Looks good to me.
Acked-by: Jay Jayatheerthan
> -Original Message-
> From: David Marchand
> Sent: Thursday, July 28, 2022 8:57 PM
> To: dev@dpdk.org
> Cc: Chautru, Nicolas ; Parav Pandit
> ; Xueming Li ; Hemant
> Agrawal ; Sachin Saxena ;
> Stephen Hemminger ;
> Long Li ; Zhang, Roy
Hi Thomas,
Sorry so long to response your email.
> -Original Message-
> From: Thomas Monjalon
> Sent: Thursday, July 7, 2022 5:05 PM
> To: Wu, WenxuanX
> Cc: andrew.rybche...@oktetlabs.ru; Li, Xiaoyun ;
> ferruh.yi...@xilinx.com; Singh, Aman Deep ;
> dev@dpdk.org; Zhang, Yuying ; Zhang,
Hi Andrew,
Apologies for the delay in getting back to you.
> -Original Message-
> From: Andrew Rybchenko
> Sent: Friday, July 8, 2022 11:01 PM
> To: Wu, WenxuanX ; tho...@monjalon.net; Li,
> Xiaoyun ; ferruh.yi...@xilinx.com; Singh, Aman Deep
> ; dev@dpdk.org; Zhang, Yuying
> ; Zhang, Qi
The rte_pcapng_write_packets() function fails when we try to write more
packets than the IOV_MAX limit. writev() system call is limited by the
IOV_MAX limit. The iovcnt argument is valid if it is greater than 0 and
less than or equal to IOV_MAX as defined in .
To avoid this problem, we can check t
As pcapng is used in the dpdk application it writes to a file.
You could repurpose it to something else, but even a pipe will not
give partial writes unless you configure the pipe as non-blocking.
Writing to a non-blocking pipe is going to have a load of other problems.
This seems like a purely
01/08/2022 09:09, Wang, YuanX:
> Hi Thomas,
>
> Sorry so long to response your email.
>
> From: Thomas Monjalon
> > 13/06/2022 12:25, wenxuanx...@intel.com:
> > > From: Wenxuan Wu
> > >
> > > This patch added new ethdev API to retrieve supported protocol header
> > > mask of a PMD, which helps
Hi, the real meat is in patch 1/2,
which fixes a segmentation fault in the OVS PMD thread when
resynchronizing with QEMU after the guest application has been killed
with SIGKILL.
This fixes an issue where the guest DPDK application is able
to crash the OVS process on the host.
Patch 2/2 is just
in virtio_dev_split we add a check for invalid nr_vec,
mainly for nr_vec == 0 (but add a check for BUF_VECTOR_MAX too),
and bail out before calling desc_to_mbuf, otherwise in desc_to_mbuf
we end up trying to memcpy from a source address buf_vec[0] that
is an uninitialized stack variable.
This shou
check when increasing vec_idx that it is still valid
in the (buf_len < dev->vhost_hlen) case too.
Signed-off-by: Claudio Fontana
---
lib/vhost/virtio_net.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/lib/vhost/virtio_net.c b/lib/vhost/virtio_net.c
index 0b8db2046e..6d
On 8/1/22 06:39, lihuisong (C) wrote:
在 2022/7/31 17:40, Andrew Rybchenko 写道:
Hi, Huisong!
On 7/29/22 05:30, lihuisong (C) wrote:
Hi Ori, and all,
For RSS flow rule, pattern item types and RSS types in action
are an inclusive relationship, and RSS types contain pattern item
types.
I disag
On 7/30/22 18:23, Claudio Fontana wrote:
> On 7/30/22 18:17, Claudio Fontana wrote:
>> Hello all,
>>
>> with the latest DPDK, openvswitch and qemu
>>
>> DPDK tag v22.07
>> openvswitch tag v2.17.1
>> qemu v7.1-git 22.07.2022
>>
>> and a DPDK setup which involves also an ubuntu guest with DPDK 16.11
On 8/1/22 13:57, Claudio Fontana wrote:
> On 7/30/22 18:23, Claudio Fontana wrote:
>> On 7/30/22 18:17, Claudio Fontana wrote:
>>> Hello all,
>>>
>>> with the latest DPDK, openvswitch and qemu
>>>
>>> DPDK tag v22.07
>>> openvswitch tag v2.17.1
>>> qemu v7.1-git 22.07.2022
>>>
>>> and a DPDK setup
Presently, rx metadata is sent to PMD by default, leading
to a performance drop as processing for the same in rx path
takes extra cycles.
Hence, introducing command line argument, 'rx-metadata' to
control passing rx metadata to PMD. By default it’s disabled.
ci: skip_checkpatch
Signed-off-by: Ha
Presently, rx metadata is sent to PMD by default, leading
to a performance drop as processing for the same in rx path
takes extra cycles.
Hence, introducing command line argument, 'rx-metadata' to
control passing rx metadata to PMD. By default it’s disabled.
Signed-off-by: Hanumanth Pothula
---
On Wed, Jul 27, 2022 at 8:10 AM Zhirun Yan wrote:
>
> For __rte_node_enqueue_prologue(), If the number of objs is more than
> the node->size * 2, the extra objs will write out of bounds memory.
> It should use __rte_node_stream_alloc_size() to request enough memory.
>
> And for rte_node_next_strea
On 8/1/22 16:27, lihuisong (C) wrote:
在 2022/8/1 19:53, Andrew Rybchenko 写道:
On 8/1/22 06:39, lihuisong (C) wrote:
在 2022/7/31 17:40, Andrew Rybchenko 写道:
Hi, Huisong!
On 7/29/22 05:30, lihuisong (C) wrote:
Hi Ori, and all,
For RSS flow rule, pattern item types and RSS types in action
are
On 7/21/22 06:24, Ding, Xuan wrote:
Hi Andrew,
-Original Message-
From: Andrew Rybchenko
Sent: 2022年7月8日 23:01
To: Wu, WenxuanX ; tho...@monjalon.net; Li,
Xiaoyun ; ferruh.yi...@xilinx.com; Singh, Aman Deep
; dev@dpdk.org; Zhang, Yuying
; Zhang, Qi Z ;
jerinjac...@gmail.com
Cc: step...
On Mon, 1 Aug 2022 10:40:56 +0200
Mário Kuka wrote:
> The rte_pcapng_write_packets() function fails when we try to write more
> packets than the IOV_MAX limit. writev() system call is limited by the
> IOV_MAX limit. The iovcnt argument is valid if it is greater than 0 and
> less than or equal to
On Thu, Jul 28, 2022 at 10:00 AM Bruce Richardson
wrote:
>
> On Thu, Jul 28, 2022 at 05:26:37PM +0200, David Marchand wrote:
> > Make rte_driver opaque for non internal users.
> > This will make extending this object possible without breaking the ABI.
> >
> > Introduce a new driver header and move
Hi Hanumanth,
Thanks for expanding support for the NIC-to-PMD Rx metadata feature.
I do not object the idea of the patch, it looks aceeptable.
However, please find my comments below.
On Mon, 1 Aug 2022, Hanumanth Pothula wrote:
Presently, rx metadata is sent to PMD by default, leading
to a per
This is an alternative, more general fix compared with PATCH v1.
The series fixes a segmentation fault in the OVS PMD thread when
resynchronizing with QEMU after the guest application has been killed
with SIGKILL (patch 1/2),
The segmentation fault can be caused by the guest DPDK application,
whi
in virtio_dev_split we cannot currently call desc_to_mbuf with
nr_vec == 0, or we end up trying to rte_memcpy from a source address
buf_vec[0] that is an uninitialized stack variable.
Improve this in general by having desc_to_mbuf and mbuf_to_desc
return -1 when called with an invalid nr_vec == 0,
check when increasing vec_idx that it is still valid
in the (buf_len < dev->vhost_hlen) case too.
Signed-off-by: Claudio Fontana
---
lib/vhost/virtio_net.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/lib/vhost/virtio_net.c b/lib/vhost/virtio_net.c
index 8d0d223983..22
This is an alternative, more general fix compared with PATCH v1,
and fixes style issues in v2.
The series fixes a segmentation fault in the OVS PMD thread when
resynchronizing with QEMU after the guest application has been killed
with SIGKILL (patch 1/2),
The segmentation fault can be caused by t
in virtio_dev_split we cannot currently call desc_to_mbuf with
nr_vec == 0, or we end up trying to rte_memcpy from a source address
buf_vec[0] that is an uninitialized stack variable.
Improve this in general by having desc_to_mbuf and mbuf_to_desc
return -1 when called with an invalid nr_vec == 0,
check when increasing vec_idx that it is still valid
in the (buf_len < dev->vhost_hlen) case too.
Tested-by: Claudio Fontana
Signed-off-by: Claudio Fontana
---
lib/vhost/virtio_net.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/lib/vhost/virtio_net.c b/lib/vhost/virti
On Tue, 2 Aug 2022 02:49:37 +0200
Claudio Fontana wrote:
> Tested-by: Claudio Fontana
> Signed-off-by: Claudio Fontana
Minor nit having both tags by same author is redundant and unnecessary.
Hi Ke,
Please update commit log.
LGTM.
> -Original Message-
> From: Zhang, Ke1X
> Sent: Friday, July 15, 2022 5:04 PM
> To: tho...@monjalon.net; Zhang, Yuying ; Xing,
> Beilei ; ferruh.yi...@xilinx.com; Zhou, YidingX
> ; dev@dpdk.org
> Cc: Zhang, Ke1X ; sta...@dpdk.org
> Subject: [PATCH
On Tue, Aug 2, 2022 at 8:33 AM Mao YingMing wrote:
>
Thanks for the patch
> From: maoyingming
>
> VFs's "bp->leds" is allways null, check bp->leds is
> not null before use bp->leds->num_leds.
Typo in 'always'
You can just say ' For VFs , bp->leds is uninitialized'
>
> segfault backtrace in trex p
From: maoyingming
VFs's "bp->leds" is allways null, check bp->leds is
not null before use bp->leds->num_leds.
segfault backtrace in trex program when use VF:
11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true)
10: bnxt_dev_led_on_op (dev=0x22d7780 )
9: rte_eth_led_on (port_id=0)
8: DpdkTR
Hi Luca,
The dpdk 21.11.2-rc1 test result from Red Hat looks good.
We tested below 17 scenarios and all got PASS on RHEL8:
- Guest with device assignment(PF) throughput testing(1G hugepage size):
PASS
- Guest with device assignment(PF) throughput testing(2M hugepage size)
: PASS
-
For VFs, bp->leds is uninitialized, check bp->leds is
not null before use bp->leds->num_leds.
segfault backtrace in trex program when use VF:
11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true)
10: bnxt_dev_led_on_op (dev=0x22d7780 )
9: rte_eth_led_on (port_id=0)
8: DpdkTRexPortAttr::set_le
On Tue, Aug 2, 2022 at 9:17 AM Mao YingMing wrote:
>
> For VFs, bp->leds is uninitialized, check bp->leds is
> not null before use bp->leds->num_leds.
Sorry for missing this the first time around, Rephrase it to say
'before checking for bp->leds->num_leds'
> segfault backtrace in trex program when
For VFs, bp->leds is uninitialized, check bp->leds is
not null before checking for bp->leds->num_leds.
segfault backtrace in trex program when use VF:
11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true)
10: bnxt_dev_led_on_op (dev=0x22d7780 )
9: rte_eth_led_on (port_id=0)
8: DpdkTRexPortAtt
On Tue, Aug 2, 2022 at 9:38 AM Mao YingMing wrote:
>
> For VFs, bp->leds is uninitialized, check bp->leds is
> not null before checking for bp->leds->num_leds.
>
> segfault backtrace in trex program when use VF:
> 11: bnxt_hwrm_port_led_cfg (bp=0x23ffb2140, led_on=true)
> 10: bnxt_dev_led_on_op (d
Thans for your review.
Mao Yingming
-邮件原件-
发件人: Somnath Kotur
发送时间: 2022年8月2日 12:09
收件人: Mao,Yingming
抄送: dev@dpdk.org; Ajit Khaparde ; Kalesh AP
; Thomas Monjalon
主题: Re: [PATCH v3] net/bnxt: fix null pointer dereference in
bnxt_hwrm_port_led_cfg()
O
36 matches
Mail list logo
