* re: virtio_net: support device stats
@ 2024-10-14 9:39 Colin King (gmail)
2024-10-14 9:47 ` Michael S. Tsirkin
0 siblings, 1 reply; 5+ messages in thread
From: Colin King (gmail) @ 2024-10-14 9:39 UTC (permalink / raw)
To: Xuan Zhuo
Cc: Jason Wang, Paolo Abeni, Michael S. Tsirkin",
netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Hi,
Static analysis on Linux-next has detected a potential issue with the
following commit:
commit 941168f8b40e50518a3bc6ce770a7062a5d99230
Author: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Fri Apr 26 11:39:24 2024 +0800
virtio_net: support device stats
The issue is in function virtnet_stats_ctx_init, in
drivers/net/virtio_net.c as follows:
if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_CVQ) {
queue_type = VIRTNET_Q_TYPE_CQ;
ctx->bitmap[queue_type] |= VIRTIO_NET_STATS_TYPE_CVQ;
ctx->desc_num[queue_type] +=
ARRAY_SIZE(virtnet_stats_cvq_desc);
ctx->size[queue_type] += sizeof(struct
virtio_net_stats_cvq);
}
ctx->bitmap is declared as a u32 however it is being bit-wise or'd with
VIRTIO_NET_STATS_TYPE_CVQ and this is defined as 1 << 32:
include/uapi/linux/virtio_net.h:#define VIRTIO_NET_STATS_TYPE_CVQ
(1ULL << 32)
..and hence the bit-wise or operation won't set any bits in ctx->bitmap
because 1ULL < 32 is too wide for a u32. I suspect ctx->bitmap should be
declared as u64.
Colin
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: virtio_net: support device stats
2024-10-14 9:39 virtio_net: support device stats Colin King (gmail)
@ 2024-10-14 9:47 ` Michael S. Tsirkin
2024-10-14 10:14 ` Colin King (gmail)
2024-10-15 9:55 ` Xuan Zhuo
0 siblings, 2 replies; 5+ messages in thread
From: Michael S. Tsirkin @ 2024-10-14 9:47 UTC (permalink / raw)
To: Colin King (gmail)
Cc: Xuan Zhuo, Jason Wang, Paolo Abeni, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
On Mon, Oct 14, 2024 at 10:39:26AM +0100, Colin King (gmail) wrote:
> Hi,
>
> Static analysis on Linux-next has detected a potential issue with the
> following commit:
>
> commit 941168f8b40e50518a3bc6ce770a7062a5d99230
> Author: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
> Date: Fri Apr 26 11:39:24 2024 +0800
>
> virtio_net: support device stats
>
>
> The issue is in function virtnet_stats_ctx_init, in drivers/net/virtio_net.c
> as follows:
>
> if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_CVQ) {
> queue_type = VIRTNET_Q_TYPE_CQ;
>
> ctx->bitmap[queue_type] |= VIRTIO_NET_STATS_TYPE_CVQ;
> ctx->desc_num[queue_type] +=
> ARRAY_SIZE(virtnet_stats_cvq_desc);
> ctx->size[queue_type] += sizeof(struct
> virtio_net_stats_cvq);
> }
>
>
> ctx->bitmap is declared as a u32 however it is being bit-wise or'd with
> VIRTIO_NET_STATS_TYPE_CVQ and this is defined as 1 << 32:
>
> include/uapi/linux/virtio_net.h:#define VIRTIO_NET_STATS_TYPE_CVQ (1ULL <<
> 32)
>
> ..and hence the bit-wise or operation won't set any bits in ctx->bitmap
> because 1ULL < 32 is too wide for a u32.
Indeed. Xuan Zhuo how did you test this patch?
> I suspect ctx->bitmap should be
> declared as u64.
>
> Colin
>
>
In fact, it is read into a u64:
u64 offset, bitmap;
....
bitmap = ctx->bitmap[queue_type];
we'll have to reorder fields to avoid wasting memory.
Like this I guess:
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Colin, can you confirm pls?
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c6af18948092..ef221429f784 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -4111,12 +4111,12 @@ struct virtnet_stats_ctx {
/* Used to calculate the offset inside the output buffer. */
u32 desc_num[3];
- /* The actual supported stat types. */
- u32 bitmap[3];
-
/* Used to calculate the reply buffer size. */
u32 size[3];
+ /* The actual supported stat types. */
+ u64 bitmap[3];
+
/* Record the output buffer. */
u64 *data;
};
--
MST
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: virtio_net: support device stats
2024-10-14 9:47 ` Michael S. Tsirkin
@ 2024-10-14 10:14 ` Colin King (gmail)
2024-10-18 7:25 ` Jason Wang
2024-10-15 9:55 ` Xuan Zhuo
1 sibling, 1 reply; 5+ messages in thread
From: Colin King (gmail) @ 2024-10-14 10:14 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Xuan Zhuo, Jason Wang, Paolo Abeni, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
On 14/10/2024 10:47, Michael S. Tsirkin wrote:
> On Mon, Oct 14, 2024 at 10:39:26AM +0100, Colin King (gmail) wrote:
>> Hi,
>>
>> Static analysis on Linux-next has detected a potential issue with the
>> following commit:
>>
>> commit 941168f8b40e50518a3bc6ce770a7062a5d99230
>> Author: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
>> Date: Fri Apr 26 11:39:24 2024 +0800
>>
>> virtio_net: support device stats
>>
>>
>> The issue is in function virtnet_stats_ctx_init, in drivers/net/virtio_net.c
>> as follows:
>>
>> if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_CVQ) {
>> queue_type = VIRTNET_Q_TYPE_CQ;
>>
>> ctx->bitmap[queue_type] |= VIRTIO_NET_STATS_TYPE_CVQ;
>> ctx->desc_num[queue_type] +=
>> ARRAY_SIZE(virtnet_stats_cvq_desc);
>> ctx->size[queue_type] += sizeof(struct
>> virtio_net_stats_cvq);
>> }
>>
>>
>> ctx->bitmap is declared as a u32 however it is being bit-wise or'd with
>> VIRTIO_NET_STATS_TYPE_CVQ and this is defined as 1 << 32:
>>
>> include/uapi/linux/virtio_net.h:#define VIRTIO_NET_STATS_TYPE_CVQ (1ULL <<
>> 32)
>>
>> ..and hence the bit-wise or operation won't set any bits in ctx->bitmap
>> because 1ULL < 32 is too wide for a u32.
>
> Indeed. Xuan Zhuo how did you test this patch?
>
>> I suspect ctx->bitmap should be
>> declared as u64.
>>
>> Colin
>>
>>
>
> In fact, it is read into a u64:
>
> u64 offset, bitmap;
> ....
> bitmap = ctx->bitmap[queue_type];
>
> we'll have to reorder fields to avoid wasting memory.
> Like this I guess:
>
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
>
> Colin, can you confirm pls?
Fix looks sane to be, with u64 bitmap[3] struct size field re-ordering
does not seem to make any difference on x86-64 (64 bytes) and i586 (56
bytes) when I compiled with gcc-12, gcc-14 and clang-20.
I can't functionally test this though (not sure how).
Reviewed-by: Colin Ian King <colin.king@gmail.com>
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index c6af18948092..ef221429f784 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -4111,12 +4111,12 @@ struct virtnet_stats_ctx {
> /* Used to calculate the offset inside the output buffer. */
> u32 desc_num[3];
>
> - /* The actual supported stat types. */
> - u32 bitmap[3];
> -
> /* Used to calculate the reply buffer size. */
> u32 size[3];
>
> + /* The actual supported stat types. */
> + u64 bitmap[3];
> +
> /* Record the output buffer. */
> u64 *data;
> };
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: virtio_net: support device stats
2024-10-14 10:14 ` Colin King (gmail)
@ 2024-10-18 7:25 ` Jason Wang
0 siblings, 0 replies; 5+ messages in thread
From: Jason Wang @ 2024-10-18 7:25 UTC (permalink / raw)
To: Colin King (gmail)
Cc: Michael S. Tsirkin, Xuan Zhuo, Paolo Abeni,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org
On Mon, Oct 14, 2024 at 6:15 PM Colin King (gmail)
<colin.i.king@gmail.com> wrote:
>
> On 14/10/2024 10:47, Michael S. Tsirkin wrote:
> > On Mon, Oct 14, 2024 at 10:39:26AM +0100, Colin King (gmail) wrote:
> >> Hi,
> >>
> >> Static analysis on Linux-next has detected a potential issue with the
> >> following commit:
> >>
> >> commit 941168f8b40e50518a3bc6ce770a7062a5d99230
> >> Author: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
> >> Date: Fri Apr 26 11:39:24 2024 +0800
> >>
> >> virtio_net: support device stats
> >>
> >>
> >> The issue is in function virtnet_stats_ctx_init, in drivers/net/virtio_net.c
> >> as follows:
> >>
> >> if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_CVQ) {
> >> queue_type = VIRTNET_Q_TYPE_CQ;
> >>
> >> ctx->bitmap[queue_type] |= VIRTIO_NET_STATS_TYPE_CVQ;
> >> ctx->desc_num[queue_type] +=
> >> ARRAY_SIZE(virtnet_stats_cvq_desc);
> >> ctx->size[queue_type] += sizeof(struct
> >> virtio_net_stats_cvq);
> >> }
> >>
> >>
> >> ctx->bitmap is declared as a u32 however it is being bit-wise or'd with
> >> VIRTIO_NET_STATS_TYPE_CVQ and this is defined as 1 << 32:
> >>
> >> include/uapi/linux/virtio_net.h:#define VIRTIO_NET_STATS_TYPE_CVQ (1ULL <<
> >> 32)
> >>
> >> ..and hence the bit-wise or operation won't set any bits in ctx->bitmap
> >> because 1ULL < 32 is too wide for a u32.
> >
> > Indeed. Xuan Zhuo how did you test this patch?
> >
> >> I suspect ctx->bitmap should be
> >> declared as u64.
> >>
> >> Colin
> >>
> >>
> >
> > In fact, it is read into a u64:
> >
> > u64 offset, bitmap;
> > ....
> > bitmap = ctx->bitmap[queue_type];
> >
> > we'll have to reorder fields to avoid wasting memory.
> > Like this I guess:
> >
> > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> >
> > Colin, can you confirm pls?
>
> Fix looks sane to be, with u64 bitmap[3] struct size field re-ordering
> does not seem to make any difference on x86-64 (64 bytes) and i586 (56
> bytes) when I compiled with gcc-12, gcc-14 and clang-20.
>
> I can't functionally test this though (not sure how).
>
> Reviewed-by: Colin Ian King <colin.king@gmail.com>
Maybe it's time to ask for a prototype for each virtio-net feature
before it can be merged into the spec.
Thanks
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: virtio_net: support device stats
2024-10-14 9:47 ` Michael S. Tsirkin
2024-10-14 10:14 ` Colin King (gmail)
@ 2024-10-15 9:55 ` Xuan Zhuo
1 sibling, 0 replies; 5+ messages in thread
From: Xuan Zhuo @ 2024-10-15 9:55 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Jason Wang, Paolo Abeni, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org, Colin King (gmail)
On Mon, 14 Oct 2024 05:47:41 -0400, "Michael S. Tsirkin" <mst@redhat.com> wrote:
> On Mon, Oct 14, 2024 at 10:39:26AM +0100, Colin King (gmail) wrote:
> > Hi,
> >
> > Static analysis on Linux-next has detected a potential issue with the
> > following commit:
> >
> > commit 941168f8b40e50518a3bc6ce770a7062a5d99230
> > Author: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
> > Date: Fri Apr 26 11:39:24 2024 +0800
> >
> > virtio_net: support device stats
> >
> >
> > The issue is in function virtnet_stats_ctx_init, in drivers/net/virtio_net.c
> > as follows:
> >
> > if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_CVQ) {
> > queue_type = VIRTNET_Q_TYPE_CQ;
> >
> > ctx->bitmap[queue_type] |= VIRTIO_NET_STATS_TYPE_CVQ;
> > ctx->desc_num[queue_type] +=
> > ARRAY_SIZE(virtnet_stats_cvq_desc);
> > ctx->size[queue_type] += sizeof(struct
> > virtio_net_stats_cvq);
> > }
> >
> >
> > ctx->bitmap is declared as a u32 however it is being bit-wise or'd with
> > VIRTIO_NET_STATS_TYPE_CVQ and this is defined as 1 << 32:
> >
> > include/uapi/linux/virtio_net.h:#define VIRTIO_NET_STATS_TYPE_CVQ (1ULL <<
> > 32)
> >
> > ..and hence the bit-wise or operation won't set any bits in ctx->bitmap
> > because 1ULL < 32 is too wide for a u32.
>
> Indeed. Xuan Zhuo how did you test this patch?
In our machines, cvq type is not supported. ^_^
Why gcc has not warning for this, or I missed it.
>
> > I suspect ctx->bitmap should be
> > declared as u64.
> >
> > Colin
> >
> >
>
> In fact, it is read into a u64:
>
> u64 offset, bitmap;
> ....
> bitmap = ctx->bitmap[queue_type];
>
> we'll have to reorder fields to avoid wasting memory.
> Like this I guess:
>
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Thanks
>
> Colin, can you confirm pls?
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index c6af18948092..ef221429f784 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -4111,12 +4111,12 @@ struct virtnet_stats_ctx {
> /* Used to calculate the offset inside the output buffer. */
> u32 desc_num[3];
>
> - /* The actual supported stat types. */
> - u32 bitmap[3];
> -
> /* Used to calculate the reply buffer size. */
> u32 size[3];
>
> + /* The actual supported stat types. */
> + u64 bitmap[3];
> +
> /* Record the output buffer. */
> u64 *data;
> };
> --
> MST
>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2024-10-18 7:26 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-10-14 9:39 virtio_net: support device stats Colin King (gmail)
2024-10-14 9:47 ` Michael S. Tsirkin
2024-10-14 10:14 ` Colin King (gmail)
2024-10-18 7:25 ` Jason Wang
2024-10-15 9:55 ` Xuan Zhuo
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).