Hi,
> -Original Message-
> From: dev On Behalf Of Raslan Darawsheh
> Sent: Monday, October 26, 2020 1:42 PM
> To: dev@dpdk.org
> Cc: Matan Azrad ; Slava Ovsiienko
> ; Shy Shyman
> Subject: [dpdk-dev] [PATCH] mlx5: add ConnectX-7 and Bluefield-3 device
> IDs
>
> This adds the ConnectX-7
Hi
I am trying to build dpdk-next-net code based Kunpeng920(ARM64) platform used
meson build. However, it is build fail.
The libtmp_rte_event_octeontx2 build error happened. The output as follows:
[1986/2298] Compiling C object
drivers/libtmp_rte_event_octeontx2.a.p/event_octeontx2_otx2_tim_
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Tuesday, October 27, 2020 4:54 AM
> To: Juraj Linkeš ; bruce.richard...@intel.com;
> Ruifeng Wang ; Phil Yang ;
> vcchu...@amazon.com; Dharmik Thakkar ;
> jerinjac...@gmail.com; hemant.agra...@nxp.com
> Cc: dev@dpdk.org; nd ; Honn
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Tuesday, October 27, 2020 4:54 AM
> To: Juraj Linkeš ; bruce.richard...@intel.com;
> Ruifeng Wang ; Phil Yang ;
> vcchu...@amazon.com; Dharmik Thakkar ;
> jerinjac...@gmail.com; hemant.agra...@nxp.com
> Cc: dev@dpdk.org; nd ; Honn
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Tuesday, October 27, 2020 5:56 AM
> To: Juraj Linkeš ; bruce.richard...@intel.com;
> Ruifeng Wang ; Phil Yang ;
> vcchu...@amazon.com; Dharmik Thakkar ;
> jerinjac...@gmail.com; hemant.agra...@nxp.com
> Cc: dev@dpdk.org; nd ; Honn
> -Original Message-
> From: Wang, Haiyue
> Sent: Tuesday, October 27, 2020 1:22 PM
> To: Guo, Jia ; Wu, Jingjing ;
> Zhang, Qi Z ; Xing, Beilei ;
> Yigit, Ferruh
> Cc: dev@dpdk.org; Richardson, Bruce
> Subject: RE: [PATCH v10] net/iavf: support flex desc metadata extraction
>
> > ---
Add the needed fields for virtq DevX object to read the error state.
Acked-by: Matan Azrad
Signed-off-by: Xueming Li
---
drivers/common/mlx5/mlx5_devx_cmds.c | 3 +++
drivers/common/mlx5/mlx5_devx_cmds.h | 1 +
drivers/common/mlx5/mlx5_prm.h | 9 +++--
3 files changed, 11 insertions(+
When hardware error happens, vdpa didn't get such information and leave
driver in silent: working state but no response.
This patch subscribes firmware virtq error event and try to recover max
3 times in 3 seconds, stop virtq if max retry number reached.
When error happens, PMD log in warning lev
On Tue, Oct 27, 2020 at 8:43 AM oulijun wrote:
>
> Hi
> I am trying to build dpdk-next-net code based Kunpeng920(ARM64) platform
> used meson build. However, it is build fail.
> The libtmp_rte_event_octeontx2 build error happened. The output as follows:
>
> [1986/2298] Compiling C object
> d
Hi,
> -Original Message-
> From: dev On Behalf Of Leyi Rong
> Sent: Friday, October 23, 2020 7:14 AM
> To: qi.z.zh...@intel.com; wenzhuo...@intel.com;
> burce.richard...@intel.com
> Cc: dev@dpdk.org; Leyi Rong ; Bruce Richardson
>
> Subject: [dpdk-dev] [PATCH v4 1/3] net/ice: add AVX512
> -Original Message-
> From: dev On Behalf Of Ali Alnubani
> Sent: Tuesday, October 27, 2020 10:33 AM
> To: Leyi Rong ; qi.z.zh...@intel.com;
> wenzhuo...@intel.com; burce.richard...@intel.com
> Cc: dev@dpdk.org; Bruce Richardson
> Subject: Re: [dpdk-dev] [PATCH v4 1/3] net/ice: add AVX51
On Mon, Oct 26, 2020 at 3:20 PM Thomas Monjalon wrote:
>
> The CFA code from mark was stored in the deprecated mbuf field udata64.
> It is moved to a dynamic field in order to allow removal of udata64.
>
> Note: the new field has 32 bits, smaller than udata64.
>
> Signed-off-by: Thomas Monjalon
27/10/2020 05:18, Dharmik Thakkar:
>
> > On Oct 22, 2020, at 4:25 PM, Honnappa Nagarahalli
> > wrote:
> >
> > + Dharmik
> >
> > From: Thomas Monjalon
> >> 14/10/2020 11:24, Ruifeng Wang:
> >>> +* Build DPDK with meson option ``-
> >> Darmv8_crypto_dir=``:
> >>> +
> >>> +meson -Darmv8_crypto_d
On Mon, Oct 26, 2020 at 5:40 PM Andrew Rybchenko
wrote:
> Is it a known issue that dpdk-next-net build is broken on Ubuntu 18.04?
I can't reproduce it in Travis, but I noticed the meson part of the
ice driver needed adjustments.
http://inbox.dpdk.org/dev/dm6pr12mb461871dfaa1d279933f9356bda...@dm6
27/10/2020 03:46, wangyunjian:
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > 26/10/2020 13:33, Andrew Rybchenko:
> > > On 10/26/20 3:29 PM, Thomas Monjalon wrote:
> > > > 26/10/2020 13:24, wangyunjian:
> > > >> From: Yunjian Wang
> > > >>
> > > >> The ethdev port id should be 16 bits no
On Mon, Oct 26, 2020 at 3:56 PM Ferruh Yigit wrote:
> Series applied to dpdk-next-net/main, thanks.
>
>
>
> 57/58 & 58/58 has been distributed to the relevant commits while merging,
> please
> double check them in the upstream repo.
>
> 56/58 & 53/58 has been dropped while merging.
> 56/58 need
@Ferruh: this patch is tested with both PIO and MMIO bar using testpmd
and start tx_first.
vfio/igb_uio tested with MMIO bar (uio_pci_generic doesn't work with
msix, so it isn't tested)
uio_pci_generic tested with PIO bar (igb_uio has unknown symbols, not
tested).
Weird igb_uio doens't hav
27/10/2020 03:01, Wang, Haiyue:
> From: Thomas Monjalon
> For ixgbe PMD,
>
> Acked-by: Haiyue Wang
>
> But I feel that 'rte_security_dynfield' name is too generic, can it be
> more specific about what the field is used for ?
>
> Like below ;-)
>
> #define RTE_SECURITY_DEV_METADATA(m) \
>
Hi,
> -Original Message-
> From: dev On Behalf Of Andrew Rybchenko
> Sent: Tuesday, October 20, 2020 11:48 AM
> To: y...@solarflare.com
> Cc: dev@dpdk.org; Ivan Malov
> Subject: [dpdk-dev] [PATCH 14/62] common/sfc_efx/base: add action set
> spec init/fini APIs
>
> From: Ivan Malov
>
>
This patch initializes a local parameter in async data path to avoid
compiler warnings.
Fixes: cd6760da1076 ("vhost: introduce async enqueue for split ring")
Signed-off-by: Patrick Fu
---
lib/librte_vhost/virtio_net.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librt
> -Original Message-
> From: dev On Behalf Of Juraj Linkeš
> Sent: Tuesday, October 27, 2020 9:11 AM
> To: Honnappa Nagarahalli ;
> bruce.richard...@intel.com; Ruifeng Wang ; Phil
> Yang ; vcchu...@amazon.com; Dharmik Thakkar
> ; jerinjac...@gmail.com;
> hemant.agra...@nxp.com
> Cc: dev@
Hi,
Sent my report to v1 the first time by mistake.
> -Original Message-
> From: dev On Behalf Of Andrew Rybchenko
> Sent: Tuesday, October 20, 2020 12:13 PM
> To: dev@dpdk.org
> Cc: Ivan Malov
> Subject: [dpdk-dev] [PATCH v2 14/62] common/sfc_efx/base: add action set
> spec init/fini AP
Hi,
>
> Acked-by: Ori Kam
> Thanks,
> Ori
Kind reminder to all maintainers.
Hi,
> -Original Message-
> From: g...@marvell.com
> Sent: Thursday, October 22, 2020 10:11 AM
> To: or...@nvidia.com
> Cc: tho...@monjalon.net; Guy Kaneti ; Smadar Fuks
> ; Dovrat Zifroni ;
> dev@dpdk.org
> Subject: [PATCH v1] app/regex: add job context
>
> From: Guy Kaneti
>
> Store m
update the profile_ov config file to support dynamic
update of subport rate.
Signed-off-by: Savinay Dharmappa
---
v2: renamed the commit subject.
---
examples/qos_sched/profile_ov.cfg | 3 +++
1 file changed, 3 insertions(+)
diff --git a/examples/qos_sched/profile_ov.cfg
b/examples/qos_sched/p
Hi,
This patchset update the bcmfs.rst file.
Regards,
Vikas
Vikas Gupta (1):
crypto/bcmfs: update documentation file
doc/guides/cryptodevs/bcmfs.rst | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
--
2.17.1
Update bcmfs.rst file with supported features and devices.
Signed-off-by: Vikas Gupta
Reviewed-by: Ajit Khaparde
Reviewed-by: JP Lee
---
doc/guides/cryptodevs/bcmfs.rst | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/doc/guides/cryptodevs/bcmfs.rst b/doc/gui
Hi Thomas,
On Mon, Oct 26, 2020 at 11:20:02PM +0100, Thomas Monjalon wrote:
> The node_mbuf_priv1 was stored in the deprecated mbuf field udata64.
> It is moved to a dynamic field in order to allow removal of udata64.
>
> Signed-off-by: Thomas Monjalon
> ---
> lib/librte_node/ip4_lookup.c |
27/10/2020 10:32, Olivier Matz:
> On Mon, Oct 26, 2020 at 11:20:02PM +0100, Thomas Monjalon wrote:
> > +static const struct rte_mbuf_dynfield node_mbuf_priv1_dynfield_desc = {
> > + .name = "rte_node_dynfield_priv1",
> > + .size = sizeof(struct node_mbuf_priv1 *),
> > + .align = __alignof__(s
> -Original Message-
> From: Dharmappa, Savinay
> Sent: Tuesday, October 27, 2020 9:22 AM
> To: Dumitrescu, Cristian ; Singh, Jasvinder
> ; dev@dpdk.org
> Cc: Dharmappa, Savinay
> Subject: [PATCH v2] examples/qos_sched: update profile oversubscribe
> config file
>
> update the profile
On Mon, Oct 26, 2020 at 11:20:03PM +0100, Thomas Monjalon wrote:
> The device-specific metadata was stored in the deprecated field udata64.
> It is moved to a dynamic mbuf field in order to allow removal of udata64.
>
> Signed-off-by: Thomas Monjalon
> ---
> doc/guides/prog_guide/rte_security.rs
From: Alvin Zhang
Update reading offload flags of last two of four packets.
Fixes: 1162f5a0ef31 ("net/iavf: support flexible Rx descriptor in SSE path")
Cc: sta...@dpdk.org
Signed-off-by: Alvin Zhang
---
drivers/net/iavf/iavf_rxtx_vec_sse.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
On Mon, Oct 26, 2020 at 11:20:04PM +0100, Thomas Monjalon wrote:
> The test worker_loopback used the deprecated mbuf field udata64.
> It is moved to a dynamic field in order to allow removal of udata64.
>
> Signed-off-by: Thomas Monjalon
> Reviewed-by: Andrew Rybchenko
> ---
> drivers/event/sw/
This commit replaces mlx5_malloc and mlx5_free calls with Linux calls
malloc and free in file mlx5_glue.c.
The current mlx5_malloc calls have no flags, alignment or socket
selection, so they are equivalent to calling malloc. Rdma core itself
is using malloc. When using mlx5_malloc the glue librar
On Mon, Oct 26, 2020 at 09:09:31AM +0100, David Marchand wrote:
> On Mon, Oct 26, 2020 at 8:13 AM Rong, Leyi wrote:
> > Would prefer using __AVX512F__ instead of RTE_MACHINE_CPUFLAG_AVX512F here
> > rather than remove the RTE_MACHINE_CPUFLAG_ macro directly to check the CPU
> > capability.
> > S
On 10/26/2020 7:12 AM, Rong, Leyi wrote:
-Original Message-
From: David Marchand
Sent: Monday, October 26, 2020 12:24 AM
To: Rong, Leyi ; Zhang, Qi Z ; Yigit,
Ferruh
Cc: Lu, Wenzhuo ; dev ; Richardson,
Bruce ; Thomas Monjalon
Subject: Re: [dpdk-dev] [PATCH v4 1/3] net/ice: add AVX512
On Mon, Sep 14, 2020 at 01:05:11PM +0200, Morten Brørup wrote:
> Updated description of rte_eth_rx_burst() to reflect what drivers,
> when using vector instructions, expect from nb_pkts.
>
> Also discussed on the mailing list here:
> http://inbox.dpdk.org/dev/98cbd80474fa8b44bf855df32c47dc35c61...
On Fri, Oct 23, 2020 at 10:07:17AM +, Juraj Linkeš wrote:
>
>
> > -Original Message-
> > From: Bruce Richardson
> > Sent: Wednesday, October 21, 2020 4:28 PM
> > To: Juraj Linkeš
> > Cc: ruifeng.w...@arm.com; honnappa.nagaraha...@arm.com;
> > phil.y...@arm.com; vcchu...@amazon.com;
On Mon, Oct 26, 2020 at 11:20:06PM +0100, Thomas Monjalon wrote:
> The CFA code from mark was stored in the deprecated mbuf field udata64.
> It is moved to a dynamic field in order to allow removal of udata64.
>
> Note: the new field has 32 bits, smaller than udata64.
>
> Signed-off-by: Thomas Mo
Intrinsic function __mm512_bsrli_epi128 should be used in
the environment which supports AVX512BW, so adds check for
this flag.
Fixes: 5dd3b8f3af34 ("net/ice: add AVX512 vector path")
Signed-off-by: Leyi Rong
---
drivers/net/ice/ice_rxtx.c | 12 ++--
drivers/net/ice/meson.build | 13 ++
On Mon, Oct 26, 2020 at 11:20:07PM +0100, Thomas Monjalon wrote:
> The segment count, used for MSS guess,
> was stored in the deprecated mbuf field udata64.
> It is moved to a dynamic field in order to allow removal of udata64.
>
> Signed-off-by: Thomas Monjalon
After seeing this commit, I wonde
On 10/27/2020 6:23 AM, Jiawen Wu wrote:
Supports check the status of Rx and Tx descriptors.
Signed-off-by: Jiawen Wu
for series,
Reviewed-by: Ferruh Yigit
Series applied to dpdk-next-net/main, thanks.
On 10/27/2020 6:24 AM, Jiawen Wu wrote:
Replace the 'rte_panic()' with an error return.
Also change the type of the calling function.
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/txgbe_ethdev.h | 2 +-
drivers/net/txgbe/txgbe_pf.c | 12
2 files changed, 9 insertions(+),
On 10/27/2020 6:24 AM, Jiawen Wu wrote:
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/{rte_pmd_txgbe_version.map => version.map} | 0
1 file changed, 0 insertions(+), 0 deletions(-)
rename drivers/net/txgbe/{rte_pmd_txgbe_version.map => version.map} (100%)
diff --git a/drivers/net/txgbe/r
On Mon, Oct 26, 2020 at 11:20:13PM +0100, Thomas Monjalon wrote:
> As announced in the deprecation note, the field userdata / udata64
> is removed to give more space to the dynamic fields.
>
> Signed-off-by: Thomas Monjalon
Acked-by: Olivier Matz
> -Original Message-
> From: David Marchand
> Sent: Thursday, October 22, 2020 9:08 AM
> To: dev@dpdk.org
> Cc: Van Haaren, Harry
> Subject: [PATCH] service: separate dump and reset code
>
> No functional change intended.
>
> service_dump_calls_per_lcore() was always called with a 0 res
On 10/26/2020 5:38 PM, Slava Ovsiienko wrote:
Hi, Ferruh
PSB
-Original Message-
From: Ferruh Yigit
Sent: Monday, October 26, 2020 19:04
To: Raslan Darawsheh ; Slava Ovsiienko
; dev@dpdk.org
Cc: NBU-Contact-Thomas Monjalon ; Matan Azrad
; Alexander Kozyrev ; Ori Kam
Subject: Re: [dpdk
Hi Thomas,
all is fixed in v10.
Regards
Liang
On 24 Oct 22:49, Thomas Monjalon wrote:
> 24/10/2020 01:06, Liang Ma:
> > Update programmer's guide to document PMD power management usage.
> >
> > Signed-off-by: Liang Ma
> > Signed-off-by: Anatoly Burakov
> > Acked-by: David Hunt
> > ---
> >
thanks for your information.
Sorry for that.
All related maintainer(include other NIC PMD) will be Cced in v10.
> You keep forgetting Cc ethdev maintainers
> (it is automatic when using --cc-cmd devtools/get-maintainer.sh).
> As a result we still don't have any feedback from Ferruh and Andrew.
On Fri, Oct 23, 2020 at 04:48:05PM +0200, Juraj Linkeš wrote:
> Add an option to automatically discover the host's numa and cpu counts
> and use those values for a non cross-build.
> Give users the option to override the per-arch default values or values
> from cross files by specifying them on the
On Mon, Oct 26, 2020 at 06:20:51AM +0100, Thomas Monjalon wrote:
> In examples, DPDK header files are external,
> so they must be enclosed with angle brackets, not quotes.
>
> Signed-off-by: Thomas Monjalon
> ---
> examples/bbdev_app/main.c | 30 +-
> example
On Mon, Oct 26, 2020 at 06:20:52AM +0100, Thomas Monjalon wrote:
> Since the kernel module is not part of EAL anymore,
> there is no need to have the common KNI header file in EAL.
> The file rte_kni_common.h is moved to librte_kni.
>
> Signed-off-by: Thomas Monjalon
> ---
> kernel/linux/kni/mes
On 10/27/2020 8:48 AM, David Marchand wrote:
On Mon, Oct 26, 2020 at 3:56 PM Ferruh Yigit wrote:
Series applied to dpdk-next-net/main, thanks.
57/58 & 58/58 has been distributed to the relevant commits while merging, please
double check them in the upstream repo.
56/58 & 53/58 has been drop
On 10/27/2020 2:39 AM, Jiawen Wu wrote:
On Monday, October 26, 2020 10:56 PM, Ferruh Yigit wrote:
On 10/19/2020 9:53 AM, Jiawen Wu wrote:
v4: Add log type introduction,
add txgbe support in release note.
v3: Merge some recent changes in the main repo,
fix the checkpatch warnings.
v2
On 10/27/2020 9:13 AM, Ali Alnubani wrote:
Hi,
Sent my report to v1 the first time by mistake.
-Original Message-
From: dev On Behalf Of Andrew Rybchenko
Sent: Tuesday, October 20, 2020 12:13 PM
To: dev@dpdk.org
Cc: Ivan Malov
Subject: [dpdk-dev] [PATCH v2 14/62] common/sfc_efx/base:
On Tue, Oct 27, 2020 at 12:36 PM Ferruh Yigit wrote:
> I will fix it in the nex-net as following:
>
> diff --git a/doc/guides/nics/index.rst b/doc/guides/nics/index.rst
> index da5d85bcd0..3443617755 100644
> --- a/doc/guides/nics/index.rst
> +++ b/doc/guides/nics/index.rst
> @@ -57,6 +5
On Mon, Oct 26, 2020 at 8:35 PM Ray Kinsella wrote:
>
> Updated references to abi versions in the contributors guide.
> Fixed an inaccurate reference to a symbol in the policy.
>
> Signed-off-by: Ray Kinsella
> Reviewed-by: David Marchand
Applied, thanks Ray.
--
David Marchand
On Mon, Oct 26, 2020 at 03:30:27PM +0100, Felix Moessbauer wrote:
> This patch introduces a meson option to disable the AVX2 support.
> If the build should be for a target without AVX2 support, the
> know can be turned to false, even if the compiler supports AVX2.
>
> Signed-off-by: Felix Moessbau
Hi Bruce,
Yes it is for pktgen. I just updated the title (hope that this will be picked
up by patchwork).
BTW: That's something to mention in the contributing guide of Pktgen as well.
Best regards,
Felix
> -Original Message-
> From: Bruce Richardson
> Sent: Tuesday, October 27, 2020 12
> -Original Message-
> From: Wang, Haiyue
> Sent: Tuesday, October 27, 2020 1:22 PM
> To: Guo, Jia ; Wu, Jingjing ;
> Zhang, Qi Z ; Xing, Beilei ;
> Yigit,
> Ferruh
> Cc: dev@dpdk.org; Richardson, Bruce
> Subject: RE: [PATCH v10] net/iavf: support flex desc metadata extraction
>
> >
Hello Felix,
On Tue, Oct 27, 2020 at 12:51 PM Moessbauer, Felix
wrote:
>
> Hi Bruce,
>
> Yes it is for pktgen. I just updated the title (hope that this will be picked
> up by patchwork).
> BTW: That's something to mention in the contributing guide of Pktgen as well.
Afaik, patchwork won't catch
On 10/27/2020 10:19 AM, Leyi Rong wrote:
Intrinsic function __mm512_bsrli_epi128 should be used in
the environment which supports AVX512BW, so adds check for
this flag.
Fixes: 5dd3b8f3af34 ("net/ice: add AVX512 vector path")
Signed-off-by: Leyi Rong
Squashed into relevant commit in next-net,
On 10/27/20 2:39 PM, Ferruh Yigit wrote:
> On 10/27/2020 9:13 AM, Ali Alnubani wrote:
>> Hi,
>> Sent my report to v1 the first time by mistake.
>>
>>> -Original Message-
>>> From: dev On Behalf Of Andrew Rybchenko
>>> Sent: Tuesday, October 20, 2020 12:13 PM
>>> To: dev@dpdk.org
>>> Cc: Iv
Hi,
> -Original Message-
> From: Li Zhang
> Sent: Monday, September 28, 2020 9:56 AM
> To: Dekel Peled ; Ori Kam ; Slava
> Ovsiienko ; Matan Azrad
> Cc: dev@dpdk.org; NBU-Contact-Thomas Monjalon
> ; Raslan Darawsheh
> Subject: [PATCH V1 1/1] net/mlx5: support item type error message in
Remove duplicate typedef for efx_mae_actions_t.
Fixes: 57f9eb33d32c ("common/sfc_efx/base: add action set spec init/fini APIs")
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/base/efx_impl.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/common/sfc_e
On Mon, Oct 19, 2020 at 11:42 AM David Marchand
wrote:
>
> The glue libraries are tightly bound to the mlx drivers of a dpdk version
> and are packaged with them.
>
> Keeping a separate ABI version prevents us from installing two versions of
> dpdk.
> Maintaining this separate version just adds co
On Thu, Oct 22, 2020 at 10:08 AM David Marchand
wrote:
>
> No functional change intended.
>
> service_dump_calls_per_lcore() was always called with a 0 reset flag.
> service_dump_one() was called with either a 0 reset flag or a NULL
> FILE pointer.
>
> We can split the code for readability sake.
>
This patch set contains multiple-thread flow operations support
for the flow objects.
Suanming Mou (19):
net/mlx5: use thread safe index pool for flow objects
net/mlx5: make meter action thread safe
net/mlx5: make VLAN network interface thread safe
net/mlx5: create global jump action
net
As mlx5 PMD is changed to be thread safe, all the flow-related
sub-objects inside the PMD should be thread safe. This commit
changes the index memory pools' lock configuration to be enabled.
That makes the index pool be thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/
From: Xueming Li
To make indexed pool to be used as ID generator, this patch allows entry
size to be zero.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5_utils.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_utils.c b/
From: Xueming Li
As part of multi-thread flow support, this patch moves flow intermediate
data to thread specific, makes them a flow workspace. The workspace is
allocated per thread, destroyed along with thread life-cycle.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/l
From: Xueming Li
The ID generation API used an integer pool to save released ID, To
support multiple flow, it has to be enhanced to be thread safe.
Indexed pool could be used to generate unique ID by setting size of pool
entry to zero. Since bitmap is used, an extra benefits is saving memory
to
From: Xueming Li
Hairpin flow matching required a unique flow ID for matching.
This patch reuses flow ID as hairpin flow ID, this will save some code
to generate a separate hairpin ID, also saves flow memory by removing
hairpin ID.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/n
From: Xueming Li
To support multi-thread flow operations, this patch introduces list lock
for the rte_flow list manages all the rte_flow handlers.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 1 +
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/
This commit adds the spinlock for the meter action to make it be thread
safe. Atomic reference counter in all is not enough as the meter action
should be created synchronized with reference counter increment. With
only atomic reference counter, even the counter is increased, the action
may still no
This commit protects the VLAN VM workaround area using a spinlock
in multiple-thread flow insertion to make it thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/common/mlx5/linux/mlx5_nl.h | 1 +
drivers/net/mlx5/linux/mlx5_vlan_os.c | 5 +
2 files changed, 6 inse
This commit changes the jump action in table to be created with table
creation in advanced. In this case, the jump action is safe to be used
in multiple thread. The jump action will be destroyed when table is not
used anymore and released.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
d
This commit creates the global drop action for flows instead of
maintain it in flow insertion time. The uniqueu global drop action
makes it thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 +
drivers/net/mlx5/mlx5.c| 2 +
This commit creates the global default miss action instead of maintain
it in flow insertion time. This makes the action to be thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7
drivers/net/mlx5/mlx5.h | 9 +---
drivers/net/ml
As flow table is shared between the ports in the same shared IB device,
flow table may be created by one port and released by other port.
Currently, the tunnel offloading active check in flow table release is
based on the port which release the flow table. Since the flow table
create port and rele
From: Xueming Li
In order to support hash list concurrent access, adding next:
1. List level read/write lock.
2. Entry reference counter.
3. Entry create/match/remove callback.
4. Remove insert/lookup/remove function which are not thread safe.
5. Add register/unregister function to support entry
From: Xueming Li
To support multi-thread flow insertion/removal, this patch uses thread
safe hash list API for flow table cache hash list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5.c | 102 +++-
drivers/net/mlx5/mlx5.h | 2 +-
driv
All table, tag, header modify, header reformat are supported only on DV
mode. For the OFED version doesn't support these, create the related
redundant DV resources waste the memory.
Add the code section in the HAVE_IBV_FLOW_DV_SUPPORT macro to avoid the
redundant resources allocation.
Fixes: 2eb4
From: Xueming Li
To support multi-thread flow insertion, this patch updates flow tag list
to use thread safe hash list with write-most mode.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 ++-
drivers/net/mlx5/mlx5_flow.h | 5 +++
drivers/net/ml
After non-cache mode feature was implemented, the flows can only be
created when port started. No need to check if the mreg flows are
created in port stopped status, and apply the mreg flows after port
start will also never happen.
This commit removed the relevant not used mreg copy code.
Signed-
From: Xueming Li
To support multi-thread flow insertion, this patch updates flow modify
action list to use thread safe hash list with write-most mode.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 +-
drivers/net/mlx5/mlx5_flow.h | 14 ++-
dri
From: Xueming Li
To support multi-thread flow insertion, this patch updates metadata copy
flow list to use thread safe hash list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 +-
drivers/net/mlx5/mlx5_flow.c | 162 ++---
From: Xueming Li
New API of linked list for cache:
- Optimized for small amount cache list.
- Optimized for read-most list.
- Thread safe.
- Since number of entries are limited, entries allocated by API.
- For dynamic entry size, pass 0 as entry size, then the creation
callback allocate the entry
To support multi-thread flow insertion, this patch updates flow header
reformat action list to use thread safe hash list with write-most mode.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 +-
drivers/net/mlx5/mlx5_flow.h | 7 ++
drivers/net/
This commit applies the cache linked list to Rx queue to make it thread
safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 +
drivers/net/mlx5/mlx5.c| 1 +
drivers/net/mlx5/mlx5.h| 28 +++-
drivers/net/mlx5/mlx5_flow.h
When create shared RSS hrxq, the hrxq will be created directly, no hrxq
will be reused.
In this case, add the shared RSS hrxq to the queue list is redundant.
And it also hurts the generic queue lookup.
This commit avoids add the shared RSS hrxq to the queue list.
Signed-off-by: Suanming Mou
Ack
From: Xueming Li
To support multi-thread flow insertion, this patch converts push VLAN
action cache list to thread safe cache list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 ++
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_f
From: Xueming Li
To support multi-thread flow insertion, this path converts matcher list
to use thread safe cache list API.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5.h | 3 +
drivers/net/mlx5/mlx5_flow.h| 15 ++-
drivers/net/mlx5/mlx5_flow_dv.c |
Currently, sample flow need to prepare and register the sub-actions
before sample action is created.
Once the same sample action exists, the sub-actions registered by
the second flow should be released, or these sub-actions will be
leaked. Since the exist sample action only release these same
sub-
From: Xueming Li
To support multi-thread flow insertion, this patch convert port id
action cache list to thread safe cache list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 ++
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_flow
This commit uses cache list to make sample and mirror action thread
safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 14 ++
drivers/net/mlx5/mlx5.h | 5 +-
drivers/net/mlx5/mlx5_flow.h | 28 ++-
drivers/net/mlx5/mlx5_flow_dv.c | 39
Currently, the sample action resource already has ft_type to indicate
the action domain attribute, the extra flow attributes parameter can
be optimized.
This commit uses action resource ty_type as domain attribute instead of
the flow attribute.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
This commit uses spinlock to protect the tunnel hub list in multiple
thread.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5_flow.c | 20 +---
drivers/net/mlx5/mlx5_flow.h | 1 +
2 files changed, 18 insertions(+), 3 deletions(-)
diff --git a/drivers
In previous commits the hash list objects have been converted
to new thread safe hash list. The legacy hash list code can be
removed now.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5_utils.c | 38 -
drivers/net/mlx5/mlx5_utils.h | 66 --
To support multi-thread flow insertion, this patch updates tunnel
offloading hash table to use thread safe hash list.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5_flow.c| 92 ++---
drivers/net/mlx5/mlx5_flow_dv.c | 11 +
1 - 100 of 385 matches
Mail list logo
