23/04/2020 01:51, Ranjit Menon:
> On 4/22/2020 12:27 AM, tal...@mellanox.com wrote:
> > From: Tal Shnaiderman
> >
> > Move common functions between Unix and Windows to eal_config.c.
>
> Like other files in common, we should call this eal_common_config.c
I am not sure about the interest of repea
>-Original Message-
>From: Gaëtan Rivet
>Sent: Wednesday, April 22, 2020 3:09 PM
>To: Sunil Kumar Kori
>Cc: step...@networkplumber.org; david.march...@redhat.com; Jerin Jacob
>Kollanukkaran ; dev@dpdk.org
>Subject: Re: [EXT] Re: [dpdk-dev] [PATCH v3 1/1] bus/pci: optimise scanning
>with w
>
> Hi Konstantin,
>
> These are data path ops and so it will be better if we can avoid such checks
> in the datapath. The same is done in ethdev also.
AFAIK, get_userdata is an *optional* dev-ops function that can be used by
data-path.
So far there was no strict requirement for the rte_secur
W dniu 23.04.2020 o 02:11, Ananyev, Konstantin pisze:
> Actually looking at app/test/test_security.c
> I also see a few '#ifdef RTE_DEBUG's.
> Let say:
>
> +static int
> +test_get_userdata_inv_context(void)
> +{
> +#ifdef RTE_DEBUG
> + uint64_t md = 0xDEADBEEF;
> +
> + void *ret = rte
W dniu 23.04.2020 o 06:07, Anoob Joseph pisze:
> Hi Konstantin,
>
> These are data path ops and so it will be better if we can avoid such checks
> in the datapath. The same is done in ethdev also.
>
> https://protect2.fireeye.com/url?k=d44931cf-89d2cdac-d448ba80-0cc47a31cdbc-8281a62b4c91d848&q=1
W dniu 23.04.2020 o 09:54, Ananyev, Konstantin pisze:
>> Hi Konstantin,
>>
>> These are data path ops and so it will be better if we can avoid such checks
>> in the datapath. The same is done in ethdev also.
> AFAIK, get_userdata is an *optional* dev-ops function that can be used by
> data-pat
On 4/23/20 2:30 PM, Marvin Liu wrote:
> Introduce free threshold setting in Rx queue, default value of it is 32.
> Limiated threshold size to multiple of four as only vectorized packed Rx
s/Limiated/Limit the/
> function will utilize it. Virtio driver will rearm Rx queue when more
> than rx_fre
> -Original Message-
> From: Lukasz Wojciechowski
> Sent: Thursday, April 23, 2020 9:06 AM
> To: Ananyev, Konstantin ; Anoob Joseph
> ; dev@dpdk.org
> Cc: akhil.go...@nxp.com; Doherty, Declan ;
> techbo...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] security: fix crash at accessing
> n
> >> Hi Konstantin,
> >>
> >> These are data path ops and so it will be better if we can avoid such
> >> checks in the datapath. The same is done in ethdev also.
> > AFAIK, get_userdata is an *optional* dev-ops function that can be used by
> > data-path.
> > So far there was no strict requireme
Many thanks Bruce,
Your suggestion is very good, I will try to do it
>-Original Message-
>From: Bruce Richardson [mailto:bruce.richard...@intel.com]
>Sent: Wednesday, April 22, 2020 7:12 PM
>To: Zhang, XuemingX
>Cc: dev@dpdk.org; tho...@monjalon.net; Chen, Zhaoyan
>; Ma, LihongX
>Subj
> W dniu 23.04.2020 o 06:07, Anoob Joseph pisze:
> > Hi Konstantin,
> >
> > These are data path ops and so it will be better if we can avoid such
> > checks in the datapath. The same is done in ethdev also.
> >
> > https://protect2.fireeye.com/url?k=d44931cf-89d2cdac-d448ba80-0cc47a31cdbc-
> 8281
On 4/23/20 2:30 PM, Marvin Liu wrote:
> Previously, virtio split ring vectorized path is enabled as default.
s/is/was/
s/as/by/
> This is not suitable for everyone because of that path not follow virtio
s/because of that path not follow/because that path does not follow the/
> spec. Add new
When having multiple VLANs in the packet, parse_ethernet
is cabable of parsing only the first vlan.
add parsing for mutliple VLAN headers in the packet.
Fixes: 51f694dd40f5 ("app/testpmd: rework checksum forward engine")
Cc: sta...@dpdk.org
Signed-off-by: Raslan Darawsheh
Acked-by: Ori Kam
Ack
On 4/23/20 2:31 PM, Marvin Liu wrote:
> Ring initialzation is different when inorder feature negotiated. This
s/initialzation/initialization/
> action should dependent on negotiated feature bits.
>
> Signed-off-by: Marvin Liu
>
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
> -Original Message-
> From: Maxime Coquelin
> Sent: Thursday, April 23, 2020 4:34 PM
> To: Liu, Yong ; Ye, Xiaolong ;
> Wang, Zhihong
> Cc: Van Haaren, Harry ; dev@dpdk.org
> Subject: Re: [PATCH v8 2/9] net/virtio: enable vectorized path
>
>
>
> On 4/23/20 2:30 PM, Marvin Liu wrote
On 4/23/20 10:46 AM, Liu, Yong wrote:
>
>
>> -Original Message-
>> From: Maxime Coquelin
>> Sent: Thursday, April 23, 2020 4:34 PM
>> To: Liu, Yong ; Ye, Xiaolong ;
>> Wang, Zhihong
>> Cc: Van Haaren, Harry ; dev@dpdk.org
>> Subject: Re: [PATCH v8 2/9] net/virtio: enable vectorized p
Hi Raslan,
> -Original Message-
> From: Raslan Darawsheh
> Sent: Thursday, April 23, 2020 9:41 AM
> To: Iremonger, Bernard ; Wu, Jingjing
> ; Lu, Wenzhuo
> Cc: Yigit, Ferruh ; dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH v2] app/testpmd: add parsing for multiple VLAN headers
Might be
Hi,
> -Original Message-
> From: Iremonger, Bernard
> Sent: Thursday, April 23, 2020 12:00 PM
> To: Raslan Darawsheh ; Wu, Jingjing
> ; Lu, Wenzhuo
> Cc: Yigit, Ferruh ; dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH v2] app/testpmd: add parsing for multiple VLAN headers
>
> Hi Ras
When having QINQ VLAN headers in the packet, parse_ethernet
is cabable of parsing only the first vlan.
add parsing for QINQ VLAN headers in the packet.
Fixes: 51f694dd40f5 ("app/testpmd: rework checksum forward engine")
Cc: sta...@dpdk.org
Signed-off-by: Raslan Darawsheh
Acked-by: Ori Kam
Acke
On 2020-04-23 09:27 GMT+0200 Thomas Monjalon wrote:
> 23/04/2020 01:51, Ranjit Menon:
> > On 4/22/2020 12:27 AM, tal...@mellanox.com wrote:
> > > From: Tal Shnaiderman
> > >
> > > Move common functions between Unix and Windows to eal_config.c.
> >
> > Like other files in common, we should ca
Hi Konstantin,
Please see inline.
Thanks,
Anoob
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Thursday, April 23, 2020 1:24 PM
> To: Anoob Joseph ; dev@dpdk.org
> Cc: akhil.go...@nxp.com; Doherty, Declan ;
> techbo...@dpdk.org
> Subject: [EXT] RE: [dpdk-dev] [PATCH] security:
> -Original Message-
> From: Maxime Coquelin
> Sent: Thursday, April 23, 2020 4:50 PM
> To: Liu, Yong ; Ye, Xiaolong ;
> Wang, Zhihong
> Cc: Van Haaren, Harry ; dev@dpdk.org
> Subject: Re: [PATCH v8 2/9] net/virtio: enable vectorized path
>
>
>
> On 4/23/20 10:46 AM, Liu, Yong wrote
On Wed, 2020-04-22 at 17:35 -0300, Dan Gora wrote:
> On Wed, Apr 22, 2020 at 5:14 PM Mattias Rönnblom
> wrote:
> > On 2020-04-22 19:44, Dan Gora wrote:
> > > On Wed, Apr 22, 2020 at 5:28 AM Mattias Rönnblom
> > > wrote:
> > > > On 2020-04-21 21:54, Dan Gora wrote:
> > > > > The getentropy() funct
Clarify the reasons behind the three part version numbering scheme.
Documents the fixes made in f26c2b3.
Signed-off-by: Ray Kinsella
Signed-off-by: Bruce Richardson
---
doc/guides/contributing/abi_policy.rst | 3 ++-
doc/guides/rel_notes/release_20_05.rst | 12
2 files changed, 14
On Thu, 2020-04-09 at 11:37 +0200, Thomas Monjalon wrote:
> 09/04/2020 11:19, Bruce Richardson:
> > On Wed, Apr 08, 2020 at 08:03:26PM +0200, Thomas Monjalon wrote:
> > > 08/04/2020 18:49, Ciara Power:
> > > > This patchset extensively reworks the telemetry library adding new
> > > > functionality
23/04/2020 12:30, Luca Boccassi:
> On Thu, 2020-04-09 at 11:37 +0200, Thomas Monjalon wrote:
> > 09/04/2020 11:19, Bruce Richardson:
> > > On Wed, Apr 08, 2020 at 08:03:26PM +0200, Thomas Monjalon wrote:
> > > > 08/04/2020 18:49, Ciara Power:
> > > > > This patchset extensively reworks the telemetr
23/04/2020 11:06, Dmitry Kozlyuk:
> On 2020-04-23 09:27 GMT+0200 Thomas Monjalon wrote:
> > 23/04/2020 01:51, Ranjit Menon:
> > > On 4/22/2020 12:27 AM, tal...@mellanox.com wrote:
> > > > From: Tal Shnaiderman
> > > >
> > > > Move common functions between Unix and Windows to eal_config.c.
> >
> > >
> > > These are data path ops and so it will be better if we can avoid such
> > > checks in
> > the datapath. The same is done in ethdev also.
> >
> > AFAIK, get_userdata is an *optional* dev-ops function that can be used by
> > data-
> > path.
> > So far there was no strict requirement
On Wed, Apr 22, 2020 at 02:16:57PM +0200, Thomas Monjalon wrote:
> 22/04/2020 14:01, Neil Horman:
> > On Tue, Apr 21, 2020 at 11:42:42PM +0200, Thomas Monjalon wrote:
> > > 21/04/2020 20:56, Neil Horman:
> > > > On Tue, Apr 21, 2020 at 01:46:43PM +0200, Thomas Monjalon wrote:
> > > > > 21/04/2020 1
On Wed, Apr 22, 2020 at 02:18:05PM +0200, Thomas Monjalon wrote:
> 22/04/2020 14:07, Neil Horman:
> > On Wed, Apr 22, 2020 at 12:43:44PM +0100, Ray Kinsella wrote:
> > > On 21/04/2020 22:42, Thomas Monjalon wrote:
> > > > 21/04/2020 20:56, Neil Horman:
> > > >> On Tue, Apr 21, 2020 at 01:46:43PM +0
Hi Konstantin,
Please see inline.
Thanks,
Anoob
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Thursday, April 23, 2020 4:25 PM
> To: Anoob Joseph ; dev@dpdk.org; Lukasz
> Wojciechowski
> Cc: akhil.go...@nxp.com; Doherty, Declan
> Subject: [EXT] RE: [dpdk-dev] [PATCH] securi
On Thu, 2020-04-23 at 12:44 +0200, Thomas Monjalon wrote:
> 23/04/2020 12:30, Luca Boccassi:
> > On Thu, 2020-04-09 at 11:37 +0200, Thomas Monjalon wrote:
> > > 09/04/2020 11:19, Bruce Richardson:
> > > > On Wed, Apr 08, 2020 at 08:03:26PM +0200, Thomas Monjalon wrote:
> > > > > 08/04/2020 18:49, C
On 23-Apr-20 7:36 AM, Feng Li wrote:
Hi,
I have tested as follows, the core dump file is ~ 200KB.
It should generate one core dump file each crash.
#include
#include
#include
#include
#include
#include
int main(int argc, char** argv) {
// FIXME(fengli): X
uint64_t size = 1<<30;
void*
Update QAT pmd to support AES-256 DOCSIS
Signed-off-by: Mairtin o Loingsigh
---
doc/guides/rel_notes/release_20_05.rst | 4
1 file changed, 4 insertions(+)
diff --git a/doc/guides/rel_notes/release_20_05.rst
b/doc/guides/rel_notes/release_20_05.rst
index 7f2049a0f..5e81c1964 100644
--- a/
On 2020-04-22 22:35, Dan Gora wrote:
> On Wed, Apr 22, 2020 at 5:14 PM Mattias Rönnblom
> wrote:
>> On 2020-04-22 19:44, Dan Gora wrote:
>>> On Wed, Apr 22, 2020 at 5:28 AM Mattias Rönnblom
>>> wrote:
On 2020-04-21 21:54, Dan Gora wrote:
> The getentropy() function was introduced into gl
On 4/22/20 10:37 AM, Fady Bader wrote:
> The code didn't compile when using exported mempool functions under windows.
>
> compilation error logs:
> rte_mempool_exports.def : error LNK2001:
> unresolved external symbol rte_mempool_cache_flush
> rte_mempool_exports.def : error LNK2001:
> unresolved
Hi Anoob/Konstantin,
> >
> > Check that ops->get_userdata is a valid function pointer will be compiled
> > out.
> > So PMDs that don't implement this function will crash in
> > rte_security_get_userdata().
> > In our particular case - ixgbe.
> > Same story with rte_security_set_pkt_metadata() - s
1. ln_en bit should not be turned on, since we only support Rx VEB.
2. lan_en bit need to be turn on for a DCF switch rule, otherwise
any Tx packet that hit on a rule will be dropped.
Fixes: fed0c5ca5f19 ("net/ice/base: support programming a new switch recipe")
Signed-off-by: Qi Zhang
---
v2:
Adding per core packet handling stats to analyze traffic distribution
when multiple cores are engaged.
Since aggregating the packet stats across cores would affect
performance, keeping the feature disabled using compile time flags.
Signed-off-by: Anoob Joseph
---
v2:
* Added lookup failure case
Minutes 23 April 2020
-
Agenda:
* Release Dates
* Subtrees
Participants:
* Arm
* Debian/Microsoft
* Intel
* Marvell
* Mellanox
* NXP
* Red Hat
Release Dates
-
* v20.05 dates:
* Integration/Merge/RC1 pushed to *Friday 24 April 2020*
* Release:
W dniu 23.04.2020 o 14:55, Akhil Goyal pisze:
> Hi Anoob/Konstantin,
>>> Check that ops->get_userdata is a valid function pointer will be compiled
>>> out.
>>> So PMDs that don't implement this function will crash in
>>> rte_security_get_userdata().
>>> In our particular case - ixgbe.
>>> Same s
Hi Vladimir,
Apologies for late review.
My comments below.
> K32V64 hash is a hash table that supports 32 bit keys and 64 bit values.
> This table is hash function agnostic so user must provide
> precalculated hash signature for add/delete/lookup operations.
>
> Signed-off-by: Vladimir Medvedki
>
> Few of the NIST TDES test files don't contain TDES string.
> Added indicators to identify such files. These indicators are part of only
> NIST
> TDES test vector files.
>
> Fixes: 527cbf3d5ee3 ("examples/fips_validation: support TDES parsing")
>
> Signed-off-by: Archana Muniganti
> Signed-
On 4/13/20 5:21 PM, xiangxia.m@gmail.com wrote:
> From: Tonghao Zhang
>
> The order of mempool initiation affects mempool index in the
> rte_mempool_ops_table. For example, when building APPs with:
>
> $ gcc -lrte_mempool_bucket -lrte_mempool_ring ...
>
> The "bucket" mempool will be regist
When adding support for DOCSIS AES-256,
when setting the cipher parameters, all key sizes
were accepted, but only 128-bit and 256-bit keys
are supported.
Fixes: 124d04b43743 ("crypto/aesni_mb: support DOCSIS AES-256")
Signed-off-by: Pablo de Lara
Acked-by: Mairtin o Loingsigh
---
v2:
- Fixed c
Hi Akhil,
>
> Hi Anoob/Konstantin,
> > >
> > > Check that ops->get_userdata is a valid function pointer will be compiled
> > > out.
> > > So PMDs that don't implement this function will crash in
> > > rte_security_get_userdata().
> > > In our particular case - ixgbe.
> > > Same story with rte_s
>
> Hi Akhil,
>
> >
> > Hi Anoob/Konstantin,
> > > >
> > > > Check that ops->get_userdata is a valid function pointer will be
> > > > compiled
> out.
> > > > So PMDs that don't implement this function will crash in
> > > > rte_security_get_userdata().
> > > > In our particular case - ixgbe.
>
On Wed, Apr 22, 2020 at 9:04 PM wrote:
> This patch set contains
>
>
> # The native implementation of common trace format(CTF)[1] based tracer
> # Public API to create the trace points.
> # Add tracepoints to eal, ethdev, mempool, eventdev and cryptodev
> library for traci
Caught by OBS on armv7:
In file included from .../lib/librte_eal/include/rte_string_fns.h:21,
from .../lib/librte_kvargs/rte_kvargs.c:9:
.../lib/librte_eal/include/rte_common.h:67:37: error: expected '=', ',',
';', 'asm' or '__attribute__' before '__rte_aligned'
67 | typedef u
Implemented the needed Windows eal timer functions.
Signed-off-by: Fady Bader
---
lib/librte_eal/common/meson.build | 1 +
lib/librte_eal/windows/eal.c| 6 +++
lib/librte_eal/windows/eal_timer.c | 67 +
lib/librte_eal/windows/include/rte_o
This patchset splits OS dependent EAL timer functions and implements them for
windows.
Depends-on: series-9374 ("Windows basic memory management")
Fady Bader (2):
timer: move from common to Unix directory
timer: support EAL functions on Windows
lib/librte_eal/common/eal_common_timer.c | 22
Eal common timer doesn't compile under Windows.
Compilation log:
error LNK2019:
unresolved external symbol nanosleep referenced in function rte_delay_us_sleep
error LNK2019:
unresolved external symbol get_tsc_freq referenced in function set_tsc_freq
error LNK2019:
unresolved external symbol sleep
From: Kalesh AP
Please apply.
Kalesh AP (1):
net/bnxt: fix to reset VNIC rxq count on VNIC free
Rahul Gupta (1):
net/bnxt: fix for memleak during queue restart
drivers/net/bnxt/bnxt_ethdev.c | 2 ++
drivers/net/bnxt/bnxt_hwrm.c | 12
drivers/net/bnxt/bnxt_rxr.c| 44 +++
From: Kalesh AP
bnxt_free_one_vnic and bnxt_setup_one_vnic are called on configuring
port vlan stripping. bnxt_setup_one_vnic keeps incrementing the
vnic rx_queue_cnt. Fix to reset vnic rx_queue_cnt in bnxt_free_one_vnic.
Fixes: cfadfee41ed1 ("net/bnxt: fix VLAN strip")
Cc: sta...@dpdk.org
Sign
From: Rahul Gupta
During port 0 rxq 1 start ie queue start,
bnxt_free_hwrm_rx_ring() we are clearing the pointers to mbuf array.
Due to this we overwrite the queue with fresh mbuf allocations
causing previously allocated mbufs to leak.
Add a check before allocating mbuf to replenish only empty mb
> >
> > Hi Akhil,
> >
> > >
> > > Hi Anoob/Konstantin,
> > > > >
> > > > > Check that ops->get_userdata is a valid function pointer will be
> > > > > compiled
> > out.
> > > > > So PMDs that don't implement this function will crash in
> > > > > rte_security_get_userdata().
> > > > > In our parti
Valid checks for optional function pointers inside dev-ops
were disabled by undefined macro.
Fixes: b6ee98547847 ("security: fix verification of parameters")
Cc: sta...@dpdk.org
Signed-off-by: Konstantin Ananyev
---
lib/librte_security/rte_security.c | 9 ++---
1 file changed, 6 insertions(
On 2020-04-23 17:43 GMT+0300 Fady Bader wrote:
[snip]
> diff --git a/lib/librte_eal/windows/eal_timer.c
> b/lib/librte_eal/windows/eal_timer.c
> new file mode 100644
> index 0..73eaff948
> --- /dev/null
> +++ b/lib/librte_eal/windows/eal_timer.c
> @@ -0,0 +1,67 @@
> +/* SPDX-License-Identi
Tested-by: Wang, Yinan
> -Original Message-
> From: dev On Behalf Of Marvin Liu
> Sent: 2020年4月23日 20:31
> To: maxime.coque...@redhat.com; Ye, Xiaolong ;
> Wang, Zhihong
> Cc: Van Haaren, Harry ; dev@dpdk.org; Liu,
> Yong
> Subject: [dpdk-dev] [PATCH v8 0/9] add packed ring vectorized
Add new test-case to measure performance of
ipsec data-path functions.
Signed-off-by: Savinay Dharmappa
---
MAINTAINERS| 2 +-
app/test/Makefile | 2 +-
app/test/meson.build | 2 +
app/test/test_ipsec_perf.c | 614 +
4 file
23/04/2020 16:24, David Marchand:
> Caught by OBS on armv7:
>
> In file included from .../lib/librte_eal/include/rte_string_fns.h:21,
> from .../lib/librte_kvargs/rte_kvargs.c:9:
> .../lib/librte_eal/include/rte_common.h:67:37: error: expected '=', ',',
> ';', 'asm' or '__attribu
Avoid dump all mapped memory to a core dump file when crash.
Otherwise it will very large and it's hard to analyze with gdb.
In my test, it will dump 128GiB memory to a core dump file when integrated
to spdk with default configuration.
Signed-off-by: Li Feng
---
lib/librte_eal/common/eal_common
> Valid checks for optional function pointers inside dev-ops
> were disabled by undefined macro.
>
> Fixes: b6ee98547847 ("security: fix verification of parameters")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Konstantin Ananyev
> ---
Acked-by: Akhil Goyal
Anoob,
Do you have any concerns over
Hi Akhil,
I have my concerns over unwanted checks in the datapath. Something that crypto
enqueue/dequeue APIs are not doing is being enforced on other APIs. As
Konstantin had suggested, PMDs (IXGBE here) could define a function which
returns -ENOTSUP and it would have been win-win for everyone.
Hi Anoob,
>
> Hi Akhil,
>
> I have my concerns over unwanted checks in the datapath. Something that
> crypto enqueue/dequeue APIs are not doing is being enforced on other APIs. As
> Konstantin had suggested, PMDs (IXGBE here) could define a function which
> returns -ENOTSUP and it would have been
Hi everyone,
> >
> > On 2020-04-16 12:18, Medvedkin, Vladimir wrote:
> > > Hi Mattias,
> > >
> > > -Original Message-
> > > From: Mattias Rönnblom
> > > Sent: Wednesday, April 15, 2020 7:52 PM
> > > To: Medvedkin, Vladimir ; dev@dpdk.org
> > > Cc: Ananyev, Konstantin ; Wang, Yipeng1
> >
After re-enabling checks for non-implemneted ops in non-debug mode
in librte_security set_pkt_metadata and get_userdata functions,
tests verifying proper work of tests can be enabled also.
Signed-off-by: Lukasz Wojciechowski
---
app/test/test_security.c | 8
1 file changed, 8 deletions(
W dniu 23.04.2020 o 18:14, Akhil Goyal pisze:
> Hi Anoob,
>> Hi Akhil,
>>
>> I have my concerns over unwanted checks in the datapath. Something that
>> crypto enqueue/dequeue APIs are not doing is being enforced on other APIs. As
>> Konstantin had suggested, PMDs (IXGBE here) could define a functio
The rte_atomic ops and rte_smp barriers enforce DMB barriers on aarch64.
Using c11 atomics with explicit memory ordering instead of the rte_atomic
ops and rte_smp barriers for inter-threads synchronization can uplift the
performance on aarch64 and no performance loss on x86.
This patchset contains
From: Honnappa Nagarahalli
The logic to identify if the MT unsafe service is running on another
core can return -EBUSY spuriously. In such cases, running the service
becomes costlier than using atomic operations. Assume that the
application passes the right parameters and reduces the number of
in
From: Honnappa Nagarahalli
The MT unsafe service might get configured to run on another core
while the service is running currently. This might result in the
MT unsafe service running on multiple cores simultaneously. Use
'execute_lock' always when the service is MT unsafe.
Fixes: e9139a32f6e8 (
On 4/23/2020 3:48 AM, Thomas Monjalon wrote:
23/04/2020 11:06, Dmitry Kozlyuk:
On 2020-04-23 09:27 GMT+0200 Thomas Monjalon wrote:
23/04/2020 01:51, Ranjit Menon:
On 4/22/2020 12:27 AM, tal...@mellanox.com wrote:
From: Tal Shnaiderman
Move common functions between Unix and Windows to eal_co
clean up rte prefix from static functions.
remove unused parameter for service_dump_one function.
Signed-off-by: Phil Yang
Reviewed-by: Honnappa Nagarahalli
---
lib/librte_eal/common/rte_service.c | 34 +++---
1 file changed, 11 insertions(+), 23 deletions(-)
diff -
The service id validation is duplicated, remove the redundant code
in the calling functions.
Signed-off-by: Phil Yang
Reviewed-by: Honnappa Nagarahalli
---
lib/librte_eal/common/rte_service.c | 28 ++--
1 file changed, 6 insertions(+), 22 deletions(-)
diff --git a/lib/l
The num_mapped_cores is used as a statistics. Use c11 atomics with
RELAXED ordering for num_mapped_cores instead of rte_atomic ops which
enforce unnessary barriers on aarch64.
Replace execute_lock operations to spinlock_try_lock to avoid duplicate
code.
Signed-off-by: Phil Yang
Reviewed-by: Honn
The runstate, comp_runstate and app_runstate are used as guard variables
in the service core lib. To guarantee the inter-threads visibility of
these guard variables, it uses rte_smp_r/wmb. This patch use c11 atomic
built-ins to relax these barriers.
Signed-off-by: Phil Yang
Reviewed-by: Honnappa
On 23-Apr-20 4:43 PM, Li Feng wrote:
Avoid dump all mapped memory to a core dump file when crash.
Otherwise it will very large and it's hard to analyze with gdb.
In my test, it will dump 128GiB memory to a core dump file when integrated
to spdk with default configuration.
Suggested rewording:
The rarp packet broadcast flag is synchronized with rte_atomic_XX APIs
which is a full barrier, DMB, on aarch64. This patch optimized it with
c11 atomic one-way barrier.
Signed-off-by: Phil Yang
Reviewed-by: Gavin Hu
Reviewed-by: Honnappa Nagarahalli
Reviewed-by: Joyce Kong
---
v2:
split from
For SA outbound packets, rte_atomic64_add_return is used to generate
SQN atomically. This introduced an unnecessary full barrier by calling
the '__sync' builtin implemented rte_atomic_XX API on aarch64. This
patch optimized it with c11 atomic and eliminated the expensive barrier
for aarch64.
Signe
On Thu, Apr 23, 2020 at 9:36 AM Mattias Rönnblom
wrote:
> >>
> >> /dev/urandom is basically only a different interface to the same
> >> underlying mechanism.
> >>
> >> Such an alternative would look something like:
> >>
> >> static int
> >> getentropy(void *buffer, size_t length)
> >> {
> >>
On Thu, Apr 23, 2020 at 12:59 PM Luca Boccassi wrote:
> > >
> > > /dev/urandom is basically only a different interface to the same
> > > underlying mechanism.
>
> This is not the whole story though - while the end result when all
> works is the same, there are important differences in getting ther
On Wed, Apr 22, 2020 at 11:39 PM Stephen Hemminger
wrote:
>
> On Wed, 22 Apr 2020 20:42:54 -0300
> Dan Gora wrote:
>
> > + fd = open("/dev/urandom", O_RDONLY);
> > + if (fd < 0) {
> > + errno = ENODEV;
> > + return -1;
> > + }
> > +
> > + end = start + leng
On Thu, Apr 23, 2020 at 10:47 PM Phil Yang wrote:
>
> For SA outbound packets, rte_atomic64_add_return is used to generate
> SQN atomically. This introduced an unnecessary full barrier by calling
> the '__sync' builtin implemented rte_atomic_XX API on aarch64. This
> patch optimized it with c11 at
>
> For SA outbound packets, rte_atomic64_add_return is used to generate
> SQN atomically. This introduced an unnecessary full barrier by calling
> the '__sync' builtin implemented rte_atomic_XX API on aarch64. This
> patch optimized it with c11 atomic and eliminated the expensive barrier
> for aa
Specified pattern may be translated in different manner.
For example the pattern "eth / ipv4" can be translated to match
untagged packets only, since the pattern doesn't specify a vlan item.
It can also be translated to match both tagged and untagged packets,
for the same reason.
This patch updates
Meeting notes for the DPDK technical board meeting held on 2020-04-22
Attendees:
- Bruce Richardson
- Ferruh Yigit
- Hemant Agrawal
- Honnappa Nagarahalli (Chair)
- Jerin Jacob
- Kevin Traynor
- Konstantin Ananyev
- Maxime Coquelin
On Thu, Apr 23, 2020 at 6:34 PM Burakov, Anatoly
wrote:
> > diff --git a/lib/librte_eal/common/eal_common_memory.c
> > b/lib/librte_eal/common/eal_common_memory.c
> > index cc7d54e0c..2d9564b28 100644
> > --- a/lib/librte_eal/common/eal_common_memory.c
> > +++ b/lib/librte_eal/common/eal_common_m
Hi Erik,
> Subject: [PATCH v2] lib/timer: relax barrier for status update
>
> Volatile has no ordering semantics. The rte_timer structure defines timer
> status as a volatile variable and uses the rte_r/wmb barrier to guarantee
> inter-thread visibility.
>
> This patch optimized the volatile ope
On Thu, Apr 23, 2020 at 7:46 AM Kalesh A P <
kalesh-anakkur.pura...@broadcom.com> wrote:
> From: Kalesh AP
>
> Please apply.
>
Applied to dpdk-next-net-brcm. Thanks
>
> Kalesh AP (1):
> net/bnxt: fix to reset VNIC rxq count on VNIC free
>
> Rahul Gupta (1):
> net/bnxt: fix for memleak durin
Hi Honnappa,
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Thursday, April 23, 2020 3:06 PM
> To: Phil Yang ; Carrillo, Erik G
> ; rsanf...@akamai.com; dev@dpdk.org
> Cc: tho...@monjalon.net; david.march...@redhat.com; Ananyev,
> Konstantin ; jer...@marvell.com;
> hemant.agra.
This patch set introduced vectorized path for packed ring.
The size of packed ring descriptor is 16Bytes. Four batched descriptors
are just placed into one cacheline. AVX512 instructions can well handle
this kind of data. Packed ring TX path can fully transformed into
vectorized path. Packed ring
Ring initialization is different when inorder feature negotiated. This
action should dependent on negotiated feature bits.
Signed-off-by: Marvin Liu
Reviewed-by: Maxime Coquelin
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 94ba7a3ec..e450477e8 100644
--
Introduce free threshold setting in Rx queue, its default value is 32.
Limit the threshold size to multiple of four as only vectorized packed
Rx function will utilize it. Virtio driver will rearm Rx queue when
more than rx_free_thresh descs were dequeued.
Signed-off-by: Marvin Liu
Reviewed-by: Ma
Optimize packed ring Rx path with SIMD instructions. Solution of
optimization is pretty like vhost, is that split path into batch and
single functions. Batch function is further optimized by AVX512
instructions. Also pad desc extra structure to 16 bytes aligned, thus
four elements will be saved in
Optimize packed ring Tx path alike Rx path. Split Tx path into batch and
single Tx functions. Batch function is further optimized by AVX512
instructions.
Signed-off-by: Marvin Liu
diff --git a/drivers/net/virtio/virtio_ethdev.h
b/drivers/net/virtio/virtio_ethdev.h
index 5c112cac7..b7d52d497 100
Add new devarg for virtio user device vectorized path selection. By
default vectorized path is disabled.
Signed-off-by: Marvin Liu
diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index 902a1f0cf..d59add23e 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.
Move xmit offload and packed ring xmit enqueue function to header file.
These functions will be reused by packed ring vectorized Tx function.
Signed-off-by: Marvin Liu
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index c9b6e7844..cf18fe564 100644
--- a/drivers
Previously, virtio split ring vectorized path was enabled by default.
This is not suitable for everyone because that path dose not follow
virtio spec. Add new devarg for virtio vectorized path selection. By
default vectorized path is disabled.
Signed-off-by: Marvin Liu
diff --git a/doc/guides/ni
Document packed virtqueue vectorized path selection logic in virtio net
PMD.
Signed-off-by: Marvin Liu
diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index d59add23e..dbcf49ae1 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -482,6 +482,13 @@ acc
Rewrite vectorized path selection logic. Default setting comes from
vectorized devarg, then checks each criteria.
Packed ring vectorized path need:
AVX512F and required extensions are supported by compiler and host
VERSION_1 and IN_ORDER features are negotiated
mergeable feature is not
1 - 100 of 121 matches
Mail list logo
