On Tue, Jul 4, 2023 at 5:56 PM David Howells <dhowells@xxxxxxxxxx> wrote:
> af_alg_sendmsg() takes data-to-be-copied that's provided by write(),
> send(), sendmsg() and similar into pages that it allocates and will merge
> new data into the last page in the list, based on the value of ctx->merge.
>
> Now that af_alg_sendmsg() accepts MSG_SPLICE_PAGES, it adds spliced pages
> directly into the list and then incorrectly appends data to them if there's
> space left because ctx->merge says that it can. This was cleared by
> af_alg_sendpage(), but that got lost.
>
> Fix this by skipping the merge if MSG_SPLICE_PAGES is specified and
> clearing ctx->merge after MSG_SPLICE_PAGES has added stuff to the list.
>
> Fixes: bf63e250c4b1 ("crypto: af_alg: Support MSG_SPLICE_PAGES")
> Reported-by: Ondrej Mosnáček <omosnacek@xxxxxxxxx>
> Link: https://lore.kernel.org/r/CAAUqJDvFuvms55Td1c=XKv6epfRnnP78438nZQ-JKyuCptGBiQ@xxxxxxxxxxxxxx/
> Signed-off-by: David Howells <dhowells@xxxxxxxxxx>
> cc: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
> cc: Paolo Abeni <pabeni@xxxxxxxxxx>
> cc: "David S. Miller" <davem@xxxxxxxxxxxxx>
> cc: Eric Dumazet <edumazet@xxxxxxxxxx>
> cc: Jakub Kicinski <kuba@xxxxxxxxxx>
> cc: linux-crypto@xxxxxxxxxxxxxxx
> cc: netdev@xxxxxxxxxxxxxxx
> ---
> crypto/af_alg.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
Thanks for the fix! I can confirm that it fixes the reported issue.
There remains some kernel panic on s390x that I hadn't noticed in the
results earlier, but that's probably a different issue. I'll
investigate and send a report/patch when I have more information.
