We returns from vmci_transport_dgram_enqueue() with error
if memcpy goes wrong
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
Signed-off-by: Artem Chernyshev <[email protected]>
---
net/vmw_vsock/vmci_transport.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 842c94286d31..7994090e0314 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -1711,7 +1711,8 @@ static int vmci_transport_dgram_enqueue(
if (!dg)
return -ENOMEM;
- memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
+ if (memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len))
+ return -EFAULT;
dg->dst = vmci_make_handle(remote_addr->svm_cid,
remote_addr->svm_port);
--
2.30.3
> On Dec 2, 2022, at 2:58 PM, Artem Chernyshev <[email protected]> wrote:
>
> We returns from vmci_transport_dgram_enqueue() with error
> if memcpy goes wrong
Thanks for the patch.
Nit: could you please update the description? Maybe something like -
vmci_transport_dgram_enqueue() does not check the return value
of memcpy_from_msg(). Return with an error if the memcpy fails.
>
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
>
> Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
> Signed-off-by: Artem Chernyshev <[email protected]>
> ---
> net/vmw_vsock/vmci_transport.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
> index 842c94286d31..7994090e0314 100644
> --- a/net/vmw_vsock/vmci_transport.c
> +++ b/net/vmw_vsock/vmci_transport.c
> @@ -1711,7 +1711,8 @@ static int vmci_transport_dgram_enqueue(
> if (!dg)
> return -ENOMEM;
>
> - memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
> + if (memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len))
Need to free dg using kfree() before returning.
> + return -EFAULT;
>
> dg->dst = vmci_make_handle(remote_addr->svm_cid,
> remote_addr->svm_port);
> --
> 2.30.3
>
Hi,
On Sat, Dec 03, 2022 at 01:17:33AM +0000, Vishnu Dasa wrote:
>
>
> > On Dec 2, 2022, at 2:58 PM, Artem Chernyshev <[email protected]> wrote:
> >
> > We returns from vmci_transport_dgram_enqueue() with error
> > if memcpy goes wrong
>
> Thanks for the patch.
>
> Nit: could you please update the description? Maybe something like -
> vmci_transport_dgram_enqueue() does not check the return value
> of memcpy_from_msg(). Return with an error if the memcpy fails.
>
> >
> > Found by Linux Verification Center (linuxtesting.org) with SVACE.
> >
> > Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
> > Signed-off-by: Artem Chernyshev <[email protected]>
> > ---
> > net/vmw_vsock/vmci_transport.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
> > index 842c94286d31..7994090e0314 100644
> > --- a/net/vmw_vsock/vmci_transport.c
> > +++ b/net/vmw_vsock/vmci_transport.c
> > @@ -1711,7 +1711,8 @@ static int vmci_transport_dgram_enqueue(
> > if (!dg)
> > return -ENOMEM;
> >
> > - memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
> > + if (memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len))
>
> Need to free dg using kfree() before returning.
>
> > + return -EFAULT;
> >
> > dg->dst = vmci_make_handle(remote_addr->svm_cid,
> > remote_addr->svm_port);
> > --
> > 2.30.3
> >
>
Thanks for review. I'll fix flaws in a patch ASAP
Artem
vmci_transport_dgram_enqueue() does not check the return value
of memcpy_from_msg(). Return with an error if the memcpy fails.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
Signed-off-by: Artem Chernyshev <[email protected]>
---
V1->V2 Fix memory leaking and updates for description
net/vmw_vsock/vmci_transport.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 842c94286d31..c94c3deaa09d 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -1711,7 +1711,10 @@ static int vmci_transport_dgram_enqueue(
if (!dg)
return -ENOMEM;
- memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
+ if (memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len)) {
+ kfree(dg);
+ return -EFAULT;
+ }
dg->dst = vmci_make_handle(remote_addr->svm_cid,
remote_addr->svm_port);
--
2.30.3
On Sat, Dec 03, 2022 at 11:33:12AM +0300, Artem Chernyshev wrote:
>vmci_transport_dgram_enqueue() does not check the return value
>of memcpy_from_msg(). Return with an error if the memcpy fails.
>
>Found by Linux Verification Center (linuxtesting.org) with SVACE.
>
>Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
>Signed-off-by: Artem Chernyshev <[email protected]>
>---
>V1->V2 Fix memory leaking and updates for description
>
> net/vmw_vsock/vmci_transport.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
>diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
>index 842c94286d31..c94c3deaa09d 100644
>--- a/net/vmw_vsock/vmci_transport.c
>+++ b/net/vmw_vsock/vmci_transport.c
>@@ -1711,7 +1711,10 @@ static int vmci_transport_dgram_enqueue(
> if (!dg)
> return -ENOMEM;
>
>- memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
>+ if (memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len)) {
>+ kfree(dg);
>+ return -EFAULT;
Since memcpy_from_msg() is a wrapper of copy_from_iter_full() that
simply returns -EFAULT in case of an error, perhaps it would be better
here to return the value of memcpy_from_msg() instead of wiring the
error.
However in the end the behavior is the same, so even if you don't want
to change it I'll leave my R-b:
Reviewed-by: Stefano Garzarella <[email protected]>
Thanks,
Stefano
Hi,
On Mon, Dec 05, 2022 at 10:47:36AM +0100, Stefano Garzarella wrote:
> On Sat, Dec 03, 2022 at 11:33:12AM +0300, Artem Chernyshev wrote:
> > vmci_transport_dgram_enqueue() does not check the return value
> > of memcpy_from_msg(). Return with an error if the memcpy fails.
> >
> > Found by Linux Verification Center (linuxtesting.org) with SVACE.
> >
> > Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
> > Signed-off-by: Artem Chernyshev <[email protected]>
> > ---
> > V1->V2 Fix memory leaking and updates for description
> >
> > net/vmw_vsock/vmci_transport.c | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
> > index 842c94286d31..c94c3deaa09d 100644
> > --- a/net/vmw_vsock/vmci_transport.c
> > +++ b/net/vmw_vsock/vmci_transport.c
> > @@ -1711,7 +1711,10 @@ static int vmci_transport_dgram_enqueue(
> > if (!dg)
> > return -ENOMEM;
> >
> > - memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
> > + if (memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len)) {
> > + kfree(dg);
> > + return -EFAULT;
>
> Since memcpy_from_msg() is a wrapper of copy_from_iter_full() that simply
> returns -EFAULT in case of an error, perhaps it would be better here to
> return the value of memcpy_from_msg() instead of wiring the error.
>
> However in the end the behavior is the same, so even if you don't want to
> change it I'll leave my R-b:
>
> Reviewed-by: Stefano Garzarella <[email protected]>
>
> Thanks,
> Stefano
Thank you for review. Sure, I will change that in V3
Artem
vmci_transport_dgram_enqueue() does not check the return value
of memcpy_from_msg(). Return with an error if the memcpy fails.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
Signed-off-by: Artem Chernyshev <[email protected]>
---
V1->V2 Fix memory leaking and updates for description
V2->V3 Return the value of memcpy_from_msg()
net/vmw_vsock/vmci_transport.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 842c94286d31..36eb16a40745 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -1711,7 +1711,11 @@ static int vmci_transport_dgram_enqueue(
if (!dg)
return -ENOMEM;
- memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
+ err = memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
+ if (err) {
+ kfree(dg);
+ return err;
+ }
dg->dst = vmci_make_handle(remote_addr->svm_cid,
remote_addr->svm_port);
--
2.30.3
On Mon, Dec 05, 2022 at 02:52:00PM +0300, Artem Chernyshev wrote:
>vmci_transport_dgram_enqueue() does not check the return value
>of memcpy_from_msg(). Return with an error if the memcpy fails.
>
>Found by Linux Verification Center (linuxtesting.org) with SVACE.
>
>Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
>Signed-off-by: Artem Chernyshev <[email protected]>
>---
>V1->V2 Fix memory leaking and updates for description
>V2->V3 Return the value of memcpy_from_msg()
>
> net/vmw_vsock/vmci_transport.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
Reviewed-by: Stefano Garzarella <[email protected]>
> On Dec 5, 2022, at 3:52 AM, Artem Chernyshev <[email protected]> wrote:
>
> vmci_transport_dgram_enqueue() does not check the return value
> of memcpy_from_msg(). Return with an error if the memcpy fails.
I think we can add some more information in the description. Sorry, I
should've said this earlier.
vmci_transport_dgram_enqueue() does not check the return value
of memcpy_from_msg(). If memcpy_from_msg() fails, it is possible that
uninitialized memory contents are sent unintentionally instead of user's
message in the datagram to the destination. Return with an error if
memcpy_from_msg() fails.
>
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
>
> Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
> Signed-off-by: Artem Chernyshev <[email protected]>
Thanks, Artem! This version looks good to me modulo my suggestion
about the description above.
Reviewed-by: Vishnu Dasa <[email protected]>
Regards,
Vishnu
> ---
> V1->V2 Fix memory leaking and updates for description
> V2->V3 Return the value of memcpy_from_msg()
>
> net/vmw_vsock/vmci_transport.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
> index 842c94286d31..36eb16a40745 100644
> --- a/net/vmw_vsock/vmci_transport.c
> +++ b/net/vmw_vsock/vmci_transport.c
> @@ -1711,7 +1711,11 @@ static int vmci_transport_dgram_enqueue(
> if (!dg)
> return -ENOMEM;
>
> - memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
> + err = memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
> + if (err) {
> + kfree(dg);
> + return err;
> + }
>
> dg->dst = vmci_make_handle(remote_addr->svm_cid,
> remote_addr->svm_port);
> --
> 2.30.3
>
Hi,
On Mon, Dec 05, 2022 at 11:03:47PM +0000, Vishnu Dasa wrote:
>
> > On Dec 5, 2022, at 3:52 AM, Artem Chernyshev <[email protected]> wrote:
> >
> > vmci_transport_dgram_enqueue() does not check the return value
> > of memcpy_from_msg(). Return with an error if the memcpy fails.
>
> I think we can add some more information in the description. Sorry, I
> should've said this earlier.
>
> vmci_transport_dgram_enqueue() does not check the return value
> of memcpy_from_msg(). If memcpy_from_msg() fails, it is possible that
> uninitialized memory contents are sent unintentionally instead of user's
> message in the datagram to the destination. Return with an error if
> memcpy_from_msg() fails.
>
> >
> > Found by Linux Verification Center (linuxtesting.org) with SVACE.
> >
> > Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
> > Signed-off-by: Artem Chernyshev <[email protected]>
>
> Thanks, Artem! This version looks good to me modulo my suggestion
> about the description above.
>
> Reviewed-by: Vishnu Dasa <[email protected]>
>
> Regards,
> Vishnu
>
No problem, I'll change description in v4
Thanks,
Artem
vmci_transport_dgram_enqueue() does not check the return value
of memcpy_from_msg(). If memcpy_from_msg() fails, it is possible that
uninitialized memory contents are sent unintentionally instead of user's
message in the datagram to the destination. Return with an error if
memcpy_from_msg() fails.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 0f7db23a07af ("vmci_transport: switch ->enqeue_dgram, ->enqueue_stream and ->dequeue_stream to msghdr")
Signed-off-by: Artem Chernyshev <[email protected]>
Reviewed-by: Stefano Garzarella <[email protected]>
Reviewed-by: Vishnu Dasa <[email protected]>
---
V1->V2 Fix memory leaking and updates for description
V2->V3 Return the value of memcpy_from_msg()
V3->V4 Updates for description
net/vmw_vsock/vmci_transport.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 842c94286d31..36eb16a40745 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -1711,7 +1711,11 @@ static int vmci_transport_dgram_enqueue(
if (!dg)
return -ENOMEM;
- memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
+ err = memcpy_from_msg(VMCI_DG_PAYLOAD(dg), msg, len);
+ if (err) {
+ kfree(dg);
+ return err;
+ }
dg->dst = vmci_make_handle(remote_addr->svm_cid,
remote_addr->svm_port);
--
2.30.3
Hello:
This patch was applied to netdev/net.git (master)
by David S. Miller <[email protected]>:
On Tue, 6 Dec 2022 09:58:34 +0300 you wrote:
> vmci_transport_dgram_enqueue() does not check the return value
> of memcpy_from_msg(). If memcpy_from_msg() fails, it is possible that
> uninitialized memory contents are sent unintentionally instead of user's
> message in the datagram to the destination. Return with an error if
> memcpy_from_msg() fails.
>
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
>
> [...]
Here is the summary with links:
- [v4] net: vmw_vsock: vmci: Check memcpy_from_msg()
https://git.kernel.org/netdev/net/c/44aa5a6dba82
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html