Re: [PATCH] smb: client: change allocation requirements in DUP_CTX_STR macro

[Steve French]

merged into cifs-2.6.git for-next pending additional testing and review

On Tue, May 19, 2026 at 8:53 PM Fredric Cover
<fredric.cover.lkernel@xxxxxxxxx> wrote:
>
> Hi Steve,
>
> Following up on this patch. This logic should remain sound and still
> merge properly.
> Let me know if you need any modifications.
>
> Thanks,
> Fredric Cover
>
> On Wed, May 13, 2026 at 1:20 PM Fredric Cover
> <fredric.cover.lkernel@xxxxxxxxx> wrote:
> >
> > From: Fredric Cover <fredric.cover.lkernel@xxxxxxxxx>
> >
> > Currently, the macro DUP_CTX_STR allocates new_ctx->field using
> > GFP_ATOMIC. DUP_CTX_STR is only used in smb3_fs_context_dup(), which
> > is never called in an atomic context. Using GFP_ATOMIC puts unnecessary
> > pressure on emergency memory pools.
> >
> > Change GFP_ATOMIC to GFP_KERNEL.
> >
> > Signed-off-by: Fredric Cover <fredric.cover.lkernel@xxxxxxxxx>
> > ---
> >  fs/smb/client/fs_context.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c
> > index b9544eb03..a3ccc06c0 100644
> > --- a/fs/smb/client/fs_context.c
> > +++ b/fs/smb/client/fs_context.c
> > @@ -420,7 +420,7 @@ static int parse_symlink_flavor(struct fs_context *fc, char *value,
> >  #define DUP_CTX_STR(field)                                             \
> >  do {                                                                   \
> >         if (ctx->field) {                                               \
> > -               new_ctx->field = kstrdup(ctx->field, GFP_ATOMIC);       \
> > +               new_ctx->field = kstrdup(ctx->field, GFP_KERNEL);       \
> >                 if (new_ctx->field == NULL) {                           \
> >                         smb3_cleanup_fs_context_contents(new_ctx);      \
> >                         return -ENOMEM;                                 \
> > --
> > 2.53.0
> >
>


-- 
Thanks,

Steve