This patch add the inflight message description and
the inflight share fd protocol feature flag.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
v1 - specify the APIs are split-ring only
v2 - fix APIs and judge split or packed
v3 - add rte_vhost_ p
This patch shows how to checkout the inflight ring and construct
the resubmit information also include destroying resubmit info.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost.h | 19
lib/librte_vhost/vhost.c
This patch introduces two new messages VHOST_USER_GET_INFLIGHT_FD
and VHOST_USER_SET_INFLIGHT_FD to support transferring a shared
buffer between qemu and backend.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/vhost.h | 7
This patches introduces two new messages VHOST_USER_GET_INFLIGHT_FD
and VHOST_USER_SET_INFLIGHT_FD to support transferring a shared
buffer between qemu and backend.
Now It can both support split and packed ring. The example code show
how these APIs work. The test has passed.
How to test the exampl
This patch introduces three APIs to operate the inflight
ring. Three APIs are set, set last and clear. It includes
split and packed ring.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost.h | 116 +++
lib/librt
This patch add the packed ring in the rte_vhost_vring.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost.h | 15 ---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/lib/librte_vhost/rte_vhost.h b
This patch fix two APIs so user can get the packed ring
support.
Fixes: fceec568 (vhost: add packed ring)
Cc: sta...@dpdk.org
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/vhost.c | 68 +---
This patch adds the inflight queue region structure include
the split and packed.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost.h | 43
1 file changed, 43 insertions(+)
diff --git
This patch introduces two APIs. one is for getting inflgiht
ring and the other is for getting base.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost.h | 40
lib/librte_vhost/vhost.c | 59 +
add all the added experimental APIs which are in
the rte_vhost.h to the rte_vhost_version.map
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost_version.map | 9 +
1 file changed, 9 insertions(+)
diff --git a/lib/l
This patch introduces an API for getting virtqueue format.
Signed-off-by: Lin Li
Signed-off-by: Xun Ni
Signed-off-by: Yu Zhang
Signed-off-by: Jin Yu
---
lib/librte_vhost/rte_vhost.h | 13 +
lib/librte_vhost/vhost.c | 12
2 files changed, 25 insertions(+)
diff --g
A vhost-user-blk example that support inflight feature. It uses the
new APIs that introduced in the first patch, so It can show how there
APIs work to support inflight feature.
Signed-off-by: Jin Yu
---
V1 - add the case.
V2 - add the rte_vhost prefix.
V3 - add packed ring support
---
examples/v
patch 1/5 minor code clean
patch 2/5 adds devargs to control pipeline mode
patch 3/5 rework for ice generic flow framework
patch 4/5 add rte_flow pattern list
patch 5/5 rework switch filter on new framework
The patchset depend on the following patches in patchwork:
(1) http://patches.dpdk.org/patc
The patch reworks the generic flow API (rte_flow) implementation.
It introduces an abstract layer which provides a unified interface
for low-level filter engine (switch, fdir, hash) to register supported
patterns and actions and implement flow validate/create/destroy/flush/
query activities.
The p
The patch removes redundant code and cleans up some wrong indentations.
Signed-off-by: Ying Wang
---
drivers/net/ice/ice_generic_flow.c | 39 +++---
1 file changed, 19 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ice/ice_generic_flow.c
b/drivers/net/i
From: wei zhao
The patch reworks packet process engine's binary classifier
(switch) for the new framework. It also adds support for new
packet type like PPPoE for switch filter.
Signed-off-by: Wei Zhao
---
doc/guides/rel_notes/release_19_11.rst |1 +
drivers/net/ice/ice_switch_filter.c
From: Qiming Yang
Added a devarg to control the mode in generic flow API.
We use none-pipeline mode by default.
Signed-off-by: Qiming Yang
---
doc/guides/nics/ice.rst| 19 +++
doc/guides/rel_notes/release_19_11.rst | 2 ++
drivers/net/ice/ice_ethdev.c
The patch adds a manifest for all patterns which can be selected
by low level filter engine to compose their supported list.
Signed-off-by: Ying Wang
---
drivers/net/ice/ice_generic_flow.c | 1287
drivers/net/ice/ice_generic_flow.h | 200 ++
2 files chan
Yes, I will do this way in v2.
-Original Message-
From: Wang, Xiao W
Sent: Thursday, September 12, 2019 3:42 PM
To: Pei, Andy ; dev@dpdk.org
Cc: Xu, Rosen ; Ye, Xiaolong ; Bie,
Tiwei ; Liang, Cunming
Subject: RE: [PATCH 3/4] vhost: call vDPA callback at the end of vring enable
handler
Even though the device reset is successful, disabling promiscuous mode
might not always succeed, causing enabling it after reset to
fail. This would happen when the kernel driver requires a reset of the
VF.
This patch resets the internal state, so next time promiscuous mode is
configured it will b
This patch series adds an option to make use of link time optimization
(if compiler has support for it). It is split as follows:
- 1st patch (build) is the enablement
- remaining patches are fixes for the warnings produced by the compiler
and they are split by directory/subsystem so their mainta
This patch adds an option to enable link time optimization. In addition
to LTO option itself (-flto) fat-lto-objects are being used. This is
because during the build pmdinfogen scans the generated ELF objects to
find this_pmd_name* symbol in symbol table. Without fat-lto-objects gcc
produces ELF
Fix the logic for the case of event queue allowing all schedule types.
Signed-off-by: Andrzej Ostruszka
---
lib/librte_eventdev/rte_event_timer_adapter.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/librte_eventdev/rte_event_timer_adapter.c
b/lib/librte_eventd
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
app/test-eventdev/test_perf_common.c | 2 +-
app/test-eventdev/test_pipeline_common.c | 4 ++--
2 file
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
drivers/event/octeontx2/otx2_tim_worker.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --gi
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
drivers/net/e1000/base/e1000_82543.c | 2 +-
drivers/net/e1000/base/e1000_ich8lan.c | 2 +-
drivers/net/
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
app/test/test_hash_readwrite.c | 2 +-
app/test/test_link_bonding_mode4.c | 10 --
app/test/t
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
drivers/net/i40e/i40e_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
drivers/net/ifc/ifcvf_vdpa.c | 14 +-
1 file changed, 9 insertions(+), 5 deletions(-)
diff --
During LTO build compiler reports some 'false positive' warnings about
variables being possibly used uninitialized. This patch silences these
warnings.
Signed-off-by: Andrzej Ostruszka
---
drivers/net/qede/base/ecore_mcp.c | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
dif
This patch fixes 'maybe-uninitialized' warnings reported by compiler
when using LTO.
Signed-off-by: Andrzej Ostruszka
---
drivers/net/dpaa2/base/dpaa2_hw_dpni.c | 1 +
drivers/net/dpaa2/mc/dpkg.c| 2 +-
drivers/net/dpaa2/mc/dpni.c| 9 ++---
3 files changed, 8 insertio
From: Haiyue Wang
Set the RXDID with flex descriptor type by default, change the Rx
function to support new descriptor handling.
Signed-off-by: Haiyue Wang
---
drivers/net/ice/ice_rxtx.c | 236 +
1 file changed, 111 insertions(+), 125 deletions(-)
diff --gi
From: Haiyue Wang
The ice has the feature to extract protocol fields into flex descriptor
by programming per queue. Currently, the ice PMD will put the protocol
fields into rte_mbuf::udata64 with different type format. Application
can access the protocol fields quickly.
Signed-off-by: Haiyue Wan
This patchset enable Rx flexible descriptor for ice PMD
in both normal path and vector path.
Depends on shared code update patchset.
---
v3:
- Parse the ‘proto_xtr’ devargs before CVL HW initialization and
save it for later accessing.
- Merge shuffle ops on vlan and rss flag on vector path.
v2:
From: Haiyue Wang
The Rx flex descriptor has 16B and 32B size, with different
field definitions compared to legacy type.
Signed-off-by: Haiyue Wang
---
drivers/net/ice/ice_rxtx.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ice/ice_rxtx.h b/drivers/net/ice/ice_rxtx.h
index
From: Wenzhuo Lu
With this path, the flexible descriptor is supported
in SSE path. And the legacy descriptor is not supported.
Signed-off-by: Wenzhuo Lu
---
drivers/net/ice/ice_rxtx_vec_sse.c | 239 +
1 file changed, 110 insertions(+), 129 deletions(-)
diff --git a
Switch to Rx flexible descriptor format instead of legacy
descriptor format.
Signed-off-by: Leyi Rong
---
drivers/net/ice/ice_rxtx_vec_avx2.c | 224 ++--
1 file changed, 109 insertions(+), 115 deletions(-)
diff --git a/drivers/net/ice/ice_rxtx_vec_avx2.c
b/drivers/net/i
>
> The kernel only consumes Tx packets if we have some Rx traffic on specified
> queue or we have called send(). So we need to issue a send() even when the
> allocation fails so that kernel will start to consume packets again.
>
> Commit 45bba02c95b0 ("net/af_xdp: support need wakeup feature") b
> Let's assume port already stopped before calling the reset, reset will cause
> port
> to be started.
Indeed, automatically start looks strange in this case. User explicitly do the
"port start" command should be better.
> I am for user explicitly do the stop, reset and start commands, instead
This is in preparation for multiqueue enabling for vDPA devices.
Signed-off-by: Xiaolong Ye
Signed-off-by: Andy Pei
---
drivers/net/ifc/base/ifcvf.c | 1 +
drivers/net/ifc/base/ifcvf.h | 2 ++
2 files changed, 3 insertions(+)
diff --git a/drivers/net/ifc/base/ifcvf.c b/drivers/net/ifc/base/ifc
vDPA's set_vring_state callback would need to know the virtqueues'
enable status to configure the hardware.
Signed-off-by: Xiaolong Ye
Signed-off-by: Andy Pei
---
v2:
add nr_active_vring as a parameter to ops function set_vring_state in
case of callback in set_vring_state() and avoid exposing
Enable mutliqueue support for ifcvf devices by setting the mutliqueue
configuration space.
Signed-off-by: Xiaolong Ye
Signed-off-by: Andy Pei
---
drivers/net/ifc/base/ifcvf.c | 8
drivers/net/ifc/base/ifcvf.h | 5 -
drivers/net/ifc/ifcvf_vdpa.c | 30 --
Hi Aaron,
I try to solve this by avoiding exposing this new API.
Hope this works.
BRs,
-Original Message-
From: Aaron Conole [mailto:acon...@redhat.com]
Sent: Friday, September 6, 2019 9:26 PM
To: Pei, Andy
Cc: dev@dpdk.org; Xu, Rosen ; Ye, Xiaolong
; Bie, Tiwei ; Wang, Xiao W
Subje
Hello,
Do you have a reference for the patches for bug 1 and 2? Have they been
sent to the mailing list?
On Thu, 2019-09-12 at 08:52 +, Peng, Yuan wrote:
> Update the latest test status here.
>
> -Original Message-
> From: Yu, PingX
> Sent: Thursday, September 12, 2019 2:37 PM
> To:
On 9/5/19 3:25 PM, Chas Williams wrote:
[...]
>> @@ -578,7 +578,9 @@ bond_get_update_timeout_ms(void)
>> {
>> struct rte_eth_bond_8023ad_conf conf;
>>
>> -rte_eth_bond_8023ad_conf_get(test_params.bonded_port_id, &conf);
>> +if (rte_eth_bond_8023ad_conf_get(test_params.bonded_port_
This patch fixes the aesni-gcm cryptodev documentation by
filling the lacked unsupported chained mbuf description.
Fixes: 6f16aab09a91 ("crypto/aesni_gcm: migrate to Multi-buffer library")
Cc: pablo.de.lara.gua...@intel.com
Cc: sta...@dpdk.org
Signed-off-by: Fan Zhang
---
doc/guides/cryptodevs/
Ting Xu writes:
> This patch adds the feature that supports loading DDP package
> according to the device serial number. Prior to loading the
> default DDP package (ice.pkg), the driver will check for the
> presence of a device-specific DDP package with the name containing
> 64-bit PCIe Device Se
JinYu writes:
> This patch introduces two new messages VHOST_USER_GET_INFLIGHT_FD
> and VHOST_USER_SET_INFLIGHT_FD to support transferring a shared
> buffer between qemu and backend.
>
> Signed-off-by: Lin Li
> Signed-off-by: Xun Ni
> Signed-off-by: Yu Zhang
> Signed-off-by: Jin Yu
> ---
> l
Add pointer null check before dereference.
Coverity issue: 344976
Fixes: 12f92a51 ("raw/ifpga/base: fix retimer link status")
Cc: sta...@dpdk.org
Signed-off-by: Tianfei zhang
---
drivers/raw/ifpga/base/opae_eth_group.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/driver
When IOMMU is enabled the incoming log address is in IOVA space. In that
case, look in IOTLB table and translate the resulting HVA to GPA.
If IOMMU is not enabled, the incoming log address is already a GPA so no
transformation is needed.
This change makes page logging work when IOVA_VA is selecte
Adding relevant maintainers. Sorry for the slip
Adrian
On 9/6/19 2:50 PM, Adrian Moreno wrote:
> Only the mapping of the vring addresses is being ensured. This causes
> errors when the vring size is larger than the IOTLB page size. E.g:
> queue sizes > 256 for 4K IOTLB pages
>
> Ensure the entir
CCing Bruce, maintainer of the meson build system.
It's a simple patch to make meson behavior consistent with make behavior.
CONFIG_RTE_KNI_PREEMPT_DEFAULT is set to "y" in config/common_base,
so we need to set RTE_KNI_PREEMPT_DEFAULT to 1 in meson.
Best regards,
Igor
On Mon, Sep 16, 2019 at 1:
On Tue, Sep 17, 2019 at 10:52:24PM +0800, JinYu wrote:
> This patch add the inflight message description and
> the inflight share fd protocol feature flag.
>
> Signed-off-by: Lin Li
> Signed-off-by: Xun Ni
> Signed-off-by: Yu Zhang
> Signed-off-by: Jin Yu
> ---
Please keep Maxime's reviewed-b
> -Original Message-
> From: Bie, Tiwei
> Sent: Wednesday, September 18, 2019 9:30 AM
> To: Yu, Jin
> Cc: dev@dpdk.org; Liu, Changpeng ;
> maxime.coque...@redhat.com; Wang, Zhihong ;
> Lin Li ; Xun Ni ; Yu Zhang
>
> Subject: Re: [PATCH v7 01/11] vhost: add the inflight description
>
>
On Wed, Sep 11, 2019 at 12:14:45AM +0800, Marvin Liu wrote:
> When doing xmit in-order enqueue, packets are buffered and then flushed
> into avail ring. Buffered packets can be dropped due to insufficient
> space. Moving stats update action just after successful avail ring
> updates can guarantee c
On Wed, Sep 11, 2019 at 12:14:46AM +0800, Marvin Liu wrote:
> Check whether space are enough before burst enqueue operation. If more
> space is needed, will try to cleanup used descriptors for space on
> demand. It can give more chances to free used descriptors, thus will
> help RFC2544 performance
> -Original Message-
> From: Bie, Tiwei
> Sent: Wednesday, September 18, 2019 10:35 AM
> To: Liu, Yong
> Cc: maxime.coque...@redhat.com; dev@dpdk.org
> Subject: Re: [PATCH v2 1/2] net/virtio: update stats when in order xmit
> done
>
> On Wed, Sep 11, 2019 at 12:14:45AM +0800, Marvin Liu
Thanks for comments, will update in next patch.
> -Original Message-
> From: Bie, Tiwei
> Sent: Wednesday, September 18, 2019 10:44 AM
> To: Liu, Yong
> Cc: maxime.coque...@redhat.com; dev@dpdk.org
> Subject: Re: [PATCH v2 2/2] net/virtio: on demand cleanup when doing in
> order xmit
>
>
On Wed, Sep 18, 2019 at 11:19:03AM +0800, Liu, Yong wrote:
> > -Original Message-
> > From: Bie, Tiwei
> > Sent: Wednesday, September 18, 2019 10:35 AM
> > To: Liu, Yong
> > Cc: maxime.coque...@redhat.com; dev@dpdk.org
> > Subject: Re: [PATCH v2 1/2] net/virtio: update stats when in order
Explicitly document that the MAC/VLAN filtering in virtio
is best effort to help users understand why unwanted packets
could still arrive.
Signed-off-by: Tiwei Bie
---
doc/guides/nics/virtio.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/guides/nics/virtio.rst b/doc/
Hi Joyce,
I test performance impact of your patch set with code base commit id:
d03d8622db48918d14bfe805641b1766ecc40088, after applying your v4 patch set ,
packed ring shows small performance drop as below:
PVP vhost/virtio 1c1q test commit:d03d8622db48918 apply v4 patch
set
On Tue, Sep 17, 2019 at 10:52:31PM +0800, JinYu wrote:
> This patch fix two APIs so user can get the packed ring
> support.
>
> Fixes: fceec568 (vhost: add packed ring)
> Cc: sta...@dpdk.org
I changed my mind, as the packed ring related fields are just
added to rte_vhost_vring in this series, we
On Tue, Sep 17, 2019 at 10:52:33PM +0800, JinYu wrote:
> add all the added experimental APIs which are in
> the rte_vhost.h to the rte_vhost_version.map
Symbols should be added to the .map file in the same
patch which introduces them. Otherwise that patch will
break the build.
Thanks,
Tiwei
>
>
> -Original Message-
> From: Bie, Tiwei
> Sent: Wednesday, September 18, 2019 2:05 PM
> To: Yu, Jin
> Cc: dev@dpdk.org; Liu, Changpeng ;
> maxime.coque...@redhat.com; Wang, Zhihong ;
> Lin Li ; Xun Ni ; Yu Zhang
>
> Subject: Re: [PATCH v7 10/11] vhost: add APIs to rte_vhost version map
>
Hi Marcin,
Sorry for the late response. But how do you plan to handle "inline protocol"
processed packets?
Thanks,
Anoob
> -Original Message-
> From: dev On Behalf Of Marcin Smoczynski
> Sent: Wednesday, September 4, 2019 7:47 PM
> To: konstantin.anan...@intel.com; akhil.go...@nxp.com
The rdma core library can map doorbell register in two ways,
depending on the environment variable "MLX5_SHUT_UP_BF":
- as regular cached memory, the variable is either missing or
set to zero. This type of mapping may cause the significant
doorbell register writing latency and requires e
The txq_uar_init() routine uses the uninitialized uar_mmap_offset
field in 32-bit configurations due to this field is initialized
after txq_uar_init() call.
Fixes: 120dc4a7dcd3 ("net/mlx5: remove device register remap")
Signed-off-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_txq.c | 2 +-
67 matches
Mail list logo
