From: Jan Harkes <jaharkes@cs.cmu.edu>
To: Kees Cook <keescook@chromium.org>
Cc: linux-kernel@vger.kernel.org, linux-hardening@vger.kernel.org
Subject: Re: [PATCH] fs/coda: Do not use partially allocated struct
Date: Fri, 20 May 2022 13:11:25 -0400 [thread overview]
Message-ID: <20220520171125.7u7q4ih5udkykv6b@cs.cmu.edu> (raw)
In-Reply-To: <20220520165922.2140450-1-keescook@chromium.org>
Looks good to me.
The old code was trying to be too smart for its own good and the type
gets lost by the uc_data = (void *) cast anyway.
Jan
On Fri, May 20, 2022 at 09:59:22AM -0700, Kees Cook wrote:
> GCC 12 does not like seeing a partially allocated structure being used,
> especially when a pointer is being passed out of function scope. Since
> only the struct coda_in_hdr member of union inputArgs is being allocated
> and used, just replace union inputArgs with struct coda_in_hdr.
>
> ../fs/coda/upcall.c: In function 'coda_upcall':
> ../fs/coda/upcall.c:801:22: warning: array subscript 'union inputArgs[0]' is partly outside array bounds of 'unsigned char[20]' [-Warray-bounds]
> 801 | sig_inputArgs->ih.opcode = CODA_SIGNAL;
> | ^~
> In file included from ../include/linux/fs.h:45,
> from ../include/linux/huge_mm.h:8,
> from ../include/linux/mm.h:700,
> from ../fs/coda/upcall.c:22:
> In function 'kvmalloc',
> inlined from 'kvzalloc' at ../include/linux/slab.h:758:9,
> inlined from 'coda_upcall' at ../fs/coda/upcall.c:794:18:
> ../include/linux/slab.h:750:16: note: object of size 20 allocated by 'kvmalloc_node'
> 750 | return kvmalloc_node(size, flags, NUMA_NO_NODE);
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ../fs/coda/upcall.c: In function 'coda_upcall':
> ../fs/coda/upcall.c:802:22: warning: array subscript 'union inputArgs[0]' is partly outside array bounds of 'unsigned char[20]' [-Warray-bounds]
> 802 | sig_inputArgs->ih.unique = req->uc_unique;
> | ^~
> In function 'kvmalloc',
> inlined from 'kvzalloc' at ../include/linux/slab.h:758:9,
> inlined from 'coda_upcall' at ../fs/coda/upcall.c:794:18:
> ../include/linux/slab.h:750:16: note: object of size 20 allocated by 'kvmalloc_node'
> 750 | return kvmalloc_node(size, flags, NUMA_NO_NODE);
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Cc: Jan Harkes <jaharkes@cs.cmu.edu>
> Cc: coda@cs.cmu.edu
> Cc: codalist@coda.cs.cmu.edu
> Signed-off-by: Kees Cook <keescook@chromium.org>
> ---
> fs/coda/upcall.c | 20 ++++++++++----------
> 1 file changed, 10 insertions(+), 10 deletions(-)
>
> diff --git a/fs/coda/upcall.c b/fs/coda/upcall.c
> index 59f6cfd06f96..21e4f5f446b2 100644
> --- a/fs/coda/upcall.c
> +++ b/fs/coda/upcall.c
> @@ -711,7 +711,7 @@ static int coda_upcall(struct venus_comm *vcp,
> union inputArgs *buffer)
> {
> union outputArgs *out;
> - union inputArgs *sig_inputArgs;
> + struct coda_in_hdr *ih;
> struct upc_req *req = NULL, *sig_req;
> int error;
>
> @@ -791,22 +791,22 @@ static int coda_upcall(struct venus_comm *vcp,
> sig_req = kmalloc(sizeof(struct upc_req), GFP_KERNEL);
> if (!sig_req) goto exit;
>
> - sig_inputArgs = kvzalloc(sizeof(struct coda_in_hdr), GFP_KERNEL);
> - if (!sig_inputArgs) {
> + ih = kvzalloc(sizeof(*ih), GFP_KERNEL);
> + if (!ih) {
> kfree(sig_req);
> goto exit;
> }
>
> error = -EINTR;
> - sig_inputArgs->ih.opcode = CODA_SIGNAL;
> - sig_inputArgs->ih.unique = req->uc_unique;
> + ih->opcode = CODA_SIGNAL;
> + ih->unique = req->uc_unique;
>
> sig_req->uc_flags = CODA_REQ_ASYNC;
> - sig_req->uc_opcode = sig_inputArgs->ih.opcode;
> - sig_req->uc_unique = sig_inputArgs->ih.unique;
> - sig_req->uc_data = (void *)sig_inputArgs;
> - sig_req->uc_inSize = sizeof(struct coda_in_hdr);
> - sig_req->uc_outSize = sizeof(struct coda_in_hdr);
> + sig_req->uc_opcode = ih->opcode;
> + sig_req->uc_unique = ih->unique;
> + sig_req->uc_data = (void *)ih;
> + sig_req->uc_inSize = sizeof(*ih);
> + sig_req->uc_outSize = sizeof(*ih);
>
> /* insert at head of queue! */
> list_add(&(sig_req->uc_chain), &vcp->vc_pending);
> --
> 2.32.0
>
>
prev parent reply other threads:[~2022-05-20 17:11 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-20 16:59 [PATCH] fs/coda: Do not use partially allocated struct Kees Cook
2022-05-20 17:11 ` Jan Harkes [this message]
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=20220520171125.7u7q4ih5udkykv6b@cs.cmu.edu \
--to=jaharkes@cs.cmu.edu \
--cc=keescook@chromium.org \
--cc=linux-hardening@vger.kernel.org \
--cc=linux-kernel@vger.kernel.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).