In async data path, when vring state changes or device is destroyed,
it is necessary to know the number of inflight packets in DMA engine.
This patch provides a thread unsafe API to return the number of
inflight packets for a vhost queue without using any lock.
Signed-off-by: Xuan Ding
---
doc
In async data path, call rte_vhost_async_get_inflight_thread_unsafe()
API to directly return the number of inflight packets instead of
maintaining a local variable.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 31 +--
examples/vhost/main.h | 1 -
2 files
nable IOMMU for async vhost")
Signed-off-by: Xuan Ding
---
v4:
* Remove unnessary ENOSUP check.
* Adjust return type.
v3:
* Fix commit title.
v2:
* Fix a format issue.
---
lib/vhost/vhost.h | 1 +
lib/vhost/vhost_user.c | 111 -
2 files c
DPDK 21.11 adds vfio support for DMA device in vhost. This patch
updates recommended IOVA mode in async datapath.
Signed-off-by: Xuan Ding
---
doc/guides/prog_guide/vhost_lib.rst | 9 +
1 file changed, 9 insertions(+)
diff --git a/doc/guides/prog_guide/vhost_lib.rst
b/doc/guides
From: Xuan Ding
When choosing IOVA as PA mode, IOVA is likely to be discontinuous,
which requires page by page mapping for DMA devices. To be consistent,
this patch implements page by page mapping instead of mapping at the
region granularity for both IOVA as VA and PA mode.
Fixes: 7c61fa08b716
From: Xuan Ding
This patch renames the host_phys_addr to host_iova in guest_page
struct. The host_phys_addr is iova, it depends on the DPDK
IOVA mode.
Signed-off-by: Xuan Ding
---
lib/vhost/vhost.h | 10 +-
lib/vhost/vhost_user.c | 24
lib/vhost
From: Xuan Ding
Hi everyone,
The presence of an asynchronous path allows applications to offload memory
copies to DMA engine, so as to save CPU cycles and improve the copy
performance. This patch set is a draft implementation for split
ring in vhost async dequeue data path. The code is based on
From: Xuan Ding
This patch implements asynchronous dequeue data path for vhost split
ring, with dmadev library integrated.
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
lib/vhost/rte_vhost_async.h | 29 +++
lib/vhost/version.map | 1 +
lib/vhost/vhost.h | 1
From: Xuan Ding
This patch adds the use case for async dequeue API. Vswitch can
leverage DMA device to accelerate vhost async dequeue path.
Signed-off-by: Wenwu Ma
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/main.c
Only allow to create symmetric rule for L3/L4.
Fixes: 38d632cbdc88("net/ice: refactor PF RSS")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
---
drivers/net/ice/ice_hash.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/ice/ice_hash.c b/drivers/net/ice/ice_ha
Only allow to create symmetric rule for L3/L4.
Fixes: 91f27b2e39ab("net/iavf: refactor RSS")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
---
drivers/net/iavf/iavf_hash.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/iavf/iavf_hash.c b/drivers/net/iavf/i
Add advanced RSS offloads check due to some legacy driver(kernel/DPDK PF)
does not support virtual channel command VIRTCHNL_OP_RSS_HENA with
hena = 0 and VIRTCHNL_OP_ADD_RSS_CFG.
Fixes: 95f2f0e9fc2a6("net/iavf: improve default RSS")
Signed-off-by: Xuan Ding
---
drivers/net/iavf/iav
For pattern MAC_PPPOE_IPV4/6, add ETH_RSS_ETH into insput_set_mask
to fix rss rule cannot be created when set eth as rss type.
Fixes: 0d84f86c3022("net/ice: fix GTPU header parsing")
Signed-off-by: Xuan Ding
---
drivers/net/ice/ice_hash.c | 18 +-
1 file changed, 9
For pattern MAC_PPPOE_IPV4/6, add ETH_RSS_ETH into insput_set_mask
to fix rss rule cannot be created when set eth as rss type.
Fixes: 0d84f86c3022("net/ice: fix GTPU header parsing")
Signed-off-by: Xuan Ding
---
v2:
* Refine the commit title.
* Adjust the indentation by addi
From: Xuan Ding
This patchset introduces a new API rte_vhost_async_dma_unconfigure()
to help user to manually free the DMA vchannel finished to use.
Note: this API should be called after async channel unregister.
v2:
* Add spinlock protection.
* Fix a memory leak issue.
* Refine the doc.
Xuan
From: Xuan Ding
This patch adds a new API rte_vhost_async_dma_unconfigure() to unconfigure
DMA vchannels in vhost async data path.
Lock protection are also added to protect DMA vchannels configuration and
unconfiguration from concurrent calls.
Signed-off-by: Xuan Ding
---
doc/guides
From: Xuan Ding
This patch applies rte_vhost_async_dma_unconfigure() API
to manually free 'dma_copy_track' array instead of waiting
until the program ends to be released.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 7 +++
1 file changed, 7 insertions(+)
diff --git
check the status of region mapping.
* Fix one typo.
Xuan Ding (2):
vfio: allow partially unmapping adjacent memory
vhost: enable IOMMU for async vhost
lib/eal/linux/eal_vfio.c | 338 ++-
lib/vhost/vhost.h| 4 +
lib/vhost/vhost_user.c | 116
rt partial unmapping. With this change,
we will still be able to merge adjacent segments, but only if they are
of the same size. If we keep with our above example, adjacent segments A
and B will be stored as separate segments if they are of different
sizes.
Signed-off-by: Anatoly Burakov
Signed-of
The use of IOMMU has many advantages, such as isolation and address
translation. This patch extends the capbility of DMA engine to use
IOMMU if the DMA engine is bound to vfio.
When set memory table, the guest memory will be mapped
into the default container of DPDK.
Signed-off-by: Xuan Ding
kernel driver.
* Add a flag to check the status of region mapping.
* Fix one typo.
Xuan Ding (2):
vfio: allow partially unmapping adjacent memory
vhost: enable IOMMU for async vhost
lib/eal/linux/eal_vfio.c | 338 ++-
lib/vhost/vhost.h| 4 +
lib
rt partial unmapping. With this change,
we will still be able to merge adjacent segments, but only if they are
of the same size. If we keep with our above example, adjacent segments A
and B will be stored as separate segments if they are of different
sizes.
Signed-off-by: Anatoly Burakov
Signed-of
The use of IOMMU has many advantages, such as isolation and address
translation. This patch extends the capbility of DMA engine to use
IOMMU if the DMA engine is bound to vfio.
When set memory table, the guest memory will be mapped
into the default container of DPDK.
Signed-off-by: Xuan Ding
From: Xuan Ding
This patchset introduces a new API rte_vhost_async_dma_unconfigure()
to help user to manually free the DMA vchannel finished to use.
Note: this API should be called after async channel unregister.
v3:
* Rebase to latest DPDK.
* Refine some descriptions in the doc.
* Fix one bug
From: Xuan Ding
This patch adds a new API rte_vhost_async_dma_unconfigure() to unconfigure
DMA vchannels in vhost async data path.
Lock protection are also added to protect DMA vchannels configuration and
unconfiguration from concurrent calls.
Signed-off-by: Xuan Ding
---
doc/guides
From: Xuan Ding
This patch applies rte_vhost_async_dma_unconfigure() API
to manually free DMA vchannels instead of waiting
until the program ends to be released.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 45 ++-
examples/vhost/main.h | 1
From: Xuan Ding
This patchset introduces a new API rte_vhost_async_dma_unconfigure()
to help user to manually free DMA vchannels finished to use.
Note: this API should be called after async channel unregister.
v4:
* Rebase to 22.11 rc1.
* Fix the usage of 'dma_ref_count' to mak
From: Xuan Ding
Add a new API rte_vhost_async_dma_unconfigure() to unconfigure DMA
vchannels in vhost async data path. Lock protection are also added
to protect DMA vchannels configuration and unconfiguration
from concurrent calls.
Signed-off-by: Xuan Ding
---
doc/guides/prog_guide
From: Xuan Ding
This patch applies rte_vhost_async_dma_unconfigure() to manually
free DMA vchannels. Before unconfiguration, need make sure the
specified DMA device is no longer used by any vhost ports.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 38
Since packed indirect descriptors are added and initialized when
initializing vring, the reconnection path also needs to be considered.
Fixes: 381f39ebb78a ("net/virtio: fix packed ring indirect descricptors setup")
Cc: sta...@dpdk.org
Cc: yong@intel.com
Signed-off-by:
-by-page.
[1] https://mails.dpdk.org/archives/dev/2021-July/213493.html
Signed-off-by: Xuan Ding
---
doc/guides/rel_notes/deprecation.rst | 3 +++
1 file changed, 3 insertions(+)
diff --git a/doc/guides/rel_notes/deprecation.rst
b/doc/guides/rel_notes/deprecation.rst
index 76a4abfd6b..272ffa993e
-by-page.
[1] https://mails.dpdk.org/archives/dev/2021-July/213493.html
Signed-off-by: Xuan Ding
---
doc/guides/rel_notes/deprecation.rst | 3 +++
1 file changed, 3 insertions(+)
diff --git a/doc/guides/rel_notes/deprecation.rst
b/doc/guides/rel_notes/deprecation.rst
index 76a4abfd6b..1234420caf
This series adds support for DMA device to use vfio.
The first patch is to extend current vfio dma mapping API to
allow partial unmapping for adjacent memory if the
platform does not support partial unmapping.
Xuan Ding (2):
vfio: allow partially unmapping adjacent memory
vhost: enable IOMMU
The use of IOMMU has many advantages, such as isolation and address
translation. This patch extends the capbility of DMA engine to use
IOMMU if the DMA device is bound to vfio.
When set memory table, the guest memory will be mapped
into the default container of DPDK.
Signed-off-by: Xuan Ding
rt partial unmapping. With this change,
we will still be able to merge adjacent segments, but only if they are
of the same size. If we keep with our above example, adjacent segments A
and B will be stored as separate segments if they are of different
sizes.
Signed-off-by: Xuan Ding
---
lib
In async data path, when vring state changes, it is necessary to
know the number of inflight packets in DMA engine. This patch
provides a thread unsafe API to return the number of inflight
packets without using any lock.
Signed-off-by: Xuan Ding
---
doc/guides/prog_guide/vhost_lib.rst| 5
From: Xuan Ding
RTE_ETH_RX_OFFLOAD_HEADER_SPLIT offload was introduced some time ago to
substitute bit-field header_split in struct rte_eth_rxmode. It allows
to enable per-port header split offload with the header size controlled
using split_hdr_size in the same structure.
Right now, no single
From: Xuan Ding
As announced in the deprecation note, this patch removes the Rx offload
flag 'RTE_ETH_RX_OFFLOAD_HEADER_SPLIT' and 'split_hdr_size' field from
the structure 'rte_eth_rxmode'. User can still use
`RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT` for per-queue
From: Xuan Ding
As announced in the deprecation note, this patch removes the Rx offload
flag 'RTE_ETH_RX_OFFLOAD_HEADER_SPLIT' and 'split_hdr_size' field from
the structure 'rte_eth_rxmode'. Meanwhile, the place where the examples
and apps initialize the 's
From: Xuan Ding
This patchset introduces a new API rte_vhost_async_dma_unconfigure()
to help user to manually free the DMA vchannel finished to use.
Note: this API should be called after async channel unregister.
Xuan Ding (2):
vhost: introduce DMA vchannel unconfiguration
example/vhost
From: Xuan Ding
This patch adds a new API rte_vhost_async_dma_unconfigure() to unconfigure
DMA vchannels in vhost async data path.
Signed-off-by: Xuan Ding
---
doc/guides/prog_guide/vhost_lib.rst| 5
doc/guides/rel_notes/release_22_11.rst | 2 ++
lib/vhost/rte_vhost_async.h
From: Xuan Ding
This patch uses rte_vhost_async_dma_unconfigure() API
to manually free 'dma_coy_track' array rather than wait for
the program to finish before being freed.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 7 +++
1 file changed, 7 insertions(+)
diff --git
From: Xuan Ding
This patchset fixes the issue of incorrect DMA mapping in PA mode.
Due to the ambiguity of host_phys_addr naming in the guest page
struct, rename it to host_iova.
v3:
* Fix some format issues.
v2:
* Change the order of patch.
Xuan Ding (2):
vhost: fix field naming in guest
From: Xuan Ding
This patch renames the host_phys_addr to host_iova in guest_page
struct. The host_phys_addr is iova, it depends on the DPDK
IOVA mode.
Fixes: e246896178e6 ("vhost: get guest/host physical address mappings")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
Reviewed-
From: Xuan Ding
When choosing IOVA as PA mode, IOVA is likely to be discontinuous,
which requires page by page mapping for DMA devices. To be consistent,
this patch implements page by page mapping instead of mapping at the
region granularity for both IOVA as VA and PA mode.
Fixes: 7c61fa08b716
From: Xuan Ding
This patchset fixes the issue of incorrect DMA mapping in PA mode.
Due to the ambiguity of host_phys_addr naming in the guest page
struct, rename it to host_iova.
v4:
* Fix the patch timezone.
v3:
* Fix some format issues.
v2:
* Change the order of patch.
Xuan Ding (2
From: Xuan Ding
This patch renames the host_phys_addr to host_iova in guest_page
struct. The host_phys_addr is iova, it depends on the DPDK
IOVA mode.
Fixes: e246896178e6 ("vhost: get guest/host physical address mappings")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
Reviewed-
From: Xuan Ding
When choosing IOVA as PA mode, IOVA is likely to be discontinuous,
which requires page by page mapping for DMA devices. To be consistent,
this patch implements page by page mapping instead of mapping at the
region granularity for both IOVA as VA and PA mode.
Fixes: 7c61fa08b716
From: Xuan Ding
This patchset introduces an unsafe API to get the number of inflight
packets in DMA engine in some situations. Like vring state changes or
device is destroyed. Compared with rte_vhost_async_get_inflight(),
this is a lock free version.
Xuan Ding (2):
vhost: add unsafe API to
From: Xuan Ding
In async data path, when vring state changes or device is destroyed,
it is necessary to know the number of inflight packets in DMA engine.
This patch provides a thread unsafe API to return the number of
inflight packets for a vhost queue without using any lock.
Signed-off-by
From: Xuan Ding
In async data path, call rte_vhost_async_get_inflight_thread_unsafe()
API to directly return the number of inflight packets instead of
maintaining a local variable.
Signed-off-by: Xuan Ding
---
examples/vhost/main.c | 28 +++-
examples/vhost/main.h | 1
From: Xuan Ding
Since QEMU 5.2.0 fixes the vhost multi-queue reconnection issue
in commit f66337bdbfda ("vhost-user: save features of multiqueues
if chardev is closed"), this patch removes the previous description
from known issue.
Signed-off-by: Xuan Ding
---
doc/guides
From: Xuan Ding
The presence of an asynchronous path allows applications to offload
memory copies to DMA engine, so as to save CPU cycles and improve
the copy performance. This patch set is a draft implementation for
split ring in vhost async dequeue data path. The code is based on
latest
From: Xuan Ding
This patch implements asynchronous dequeue data path for vhost split
ring, with dmadev library integrated.
Signed-off-by: Xuan Ding
Signed-off-by: Yuan Wang
---
lib/vhost/rte_vhost_async.h | 37 ++-
lib/vhost/version.map | 1 +
lib/vhost/vhost.h | 1
From: Xuan Ding
This patch adds the use case for async dequeue API. Vswitch can
leverage DMA device to accelerate vhost async dequeue path.
Signed-off-by: Wenwu Ma
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/main.c
From: Xuan Ding
Since QEMU 5.2.0 fixes the vhost multi-queue reconnection issue
in commit f66337bdbfda ("vhost-user: save features of multiqueues
if chardev is closed"), this patch removes the previous description
from known issue.
Fixes: b37e95507e1b ("doc: add vhost multi-que
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. Splitting is usually between the
packet header that can be posted to a dedicated buffer and the packet
payload that can be posted to a different buffer. This kind of
From: Xuan Ding
Header split is useful in some scenarios, such as GPU acceleration.
The spliting will help to enable true zero copy and hence improve
the performance significantly.
This patchset enables header split in normal Rx data paths. When
Rx queue is configured with header split feature
From: Xuan Ding
This patch adds header split configuration in testpmd. The header split
feature is off by default. To enable header split, you need:
1. Configure Rx queue with rx_offload header split on.
2. Set the protocol type of header split.
Command for set header split protocol type
From: Xuan Ding
This patch adds support for header split in normal Rx data paths.
When Rx queue is configured with header split feature, packets
received will be directly splited into header and payload parts.
And the two parts will be put into different mempools.
Currently, the vectorized path
From: Xuan Ding
The presence of asynchronous path allows applications to offload
memory copies to DMA engine, so as to save CPU cycles and improve
the copy performance. This patch set is a draft implementation for
split ring in vhost async dequeue data path. The code is based on
latest enqueue
From: Xuan Ding
This patch implements asynchronous dequeue data path for vhost split
ring, with dmadev library integrated.
Signed-off-by: Xuan Ding
Signed-off-by: Yuan Wang
---
lib/vhost/rte_vhost_async.h | 37 ++-
lib/vhost/version.map | 1 +
lib/vhost/vhost.h | 1
From: Xuan Ding
This patch adds the use case for async dequeue API. Vswitch can
leverage DMA device to accelerate vhost async dequeue path.
Signed-off-by: Wenwu Ma
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/main.c
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. It is useful in some scenarios,
such as GPU acceleration. The spliting will help to enable true zero
copy and hence improve the performance significantly.
This patchset
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. The split happens after the
packet header and before the packet payload. Splitting is usually between
the packet header that can be posted to a dedicated buffer and the
From: Xuan Ding
This patch adds header split configuration in testpmd. The header split
feature is off by default. To enable header split, you need:
1. Configure Rx queue with rx_offload header split on.
2. Set the protocol type of header split.
Command for set header split protocol type
From: Xuan Ding
This patch adds support for header split in normal Rx data paths.
When the Rx queue is configured with header split for specific
protocol type, packets received will be directly splited into
header and payload parts. And the two parts will be put into
different mempools
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. It is useful in some scenarios,
such as GPU acceleration. The spliting will help to enable true zero
copy and hence improve the performance significantly.
This patchset
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. The split happens after the
packet header and before the packet payload. Splitting is usually between
the packet header that can be posted to a dedicated buffer and the
From: Xuan Ding
This patch adds header split configuration in testpmd. The header split
feature is off by default. To enable header split, you need:
1. Configure Rx queue with rx_offload header split on.
2. Set the protocol type of header split.
Command for set header split protocol type
From: Xuan Ding
This patch adds support for header split in normal Rx data paths.
When the Rx queue is configured with header split for specific
protocol type, packets received will be directly splited into
header and payload parts. And the two parts will be put into
different mempools
From: Xuan Ding
For async copy, it is unsafe to directly use the physical address.
and current address translation from GPA to HPA via SW also takes
CPU cycles, these can all benefit from IOMMU.
Since the existing DMA engine supports to use platform IOMMU,
this patch enables IOMMU for async
From: Xuan Ding
For async copy, it is unsafe to directly use the physical address.
and current address translation from GPA to HPA via SW also takes
CPU cycles, these can all benefit from IOMMU.
Since the existing DMA engine supports to use platform IOMMU,
this patch enables IOMMU for async
From: Xuan Ding
For async copy, it is unsafe to directly use the physical address.
And current address translation from GPA to HPA via SW also takes
CPU cycles, these can all benefit from IOMMU.
Since the existing DMA engine supports to use platform IOMMU,
this patch enables IOMMU for async
From: Xuan Ding
The presence of asynchronous path allows applications to offload
memory copies to DMA engine, so as to save CPU cycles and improve
the copy performance. This patch implements vhost async dequeue data
path for split ring.
This patch set is a new design and implementation of [2
From: Xuan Ding
This patch extracts the descriptors to buffers filling from
copy_desc_to_mbuf() into a dedicated function. Besides, enqueue
and dequeue path are refactored to use the same function
sync_fill_seg() for preparing batch elements, which simplies
the code without performance
From: Xuan Ding
This patch refactors vhost async enqueue path and dequeue path to use
the same function async_fill_seg() for preparing batch elements,
which simplies the code without performance degradation.
Signed-off-by: Xuan Ding
---
lib/vhost/virtio_net.c | 23 +++
1
From: Xuan Ding
This patches refactors copy_desc_to_mbuf() used by the sync
path to support both sync and async descriptor to mbuf filling.
Signed-off-by: Xuan Ding
---
lib/vhost/vhost.h | 1 +
lib/vhost/virtio_net.c | 47 --
2 files changed, 37
From: Xuan Ding
This patch implements asynchronous dequeue data path for vhost split
ring, a new API rte_vhost_async_try_dequeue_burst() is introduced.
Signed-off-by: Xuan Ding
Signed-off-by: Yuan Wang
---
doc/guides/prog_guide/vhost_lib.rst| 7 +
doc/guides/rel_notes/release_22_07.rst
From: Xuan Ding
This patch adds the use case for async dequeue API. Vswitch can
leverage DMA device to accelerate vhost async dequeue path.
Signed-off-by: Wenwu Ma
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/main.c
From: Xuan Ding
This patchset introduces an unsafe API to get the number of inflight
packets in DMA engine. It should be only used within the vhost ops
which already holds the lock. Like vring state changes or device is
destroyed. Compared with rte_vhost_async_get_inflight(), this is a
lock free
From: Xuan Ding
In async data path, when vring state changes or device is destroyed,
it is necessary to know the number of inflight packets in DMA engine.
This patch provides a thread unsafe API to return the number of
inflight packets for a vhost queue without using any lock.
Signed-off-by
From: Xuan Ding
In async data path, call rte_vhost_async_get_inflight_thread_unsafe()
API to directly return the number of inflight packets instead of
maintaining a local variable.
Signed-off-by: Xuan Ding
Reviewed-by: Maxime Coquelin
---
examples/vhost/main.c | 28
From: Xuan Ding
The presence of asynchronous path allows applications to offload memory
copies to DMA engine, so as to save CPU cycles and improve the copy
performance. This patch set implements vhost async dequeue data path
for split ring. The code is based on latest enqueue changes [1].
This
From: Xuan Ding
This patch extracts the descriptors to buffers filling from
copy_desc_to_mbuf() into a dedicated function. Besides, enqueue
and dequeue path are refactored to use the same function
sync_fill_seg() for preparing batch elements, which simplifies
the code without performance
From: Xuan Ding
This patch refactors vhost async enqueue path and dequeue path to use
the same function async_fill_seg() for preparing batch elements,
which simplifies the code without performance degradation.
Signed-off-by: Xuan Ding
---
lib/vhost/virtio_net.c | 23 +++
1
From: Xuan Ding
This patches refactors copy_desc_to_mbuf() used by the sync
path to support both sync and async descriptor to mbuf filling.
Signed-off-by: Xuan Ding
---
lib/vhost/vhost.h | 1 +
lib/vhost/virtio_net.c | 48 --
2 files changed, 38
From: Xuan Ding
This patch implements asynchronous dequeue data path for vhost split
ring, a new API rte_vhost_async_try_dequeue_burst() is introduced.
Signed-off-by: Xuan Ding
Signed-off-by: Yuan Wang
---
doc/guides/prog_guide/vhost_lib.rst| 7 +
doc/guides/rel_notes/release_22_07.rst
From: Xuan Ding
This patch adds the use case for async dequeue API. Vswitch can
leverage DMA device to accelerate vhost async dequeue path.
Signed-off-by: Wenwu Ma
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/main.c
From: Xuan Ding
Since QEMU 5.2.0 fixes the vhost split ring multi-queue reconnection
issue in commit f66337bdbfda ("vhost-user: save features of multiqueues
if chardev is closed"), this patch updates known issue to indicate
the range of affeacted QEMU versions.
Fixes: b37e95507e1b
From: Xuan Ding
The presence of asynchronous path allows applications to offload memory
copies to DMA engine, so as to save CPU cycles and improve the copy
performance. This patch set implements vhost async dequeue data path
for split ring. The code is based on latest enqueue changes [1].
This
From: Xuan Ding
This patch extracts the descriptors to buffers filling from
copy_desc_to_mbuf() into a dedicated function. Besides, enqueue
and dequeue path are refactored to use the same function
sync_fill_seg() for preparing batch elements, which simplifies
the code without performance
From: Xuan Ding
This patch refactors vhost async enqueue path and dequeue path to use
the same function async_fill_seg() for preparing batch elements,
which simplifies the code without performance degradation.
Signed-off-by: Xuan Ding
---
lib/vhost/virtio_net.c | 23 +++
1
From: Xuan Ding
This patches refactors copy_desc_to_mbuf() used by the sync
path to support both sync and async descriptor to mbuf filling.
Signed-off-by: Xuan Ding
---
lib/vhost/vhost.h | 1 +
lib/vhost/virtio_net.c | 48 --
2 files changed, 38
From: Xuan Ding
This patch implements asynchronous dequeue data path for vhost split
ring, a new API rte_vhost_async_try_dequeue_burst() is introduced.
Signed-off-by: Xuan Ding
Signed-off-by: Yuan Wang
---
doc/guides/prog_guide/vhost_lib.rst| 7 +
doc/guides/rel_notes/release_22_07.rst
From: Xuan Ding
This patch adds the use case for async dequeue API. Vswitch can
leverage DMA device to accelerate vhost async dequeue path.
Signed-off-by: Wenwu Ma
Signed-off-by: Yuan Wang
Signed-off-by: Xuan Ding
---
doc/guides/sample_app_ug/vhost.rst | 9 +-
examples/vhost/main.c
: add packed virtqueue defines")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
v2 changes:
* Renamed queue reset functions and moved them to virtqueue.c.
---
drivers/net/virtio/virtio_ethdev.c | 48 -
drivers/net/virtio/virtio_ethdev.h | 2 +
drivers/
: add packed virtqueue defines")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
v3:
* Removed an extra asterisk from a comment.
* Renamed device reset function and moved it to virtio_user_ethdev.c.
v2:
* Renamed queue reset functions and moved them to virtqueue.c.
---
drivers/net/virtio/virti
: add packed virtqueue defines")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
---
v4:
* Moved change log below '---' marker.
v3:
* Removed an extra asterisk from a comment.
* Renamed device reset function and moved it to virtio_user_ethdev.c.
v2:
* Renamed queue reset functio
acked virtqueue defines")
Cc: sta...@dpdk.org
Signed-off-by: Xuan Ding
---
v5:
* Fixed two spelling mistakes in the commit log.
* Added notice message when resetting vring.
v4:
* Moved change log below '---' marker.
v3:
* Removed an extra asterisk from a comment.
* Renamed devic
101 - 200 of 218 matches
Mail list logo
