When nci_rx_work() receives a zero-length payload packet, it should
discard the packet without exiting the loop. Instead, it should continue
processing subsequent packets.
Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet")
Closes: https://lore.kernel.org/lkml/[email protected]/T/
Reported-by: Ryosuke Yasuoka <[email protected]>
Signed-off-by: Ryosuke Yasuoka <[email protected]>
---
net/nfc/nci/core.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c
index b133dc55304c..f2ae8b0d81b9 100644
--- a/net/nfc/nci/core.c
+++ b/net/nfc/nci/core.c
@@ -1518,8 +1518,7 @@ static void nci_rx_work(struct work_struct *work)
if (!nci_plen(skb->data)) {
kfree_skb(skb);
- kcov_remote_stop();
- break;
+ continue;
}
/* Process frame */
--
2.44.0
Hi Yasuoka-san,
On Thu, May 16, 2024 at 12:17:07AM +0900, Ryosuke Yasuoka wrote:
> When nci_rx_work() receives a zero-length payload packet, it should
> discard the packet without exiting the loop. Instead, it should continue
> processing subsequent packets.
nit: I think it would be clearer to say:
.. it should not discard the packet and exit the loop. Instead, ...
>
> Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet")
> Closes: https://lore.kernel.org/lkml/[email protected]/T/
nit: I'm not sure this Closes link is adding much,
there are more changes coming, right?
> Reported-by: Ryosuke Yasuoka <[email protected]>
> Signed-off-by: Ryosuke Yasuoka <[email protected]>
> ---
> net/nfc/nci/core.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c
> index b133dc55304c..f2ae8b0d81b9 100644
> --- a/net/nfc/nci/core.c
> +++ b/net/nfc/nci/core.c
> @@ -1518,8 +1518,7 @@ static void nci_rx_work(struct work_struct *work)
>
> if (!nci_plen(skb->data)) {
> kfree_skb(skb);
> - kcov_remote_stop();
> - break;
> + continue;
> }
>
> /* Process frame */
> --
> 2.44.0
>
>
Thank you for your review and comment, Simon.
On Thu, May 16, 2024 at 09:43:48AM +0100, Simon Horman wrote:
> Hi Yasuoka-san,
>
> On Thu, May 16, 2024 at 12:17:07AM +0900, Ryosuke Yasuoka wrote:
> > When nci_rx_work() receives a zero-length payload packet, it should
> > discard the packet without exiting the loop. Instead, it should continue
> > processing subsequent packets.
>
> nit: I think it would be clearer to say:
>
> ... it should not discard the packet and exit the loop. Instead, ...
Great. I'll update commit msg like this.
> >
> > Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet")
> > Closes: https://lore.kernel.org/lkml/[email protected]/T/
>
> nit: I'm not sure this Closes link is adding much,
> there are more changes coming, right?
No. I just wanna show the URL link as a reference where this bug is
found. This URL discuss a little bit different topic as you know.
In the following discussion [1], Jakub pointed out that changing
continue statement to break is not related to the patch "Fix
uninit-value in nci_rw_work". So I posted this new small patch before
posting v5 patch for "Fix: uninit-value in nci_rw_work".
If Closes tag is not appropriate, I can remove this in this v2 patch.
What do you think?
[1] https://lore.kernel.org/all/[email protected]/
> > Reported-by: Ryosuke Yasuoka <[email protected]>
> > Signed-off-by: Ryosuke Yasuoka <[email protected]>
> > ---
> > net/nfc/nci/core.c | 3 +--
> > 1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c
> > index b133dc55304c..f2ae8b0d81b9 100644
> > --- a/net/nfc/nci/core.c
> > +++ b/net/nfc/nci/core.c
> > @@ -1518,8 +1518,7 @@ static void nci_rx_work(struct work_struct *work)
> >
> > if (!nci_plen(skb->data)) {
> > kfree_skb(skb);
> > - kcov_remote_stop();
> > - break;
> > + continue;
> > }
> >
> > /* Process frame */
> > --
> > 2.44.0
> >
> >
>
Thank you for your help.
Ryosuke
On Thu, May 16, 2024 at 06:24:54PM +0900, Ryosuke Yasuoka wrote:
> Thank you for your review and comment, Simon.
>
> On Thu, May 16, 2024 at 09:43:48AM +0100, Simon Horman wrote:
> > Hi Yasuoka-san,
> >
> > On Thu, May 16, 2024 at 12:17:07AM +0900, Ryosuke Yasuoka wrote:
> > > When nci_rx_work() receives a zero-length payload packet, it should
> > > discard the packet without exiting the loop. Instead, it should continue
> > > processing subsequent packets.
> >
> > nit: I think it would be clearer to say:
> >
> > ... it should not discard the packet and exit the loop. Instead, ...
>
> Great. I'll update commit msg like this.
>
> > >
> > > Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet")
> > > Closes: https://lore.kernel.org/lkml/[email protected]/T/
> >
> > nit: I'm not sure this Closes link is adding much,
> > there are more changes coming, right?
>
> No. I just wanna show the URL link as a reference where this bug is
> found. This URL discuss a little bit different topic as you know.
>
> In the following discussion [1], Jakub pointed out that changing
> continue statement to break is not related to the patch "Fix
> uninit-value in nci_rw_work". So I posted this new small patch before
> posting v5 patch for "Fix: uninit-value in nci_rw_work".
>
> If Closes tag is not appropriate, I can remove this in this v2 patch.
> What do you think?
Thanks, if it was me I would drop the Closes tag.
> [1] https://lore.kernel.org/all/[email protected]/
..
Hi Yasuoka-san,
On Thu, 16 May 2024 18:24:54 +0900, Ryosuke Yasuoka wrote:
> Thank you for your review and comment, Simon.
>
> On Thu, May 16, 2024 at 09:43:48AM +0100, Simon Horman wrote:
>> Hi Yasuoka-san,
>>
>> On Thu, May 16, 2024 at 12:17:07AM +0900, Ryosuke Yasuoka wrote:
>> > When nci_rx_work() receives a zero-length payload packet, it should
>> > discard the packet without exiting the loop. Instead, it should continue
>> > processing subsequent packets.
>>
>> nit: I think it would be clearer to say:
>>
>> ... it should not discard the packet and exit the loop. Instead, ...
>
> Great. I'll update commit msg like this.
>
>> >
>> > Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet")
>> > Closes: https://lore.kernel.org/lkml/[email protected]/T/
>>
>> nit: I'm not sure this Closes link is adding much,
>> there are more changes coming, right?
>
> No. I just wanna show the URL link as a reference where this bug is
> found. This URL discuss a little bit different topic as you know.
>
> In the following discussion [1], Jakub pointed out that changing
> continue statement to break is not related to the patch "Fix
> uninit-value in nci_rw_work". So I posted this new small patch before
> posting v5 patch for "Fix: uninit-value in nci_rw_work".
>
> If Closes tag is not appropriate, I can remove this in this v2 patch.
> What do you think?
I think this patch, continuing the loop after freeing skb, came from
the following discussion:
https://lore.kernel.org/lkml/Zi-vGH1ROjiv1yJ2@zeus/
In such a case, you can use Link: tag with the above URL and mention
it in the change log like "As discussed [1], ...". But this patch
small and clear, so I think the current change log is enough to convey
the intent of the changes without external link.
You may find the following document helpful:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/submitting-patches.rst?h=v6.9#n115
Thanks,
Shigeru
>
> [1] https://lore.kernel.org/all/[email protected]/
>
>> > Reported-by: Ryosuke Yasuoka <[email protected]>
>> > Signed-off-by: Ryosuke Yasuoka <[email protected]>
>> > ---
>> > net/nfc/nci/core.c | 3 +--
>> > 1 file changed, 1 insertion(+), 2 deletions(-)
>> >
>> > diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c
>> > index b133dc55304c..f2ae8b0d81b9 100644
>> > --- a/net/nfc/nci/core.c
>> > +++ b/net/nfc/nci/core.c
>> > @@ -1518,8 +1518,7 @@ static void nci_rx_work(struct work_struct *work)
>> >
>> > if (!nci_plen(skb->data)) {
>> > kfree_skb(skb);
>> > - kcov_remote_stop();
>> > - break;
>> > + continue;
>> > }
>> >
>> > /* Process frame */
>> > --
>> > 2.44.0
>> >
>> >
>>
>
> Thank you for your help.
> Ryosuke
>
>