In ubuntu 24.04, DSA's workqueue has a new driver_name file in sysfs,
DPDK needs to write a correct value to this file when configuring a device
instance using the dpdk_idxd_cfg.py script, otherwise it will fail to be
configured, this patch fixes the issue.
Signed-off-by: Wenwu Ma
-
In ubuntu 24.04, DSA's workqueue has a new driver_name file in sysfs,
DPDK needs to write a correct value to this file when configuring a device
instance using the dpdk_idxd_cfg.py script, otherwise it will fail to be
configured, this patch fixes the issue.
Signed-off-by: Wenwu Ma
---
dr
nment. The patch
chooses the former option to fix the issue.
Fixes: b36970f2e13e ("dmadev: introduce DMA device library")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v2:
- Because of performance drop, adjust the code to
no longer demand cache line alignment
v3:
- back to v1 pa
The structure rte_dma_dev needs only 8 byte alignment.
This patch replaces __rte_cache_aligned of rte_dma_dev
with __rte_aligned(8).
Fixes: b36970f2e13e ("dmadev: introduce DMA device library")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v2:
- Because of performance drop, adjus
When disabling timesync, we should clear the IGC_RXPBS_CFG_TS_EN bit
of IGC_RXPBS, the patch fixes this.
Fixes: 4f6fbbf6f17d ("net/igc: support IEEE 1588 PTP")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
drivers/net/igc/igc_ethdev.c | 2 +-
1 file changed, 1 insertion(+),
The structure rte_dma_dev needs cacheline alignment, but the return
value of malloc may not be aligned to the cacheline. Therefore,
extra memory is applied for realignment.
Fixes: b36970f2e13e ("dmadev: introduce DMA device library")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
This patch fixes two null pointer dereferences detected by
coverity scan.
Coverity issue: 414096
Fixes: 6ccef90ff5d3 ("net/ice: support VSI level bandwidth config")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v2:
- deleting rather than modifying the prints that
cause nu
This patch fixes two null pointer dereferences detected by
coverity scan.
Coverity issue: 414096
Fixes: 6ccef90ff5d3 ("net/ice: support VSI level bandwidth config")
Signed-off-by: Wenwu Ma
---
drivers/net/ice/ice_tm.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
When doing IO port mapping for legacy device in secondary process, the
region information is missing, so, we need to refill it.
Fixes: 4b741542ecde ("bus/pci: avoid depending on private kernel value")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v5:
- adding checks to vfi
The settings of the vector for link state interrupts
should be done before the initialization of the device
is completed.
Fixes: ee85024cf5f7 ("net/virtio: complete init stage at the right place")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
Tested-by: Wei Ling
Reviewed-by: Maxim
When doing IO port mapping for legacy device in secondary process, the
region information is missing, so, we need to refill it.
Fixes: 4b741542ecde ("bus/pci: avoid depending on private kernel value")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v4:
- adjusting commit log
v3:
-
When doing IO port mapping for legacy device
in secondary process, the region information
is missing, so, we need to refill it.
Fixes: 4b741542ecde ("bus/pci: avoid depending on private kernel value")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v3:
- adjusting variable settings
When doing IO port mapping for legacy device
in secondary process, the region information
is missing, so, we need to refill it.
Fixes: 4b741542ecde ("bus/pci: avoid depending on private kernel value")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v2: add release of
The settings of the vector for link state interrupts
should be done before the initialization of the device
is completed.
Fixes: ee85024cf5f7 ("net/virtio: complete init stage at the right place")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
drivers/net/virtio/virtio_eth
When doing IO port mapping for legacy device
in secondary process, the region information
is missing, so, we need to refill it.
Fixes: 4b741542ecde ("bus/pci: avoid depending on private kernel value")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
drivers/bus/pci/linux/pci_v
In vhost sync batch enqueue, the "num_buffers" of
virtio net header is not be set, but it should be
set to 1 if the mrg_rxbuf feature is turned on,
This patch fix the issue.
Fixes: ef861692c398 ("vhost: add packed ring batch enqueue")
Cc: sta...@dpdk.org
Signed-off-by: Wenw
timeout like
TCP handshake packets. Therefore, this patch use CPU to perform
small copies in vhost.
Signed-off-by: Wenwu Ma
---
v4:
* fix coding style issues
v3:
* compare threshold with entire packet length
v2:
* fix CI build error
---
lib/vhost/vhost.h | 7 ++--
lib/vhost/virtio_net.c | 73
timeout like
TCP handshake packets. Therefore, this patch use CPU to perform
small copies in vhost.
Signed-off-by: Wenwu Ma
---
v3:
* compare threshold with entire packet length
v2:
* fix CI build error
---
lib/vhost/vhost.h | 7 ++--
lib/vhost/virtio_net.c | 72
timeout like
TCP handshake packets. Therefore, this patch use CPU to perform
small copies in vhost.
Signed-off-by: Wenwu Ma
---
v2:
* fix CI build error
---
lib/vhost/vhost.h | 6 ++-
lib/vhost/virtio_net.c | 87 +++---
2 files changed, 61 insertions(+), 32
timeout like
TCP handshake packets. Therefore, this patch use CPU to perform
small copies in vhost.
Signed-off-by: Wenwu Ma
---
lib/vhost/vhost.h | 6 ++-
lib/vhost/virtio_net.c | 87 +++---
2 files changed, 61 insertions(+), 32 deletions(-)
diff --git a
In async_enqueue_pkts(), the failed pkts will
be freed before return, but, the failed pkts may be
retried later, it will cause use after free. So,
we free the failed pkts after retry.
Fixes: 1907ce4baec3 ("examples/vhost: fix retry logic on Rx path")
Cc: sta...@dpdk.org
Signed-off-by
users may perform a SW IP checksum. This patch
add the code about it.
Fixes: b7091f1dcfbc ("app/testpmd: enable the heavyweight mode TCP/IPv4 GRO")
Fixes: 52f38a2055ed ("app/testpmd: enable TCP/IPv4 VxLAN and GRE GSO")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
users may perform a SW IP checksum. This patch
add the code about it.
Fixes: b7091f1dcfbc ("app/testpmd: enable the heavyweight mode TCP/IPv4 GRO")
Fixes: 52f38a2055ed ("app/testpmd: enable TCP/IPv4 VxLAN and GRE GSO")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
users may perform a SW IP
checksum. This patch add the code about it.
Signed-off-by: Wenwu Ma
---
app/test-pmd/csumonly.c | 12
1 file changed, 12 insertions(+)
diff --git a/app/test-pmd/csumonly.c b/app/test-pmd/csumonly.c
index 8b6665d6f3..76ce58f6d0 100644
--- a/app/test-pmd/csumonly.c
In vhost_user_msg_handler(), if vhost message handling
failed, we should check whether the queue is locked and
release the lock before returning. Or, it will cause a
deadlock later.
Fixes: 7f31d4ea05ca ("vhost: fix lock on device readiness notification")
Cc: sta...@dpdk.org
Signed-off
In vhost_user_msg_handler(), if vhost message handling
failed, we should check whether the queue is locked and
release the lock before returning. Or, it will cause a
deadlock later.
Fixes: 7f31d4ea05ca ("vhost: fix lock on device readiness notification")
Cc: sta...@dpdk.org
Signed-off
In vhost_user_msg_handler(), if vhost message handling
failed, we should check whether the queue is locked and
release the lock before returning. Or, it will cause a
deadlock later.
Signed-off-by: Wenwu Ma
---
lib/vhost/vhost_user.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib
In async datapath, if the length of data buffer
is less than 256, the data will be copied by CPU
instead of DMA.
Signed-off-by: Wenwu Ma
---
lib/vhost/vhost.h | 6 ++-
lib/vhost/virtio_net.c | 96 ++
2 files changed, 73 insertions(+), 29 deletions
;)
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 83 +++
1 file changed, 29 insertions(+), 54 deletions(-)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index 68afd398bb..d94fabb060 100644
--- a/examples/vhost/main.c
+++
The default values of rx mq_mode and rx offloads for port
will cause symmetric_mp startup failure if the port do not
support rss or csum. This Patch make the app to reconfigure
the NIC without them. Only quit the app if the second
reconfiguration fails.
Signed-off-by: Wenwu Ma
---
examples
The default values of rx mq_mode and rx offloads for port
will cause symmetric_mp startup failure if the port do not
support rss or csum. This Patch make the app to reconfigure
the NIC without them. Only quit the app if the second
reconfiguration fails.
Signed-off-by: Wenwu Ma
---
examples
app startup normally.
Signed-off-by: Wenwu Ma
---
examples/multi_process/symmetric_mp/main.c | 18 --
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/examples/multi_process/symmetric_mp/main.c
b/examples/multi_process/symmetric_mp/main.c
index 050337765f..df41abe682
dpdk-vhost will fail to launch with a 40G i40e port because
there are not enough mbufs. This patch adds a new option
--total-num-mbufs, through which the user can set larger
mbuf pool to avoid this problem.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 83
The current vhost backend lacks csum offloads information,
which will cause testpmd command such as "csum set tcp hw
" to fail. This patch adds the information according
to the device features.
Signed-off-by: Wenwu Ma
---
drivers/net/vhost/rte_eth_vhost.c | 18 ++
1 fi
mples/vhost: import userspace vhost application")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
---
examples/vhost/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index bc3d71c898..36969a4
vswitch can't launch with a 40G i40e port due to device start fails
if NIC’s max queues > the default number of 128, so, we changed
the default value from 128 to 512.
Fixes: 4796ad63ba1f ("examples/vhost: import userspace vhost application")
Cc: sta...@dpdk.org
Signe
vswitch can't launch with a 40G i40e port due to device start fails
if NIC’s max queues > the default number of 128, so, we changed
the default value from 128 to 512.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
Tested-by: Yvonne Yang
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
Reviewed-by: Chenbo Xia
Tested-by: Yvonne Yang
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
Tested-by: Yinan Wang
Tested-by: Yvonne Yang
---
doc/guides/prog_guide/vhost_lib.rst| 9 +
doc/guides/rel_notes/release_21_11.rst | 3 +
lib/vhost/rte_vhost_async.h| 34 +-
lib/vhost/version.map
note.
- Update function comments.
v2:
- Removed struct async_nethdr in 1/4.
- Removed a useless function declaration in 2/4,
and fixed some coding style in 4/4.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths
examples/vhost: use a new API to query remaining ring space
the vdev
has been removed, if so, let's skip this vdev.
Fixes: a68ba8e0a6b6 ("examples/vhost: refactor vhost data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/examples/vhost/main.c b/examples/vho
;vid], we need to check
whether the vdev has been removed, if so, let's skip
this vdev.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index d0bf1f31e3..1f6f7be8e3 100644
--- a/examples/vhos
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
Tested-by: Yvonne Yang
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
Reviewed-by: Chenbo Xia
Tested-by: Yvonne Yang
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
Tested-by: Yinan Wang
Tested-by: Yvonne Yang
---
doc/guides/prog_guide/vhost_lib.rst | 9 +
lib/vhost/rte_vhost_async.h | 33 +-
lib/vhost/version.map | 3 +
lib/vhost/vhost.h
async_nethdr in 1/4.
- Removed a useless function declaration in 2/4,
and fixed some coding style in 4/4.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths
examples/vhost: use a new API to query remaining ring space
examples/vhost: support vhost async dequeue data path
vswitch can't launch with 40G FTV due to Device start fails
if NIC’s max queues > the default number of 128,
so, we changed the default value from 128 to 512.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/exampl
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 61
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
Tested-by: Yinan Wang
---
doc/guides/prog_guide/vhost_lib.rst | 9 +
lib/vhost/rte_vhost_async.h | 36 +-
lib/vhost/version.map | 3 +
lib/vhost/vhost.h | 3 +-
lib/vhost
on vhost patch from Jiayu Hu
(http://patches.dpdk.org/project/dpdk/patch/1629463466-450012-1-git-send-email-jiayu...@intel.com/)
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths
examples/vhost: use a new API to query remaining ring space
examples/vhost: support vhost
")
Cc: sta...@dpdk.org
v2:
- Delete '.' in the title.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index bc3d71c898..07fd90ec64 100644
--- a/examples/vhost/main.c
+++ b/e
")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index bc3d71c898..07fd90ec64 100644
--- a/examples/vhost/main.c
+++ b/examples/vhost/main.c
@@ -965,6 +965,7 @@ virti
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 61
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
---
doc/guides/prog_guide/vhost_lib.rst | 9 +
lib/vhost/rte_vhost_async.h | 39 +-
lib/vhost/version.map | 3 +
lib/vhost/virtio_net.c | 586
4 files
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
Reviewed-by: Maxime Coquelin
rte_vhost_async_try_dequeue_burst.
- The completed package updates the used ring directly.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths
examples/vhost: use a new API to query remaining ring space
examples/vhost: support vhost async dequeue data path
Yuan Wang (1):
vhost: support async
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 61 ++---
examples/vhost/ioat.h
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
---
doc/guides/prog_guide/vhost_lib.rst | 9 +
lib/vhost/rte_vhost_async.h | 39 +-
lib/vhost/version.map | 3 +
lib/vhost/virtio_net.c | 586
4 files
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost/ioat.c b/examples/vhost/ioat.c
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c
directly.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths
examples/vhost: use a new API to query remaining ring space
examples/vhost: support vhost async dequeue data path
Yuan Wang (1):
vhost: support async dequeue for split ring
doc/guides/prog_guide
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 61 ++---
examples/vhost/ioat.h
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
---
doc/guides/prog_guide/vhost_lib.rst | 10 +
lib/vhost/rte_vhost_async.h | 44 +-
lib/vhost/version.map | 3 +
lib/vhost/virtio_net.c | 601
4 files
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost/ioat.c b/examples/vhost/ioat.c
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c
from rte_vhost_try_dequeue_burst to
rte_vhost_async_try_dequeue_burst.
- The completed package updates the used ring directly.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths.
examples/vhost: use a new API to query remaining ring space
examples/vhost: support vhost
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 61 ++---
examples/vhost/ioat.h
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
---
doc/guides/prog_guide/vhost_lib.rst | 10 +
lib/vhost/rte_vhost_async.h | 44 ++-
lib/vhost/version.map | 3 +
lib/vhost/virtio_net.c | 579
4 files
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost/ioat.c b/examples/vhost/ioat.c
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c
performance enhancement. 160% performance improvement
for v2 vs. v1.
- Async dequeue API name change from rte_vhost_try_dequeue_burst to
rte_vhost_async_try_dequeue_burst.
- The completed package updates the used ring directly.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 61 ++---
examples/vhost/ioat.h
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
---
doc/guides/prog_guide/vhost_lib.rst | 10 +
lib/vhost/rte_vhost_async.h | 44 ++-
lib/vhost/version.map | 3 +
lib/vhost/virtio_net.c | 579
4 files
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost/ioat.c b/examples/vhost/ioat.c
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c
queue_id value before using it.
- Async dequeue performance enhancement. 160% performance improvement
for v2 vs. v1.
- Async dequeue API name change from rte_vhost_try_dequeue_burst to
rte_vhost_async_try_dequeue_burst.
- The completed package updates the used ring directly.
Wenwu Ma (3
This patch is to add vhost async dequeue data-path in vhost sample.
vswitch can leverage IOAT to accelerate vhost async dequeue data-path.
Signed-off-by: Wenwu Ma
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/ioat.c | 34 +--
examples/vhost/ioat.h
.
Signed-off-by: Yuan Wang
Signed-off-by: Jiayu Hu
Signed-off-by: Wenwu Ma
---
doc/guides/prog_guide/vhost_lib.rst | 10 +
lib/vhost/rte_vhost_async.h | 44 ++-
lib/vhost/version.map | 3 +
lib/vhost/virtio_net.c | 553
4 files
A new API for querying the remaining descriptor ring capacity
is available, so we use the new one instead of the old one.
Signed-off-by: Wenwu Ma
---
examples/vhost/ioat.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/examples/vhost/ioat.c b/examples/vhost/ioat.c
rte_vhost_try_dequeue_burst to
rte_vhost_async_try_dequeue_burst.
- The completed package updates the used ring directly.
Wenwu Ma (3):
examples/vhost: refactor vhost enqueue and dequeue datapaths.
examples/vhost: use a new API to query remaining ring space
examples/vhost: support vhost async
Previously, by judging the flag, we call different enqueue/dequeue
functions in data path.
Now, we use an ops that was initialized when Vhost was created,
so that we can call ops directly in Vhost data path without any more
flag judgment.
Signed-off-by: Wenwu Ma
---
examples/vhost/main.c
Amount of allocated memory was not enough for mempool
which cause buffer overflow when access fields of mempool
private structure in the rte_pktmbuf_priv_size function.
Fixes: 923ceaeac140 ("test/mempool: add unit test cases")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v2:
- ref
Amount of allocated memory was not enough for mempool
which cause buffer overflow when access fields of mempool
private structure in the rte_pktmbuf_priv_size function.
Fixes: 923ceaeac140 ("test/mempool: add unit test cases")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
v2:
- ref
When removing the mac filter in ice_remove_all_mac_vlan_filters(),
TAILQ_FOREACH_SAFE should be used instead of TAILQ_FOREACH,
Otherwise, it will result in a illegal pointer access.
Fixes: e0dcf94a0d7f ("net/ice: support VLAN ops")
Cc: sta...@dpdk.org
Signed-off-by: Wenwu Ma
---
d
mac_addrs in dev data should be allocated more memory space,
Otherwise, traversing mac_addrs in function eth_dev_mac_restore()
will result heap-buffer-overflow.
Signed-off-by: Wenwu Ma
---
drivers/net/ice/ice_ethdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a
When deleting the mac filter in ice_remove_all_mac_vlan_filters(),
TAILQ_FOREACH_SAFE should be used instead of TAILQ_FOREACH,
Otherwise, it will result in a illegal pointer access.
Signed-off-by: Wenwu Ma
---
drivers/net/ice/ice_ethdev.c | 7 +--
1 file changed, 5 insertions(+), 2
From: wenwu ma
When deleting the mac filter in ice_remove_all_mac_vlan_filters(),
TAILQ_FOREACH_SAFE should be used instead of TAILQ_FOREACH,
Otherwise, it will result in a null pointer access.
Signed-off-by: wenwu ma
---
drivers/net/ice/ice_ethdev.c | 7 +--
1 file changed, 5 insertions
When deleting the mac filter in ice_remove_all_mac_vlan_filters(),
TAILQ_FOREACH_SAFE should be used instead of TAILQ_FOREACH,
Otherwise, it will result in a illegal pointer access.
Signed-off-by: Wenwu Ma
---
drivers/net/ice/ice_ethdev.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions
mac_addrs in dev data should be allocated more memory space,
Otherwise, traversing mac_addrs in function eth_dev_mac_restore()
will result heap-buffer-overflow.
Signed-off-by: Wenwu Ma
---
drivers/net/ice/ice_ethdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a
The value of parameter private_data_size of function
rte_mempool_create() called in test_mempool() should not be 0,
Otherwise, the function rte_pktmbuf_priv_size() called in
rte_pktmbuf_init() will cause heap-buffer-overflow.
Signed-off-by: Wenwu Ma
---
app/test/test_mempool.c | 3 ++-
1 file
Before accessing the private data of mempool in
function rte_pktmbuf_priv_size() and rte_pktmbuf_data_room_size(),
it is necessary to determine whether the private data exists,
otherwise it will cause heap-buffer-overflow.
Signed-off-by: Wenwu Ma
---
lib/librte_mbuf/rte_mbuf.h | 6 ++
1
Before accessing the private data of mempool in
function rte_pktmbuf_priv_size() and rte_pktmbuf_data_room_size(),
it is necessary to determine whether the private data exists,
otherwise it will cause null pointer access.
Signed-off-by: Wenwu Ma
---
lib/librte_mbuf/rte_mbuf.h | 6 ++
1 file
98 matches
Mail list logo