mbuf is largest in all mbuf，because the vhost
app may access invalid memory .
Thanks.
Gaoxiang
Replied Message
| From | Xia, Chenbo |
| Date | 07/01/2022 21:05 |
| To | Gaoxiang Liu,
maxime.coque...@redhat.com |
| Cc | dev@dpdk.org,
liugaoxi...@huawei.com,
sta...@dpdk.org
Hi, Chenbo
Yes. It is the latest patch.
Replied Message
| From | Xia, Chenbo |
| Date | 06/23/2022 11:07 |
| To | Gaoxiang Liu,
maxime.coque...@redhat.com |
| Cc | dev@dpdk.org,
liugaoxi...@huawei.com,
sta...@dpdk.org |
| Subject | RE: [PATCH v2] vhost: fix avail idx update error
Hi, Chen bo, Maxime,
Please help to review the patch below.
Thanks.
Gaoxiang.
Forwarding messages
From: "Gaoxiang Liu"
Date: 2022-06-22 09:19:44
To: maxime.coque...@redhat.com,chenbo@intel.com
Cc: dev@dpdk.org,liugaoxi...@huawei.com,"Gaoxiang Liu&quo
when parsing the mbuf.
Fixes: 0fd5608ef97f ("vhost: handle mbuf allocation failure")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed other idx update errors.
---
lib/vhost/virtio_net.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a
when parsing the mbuf.
Fixes: 0fd5608ef97f ("vhost: handle mbuf allocation failure")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed other idx update errors.
---
lib/vhost/virtio_net.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a
when parsing the mbuf.
Fixes: 0fd5608ef97f ("vhost: handle mbuf allocation failure")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
lib/vhost/virtio_net.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/vhost/virtio_net.c b/lib/vhost/virtio_net.c
index
the bond port to the slave port's system to solve the
problem.
Bugzilla ID: 551
Fixes: 46fb43683679 ("bond: add mode 4")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 13 +
1
the bond port to the slave port's system to solve the
problem.
Bugzilla ID: 551
Fixes: 46fb43683679 ("bond: add mode 4")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 12 +---
1 file changed, 5 insertions(+), 7 deletions(-
thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* delete duplicate code.
v4;
* Fixed some issues.
1. total_tx_desc should use +=
2. add detailed logs
v5:
* Fixed some issues.
1. move CACHE_FLUSHTHRESH_MULTIPLIER to rte_eth_bond-8023ad.c
2.
thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* delete duplicate code.
v4;
* Fixed some issues.
1. total_tx_desc should use +=
2. add detailed logs
v5:
* Fixed some issues.
1. move CACHE_FLUSHTHRESH_MULTIPLIER to rte_eth_bond-8023a
thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* delete duplicate code.
v4;
* Fixed some issues.
1. total_tx_desc should use +=
2. add detailed logs
v5:
* Fixed some issues.
1. move CACHE_FLUSHTHRESH_MULTIPLIER to rte_eth_bond-8023ad.c
2.
thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* delete duplicate code.
v4;
* Fixed some issues.
1. total_tx_desc should use +=
2. add detailed logs
v5:
* Fixed some issues.
1. move CACHE_FLUSHTHRESH_MULTIPLIER to rte_eth_bond-8023a
Hi, Stephen, Min Hu,
I will fix the issues.
Thanks.
Gaoxiang.
At 2022-03-09 10:53:15, "Min Hu (Connor)" wrote:
>Hi, Stephen,
>
>在 2022/3/9 9:25, Stephen Hemminger 写道:
>> On Tue, 8 Mar 2022 22:24:02 +0800
>> Gaoxiang Liu wrote:
>>
>>> diff -
thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* delete duplicate code.
v4;
* Fixed some issues.
1. total_tx_desc should use +=
2. add detailed logs
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 8 +---
lib/mempool/rte_mempool.h
fuction is the only code that allocates
from a slave's private pool. It runs in the context of the interrupt
thread, and thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* delete duplicate code.
---
drivers/net/bonding/rte_eth_bond_80
fuction is the only code that allocates
from a slave's private pool. It runs in the context of the interrupt
thread, and thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 4
1 file cha
fuction is the only code that allocates
from a slave's private pool. It runs in the context of the interrupt
thread, and thus it has no mempool cache of its own.
Signed-off-by: Gaoxiang Liu
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 4
1 file changed, 4 insertions(+
se to open it according to the actual
situation.
Fixes: d72e4042c5eb ("mem: exclude unused memory from core dump")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
v3:
* Fixed review issues. madvise is replaced by eal_mem_set_dump(),
*
Hi Jerin,
I will fix the issues.
Thanks.
Gaoxiang.
回复的原邮件
| 发件人 | Jerin Jacob |
| 日期 | 2021年12月14日 20:08 |
| 收件人 | Gaoxiang Liu |
| 抄送至 | dpdk-dev、Anatoly
Burakov、liugaoxi...@huawei.com、dpdk
stable |
| 主题 | Re: [PATCH] mem: exclude used memory from core dump by config |
On Tue, Dec
hoose to open it according to the actual
situation.
Fixes: d72e4042c5eb ("mem: exclude unused memory from core dump")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fixed compile issues.
---
lib/eal/common/eal_internal_cfg.h | 1 +
lib/eal/common/eal_options.h |
hoose to open it according to the actual
situation.
Fixes: d72e4042c5eb ("mem: exclude unused memory from core dump")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
lib/eal/common/eal_internal_cfg.h | 1 +
lib/eal/common/eal_options.h | 3 ++-
lib/eal/linux/eal.c
Fixes: 4958ca3a443a ("mbuf: support dynamic fields and flags")
Signed-off-by: Gaoxiang Liu
---
lib/mbuf/rte_mbuf.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/lib/mbuf/rte_mbuf.h b/lib/mbuf/rte_mbuf.h
index dedf83c38d..e2dfd47442 100644
--- a/lib/mbuf/rte_mbuf.h
+++ b/lib/
Hi Ferruh,
liugaoxi...@huawei.com is also my email address. You can update the commit to
use it.
Thanks,
Gaoxiang
发自 网易邮箱大师
回复的原邮件
| 发件人 | Ferruh Yigit |
| 日期 | 2021年10月22日 02:56 |
| 收件人 | Gaoxiang
Liu、maxime.coque...@redhat.com、chenbo@intel.com
|
| 抄送至 |
dev@dpdk.org
To improve performance in vhost Tx/Rx, merge vhost stats loop.
eth_vhost_tx has 2 loop of send num iteraion.
It can be merge into one.
eth_vhost_rx has the same issue as Tx.
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix coding style issues.
v3:
* add __rte_always_inline to
Hi, Maxime
I agree with you.The inline should be added to
vhost_update_single_packet_xstats function.
I will fix it in [PATCH v3].
Thanks,
Gaoxiang
发自 网易邮箱大师
回复的原邮件
| 发件人 | Maxime Coquelin |
| 日期 | 2021年10月15日 20:16 |
| 收件人 | Gaoxiang
Liu、chenbo@intel.com |
| 抄送至 |
dev
Hi, Maxime, Chenbo
Please help to review the patch.
Thanks.
Gaoxiang
On 09/28/2021 09:43, Gaoxiang Liu wrote:
To improve performance in vhost Tx/Rx, merge vhost stats loop.
eth_vhost_tx has 2 loop of send num iteraion.
It can be merge into one.
eth_vhost_rx has the same issue as Tx.
Fixes
To improve performance in vhost Tx/Rx, merge vhost stats loop.
eth_vhost_tx has 2 loop of send num iteraion.
It can be merge into one.
eth_vhost_rx has the same issue as Tx.
Fixes: 4d6cf2ac93dc ("net/vhost: add extended statistics")
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix co
To improve performance in vhost Tx/Rx, merge vhost stats loop.
eth_vhost_tx has 2 loop of send num iteraion.
It can be merge into one.
eth_vhost_rx has the same issue as Tx.
Fixes: 4d6cf2ac93dc ("net/vhost: add extended statistics")
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix co
To improve performance in vhost Tx/Rx, merge vhost stats loop.
eth_vhost_tx has 2 loop of send num iteraion.
It can be merge into one.
eth_vhost_rx has the same issue as Tx.
Fixes: 4d6cf2ac93dc ("net/vhost: add extended statistics")
Signed-off-by: Gaoxiang Liu
---
drivers
permission to send emails, so I have to send patches with @163
email.
Thanks,
Gaoxiang
At 2021-09-16 21:16:11, "Xia, Chenbo" wrote:
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Thursday, September 16, 2021 8:57 PM
>> To: Xia, Chenbo ; Maxime
k.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/vhost/rte_eth_vhost.c | 25 -
lib/vhost/virtio_net.c| 21 +
2 files changed, 25 insertions(+), 21 deletions(-)
diff --git a/drivers/net/vhost/rte_eth_vhost.c
b/drivers/net/vhost/rte_eth_vhos
k.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/vhost/rte_eth_vhost.c | 25 -
lib/vhost/virtio_net.c| 21 +
2 files changed, 25 insertions(+), 21 deletions(-)
diff --git a/drivers/net/vhost/rte_eth_vhost.c
b/drivers/net/vhost/rte_eth_vho
k.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/vhost/rte_eth_vhost.c | 25 -
lib/vhost/virtio_net.c| 21 +
2 files changed, 25 insertions(+), 21 deletions(-)
diff --git a/drivers/net/vhost/rte_eth_vhost.c
b/drivers/net/vhost/rte_eth_vho
k.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/vhost/rte_eth_vhost.c | 25 -
lib/vhost/virtio_net.c| 21 +
2 files changed, 25 insertions(+), 21 deletions(-)
diff --git a/drivers/net/vhost/rte_eth_vhost.c
b/drivers/net/vhost/rte_eth_vho
Hi, chenbo, maxime.
Please help to review the patch.
The patch has been proposed a month ago.
At 2021-08-31 22:39:51, "Gaoxiang Liu" wrote:
>When virtio_init_queue returns error, the memory of vq is freed.
>But the value of hw->vqs[queue_idx] does not restore.
>If vir
Add log print of socket path in vhost_user_add_connection.
It's useful when adding a mass of socket connections,
because the information of every connection is clearer.
Fixes: 8f972312b8f4 ("vhost: support vhost-user")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
lib/vho
Hi chenbo,
I will fix the issues in [PATCH v2].
Thanks.
Gaoxiang.
At 2021-09-06 13:35:23, "Xia, Chenbo" wrote:
>Hi Gaoxiang,
>
>> -Original Message-----
>> From: Gaoxiang Liu
>> Sent: Saturday, September 4, 2021 10:28 PM
>> To: maxime.coque...@redh
and fdset_try_del returns 0.
Thread1 will continue to delete the left conn_fd.
Thanks.
Gaoxiang
On 09/06/2021 11:18, Xia, Chenbo wrote:
Hi Gaoxiang,
>
>
>From: Gaoxiang Liu
>Sent: Thursday, September 2, 2021 11:38 PM
>To: Xia, Chenbo
>Cc: maxime.coque...@redhat.com; dev
Add log print of socket path in vhost_user_add_connection.
It's useful when adding a mass of socket connections,
because the information of every connection is more clearer.
Fixes: a277c7159876 ("vhost: refactor code structure")
Cc: sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
t; in front of free memory of conn,
then avoid the race condition.
The core trace is:
Program terminated with signal 11, Segmentation fault.
Fixes: 52d874dc6705 ("vhost: fix crash on closing in client mode")
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix coding style issues.
v3:
* Add de
it will arrive the "again" label, then close vsocket->socket_fd and uplink
"path" again.
it's not necessary.
And closing socket_fd does not trigger vhost_user_server_new_connection.
Thanks.
Gaoxiang
At 2021-08-31 13:37:38, "Xia, Chenbo" wrote:
>H
a...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix spelling warning
v3:
* Add detailed log
v4:
* Update the email address
---
drivers/net/virtio/virtio_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
in
a...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix spelling warning
v3:
* Add detailed log
---
drivers/net/virtio/virtio_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
index 056830566..fc72d71cb 100644
---
t; in front of free memory of conn,
then avoid the race condition.
The core trace is:
Program terminated with signal 11, Segmentation fault.
Fixes: 52d874dc6705 ("vhost: fix crash on closing in client mode")
Signed-off-by: Gaoxiang Liu
---
v2:
* Fix coding style issues.
v3:
* Add detai
et.
On 08/16/2021 14:44, Xia, Chenbo wrote:
Hi Gaoxiang,
> -Original Message-
> From: Gaoxiang Liu
> Sent: Friday, August 13, 2021 10:23 PM
> To: maxime.coque...@redhat.com; Xia, Chenbo
> Cc: dev@dpdk.org; liugaoxi...@huawei.com; Gaoxiang Liu
t of free memory of conn,
then avoid the race condition.
The core trace is:
Program terminated with signal 11, Segmentation fault.
Fixes: 52d874dc6705 ("vhost: fix crash on closing in client mode")
Signed-off-by: Gaoxiang Liu
v2:
* Fix coding style issues.
v3:
* Add detailed log.
v
ue also exists.
The fix is to move the "fdset_try_del" in front of free memory of conn,
then avoid the race condition.
The core trace is:
Program terminated with signal 11, Segmentation fault.
Fixes: 52d874dc6705 ("vhost: fix crash on closing in client mode")
Signed-off-b
4dc6705 ("vhost: fix crash on closing in client mode")
v2:
fix coding style issues
v3:
add detailed log
Signed-off-by: Gaoxiang Liu
---
lib/vhost/socket.c | 27 ++-
1 file changed, 14 insertions(+), 13 deletions(-)
diff --git a/lib/vhost/socket.c b/lib/vhost/so
v3:
add detailed log
Signed-off-by: Gaoxiang Liu
---
lib/vhost/socket.c | 27 ++-
1 file changed, 14 insertions(+), 13 deletions(-)
diff --git a/lib/vhost/socket.c b/lib/vhost/socket.c
index 5d0d728d5..2eb8fcadd 100644
--- a/lib/vhost/socket.c
+++ b/lib/vhost/socket.c
l" in front of free memory of conn ,
then avoid the race condition.
Fixes: 52d874dc6705 ("vhost: fix crash on closing in client mode")
v2:
fix coding style issues
Signed-off-by: Gaoxiang Liu
---
lib/vhost/socket.c | 27 ++-
1 file changed, 14 insertions(+), 1
l" in front of free memory of conn ,
then avoid the race condition.
Fixes: 52d874dc6705 ("vhost: fix crash on closing in client mode")
Signed-off-by: Gaoxiang Liu
---
lib/vhost/socket.c | 26 +-
1 file changed, 13 insertions(+), 13 deletions(-)
diff --git a/
On 08/04/2021 07:45, Gaoxiang Liu wrote:
When virtio_init_queue returns error, the memory of vq is freed.
But the value of hw->vqs[queue_idx] does not restore.
If hw->vqs[queue_idx] != NULL, the memory of vq is freed again
in virtio_free_queues.
Fixes: 69c80d4ef89b ("net/virti
dpdk.org
Signed-off-by: Gaoxiang Liu
v2:
fix spelling warning
---
drivers/net/virtio/virtio_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
index 056830566..fc72d71cb 100644
--- a/drivers/net/virtio/virti
dpdk.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/virtio/virtio_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
index 056830566..fc72d71cb 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/dr
sta...@dpdk.org
Signed-off-by: Gaoxiang Liu
---
drivers/net/virtio/virtio_ethdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
index 056830566..fc72d71cb 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/dr
Free memory of interrupt handle in virtio_user_dev_uninit() to
avoid memory leak.
when virtio user dev closes, memory of interrupt handle is not freed
that is allocated in virtio_user_fill_intr_handle().
Fixes: 3d4fb6fd2505 <"net/virtio-user: support Rx interrupt">
Signed-off-
Free memory of interrupt handle in virtio_user_dev_uninit() to
avoid memory leak.
when virtio user dev closes, the "intr_handle" memory is not freed
that is allocated in virtio_user_fill_intr_handle().
Fixes: 3d4fb6fd2505 <"net/virtio-user: support Rx interrupt">
S
Free the "intr_handle" memory in virtio_user_dev_uninit() to
avoid memory leak.
when virtio user dev closes, the "intr_handle" memory is not freeed
that is alloced in virtio_user_fill_intr_handle().
Fixes: 7f468b2ebfad ("net/virtio: release port upon close"
58 matches
Mail list logo
