> -Original Message-
> From: Ferruh Yigit
<..>
> >>
> >
> > Ferruh, your suggestion ([1] + update) looks correct. I also verified it on
> mlx5 PMD.
> > Advantage: it's a generic fix for all dpdk applications using rte_flows (not
> just testpmd).
> > It reduces code.
> > With this fix the
We are including this in rte_prefetch.h since it is the most closely
related code location.rte_cldemte is similar to a prefetch hint -in reverse
Omkar Maslekar (1):
eal: add cache-line demote support
app/test/test_prefetch.c | 2 ++
doc/guides/rel_notes/release_20_11.rst
rte_cldemote is similar to a prefetch hint - in reverse. cldemote(addr)
enables software to hint to hardware that line is likely to be shared.
Useful in core-to-core communications where cache-line is likely to be
shared. ARM and PPC implementation is provided with NOP and can be added
if any equiv
> >
> > On 15/10/20 10:49 +, Juraj Linkeš wrote:
> > > Hi dpdk devs,
> > >
> > > Is there a constraint on how low RTE_MAX_LCORE can be? I'm
> > > implementing a
> > discovery mechanism that sets RTE_MAX_LCORE according to the number
> of
> > host cores, but I'm hitting errors when the values
When the value of RTE_MAX_LCORE is small, it results in the
following compilation error.
../examples/performance-thread/l3fwd-thread/main.c:2338:34: error:
iteration 4 invokes undefined behavior
[-Werror=aggressive-loop-optimizations]
Fixes: d48415e1fee3 ("examples/performance-thread: add l3fwd-t
When RTE_MAX_LCORE value is small, following compiler errors
are observed.
../app/test/test_rcu_qsbr.c:296:54: error: iteration 2 invokes
undefined behavior [-Werror=aggressive-loop-optimizations]
../app/test/test_rcu_qsbr.c:315:55: error: array subscript is above
array bounds [-Werror=array-boun
Correct my typo issue that vhost-user can't launch with client mode.
BR,
Yinan
> -Original Message-
> From: Wang, Yinan
> Sent: 2020?10?16? 13:42
> To: Maxime Coquelin ; dev@dpdk.org; Xia,
> Chenbo ; Fu, Patrick ;
> amore...@redhat.com
> Subject: RE: [dpdk-dev] [PATCH v3 3/8] net/virtio:
> -Original Message-
> From: Ferruh Yigit
> Sent: 2020年10月16日 2:57
> To: Zhang, Tianfei ; dev@dpdk.org; Xu, Rosen
> ; Huang, Wei
> Subject: Re: [dpdk-dev] [PATCH v2 3/4] raw/ifpga/base: cleanup ifpga raw
> devices when process quit
>
> On 9/28/2020 2:40 AM, Tianfei zhang wrote:
> > Fro
> -Original Message-
> From: Ferruh Yigit
> Sent: 2020年10月16日 2:57
> To: Zhang, Tianfei ; dev@dpdk.org; Xu, Rosen
> ; Huang, Wei
> Subject: Re: [dpdk-dev] [PATCH v2 2/4] raw/ifpga/base: free resources when
> destroying ifpga device
>
> On 9/28/2020 2:40 AM, Tianfei zhang wrote:
> > Fro
> -Original Message-
> From: Ferruh Yigit
> Sent: 2020年10月16日 2:57
> To: Zhang, Tianfei ; dev@dpdk.org; Xu, Rosen
> ; Huang, Wei
> Cc: sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 1/4] raw/ifpga/base: fix bug in IRQ
> functions
>
> On 9/28/2020 2:40 AM, Tianfei zhang wrote:
> >
>
> Hi Honnappa,
>
> Many thanks for the review!
>
> I'll write my answers here not inline as it would be easier to read them in
> one
> place, I think.
> So first of all I agree with you in 2 things:
> 1) all uses of statistics must be atomic and lack of that caused most of the
> problems
>
Hi Adrian,
This patch introduce a regression issue that vhost-user can't launch with
server mode, details in https://bugs.dpdk.org/show_bug.cgi?id=559.
This issue blocked amount of regression cases, could you help to take a look?
Thanks in advance!
BR,
Yinan
> -Original Message-
> Fro
https://bugs.dpdk.org/show_bug.cgi?id=559
Bug ID: 559
Summary: [dpdk-20.11]loopback_virtio_user_server_mode: failed
to start virtio-user with Virtio user server mode
Product: DPDK
Version: 20.11
Hardware: All
Hi Anatoly,
Thanks for your confirmation. Yes, you got it right now. The primary process
removing *its own* own files for some locking issue, which should not be the
case.
Just to clarify:
Removing stuff at the end of the init process is intended process. This
is because every process will creat
Hi Lukasz,
I see that in commit 8/16, the same code is changed again (updating the
counters using the RELAXED memory order). It is better to pull the statistics
changes from 8/16 into this commit.
Thanks,
Honnappa
> -Original Message-
> From: dev On Behalf Of Lukasz Wojciechows
>
> During quit_workers function distributor's main core processes some packets
> to wake up pending worker cores so they can quit.
> As quit_workers acts also as a cleanup procedure for next test case it should
> also collect these packages returned by workers'
nit ^
>
> Sanity tests with mbuf alloc and shutdown tests assume that mbufs passed
> to worker cores are freed in handlers.
> Such packets should not be returned to the distributor's main core. The only
> packets that should be returned are the packets send after completion of
> the tests in quit_work
From: murphy
When set 'csum set outer-udp hw 0' ,support for
ICE_RX_FLEX_DESC_STATUS0_XSUM_EUDPE_S,mark the packet
PKT_RX_OUTER_L4_CKSUM_BAD or PKT_RX_OUTER_L4_CKSUM_GOOD.
Fixes: dbf3c0e77a22 ("net/ice: handle Rx flex descriptor")
Fixes: 4ab7dbb0a0f6 ("net/ice: switch to Rx flexible descriptor i
> -Original Message-
> From: dev On Behalf Of Lukasz Wojciechowski
> Sent: Saturday, October 10, 2020 11:05 AM
> To: David Hunt ; Bruce Richardson
>
> Cc: dev@dpdk.org; l.wojciec...@partner.samsung.com; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH v7 09/16] test/distributor: collect ret
Add vhost async driver arguments information for vhost async data
path in vhost sample application.
Signed-off-by: Cheng Jiang
---
doc/guides/sample_app_ug/vhost.rst | 11 +++
1 file changed, 11 insertions(+)
diff --git a/doc/guides/sample_app_ug/vhost.rst
b/doc/guides/sample_app_ug/vh
This patch is to implement vhost DMA operation callbacks for CBDMA
PMD and add vhost async data-path in vhost sample. With providing
callback implementation for CBDMA, vswitch can leverage IOAT to
accelerate vhost async data-path.
Signed-off-by: Cheng Jiang
---
examples/vhost/ioat.c | 97 +++
Add release notes for vhost async data path support in vhost sample.
Signed-off-by: Cheng Jiang
---
doc/guides/rel_notes/release_20_11.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/rel_notes/release_20_11.rst
b/doc/guides/rel_notes/release_20_11.rst
index f8686a50d..8
This patch set makes vhost-vswitch be able to use vhost async APIs
for enqueue operations. Demonstrated how the application
leverage IOAT DMA channel with vhost async APIs.
We introduce two parameters to enable DMA acceleration for Tx
operations of queues:
-dma_type This parameter is used to speci
This patch is to add async vhost driver arguments parsing function
for CBDMA channel, DMA initiation function and args description.
The meson build file is changed to fix dependency problem. With
these arguments vhost device can be set to use CBDMA or CPU for
enqueue operation and bind vhost device
On Thu, Oct 15, 2020 at 4:54 PM Ferruh Yigit wrote:
>
> On 10/16/2020 12:21 AM, Ophir Munk wrote:
> >> From: Ferruh Yigit
> > On 10/15/2020 1:41 PM, Lijun Ou wrote:
> >> When start the testpmd, the pmd driver initializes the RSS
> >> configuration. Generally, the recommended RSS hash
08/10/2020 20:48, Harman Kalra:
> + /* Copy four words if lmtext = 0
> + * six words if lmtext = 1
> + * eight words if lmtext =2
> + */
> + memcpy(out, in, (4 + (2 * lmtext)) * sizeof(uint64_t));
It raises an error when compiling for PowerPC:
drivers/common/octeon
When running Sphinx through ninja, the wrapper configured in meson
redirects stdout to a log file.
It makes more important to print issues on stderr.
Some warnings generated by the conf.py were hidden because
printed on stdout. The first improvement is to print them on stderr.
The second measure
The DPDK version should appear in the top left corner of the HTML guides.
When dropping make, the variable version has been removed,
so Sphinx stopped integrating the version number.
Fixes: a4362f150204 ("doc: build without using make")
Signed-off-by: Thomas Monjalon
---
doc/guides/conf.py | 1
09/10/2020 17:00, Akhil Goyal:
> > This patchset contains support for Crypto offload on Broadcom’s
> > Stingray/Stingray2 SoCs having FlexSparc unit.
[...]
> Series applied to dpdk-next-crypto With 2 fixes as mentioned in the last
> patch.
There are 2 errors in the doc:
Warning generate_overview
Tested-by: Han, Yingya
Best Regards,
Yingya
-Original Message-
From: Richardson, Bruce
Sent: Monday, October 12, 2020 10:52 PM
To: dev@dpdk.org
Cc: Han, YingyaX ; Ananyev, Konstantin
; Tu, Lijuan ; Richardson,
Bruce
Subject: [PATCH] build: fix memcpy behaviour regression
When testin
There is no error info displayed when running flow flush
command with invalid port. This patch fixed the issue.
Fixes: 2a449871a12d ("app/testpmd: align behaviour of multi-port detach")
Cc: sta...@dpdk.org
Signed-off-by: Junyu Jiang
Reviewed-by: Suanming Mou
---
app/test-pmd/config.c | 13
On Thu, Oct 15, 2020 at 1:33 PM Thomas Monjalon wrote:
>
> 15/10/2020 22:17, Viacheslav Ovsiienko:
> > /**
> > + * Ethernet device Rx buffer segmentation capabilities.
> > + */
> > +struct rte_eth_rxseg_capa {
> > + uint16_t max_seg; /**< Maximum amount of segments to split. */
> > + uint
Add support to negotiate large VF capability, including configure VSI
queues, enable/disable queues and IRQ mapping for large VF. Use new
virtchnl opcodes and structures to help indicate up to 256 queue pairs.
Need to request additional queues from PF first if the current allocated
queues are not e
Add new virtchnl function to request additional queues from PF. Current
default queue pairs number when creating a VF is 16. In order to support
up to 256 queue pairs, enable this request queues function.
Since request queues command may return event message, modify function
iavf_read_msg_from_pf t
This patchset supports to configure up to 256 queue pairs per VF. If
large VF is supported after capability negotiation, VF will request
queues from PF as needed. New virtual channel opcodes and structures
are used to indicate 256 queues, so VF is designed to handle the new
function of configure VS
> -Original Message-
> From: David Christensen
> Sent: Friday, October 16, 2020 1:50 AM
> To: dev@dpdk.org; maxime.coque...@redhat.com; Xia, Chenbo
> ; Wang, Zhihong
> Cc: sta...@dpdk.org; David Christensen
> Subject: [PATCH v3] net/vhost: fix xstats wrong after clearing stats
>
> The P
Hi Ferruh,
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, October 16, 2020 7:03 AM
> To: Bing Zhao ; NBU-Contact-Thomas Monjalon
> ; Ori Kam ;
> arybche...@solarflare.com; m...@ashroe.eu; nhor...@tuxdriver.com;
> bernard.iremon...@intel.com; beilei.x...@intel.com;
> wenzhuo...@
Add support to negotiate large VF capability, including configure VSI
queues, enable/disable queues and IRQ mapping for large VF. Use new
virtchnl opcodes and structures to help indicate up to 256 queue pairs.
Need to request additional queues from PF first if the current allocated
queues are not e
Add new virtchnl function to request additional queues from PF. Current
default queue pairs number when creating a VF is 16. In order to support
up to 256 queue pairs, enable this request queues function.
Since request queues command may return event message, modify function
iavf_read_msg_from_pf t
This patchset supports to configure up to 256 queue pairs per VF. If
large VF is supported after capability negotiation, VF will request
queues from PF as needed. New virtual channel opcodes and structures
are used to indicate 256 queues, so VF is designed to handle the new
function of configure VS
在 2020/10/15 21:30, Andrew Rybchenko 写道:
From: Ivan Ilchenko
rte_eth_dev_stop() return value was changed from void to int,
so this patch modify usage of this function across app
according to new return type.
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
---
app/test/test_
Hi,
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, October 16, 2020 6:14 AM
> To: Suanming Mou ; Jiang, JunyuX
> ; dev@dpdk.org
> Cc: Xing, Beilei ; Lu, Wenzhuo
> ; sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] app/testpmd: fix flow flush with invalid port
>
> On 10/14/202
Add new virtchnl function to request additional queues from PF. Current
default queue pairs number when creating a VF is 16. In order to support
up to 256 queue pairs, enable this request queues function.
Since request queues command may return event message, modify function
iavf_read_msg_from_pf t
Add support to negotiate large VF capability, including configure VSI
queues, enable/disable queues and IRQ mapping for large VF. Use new
virtchnl opcodes and structures to help indicate up to 256 queue pairs.
Need to request additional queues from PF first if the current allocated
queues are not e
This patchset supports to configure up to 256 queue pairs per VF. If
large VF is supported after capability negotiation, VF will request
queues from PF as needed. New virtual channel opcodes and structures
are used to indicate 256 queues, so VF is designed to handle the new
function of configure VS
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Friday, October 16, 2020 12:16 AM
> To: Hu, Jiayu ; yang_y_yi
> Cc: dev@dpdk.org; olivier.m...@6wind.com; tho...@monjalon.net;
> yangy...@inspur.com
> Subject: RE: Re:RE: [PATCH] gso: fix free issue of mbuf gso segments attach to
Hi,
>-Original Message-
>From: Andrew Rybchenko
>Sent: Wednesday, October 14, 2020 4:29 PM
>To: Wisam Monther
>Cc: dev@dpdk.org; Ivan Ilchenko
>Subject: [PATCH 08/11] app/flow-perf: check eth dev stop status
>
>From: Ivan Ilchenko
>
>rte_eth_dev_stop() return value was changed from voi
On 10/14/20 7:57 PM, Honnappa Nagarahalli wrote:
On 13/10/2020 18:46, Michel Machado wrote:
On 10/13/20 11:41 AM, Medvedkin, Vladimir wrote:
Hi Michel,
Could you please describe a condition when LPM gets inconsistent?
As I can see if there is no free tbl8 it will return -ENOSPC.
Conside
Fixes spelling in comment and message about thread error.
Found while looking at checkpatch complaints about "thead"
Signed-off-by: Stephen Hemminger
---
lib/librte_eal/common/eal_common_proc.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/librte_eal/common/eal_comm
On 10/16/2020 12:21 AM, Ophir Munk wrote:
From: Ferruh Yigit
On 10/15/2020 1:41 PM, Lijun Ou wrote:
When start the testpmd, the pmd driver initializes the RSS
configuration. Generally, the recommended RSS hash key is used as
the default key in the driver. In addition, the default key is
differ
>
> rte_distributor_return_pkt function which is run on worker cores must wait
> for distributor core to clear handshake on retptr64 before using those
> buffers. While the handshake is set distributor core controls buffers and any
> operations on worker side might overwrite buffers which are un
> From: Ferruh Yigit
> >>> On 10/15/2020 1:41 PM, Lijun Ou wrote:
> When start the testpmd, the pmd driver initializes the RSS
> configuration. Generally, the recommended RSS hash key is used as
> the default key in the driver. In addition, the default key is
> different from t
On 10/15/2020 4:51 PM, Dekel Peled wrote:
This series implements the change proposes in RFC [1], adding dedicated
fields to ETH and VLAN items structs, to clearly define the required
characteristic of a packet, and enable precise match criteria.
testpmd CLI is updated accordingly.
Documentation i
On 10/15/2020 2:08 PM, Bing Zhao wrote:
The patches contain the following changes:
1. new APIs to bind and unbind hairpin ports in manual binding mode.
2. new API to get the hairpin peer ports list.
3. new internal APIs for PMD to pass the queue information and
configure the queue pair.
4. ne
Replace use of the term master lcore with main lcore in
the example programs.
Signed-off-by: Stephen Hemminger
---
examples/bbdev_app/main.c | 14 +--
examples/bond/main.c| 32 -
examples/distributor/main.c | 10 --
This test uses master/slave to describe the main thread
versus worker thread.
Signed-off-by: Stephen Hemminger
---
drivers/event/octeontx2/otx2_evdev_selftest.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/event/octeontx2/otx2_evdev_selftest.c
b/drivers
The term slave is only used in some comments and can be
replaced with sub devices, as done elsewhere.
Signed-off-by: Stephen Hemminger
---
drivers/net/failsafe/failsafe.c | 6 +++---
drivers/net/failsafe/failsafe_private.h | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --
Simple replacement.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
examples/ipsec-secgw/event_helper.c | 6 +++---
examples/ipsec-secgw/ipsec-secgw.c | 4 ++--
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/examples/ipsec-secgw/event_helper.c
b/examples/ipsec-s
Use main instead of master in the QoS examples.
Replace the "mst" command line argument with "mnc"
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
examples/qos_meter/main.c| 4 ++--
examples/qos_sched/args.c| 26 +-
examples/qos_sched/cmdline.c |
Replace SKIP_MASTER with SKIP_MAIN
and FOREACH_SLAVE with FOREACH_WORKER.
Signed-off-by: Stephen Hemminger
---
examples/pipeline/main.c | 2 +-
examples/pipeline/thread.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/pipeline/main.c b/examples/pipeline/main.c
Replace term master with main here.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
examples/performance-thread/l3fwd-thread/main.c | 16
examples/performance-thread/pthread_shim/main.c | 4 ++--
2 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/e
Replace master with main and slave with worker.
Fix the wording in comments.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
.../multi_process/client_server_mp/mp_server/main.c| 10 +-
examples/multi_process/simple_mp/main.c| 4 ++--
examples/multi_pr
Change references to master/slave lcore to main/worker.
For l2fwd, l3fwd and link_status examples.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
app/pdump/main.c | 2 +-
examples/ip_pipeline/thread.c| 14 +++---
examp
Replace master with main lcore.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
app/test-eventdev/evt_options.c | 2 +-
app/test-eventdev/test_order_common.c| 12 ++--
app/test-eventdev/test_perf_common.c | 16
app/test-eventdev/test_pip
For the miscellaneous tests. Replace references to master lcore with
main lcore and slave lcore with worker lcore.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
app/test-acl/main.c | 2 +-
app/test-bbdev/test_bbdev_perf.c | 16
app/test-compress-
Replace master lcore with main lcore and slave lcore with worker lcore.
Mostly automatic replacement.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
app/test/autotest_test_funcs.py | 2 +-
app/test/meson.build| 2 +-
app/test/test.c
Make sure that master lcore is not used in documentation.
Signed-off-by: Stephen Hemminger
---
doc/guides/contributing/coding_style.rst | 2 +-
doc/guides/faq/faq.rst | 6 +++---
doc/guides/howto/debug_troubleshoot.rst | 2 +-
doc/guides/linux_gsg/ea
Replace master/slave terms in this driver.
The memory interface drivers uses a client/server architecture
so change the variable names and device arguments to that.
The previous devargs are maintained for compatibility, but if
used cause a notice in the log.
Signed-off-by: Stephen Hemminger
---
Replace wording around main lcore.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
app/test-pmd/cmdline.c| 2 +-
app/test-pmd/config.c | 4 ++--
app/test-pmd/parameters.c | 2 +-
app/test-pmd/testpmd.c| 4 ++--
4 files changed, 6 insertions(+), 6 deletions(-)
diff --
The old rte_master_lcore is now replaced by rte_main_lcore.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
lib/librte_power/rte_power_empty_poll.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_power/rte_power_empty_poll.c
b/lib/librte_power/rte_
Replace use of deprecated rte_master_lcore with rte_main_lcore.
Signed-off-by: Stephen Hemminger
Acked-by: Anatoly Burakov
---
drivers/bus/dpaa/dpaa_bus.c | 2 +-
drivers/bus/pci/pci_common_uio.c| 3 ++-
drivers/bus/vmbus/vmbus_common_uio.c| 2
Add a macro that causes GCC and CLANG to emit a warning when
a deprecated macro is used.
Acked-by: Anatoly Burakov
Acked-by: Bruce Richardson
Signed-off-by: Stephen Hemminger
---
lib/librte_eal/include/rte_common.h | 11 +++
1 file changed, 11 insertions(+)
diff --git a/lib/librte_eal
Replace master lcore with main lcore and
replace slave lcore with worker lcore.
Keep the old functions and macros but mark them as deprecated
for this release.
The "--master-lcore" command line option is also deprecated
and any usage will print a warning and use "--main-lcore"
as replacement.
Ac
Update the coding style document to include a policy against
introducing new master/slave usage. This is taken from the similar
place in the Linux kernel coding style.
Acked-by: Anatoly Burakov
Signed-off-by: Stephen Hemminger
---
doc/guides/contributing/coding_style.rst | 23 ++
As announced in 20.08 release notes, the terms for lcore role
will change in 20.11 release.
This is a project wide patchset to terminolgy used for DPDK lcore.
The term "master" referring to lcore is replaced with "main" and
"slave" lcore is replaced with "worker" lcore.
This is most visible in AP
>
> Hello,
>
> On 15/10/2020 18:38, Honnappa Nagarahalli wrote:
> >
> >>
> >> On 10/14/20 7:57 PM, Honnappa Nagarahalli wrote:
> >> On 13/10/2020 18:46, Michel Machado wrote:
> >>> On 10/13/20 11:41 AM, Medvedkin, Vladimir wrote:
> Hi Michel,
>
> Could you ple
On 10/15/2020 1:41 PM, Gregory Etelson wrote:
Tunnel Offload API provides hardware independent, unified model
to offload tunneled traffic. Key model elements are:
- apply matches to both outer and inner packet headers
during entire offload procedure;
- restore outer header of partially of
15/10/2020 03:07, Suanming Mou:
> Currently, the rte_flow functions are not defined as thread safe.
> DPDK applications either call the functions in single thread or add
> locks around the functions for the critical section.
>
> For PMDs support the flow operations thread safe natively, the
> redu
rte_cldemote is similar to a prefetch hint - in reverse. cldemote(addr)
enables software to hint to hardware that line is likely to be shared.
Useful in core-to-core communications where cache-line is likely to be
shared. ARM and PPC implementation is provided with NOP and can be added
if any equiv
We are including this in rte_prefetch.h since it is the most closely
related code location.rte_cldemte is similar to a prefetch hint -in reverse
Omkar Maslekar (1):
eal: add cache-line demote support
app/test/test_prefetch.c | 2 ++
doc/guides/rel_notes/release_20_11.rst
On 10/14/2020 11:16 AM, Suanming Mou wrote:
Hi,
-Original Message-
From: dev On Behalf Of Junyu Jiang
Sent: Wednesday, October 14, 2020 4:30 PM
To: dev@dpdk.org
Cc: Beilei Xing ; Wenzhuo Lu ;
Junyu Jiang ; sta...@dpdk.org
Subject: [dpdk-dev] [PATCH] app/testpmd: fix flow flush with inv
Patches postponed to RC2:
DLB driver: http://patches.dpdk.org/patch/77484/
DLB2 driver: http://patches.dpdk.org/patch/77507/
evendev feature matrix documentation:
http://patches.dpdk.org/user/todo/dpdk/?series=8706
The following changes since commit 057d9a92f0a9aa07dd820669d4a07b325661929f:
On Thu, 15 Oct 2020 19:14:48 +0200
Thomas Monjalon wrote:
> 15/10/2020 19:08, Bruce Richardson:
> > On Thu, Oct 15, 2020 at 04:00:44PM +, Ali Alnubani wrote:
> > >We have been seeing in some cases that the DPDK forwarding performance
> > >is up to 9% lower when DPDK is built as stat
On 10/15/2020 1:43 AM, Qi Zhang wrote:
Update base code release version in readme
Signed-off-by: Qi Zhang
---
drivers/common/iavf/README | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/common/iavf/README b/drivers/common/iavf/README
index f301ff665a..5a42750465 10
On 10/15/2020 1:43 AM, Qi Zhang wrote:
Replace ETH_ALEN with VIRTCHNL_ETH_LENGTH_OF_ADDRESS.
>
Why? They have same value, righ? Can you please put the reason, and impact if
any to the commit log.
Signed-off-by: Maciej Rabeda
Signed-off-by: Qi Zhang
---
drivers/common/iavf/virtchnl.h |
On Tue, Oct 6, 2020 at 9:23 AM Olivier Matz wrote:
>
> From: Somnath Kotur
>
> Clarify the documentation of QinQ flags, and extend the meaning of the
> flag: if PKT_RX_QINQ_STRIPPED is set and PKT_RX_VLAN_STRIPPED is unset,
> only the outer VLAN is removed from packet data, but both tci are saved
On Mon, Oct 12, 2020 at 7:39 PM David Marchand
wrote:
> On Mon, Sep 28, 2020 at 2:01 AM Thomas Monjalon wrote:
> > The option RTE_EAL_ALWAYS_PANIC_ON_ERROR was off by default,
> > and not customizable with meson. It is completely removed.
> >
> > The function rte_dump_registers is a trace of the
15/10/2020 22:17, Viacheslav Ovsiienko:
> /**
> + * Ethernet device Rx buffer segmentation capabilities.
> + */
> +struct rte_eth_rxseg_capa {
> + uint16_t max_seg; /**< Maximum amount of segments to split. */
> + uint8_t multi_pools; /**< Supports receiving to multiple pools.*/
> + ui
On Thu, Oct 15, 2020 at 4:41 PM Maslekar, Omkar
wrote:
> >> diff --git a/lib/librte_eal/arm/include/rte_prefetch_32.h
> >> b/lib/librte_eal/arm/include/rte_prefetch_32.h
> >> index e53420a..28b3d48 100644
> >> --- a/lib/librte_eal/arm/include/rte_prefetch_32.h
> >> +++ b/lib/librte_eal/arm/in
On Fri, Oct 16, 2020 at 1:47 AM Viacheslav Ovsiienko
wrote:
>
> The DPDK datapath in the transmit direction is very flexible.
> An application can build the multi-segment packet and manages
> almost all data aspects - the memory pools where segments
> are allocated from, the segment lengths, the m
On Thu, Oct 15, 2020 at 12:31 PM Harry van Haaren
wrote:
>
> This commit adds new rte_prefetchX_write() variants, that suggest to the
> compiler to use a prefetch instruction with intention to write. As a
> compiler builtin, the compiler can choose based on compilation target
> what the best imple
Hi,
Evening update:
- addressed code comments
- provided the union of segmentation description with dedicated feature
structures according Jerin's proposal
- added the reporting of split limitation
With best regards, Slava
> -Original Message-
> From: Andrew Rybchenko
> Sent: Thursda
If Rx queue is configured with split feature the extended
setup with specified segment sizes and pool will be performed.
Signed-off-by: Viacheslav Ovsiienko
---
app/test-pmd/cmdline.c | 12 ++--
app/test-pmd/testpmd.c | 48 ++--
app/test-pmd/te
Add command line parameter:
--rxoffs=X[,Y]
Sets the offsets of packet segments from the beginning of the
receiving buffer if split feature is engaged. Affects only the
queues configured with split offloads (currently BUFFER_SPLIT
is supported only).
Add interactive mode command, providing the sa
The DPDK datapath in the transmit direction is very flexible.
An application can build the multi-segment packet and manages
almost all data aspects - the memory pools where segments
are allocated from, the segment lengths, the memory attributes
like external buffers, registered for DMA, etc.
In th
This patch add support for RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT
providing per queue configuration for this offload.
Signed-off-by: Viacheslav Ovsiienko
---
app/test-pmd/cmdline.c | 21 +++--
app/test-pmd/config.c | 9 +
2 files changed, 20 insertions(+), 10 deletions(-)
dif
Add command line parameter:
--rxpkts=X[,Y]
Sets the length of segments to scatter packets on receiving if split
feature is engaged. Affects only the queues configured with split
offloads (currently BUFFER_SPLIT is supported only).
Add interactive mode command:
testpmd> set rxpkts (x[,y]*)
Wher
The command line parameter --mbuf-size is updated, it can handle
the multiple values like the following:
--mbuf-size=2176,512,768,4096
specifying the creation the extra memory pools with the requested
mbuf data buffer sizes. If some buffer split feature is engaged
the extra memory pools can be us
The DPDK datapath in the transmit direction is very flexible.
An application can build the multi-segment packet and manages
almost all data aspects - the memory pools where segments
are allocated from, the segment lengths, the memory attributes
like external buffers, registered for DMA, etc.
In th
From: Long Li
If a device probe fails, the alarm is canceled and will no longer work for
previously probed devices.
Fix this by introducing a flag to track if alarm has been set. Because it's
possible that an alarm is triggered while probing is in progress that may
modify vdev_netvsc_ctx_list, i
1 - 100 of 393 matches
Mail list logo
