The compilation with CONFIG_WERROR=y is broken:
../hw/mana/device.c:88:6: error: variable 'ret' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
if (!upper_ndev) {
^~~~~~~~~~~
Fix this by assigning the ret to -ENODEV in respective condition.
Fixes: 8b184e4f1c32 ("RDMA/mana_ib: Enable RoCE on port 1")
Signed-off-by: Andy Shevchenko <[email protected]>
---
drivers/infiniband/hw/mana/device.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/infiniband/hw/mana/device.c b/drivers/infiniband/hw/mana/device.c
index fca4d0d85c64..4c45f8681e7f 100644
--- a/drivers/infiniband/hw/mana/device.c
+++ b/drivers/infiniband/hw/mana/device.c
@@ -88,6 +88,7 @@ static int mana_ib_probe(struct auxiliary_device *adev,
if (!upper_ndev) {
rcu_read_unlock();
ibdev_err(&dev->ib_dev, "Failed to get master netdev");
+ ret = -ENODEV;
goto free_ib_device;
}
ether_addr_copy(mac_addr, upper_ndev->dev_addr);
--
2.43.0.rc1.1336.g36b5255a03ac
> Subject: [PATCH v1 1/1] RDMA/mana_ib: Fix compilation error
>
> The compilation with CONFIG_WERROR=y is broken:
>
> .../hw/mana/device.c:88:6: error: variable 'ret' is used uninitialized whenever
> 'if' condition is true [-Werror,-Wsometimes-uninitialized]
> if (!upper_ndev) {
> ^~~~~~~~~~~
>
> Fix this by assigning the ret to -ENODEV in respective condition.
>
> Fixes: 8b184e4f1c32 ("RDMA/mana_ib: Enable RoCE on port 1")
> Signed-off-by: Andy Shevchenko <[email protected]>
Reviewed-by: Long Li <[email protected]>
> ---
> drivers/infiniband/hw/mana/device.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/infiniband/hw/mana/device.c
> b/drivers/infiniband/hw/mana/device.c
> index fca4d0d85c64..4c45f8681e7f 100644
> --- a/drivers/infiniband/hw/mana/device.c
> +++ b/drivers/infiniband/hw/mana/device.c
> @@ -88,6 +88,7 @@ static int mana_ib_probe(struct auxiliary_device *adev,
> if (!upper_ndev) {
> rcu_read_unlock();
> ibdev_err(&dev->ib_dev, "Failed to get master netdev");
> + ret = -ENODEV;
> goto free_ib_device;
> }
> ether_addr_copy(mac_addr, upper_ndev->dev_addr);
> --
> 2.43.0.rc1.1336.g36b5255a03ac
On Tue, Apr 23, 2024 at 11:42:58PM +0300, Andy Shevchenko wrote:
> The compilation with CONFIG_WERROR=y is broken:
>
> .../hw/mana/device.c:88:6: error: variable 'ret' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
> if (!upper_ndev) {
> ^~~~~~~~~~~
>
> Fix this by assigning the ret to -ENODEV in respective condition.
>
> Fixes: 8b184e4f1c32 ("RDMA/mana_ib: Enable RoCE on port 1")
> Signed-off-by: Andy Shevchenko <[email protected]>
> ---
> drivers/infiniband/hw/mana/device.c | 1 +
> 1 file changed, 1 insertion(+)
This was fixed in
commit f88320b698ad099a2f742adfb9f87177bfffe0c5
Author: Konstantin Taranov <[email protected]>
Date: Tue Apr 23 07:15:51 2024 -0700
RDMA/mana_ib: Fix missing ret value
Thanks,
Jason
On Fri, Apr 26, 2024 at 01:37:19PM -0300, Jason Gunthorpe wrote:
> On Tue, Apr 23, 2024 at 11:42:58PM +0300, Andy Shevchenko wrote:
> > The compilation with CONFIG_WERROR=y is broken:
> >
> > .../hw/mana/device.c:88:6: error: variable 'ret' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
> > if (!upper_ndev) {
> > ^~~~~~~~~~~
> >
> > Fix this by assigning the ret to -ENODEV in respective condition.
> >
> > Fixes: 8b184e4f1c32 ("RDMA/mana_ib: Enable RoCE on port 1")
> > Signed-off-by: Andy Shevchenko <[email protected]>
> > ---
> > drivers/infiniband/hw/mana/device.c | 1 +
> > 1 file changed, 1 insertion(+)
>
> This was fixed in
Hmm... The below patch had been sent _after_ mine. What's wrong with mine patch?
> commit f88320b698ad099a2f742adfb9f87177bfffe0c5
> Author: Konstantin Taranov <[email protected]>
> Date: Tue Apr 23 07:15:51 2024 -0700
>
> RDMA/mana_ib: Fix missing ret value
--
With Best Regards,
Andy Shevchenko
On Fri, Apr 26, 2024 at 07:50:41PM +0300, Andy Shevchenko wrote:
> On Fri, Apr 26, 2024 at 01:37:19PM -0300, Jason Gunthorpe wrote:
> > On Tue, Apr 23, 2024 at 11:42:58PM +0300, Andy Shevchenko wrote:
> > > The compilation with CONFIG_WERROR=y is broken:
> > >
> > > .../hw/mana/device.c:88:6: error: variable 'ret' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
> > > if (!upper_ndev) {
> > > ^~~~~~~~~~~
> > >
> > > Fix this by assigning the ret to -ENODEV in respective condition.
> > >
> > > Fixes: 8b184e4f1c32 ("RDMA/mana_ib: Enable RoCE on port 1")
> > > Signed-off-by: Andy Shevchenko <[email protected]>
> > > ---
> > > drivers/infiniband/hw/mana/device.c | 1 +
> > > 1 file changed, 1 insertion(+)
> >
> > This was fixed in
>
> Hmm... The below patch had been sent _after_ mine. What's wrong with mine patch?
Oh, my... Sorry, I missed PM, it was mine sent after that one!
I guess time for weekend.
--
With Best Regards,
Andy Shevchenko
On Fri, Apr 26, 2024 at 07:50:41PM +0300, Andy Shevchenko wrote:
> On Fri, Apr 26, 2024 at 01:37:19PM -0300, Jason Gunthorpe wrote:
> > On Tue, Apr 23, 2024 at 11:42:58PM +0300, Andy Shevchenko wrote:
> > > The compilation with CONFIG_WERROR=y is broken:
> > >
> > > .../hw/mana/device.c:88:6: error: variable 'ret' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
> > > if (!upper_ndev) {
> > > ^~~~~~~~~~~
> > >
> > > Fix this by assigning the ret to -ENODEV in respective condition.
> > >
> > > Fixes: 8b184e4f1c32 ("RDMA/mana_ib: Enable RoCE on port 1")
> > > Signed-off-by: Andy Shevchenko <[email protected]>
> > > ---
> > > drivers/infiniband/hw/mana/device.c | 1 +
> > > 1 file changed, 1 insertion(+)
> >
> > This was fixed in
>
> Hmm... The below patch had been sent _after_ mine. What's wrong with
> mine patch?
It wasn't
Your's:
Date: Tue, 23 Apr 2024 23:42:58 +0300 [thread overview]
Konstantin's:
Date: Tue, 23 Apr 2024 07:15:51 -0700 [thread overview]
He beat you by 6 hours, which is why I took his, yours didn't exist at
the time I took it.
Jason