_cilookup_result(
> > !(args->op_flags & XFS_DA_OP_CILOOKUP))
> > return -EEXIST;
> >
> > - args->value = kmalloc(len,
> > + args->value = kmemdup(name, len,
> > GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_RETRY_MAYFAIL);
> > if (!args->value)
> > return -ENOMEM;
> >
> > - memcpy(args->value, name, len);
> > args->valuelen = len;
> > return -EEXIST;
> > }
> > --
> > 2.43.0
> >
> >
>
--
- Andrey
Powered by blists - more mailing lists