[dpdk-dev] vhost: fix false-positive warning from clang 5
Checks
Commit Message
When compiling with clang extra warning flags, such as used by default with
meson, a warning is given in iotlb.c:
../lib/librte_vhost/iotlb.c:318:6: warning: variable 'socket' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
This is a false positive, as the socket value will be initialized by the
call to get_mempolicy in the case where the NUMA build-time flag is set,
and in cases where it is not set, "if (ret)" will always be true as ret is
initialized to -1 and never changed.
However, this is not immediately obvious, and is perhaps a little fragile,
as it will break if other code using ret is subsequently added above the
call to get_mempolicy by someone unaware of this subtle dependency.
Therefore, we can fix the warning and making the code more robust by
explicitly initializing socket to zero, and moving the extra condition
check on ret into the #ifdef alongside the call to get_mempolicy which sets
ret.
Fixes: d012d1f293f4 ("vhost: add IOTLB helper functions")
CC: Maxime Coquelin <maxime.coquelin@redhat.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
---
lib/librte_vhost/iotlb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
Comments
On 10/11/2017 11:59 AM, Bruce Richardson wrote:
> When compiling with clang extra warning flags, such as used by default with
> meson, a warning is given in iotlb.c:
>
> ../lib/librte_vhost/iotlb.c:318:6: warning: variable 'socket' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
>
> This is a false positive, as the socket value will be initialized by the
> call to get_mempolicy in the case where the NUMA build-time flag is set,
> and in cases where it is not set, "if (ret)" will always be true as ret is
> initialized to -1 and never changed.
>
> However, this is not immediately obvious, and is perhaps a little fragile,
> as it will break if other code using ret is subsequently added above the
> call to get_mempolicy by someone unaware of this subtle dependency.
> Therefore, we can fix the warning and making the code more robust by
> explicitly initializing socket to zero, and moving the extra condition
> check on ret into the #ifdef alongside the call to get_mempolicy which sets
> ret.
>
> Fixes: d012d1f293f4 ("vhost: add IOTLB helper functions")
>
> CC: Maxime Coquelin <maxime.coquelin@redhat.com>
> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
> ---
> lib/librte_vhost/iotlb.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
I agree this is a bit fragile. Thanks for handling this:
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Thanks,
Maxime
11/10/2017 12:12, Maxime Coquelin:
> On 10/11/2017 11:59 AM, Bruce Richardson wrote:
> > When compiling with clang extra warning flags, such as used by default with
> > meson, a warning is given in iotlb.c:
> >
> > ../lib/librte_vhost/iotlb.c:318:6: warning: variable 'socket' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
> >
> > This is a false positive, as the socket value will be initialized by the
> > call to get_mempolicy in the case where the NUMA build-time flag is set,
> > and in cases where it is not set, "if (ret)" will always be true as ret is
> > initialized to -1 and never changed.
> >
> > However, this is not immediately obvious, and is perhaps a little fragile,
> > as it will break if other code using ret is subsequently added above the
> > call to get_mempolicy by someone unaware of this subtle dependency.
> > Therefore, we can fix the warning and making the code more robust by
> > explicitly initializing socket to zero, and moving the extra condition
> > check on ret into the #ifdef alongside the call to get_mempolicy which sets
> > ret.
> >
> > Fixes: d012d1f293f4 ("vhost: add IOTLB helper functions")
> >
> > CC: Maxime Coquelin <maxime.coquelin@redhat.com>
> > Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
> > ---
> > lib/librte_vhost/iotlb.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
>
> I agree this is a bit fragile. Thanks for handling this:
> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
New error with this patch:
lib/librte_vhost/iotlb.c:303:6: error:
unused variable ‘ret’ [-Werror=unused-variable]
On Wed, Oct 11, 2017 at 12:51:59PM +0200, Thomas Monjalon wrote:
> 11/10/2017 12:12, Maxime Coquelin:
> > On 10/11/2017 11:59 AM, Bruce Richardson wrote:
> > > When compiling with clang extra warning flags, such as used by default with
> > > meson, a warning is given in iotlb.c:
> > >
> > > ../lib/librte_vhost/iotlb.c:318:6: warning: variable 'socket' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
> > >
> > > This is a false positive, as the socket value will be initialized by the
> > > call to get_mempolicy in the case where the NUMA build-time flag is set,
> > > and in cases where it is not set, "if (ret)" will always be true as ret is
> > > initialized to -1 and never changed.
> > >
> > > However, this is not immediately obvious, and is perhaps a little fragile,
> > > as it will break if other code using ret is subsequently added above the
> > > call to get_mempolicy by someone unaware of this subtle dependency.
> > > Therefore, we can fix the warning and making the code more robust by
> > > explicitly initializing socket to zero, and moving the extra condition
> > > check on ret into the #ifdef alongside the call to get_mempolicy which sets
> > > ret.
> > >
> > > Fixes: d012d1f293f4 ("vhost: add IOTLB helper functions")
> > >
> > > CC: Maxime Coquelin <maxime.coquelin@redhat.com>
> > > Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
> > > ---
> > > lib/librte_vhost/iotlb.c | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> >
> > I agree this is a bit fragile. Thanks for handling this:
> > Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>
> New error with this patch:
>
> lib/librte_vhost/iotlb.c:303:6: error:
> unused variable ‘ret’ [-Werror=unused-variable]
>
See it now, when the VHOST NUMA option - which is different from the EAL
one, I notice - is disabled. I'll do a v2 and remove the variable
completely as it doesn't seem necessary.
/Bruce
@@ -300,7 +300,7 @@ vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
{
char pool_name[RTE_MEMPOOL_NAMESIZE];
struct vhost_virtqueue *vq = dev->virtqueue[vq_index];
- int ret = -1, socket;
+ int ret = -1, socket = 0;
if (vq->iotlb_pool) {
/*
@@ -314,9 +314,9 @@ vhost_user_iotlb_init(struct virtio_net *dev, int vq_index)
#ifdef RTE_LIBRTE_VHOST_NUMA
ret = get_mempolicy(&socket, NULL, 0, vq, MPOL_F_NODE | MPOL_F_ADDR);
-#endif
if (ret)
socket = 0;
+#endif
rte_rwlock_init(&vq->iotlb_lock);
rte_rwlock_init(&vq->iotlb_pending_lock);