Commit d0fcc38f "vhost: improve device readiness notifications"
needs at least 2 vrings before changing the device state to
ready, this is fine for NET device but not correct for BLK
device.
The number of vring required should be based on the device
type, e.g. virtio_scsi device needs at least 3 v
At 2020-09-01 12:27:29, "Hu, Jiayu" wrote:
>Hi Yi,
>
>This patch supports UDP and VxLAN/UDP, but both are in one patch.
>It's too large, and please split it into small patches.
Jiayu, thank you so much for your great review , I'll send v2 to split it into
two patches and fix your comments. Detai
HI Bruce,
Will you please also add similar command examples in docs so that it
becomes easy for the developers to use meson?
Regards,
Hemant
-Original Message-
From: dev On Behalf Of Ma, LihongX
Sent: Tuesday, September 1, 2020 10:48 AM
To: Richardson, Bruce ; dev@dpdk.org
Cc: b
Tested-by: lihongx Ma
Before apply this patchset, set config like DRTE_LIBRTE_I40E_QUEUE_NUM_PER_VM=8
will failed,
After apply this patchset, the the meson build can work find.
Cmd like below:
meson -Denable_kmods=True -Dlibdir=lib --default-library=static
-Dexamples=vmdq_dcb -Dc_args='-DRTE_LIBR
On Mon, Aug 31, 2020 at 5:58 PM wangyunjian wrote:
>
>
> Ping for review, please.
yes. It’s in the pipeline and will be applied in a day or two.
Thanks
Ajit
>
>
>
> > -Original Message-
>
> > From: wangyunjian
>
> > Sent: Friday, July 31, 2020 8:09 PM
>
> > To: dev@dpdk.org
>
> > Cc:
Hi Yi,
This patch supports UDP and VxLAN/UDP, but both are in one patch.
It's too large, and please split it into small patches.
Thanks,
Jiayu
> -Original Message-
> From: yang_y...@163.com
> Sent: Wednesday, July 1, 2020 2:48 PM
> To: dev@dpdk.org
> Cc: Hu, Jiayu ; tho...@monjalon.net;
From: Kiran Kumar K
This patch reserves 2 bits as input selection to select Inner and
outer encapsulation level for RSS computation. It is combined with existing
ETH_RSS_* to choose Inner or outer layers.
This functionality already exists in rte_flow through level parameter in
RSS action configur
From: Kiran Kumar K
Adding support to set RSS level from ethdev config.
level-0 will requests the default behavior.
level-1 will requests RSS to be performed on the outermost packet
encapsulation level.
level-2 will request RSS to be performed on the specified inner packet
encapsulation level, fr
From: Kiran Kumar K
Add support to choose rss hash level from ethdev rss config.
Signed-off-by: Kiran Kumar K
---
drivers/net/octeontx2/otx2_ethdev.h | 2 +-
drivers/net/octeontx2/otx2_rss.c| 9 +++--
2 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/net/octeontx2/
On Tue, 1 Sep 2020 09:33:18 +0800
"Min Hu (Connor)" wrote:
> From: Huisong Li
>
> Currently, only statistics of rx/tx queues with queue_id less than
> RTE_ETHDEV_QUEUE_STAT_CNTRS can be displayed. If there is a certain
> application scenario that it needs to use 256 or more than 256 queues
> an
Friendly ping
On 2020/8/19 18:56, Wei Hu (Xavier) wrote:
This series updates CPU flags for arm64 platform.
Wei Hu (Xavier) (2):
eal/arm64: update CPU flags
test/cpuflag: add new flags for ARM64 platform
app/test/test_cpuflags.c | 39
lib/librte
From: Huisong Li
Currently, only statistics of rx/tx queues with queue_id less than
RTE_ETHDEV_QUEUE_STAT_CNTRS can be displayed. If there is a certain
application scenario that it needs to use 256 or more than 256 queues
and display all statistics of rx/tx queue. At this moment, we have to
chang
From: Huisong Li
Currently, base and nb_queue in the tc_rxq and tc_txq information
of queue and TC mapping on both TX and RX paths are uint8_t.
However, these datas will be truncated when queue number under a TC
is greater than 256. So it is necessay for base and nb_queue to
change from uint8_t t
Ping for review, please.
> -Original Message-
> From: wangyunjian
> Sent: Friday, July 31, 2020 8:09 PM
> To: dev@dpdk.org
> Cc: ajit.khapa...@broadcom.com; somnath.ko...@broadcom.com; Lilijun
> (Jerry) ; xudingke ;
> wangyunjian
> Subject: [dpdk-dev] [PATCH 0/2] fixes for bnxt driver
>
Sequences like "value = %"PRIu64 (no space before PRIu64) are parsed as
a single preprocessor token, user-defined-string-literal, in C++11
onwards. While modern compilers are smart enough to parse this properly,
GCC 9.3.0 generates warnings like:
rte_rcu_qsbr.h:555:26: warning: invalid suffix
> -Original Message-
> From: Dharmik Thakkar
> Sent: Tuesday, August 18, 2020 9:06 PM
> To: Wang, Yipeng1 ; Gobriel, Sameh
> ; Richardson, Bruce ;
> Ray Kinsella ; Neil Horman
> Cc: dev@dpdk.org; n...@arm.com; Dharmik Thakkar
>
> Subject: [RFC v2] lib/hash: integrate RCU QSBR
>
> Integr
Hi Diogo,
Thanks for your explanation.
As documented in https://developer.arm.com/documentation/ddi0487/fc B2.9.5
Load-Exclusive and Store-Exclusive instruction usage restrictions:
" Between the Load-Exclusive and the Store-Exclusive, there are no explicit
memory accesses, preloads,
direct
On Mon, 31 Aug 2020 13:23:33 +0530
Nipun Gupta wrote:
> This change adds a RX offload capability where hardware can drop the
> packets in case there is an error in the packet such as L3 checksum
> error or L4 checksum.
>
> Signed-off-by: Nipun Gupta
> Signed-off-by: Rohit Raj
> ---
> lib/libr
On Mon, 31 Aug 2020 09:35:18 +0300
Andrew Rybchenko wrote:
> > multisegment packets.
>
> I hope it will be mentioned in the feature documentation in the future,
> but
> I'm not 100% sure that it is required. See below.
> >>> I suppose there is the hierarchy:
> >>>
Hi Ferruh,
> -Original Message-
> From: Ferruh Yigit
> Sent: Monday, August 31, 2020 6:29 PM
> To: Nipun Gupta ; dev@dpdk.org
> Cc: tho...@monjalon.net; arybche...@solarflare.com; Hemant Agrawal
> ; Rohit Raj
> Subject: Re: [PATCH] ethdev: add rx offload to drop error packets
>
> On 8/3
On 8/31/2020 1:48 AM, Zhang, Qi Z wrote:
>
>
>> -Original Message-
>> From: Zhao1, Wei
>> Sent: Monday, July 13, 2020 3:48 PM
>> To: dev@dpdk.org
>> Cc: Zhang, Qi Z ; Zhao1, Wei
>> Subject: [PATCH] net/iavf: add support for multicast configuration
>>
>> This patch enable add multicast a
Hi Arek,
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Monday, August 31, 2020 7:24 AM
> To: Zhang, Roy Fan ; dev@dpdk.org
> Cc: akhil.go...@nxp.com; Trahe, Fiona ; Dybkowski,
> AdamX ; Bronowski, PiotrX
>
> Subject: RE: [dpdk-dev v7 1/4] cryptodev: add crypto data-path servi
> -Original Message-
> From: Stephen Hemminger
> Sent: Saturday, August 29, 2020 04:31
> To: Guo, Jia
> Cc: Wang, Haiyue ; Morten Brørup
> ; Yang, Qiming
> ; Xing, Beilei ; Zhao1, Wei
> ; Zhang,
> Qi Z ; Wu, Jingjing ;
> Richardson, Bruce
> ; dev@dpdk.org; Zhang, Helin
> ; Yigit, Ferr
On 8/6/2020 9:16 AM, Guinan Sun wrote:
> If the PF driver supports the new speed reporting capabilities
> then use link_event_adv instead of link_event to get the speed.
>
> Fixes: 2a73125b7041 ("i40evf: fix link info update")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Guinan Sun
> ---
> v3:
> * r
On 8/31/2020 8:53 AM, Nipun Gupta wrote:
> This change adds a RX offload capability where hardware can drop the
> packets in case there is an error in the packet such as L3 checksum
> error or L4 checksum.
>
> Signed-off-by: Nipun Gupta
> Signed-off-by: Rohit Raj
> ---
> lib/librte_ethdev/rte_e
HI Satya -
Do you mean packets are not at all seen at the output OR packets are seen but
data is not encrypted with just ESP headers added?
This is what I see with Cypto NULL device and I think is expected behavior of
NULL device.
Pravin
-Original Message-
From: users On Behalf Of An
On 8/29/2020 12:57 PM, Andrew Rybchenko wrote:
On 8/24/20 12:40 PM, Ferruh Yigit wrote:
This patch is a preparation to hide the 'struct eth_dev_ops' from
applications by moving some device operations from 'struct eth_dev_ops'
to 'struct rte_eth_dev'.
Mentioned ethdev APIs are in the data path a
Hi Arek,
Thank you very much to review.
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Monday, August 31, 2020 7:24 AM
> To: Zhang, Roy Fan ; dev@dpdk.org
> Cc: akhil.go...@nxp.com; Trahe, Fiona ; Dybkowski,
> AdamX ; Bronowski, PiotrX
>
> Subject: RE: [dpdk-dev v7 1/4] crypto
> -Original Message-
> From: Akhil Goyal
> Sent: środa, 15 lipca 2020 22:16
> To: Thomas Monjalon ; Kusztal, ArkadiuszX
>
> Cc: dev@dpdk.org; Trahe, Fiona ;
> techbo...@dpdk.org; Anoob Joseph ; Somalapuram,
> Amaranath ; Ankur Dwivedi
> ; ruifeng.w...@arm.com; De Lara Guarch, Pablo
> ;
> -Original Message-
> From: Eugenio Pérez
> Sent: Monday, August 31, 2020 3:59 PM
> To: dev@dpdk.org
> Cc: Adrian Moreno Zapata ; Xia, Chenbo
> ; Wang, Zhihong ; Jens
> Freimann ; sta...@dpdk.org; Maxime Coquelin
> ; Kevin Traynor
> Subject: [PATCH v2 1/1] vhost: fix iotlb mempool singl
On Sun, 2020-08-30 at 14:36 +, Ali Alnubani wrote:
> Hi,
>
> > -Original Message-
> > From: dev On Behalf Of Luca Boccassi
> > Sent: Tuesday, August 18, 2020 9:12 PM
> > To: sta...@dpdk.org
> > Cc: dev@dpdk.org; Abhishek Marathe ;
> > Akhil Goyal ; Ali Alnubani ;
> > benjamin.wal...@i
https://bugs.dpdk.org/show_bug.cgi?id=531
Konstantin Ananyev (konstantin.anan...@intel.com) changed:
What|Removed |Added
Resolution|--- |INVALID
Crypto session and device session private data were put back
to wrong pools.
This caused data corruption when the object was reallocated and used.
Because objects from different mempools have different element size,
and wrong mempool info caused out of bound write.
Fix the issue by putting back obj
Switched from device specific test suite to unified cryptodev test suite.
Removed the armv8 device specific test suite.
Signed-off-by: Ruifeng Wang
---
app/test/test_cryptodev.c | 19 +--
1 file changed, 1 insertion(+), 18 deletions(-)
diff --git a/app/test/test_cryptodev.c b/ap
This patch set migrates armv8 crypto PMD unit test from device
specific test suite to common crypto dev test suite.
Fixed couple of issues found during the migration.
Ruifeng Wang (3):
test/crypto: fix stat test
crypto/armv8: fix mempool object returning
crypto/armv8: move to cryptodev test
ut_setup / ut_teardown are invoked for each test case by test framework.
The call inside test_stats is unnecessary and even incorrect.
This caused double free of objects such as crypto operation structure.
Trapped the issue when RTE_LIBRTE_MEMPOOL_DEBUG was enabled.
Fix issue by removing ut_setup /
Tested-by: Huang, ZhiminX
Regards,
HuangZhiMin
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Wei Zhao
Sent: Monday, August 31, 2020 2:58 PM
To: dev@dpdk.org
Cc: sta...@dpdk.org; Zhang, Qi Z ; Zhao1, Wei
Subject: [dpdk-dev] [PATCH v4] net/ice: fix FDIR GTPU rul
> -Original Message-
> From: dev On Behalf Of Li Han
> Sent: Tuesday, August 25, 2020 11:01 AM
> To: Wu, Jingjing ; Xing, Beilei
> Cc: dev@dpdk.org; Li Han
> Subject: [dpdk-dev] [PATCH] net/iavf: fix default mac set fail issue
>
> rte_is_valid_assigned_ether_addr retrun true if the g
Bugzilla bug: 523
This behavior is only observed on packed vq + virtio-net kernel driver in the
guest, so we could make the single-consumer flag optional. However, I have not
found why I cannot see the issue in split, so the safer option is to never set
it.
Any comments?
Thanks!
v2: Modify comm
Bugzilla bug: 523
Control thread (which handles iotlb msg) and forwarding
thread both use iotlb to translate address. The former may modify the
same entry of mempool and may cause a loop in iotlb_pending_entries
list.
Fixes: d012d1f293f4 ("vhost: add IOTLB helper functions")
Signed-off-by: Eugeni
This change adds a RX offload capability where hardware can drop the
packets in case there is an error in the packet such as L3 checksum
error or L4 checksum.
Signed-off-by: Nipun Gupta
Signed-off-by: Rohit Raj
---
lib/librte_ethdev/rte_ethdev.h | 1 +
1 file changed, 1 insertion(+)
diff --git
> -Original Message-
> From: Zhang, Yuying
> Sent: Wednesday, August 26, 2020 3:45 PM
> To: dev@dpdk.org; Zhang, Qi Z ; Zhang, Yuying
>
> Subject: [PATCH v1] net/fm10k: use inclusive engineering terminology
>
> Offensive terms should be replaced with more inclusive engineering terms.
> -Original Message-
> From: Zhao1, Wei
> Sent: Monday, August 31, 2020 2:58 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Zhang, Qi Z ; Zhao1, Wei
>
> Subject: [PATCH v4] net/ice: fix FDIR GTPU rule create fail
>
> For GTPU rule without extend header the trainning packet for FDIR is
For GTPU rule without extend header the trainning packet
for FDIR is different. This patch enable these case.
Fixes: 934fd00c9389 ("net/ice/base: fix GTPU IP hash")
Cc: sta...@dpdk.org
Signed-off-by: Wei Zhao
Tested-by: Huang ZhiminX
---
v3: fix typo
v4: fix bug
---
drivers/net/ice/base/ice
> -Original Message-
> From: Guo, Jia
> Sent: Monday, August 31, 2020 1:51 PM
> To: Yang, Qiming ; Zhang, Qi Z
>
> Cc: dev@dpdk.org; Guo, Jia
> Subject: [PATCH v2] net/ice: fix the hash parser
>
> GTPU TEID hash should only be enabled when ETH_RSS_GTPU is required.
> And the hash par
45 matches
Mail list logo
