Hello,
David Marchand writes:
>> > +
>> > LIBABIGAIL_REPO=${LIBABIGAIL_REPO:-https://sourceware.org/git/libabigail.git}
>> > +LIBABIGAIL_VERSION=${LIBABIGAIL_VERSION:-libabigail-1.6}
>> > +
>> > +if [ "$(cat libabigail/VERSION 2>/dev/null)" != "$LIBABIGAIL_VERSION"
>> > ]; then
>> >
Hi Tiwei,
On 2/19/20 4:44 AM, Tiwei Bie wrote:
> On Tue, Feb 18, 2020 at 06:22:40PM +0100, Maxime Coquelin wrote:
>> Ethdev's .dev_configure callback can be called multiple
>> time during a device life-time, but Vhost makes the
>> wrong assumption that it is not the case and try to
>> setup again
With this patch it is possible to connect 2 DPDK memifs into loopback,
i.e. when they have the same id and different roles, as for example:
"--vdev=net_memif0,role=master,id=0"
"--vdev=net_memif1,role=slave,id=0"
Signed-off-by: Július Milan
---
drivers/net/memif/memif_socket.c | 30
This patch fixes the situation when there is already connected pair of
memif interfaces and another slave tries to initiate the connection with
(already occupied) master. Expected behavior is that the second slave
is refused and gets disconnect message with reason: "Already connected",
while old co
Series enables loopback for memif and fixes an issue found
during testing of loopback functionality.
Július Milan (2):
net/memif: enable loopback
net/memif: fix init handling when already connected
memif_socket.c | 44 ++--
rte_eth_memif.c |1 +
Series enables loopback for memif and fixes one issue found
during testing of loopback functionality.
Július Milan (2):
net/memif: enable loopback
net/memif: fix init handling when already connected
memif_socket.c | 44 ++--
rte_eth_memif.c |1 +
The hairpin TX/RX queue depth and packet size is fixed in the past.
When the firmware has some fix or improvement, the PMD will not
make full use of it. And also, 32 packets for a single queue will not
guarantee a good performance for hairpin flows. It will make the
stride size larger and for small
Series enables loopback for memif and fixes one issue found
during testing of loopback functionality.
Július Milan (2):
net/memif: enable loopback
net/memif: fix init handling when already connected
memif_socket.c | 44 ++--
rte_eth_memif.c |1 +
The RSS expansion feature was introduced to split RSS flows according to
the adjustment between the RSS types and the flow items.
The expansion function gets an item tree for the above adjustment from
the caller which reflects the HW needs.
The standard vxlan header next protocol is always ethern
OK James,
Thanks for sharing your own experience.
What I would need right now is to know from maintainers if this latency
behaviour is something inherent in DPDK in the particular case we are
talking about. Furthermore, I would also appreciate it if some maintainer
could tell us if there is some w
> -Original Message-
> From: Matan Azrad
> Sent: Wednesday, February 19, 2020 10:30
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; sta...@dpdk.org
> Subject: [PATCH] net/mlx5: fix L3 vxlan RSS expansion
>
> The RSS expansion feature was introduced to split RSS flows according to the
> adjustm
> -Original Message-
> From: Matan Azrad
> Sent: Wednesday, February 19, 2020 8:52
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; sta...@dpdk.org
> Subject: [PATCH] net/mlx5: fix geneve tunnel flow validation
>
> The new Mellanox devices support geneve tunnel statless offloads by native
> HW
> -Original Message-
> From: Matan Azrad
> Sent: Tuesday, February 18, 2020 14:25
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; Maxime Coquelin
>
> Subject: [PATCH] vdpa/mlx5: fix completion queue arming
>
> The mlx5 vDPA driver manages QP and CQ in order to forward the HW event
> to the gu
On 2/18/20 1:25 PM, Matan Azrad wrote:
> The mlx5 vDPA driver manages QP and CQ in order to forward the HW event
> to the guest by the callfd file descriptor for each virtq.
>
> The driver arms the CQ for the next CQE index that should be
> completed by the HW in order to create completion even
The dev_ops .get_module_info and .get_module_eeprom
were implemented in mlx5 in DPDK 19.11.
Fixes: 8a6a09f853a0 ("net/mlx5: support reading module EEPROM data")
Cc: sta...@dpdk.org
Reported-by: Gal Cohen
Signed-off-by: Thomas Monjalon
---
doc/guides/nics/features/mlx5.ini | 1 +
1 file changed
On 2/18/20 11:37 PM, Itsuro ODA wrote:
> Hi Yinan, Maxime, David,
>
> Thank you for quick test, fix and review.
> It looks good to me.
>
> The original fix has been queued to stable 19.11.1.
> It should be rejected or applied this fix too.
Thanks for the heads-up.
I added Cc: sta...@dpdk.org
Cc'ing stable as the commit it fixes was backported to v19.11 stable
branch.
On 2/18/20 6:22 PM, Maxime Coquelin wrote:
> This series fixes regression introduced in v20.02-rc3.
> First patch fixes the error path, and second one fix
> Vhost device reconfigure issue reported by Yinan.
>
> v2:
> ---
Hi Fan, Akhil,
Can you review this patch? Without this change, the app would not work for
AES-GCM decrypt.
Thanks,
Anoob
> -Original Message-
> From: Anoob Joseph
> Sent: Tuesday, February 18, 2020 4:01 PM
> To: Marko Kovacevic ; Akhil Goyal
> ; David Marchand
> Cc: Anoob Joseph ; Fan
On 17-Feb-20 9:57 AM, Kamaraj P wrote:
Hi Anatoly,
Thanks for the clarifications.
Currently we are migrating to the new DPDK 18.11 ( from 17.05). Here is
our configuration:
===
We have configured the "--legacy-mem" option and
On 2/18/20 1:25 PM, Matan Azrad wrote:
> The mlx5 vDPA driver manages QP and CQ in order to forward the HW event
> to the guest by the callfd file descriptor for each virtq.
>
> The driver arms the CQ for the next CQE index that should be
> completed by the HW in order to create completion even
On 2/18/20 6:22 PM, Maxime Coquelin wrote:
> This series fixes regression introduced in v20.02-rc3.
> First patch fixes the error path, and second one fix
> Vhost device reconfigure issue reported by Yinan.
>
> v2:
> ---
> - Fix error path order (David)
>
> Maxime Coquelin (2):
> net/vhost:
From: Zhu Tao
To bind a queue to an MSI-X interrupt, need to set some register.
The register consists of many parts, each of which has several
bits; therefore, the shift operator '<<' was used; so the operator
'<' in the code should be '<<'.
Old code adds 1 on even MSI-X interrupt vector index u
Hi ,
I put some maintainers as destination that could provide some extra
information on this issue.
I hope they can shed some light on this.
Regards
El mié., 19 feb. 2020 a las 9:29, Victor Huertas ()
escribió:
> OK James,
> Thanks for sharing your own experience.
> What I would need right now
The macros RTE_MIN and RTE_MAX can be used in DPDK applications.
Signed-off-by: Thomas Monjalon
---
app/test-crypto-perf/cperf_test_latency.c | 19 ---
1 file changed, 8 insertions(+), 11 deletions(-)
diff --git a/app/test-crypto-perf/cperf_test_latency.c
b/app/test-crypto-perf
The macros RTE_MIN and RTE_MAX can be used in DPDK applications.
This change implies fixing the sign of used_len as size_t
as defined in vhost_strcpy_pad().
Signed-off-by: Thomas Monjalon
---
examples/vhost_blk/blk.c | 4 ++--
examples/vhost_blk/vhost_blk.h| 2 --
examples/
Hi Victor,
I have no experience with ip_pipeline. I can at least say that this
latency is much higher that what you should get.
My initial thought was that you were using several pthreads bound to the
same core, but from what I read in your first mail, this is not the
case.
Do you have a simple
On 19/02/2020 10:23, Burakov, Anatoly wrote:
> On 17-Feb-20 9:57 AM, Kamaraj P wrote:
>> Hi Anatoly,
>> Thanks for the clarifications.
>>
>> Currently we are migrating to the new DPDK 18.11 ( from 17.05). Here is
>> our configuration:
>> ===
Hi Xiaojun,
1. No, don't base on the old patch. Just send new version.
2. Yes.
3. No need, just send to the same names as this version. Please remember to
add "-in-reply-to " when sending new version.
The Message-ID can be found in the email header of the previous patch or in
http://
Hi Kevin/Anatoly,
Yes we have the patch already included in our code base.
Looks like it get struck in the below piece of the code:
mapped_addr = mmap(requested_addr, (size_t)map_sz, PROT_READ,
mmap_flags, -1, 0);
Could you please share your thoughts on this?
Tha
On 19/02/2020 10:18, Maxime Coquelin wrote:
> Cc'ing stable as the commit it fixes was backported to v19.11 stable
> branch.
>
fyi - it is relevant to 18.11 also as 3d01b759d is fixing something
introduced in 18.11. I skipped 3d01b759d for now and will take the full
set together when available.
The name of the static function check_cryptodev_capablity()
is fixed for the word "capability".
There is no functional change.
The same typo is fixed in a comment in ip_fragmentation example.
Fixes: 05fe65eb66b2 ("test/ipsec: introduce functional test")
Cc: sta...@dpdk.org
Signed-off-by: Thomas
On Tue, Feb 18, 2020 at 03:06:51PM +0100, Victor Huertas wrote:
> Thanks Neil for your answer,
> I think I will try the first option you posted without touching the
> makefiles from the dpdk-stable and I will tell how it goes.
> So, as a conclusion, it seems that all the modifications I have done a
On Tue, Feb 18, 2020 at 10:50:09AM +0100, Thomas Monjalon wrote:
> 18/02/2020 10:42, Bruce Richardson:
> > On Tue, Feb 18, 2020 at 12:15:56AM +0100, Thomas Monjalon wrote:
> > > Hi,
> > >
> > > I would like to remind everybody our mistake when defining ABI versions.
> > > It has been "fixed" in th
>
> The name of the static function check_cryptodev_capablity()
> is fixed for the word "capability".
> There is no functional change.
>
> The same typo is fixed in a comment in ip_fragmentation example.
>
> Fixes: 05fe65eb66b2 ("test/ipsec: introduce functional test")
> Cc: sta...@dpdk.org
>
>
Hi,
> -Original Message-
> From: Shiri Kuzin
> Sent: Tuesday, February 18, 2020 1:53 PM
> To: dev@dpdk.org
> Cc: Matan Azrad ; Raslan Darawsheh
> ; Slava Ovsiienko ;
> sta...@dpdk.org
> Subject: [PATCH] net/mlx5: fix ipv6 type icmp6 header rewrite actions
>
> ICMPv6 header contains the c
Hi Oliver,
Thanks for your answer. I think that the most appropriate maintainer to
answer to this issue is Cristian as it is the maintainer of ip_pipeline.
In order to tell you how to reproduce the problem, you should go back to
the DPDK v17.11 and run the ip_pipeline app having a *.cfg configura
> -Original Message-
> From: Bing Zhao
> Sent: Wednesday, February 19, 2020 10:29 AM
> To: Ori Kam ; Slava Ovsiienko
>
> Cc: Raslan Darawsheh ; Matan Azrad
> ; dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH v2] net/mlx5: fix the hairpin queue capacity
>
> The hairpin TX/RX queue dept
On 2/19/2020 12:57 AM, Ye Xiaolong wrote:
> Nice doc about MTU of AF_XDP, thanks for the update.
>
> On 02/18, Ciara Loftus wrote:
>> Explain how kernel driver RX buffer sizes affect the maximum
>> MTU size in practice.
>>
>> Signed-off-by: Ciara Loftus
>> ---
>> doc/guides/nics/af_xdp.rst | 26 +
On 2/19/2020 10:11 AM, Thomas Monjalon wrote:
> The dev_ops .get_module_info and .get_module_eeprom
> were implemented in mlx5 in DPDK 19.11.
>
> Fixes: 8a6a09f853a0 ("net/mlx5: support reading module EEPROM data")
> Cc: sta...@dpdk.org
>
> Reported-by: Gal Cohen
> Signed-off-by: Thomas Monjalon
On 2/17/2020 4:13 PM, Ferruh Yigit wrote:
> Changed the ABI version to 20.0.1.
>
> Fixes: a72cf98cc93e ("net/ionic: add skeleton")
>
> Reported-by: David Marchand
> Signed-off-by: Ferruh Yigit
Applied to dpdk-next-net/master, thanks.
Coverity complains about out of bound access, which is a false positive.
The return value of the 'parse_port_list()' can't be bigger than
'maxsize' because of the logic in the function. ('value >= (int)maxsize'
check and 'marked[]' usage.)
But this is not explicitly clear, causing coverity warnin
19/02/2020 12:43, Neil Horman:
> On Tue, Feb 18, 2020 at 10:50:09AM +0100, Thomas Monjalon wrote:
> > 18/02/2020 10:42, Bruce Richardson:
> > > On Tue, Feb 18, 2020 at 12:15:56AM +0100, Thomas Monjalon wrote:
> > > > Hi,
> > > >
> > > > I would like to remind everybody our mistake when defining AB
>-Original Message-
>From: dev On Behalf Of Ye Xiaolong
>Sent: 19 February 2020 08:06
>To: Sun, GuinanX
>Cc: dev@dpdk.org; Lu, Wenzhuo ; Yang, Qiming
>; Zhang, Qi Z ; sta...@dpdk.org
>Subject: Re: [dpdk-dev] [PATCH v4] net/ixgbe: fix flow ctrl mode setting
>
>On 02/18, Guinan Sun wrote:
>
> -Original Message-
> From: Július Milan
> Sent: Wednesday, February 19, 2020 9:19 AM
> To: dev@dpdk.org; Jakub Grajciar -X (jgrajcia - PANTHEON TECH SRO at Cisco)
>
> Subject: [dpdk-dev][PATCH 2/2] net/memif: fix init handling when already
> connected
>
> This patch fixes the situation
> -Original Message-
> From: Július Milan
> Sent: Wednesday, February 19, 2020 9:19 AM
> To: dev@dpdk.org; Jakub Grajciar -X (jgrajcia - PANTHEON TECH SRO at Cisco)
>
> Subject: [dpdk-dev][PATCH 1/2] net/memif: enable loopback
>
> With this patch it is possible to connect 2 DPDK memifs i
Hi,
> -Original Message-
> From: dev On Behalf Of Matan Azrad
> Sent: Tuesday, February 18, 2020 4:45 PM
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix flow match on GRE key
>
> The RTE_FLOW_ITEM_TYPE_GRE_KEY means that GRE key shou
Hi all,
I would like to update our testing results with dpdk 20.02-rc3 from Red Hat.
PASS: All below 15 scenarios work well.
(1)Guest with device assignment(PF) throughput testing(1G hugepage size): PASS
(2)Guest with device assignment(PF) throughput testing(2M hugepage size) : PASS
(3)Gues
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, February 19, 2020 12:40 PM
> To: Lu, Wenzhuo ; Wu, Jingjing
> ; Iremonger, Bernard
> ; Lipiec, Herakliusz
> ; Govindharajan, Hariprasad
> ; Burakov, Anatoly
>
> Cc: dev@dpdk.org; Yigit, Ferruh
> Subject: [PATCH] app/testpmd: g
Hi,
> -Original Message-
> From: dev On Behalf Of Matan Azrad
> Sent: Wednesday, February 19, 2020 8:52 AM
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix geneve tunnel flow validation
>
> The new Mellanox devices support geneve tunn
On Wed, Feb 19, 2020 at 2:10 AM David Wilder wrote:
>
> This change follows the example of aarch64 Travis support
> by adding support for ppc64le gcc builds. Limitations for
> ppc64le are the same as aarch64 as described in commit
> 31bb45bcfd.
checkpatch did not catch it because of the \n (?).
On 19/02/2020 12:43, Thomas Monjalon wrote:
> 19/02/2020 12:43, Neil Horman:
>> On Tue, Feb 18, 2020 at 10:50:09AM +0100, Thomas Monjalon wrote:
>>> 18/02/2020 10:42, Bruce Richardson:
On Tue, Feb 18, 2020 at 12:15:56AM +0100, Thomas Monjalon wrote:
> Hi,
>
> I would like to rem
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, February 19, 2020 12:40 PM
> To: Lu, Wenzhuo ; Wu, Jingjing
> ; Iremonger, Bernard
> ; Lipiec, Herakliusz
> ; Govindharajan, Hariprasad
> ; Burakov, Anatoly
>
> Cc: dev@dpdk.org; Yigit, Ferruh
> Subject: [PATCH] app/testpmd
The format specifier for the 'size_t' format should be '%z'.
Also this fix enables compiling PMD for 32bit architecture.
Fixes: ba641f207642 ("net/sfc: add init on attach")
Fixes: 82faef507608 ("net/sfc: set RSS key and hash types config")
Fixes: 7803554a9e38 ("net/sfc: process RSS settings on Rx
On 2/19/20 5:04 PM, Ferruh Yigit wrote:
> The format specifier for the 'size_t' format should be '%z'.
>
> Also this fix enables compiling PMD for 32bit architecture.
>
> Fixes: ba641f207642 ("net/sfc: add init on attach")
> Fixes: 82faef507608 ("net/sfc: set RSS key and hash types config")
> Fix
> -Original Message-
> From: dev On Behalf Of Konstantin Ananyev
> Sent: Wednesday, February 19, 2020 12:23 AM
> To: dev@dpdk.org
> Cc: akhil.go...@nxp.com; Ananyev, Konstantin
>
> Subject: [dpdk-dev] [PATCH] ipsec: fix use of uninitialized variable
>
> Defects reported by coverity scan
Currently, the set VLAN ID header modify action has already got the
action pointer with offset from the header modify action array, but
the configuration saves the detail to the memory of action with the
offset again. It causes double offset to set the VLAN ID action to
the wrong place in the heade
'fm10k_rxtx_vec.c' is SSE vector instructions implementation and should
be compiled only for x86.
The vector PMD support controlled by
'CONFIG_RTE_LIBRTE_FM10K_INC_VECTOR' config option, but it is enabled by
default, safer to add x86 checks too.
Fixes: 10eb9ce8dccd ("fm10k: allow to disable vecto
On 19-Feb-20 11:16 AM, Kamaraj P wrote:
Hi Kevin/Anatoly,
Yes we have the patch already included in our code base.
Looks like it get struck in the below piece of the code:
mapped_addr = mmap(requested_addr, (size_t)map_sz, PROT_READ,
mmap_flags, -1, 0);
Could y
For header modify actions after decapsulation action, the header modify
actions should be applied to the inner layers.
Currently, it always treats the outermost layers as the corresponding
layer even though with decapsulation action.
Add the layer validation with decapsulation for the header modi
Currently, the flow validate function only validate the outermost layer
with the header modify actions. If there is decapsulation action before
the header modify action, the validation should choose the inner layer
for validation.
Add decapsulation check when validate with the header modify action
Currently, for header modify actions after tunnel decapsulation, the
header modify actions will still select the old outermost layer type.
It will cause header modify actions get the wrong layer type.
Add the tunnel decap flag to indicate the layer type to get for header
modify actions.
Fixes: 4b
The header modify actions number supported now has some limitation,
and it is decided by both driver and hardware. If the configuration
is different or the table to insert the flow is different, the result
might be different if the flow contains header modify actions.
Currently, the actual action n
Acked-by: Xiao Wang
Thanks,
Xiao
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, February 19, 2020 10:17 PM
> To: Zhang, Qi Z ; Wang, Xiao W
> ; Chen Jing D(Mark)
> Cc: dev@dpdk.org; Yigit, Ferruh ; sta...@dpdk.org
> Subject: [PATCH] net/fm10k: fix non-x86 build
>
> 'fm1
As meter flows are split into three subflows each, the prefix subflow
with meter action color the packet, the meter subflow filters out the
colored packets, the suffix subflow applies all the remaining actions
to the passed packets.
Currently, all the user defined items are matched in the prefix f
Metadata suffix subflow inherits the RSS needed hash_fields from the
prefix subflow as the suffix subflow only has the tag match item unable
to generate the full original hash_fields for RSS action.
Unfortunately, hash_fields will only be generated if flow has RSS action.
So it means the prefix fl
For flow split to several subflows, the match items maybe in the prefix
flow, and the actions are split to the suffix flow.
In this case, for the actions need the user defined match item will not
create correctly.
Copy the item layers flags to the suffix flow from prefix flow to fix
the issue.
T
On 2/19/2020 2:02 PM, Iremonger, Bernard wrote:
>
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Wednesday, February 19, 2020 12:40 PM
>> To: Lu, Wenzhuo ; Wu, Jingjing
>> ; Iremonger, Bernard
>> ; Lipiec, Herakliusz
>> ; Govindharajan, Hariprasad
>> ; Burakov, Anatoly
>>
>> Cc
Hi,
> -Original Message-
> From: Bing Zhao
> Sent: Wednesday, February 19, 2020 10:29 AM
> To: Ori Kam ; Slava Ovsiienko
>
> Cc: Raslan Darawsheh ; Matan Azrad
> ; dev@dpdk.org; sta...@dpdk.org
> Subject: [PATCH v2] net/mlx5: fix the hairpin queue capacity
>
> The hairpin TX/RX queue dep
Thanks for the suggestions. We didnt have --mlockall parameter option in
the rte_eal_init().
we have just tried the option and our application says an *unrecognized
option*.
Lets us check further on this and let you know.
Thanks,
Kamaraj
On Wed, Feb 19, 2020 at 7:53 PM Burakov, Anatoly
wrote:
>
Hi,
> -Original Message-
> From: dev On Behalf Of Matan Azrad
> Sent: Wednesday, February 19, 2020 10:30 AM
> To: dev@dpdk.org
> Cc: Slava Ovsiienko ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix L3 vxlan RSS expansion
>
> The RSS expansion feature was introduced to split
HW supports match on one Ethertype, the Ethertype following the last
VLAN tag of the packet (see PRM).
Previous patch added specific handling for packets with VLAN tag,
after setting match on Ethertype.
This patch moves the handling of packets with VLAN tag, to be done
before and instead of settin
> -Original Message-
> From: dev On Behalf Of Iremonger, Bernard
> Sent: Wednesday, February 19, 2020 2:15 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: akhil.go...@nxp.com; Ananyev, Konstantin
>
> Subject: Re: [dpdk-dev] [PATCH] ipsec: fix use of uninitialized variable
>
> > -Or
On 19-Feb-20 3:02 PM, Kamaraj P wrote:
Thanks for the suggestions. We didnt have --mlockall parameter option in
the rte_eal_init().
we have just tried the option and our application says an *unrecognized
option*.
Lets us check further on this and let you know.
Thanks,
Kamaraj
No, that's not
On 02/19, Konieczny, TomaszX wrote:
>>-Original Message-
>>From: dev On Behalf Of Ye Xiaolong
>>Sent: 19 February 2020 08:06
>>To: Sun, GuinanX
>>Cc: dev@dpdk.org; Lu, Wenzhuo ; Yang, Qiming
>>; Zhang, Qi Z ; sta...@dpdk.org
>>Subject: Re: [dpdk-dev] [PATCH v4] net/ixgbe: fix flow ctrl mo
Hi Mou
One comment inline.
From: Suanming Mou
> As meter flows are split into three subflows each, the prefix subflow with
> meter action color the packet, the meter subflow filters out the colored
> packets, the suffix subflow applies all the remaining actions to the passed
> packets.
>
> Cur
From: Suanming Mou
> Currently, the flow validate function only validate the outermost layer with
> the header modify actions. If there is decapsulation action before the header
> modify action, the validation should choose the inner layer for validation.
>
> Add decapsulation check when valida
From: Suanming Mou
> Currently, for header modify actions after tunnel decapsulation, the header
> modify actions will still select the old outermost layer type.
> It will cause header modify actions get the wrong layer type.
>
> Add the tunnel decap flag to indicate the layer type to get for h
Hi Anatoly,
Thanks for the suggestions. Yeah we have just changed in our application to
invoke mlockall() before rte_eal_init(). Looks like it does not help
either.
if (mlockall(MCL_CURRENT | MCL_FUTURE)) {
printf("Failed mlockall !! **\n");
}
ret = rte_eal_init(argc, ar
On 19-Feb-20 3:42 PM, Kamaraj P wrote:
Hi Anatoly,
Thanks for the suggestions. Yeah we have just changed in our application
to invoke mlockall() before rte_eal_init(). Looks like it does not help
either.
if (mlockall(MCL_CURRENT | MCL_FUTURE)) {
printf("Failed mlockall !!
Hi Anatoly,
Yes we are facing an issue with our custom applications.
Earlier we have tried with l2fwd DPDK application and does not see any
issue with memory initialization.
Not sure whether we missed any other options.
BTW when we tried with l2fwd application, the application does not seem to
ha
Identifier for new port may contain white list options,
and white list options will not fit into 128 from STR_TOKEN_SIZE,
instead having 4096 char from STR_MULTI_TOKEN_SIZE will provide
better and more options
Fixes: edab33b1c01d ("app/testpmd: support port hotplug")
Cc: muk...@igel.co.jp
Cc: sta.
Minutes of Technical Board meeting, 2020-02-12
Members Attending: 9/11
- Bruce Richardson
- Ferruh Yigit
- Hemant Agrawal
- Jerin Jacob
- Kevin Traynor
- Konstantin Ananyev
- Maxime Coquelin
- Olivier Matz
- Thomas Monjalon (Chair)
NOTE: Technical Board meetings happen every sec
On 2/19/2020 2:11 PM, Andrew Rybchenko wrote:
> On 2/19/20 5:04 PM, Ferruh Yigit wrote:
>> The format specifier for the 'size_t' format should be '%z'.
>>
>> Also this fix enables compiling PMD for 32bit architecture.
>>
>> Fixes: ba641f207642 ("net/sfc: add init on attach")
>> Fixes: 82faef507608
On 2/19/2020 2:31 PM, Wang, Xiao W wrote:
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Wednesday, February 19, 2020 10:17 PM
>> To: Zhang, Qi Z ; Wang, Xiao W
>> ; Chen Jing D(Mark)
>> Cc: dev@dpdk.org; Yigit, Ferruh ; sta...@dpdk.org
>> Subject: [PATCH] net/fm10k: fix non-x8
This currently serves the same purpose as the default implementation
but will be expanded for additional functionality.
Reviewed-by: Cornelia Huck
Signed-off-by: Alex Williamson
---
drivers/vfio/pci/vfio_pci.c |8
1 file changed, 8 insertions(+)
diff --git a/drivers/vfio/pci/vfio_
The VFIO_DEVICE_FEATURE ioctl is meant to be a general purpose, device
agnostic ioctl for setting, retrieving, and probing device features.
This implementation provides a 16-bit field for specifying a feature
index, where the data porition of the ioctl is determined by the
semantics for the given f
If we enable SR-IOV on a vfio-pci owned PF, the resulting VFs are not
fully isolated from the PF. The PF can always cause a denial of service
to the VF, even if by simply resetting itself. The degree to which a PF
can access the data passed through a VF or interfere with its operation
is dependen
Allow bus drivers to provide their own callback to match a device to
the user provided string.
Reviewed-by: Cornelia Huck
Signed-off-by: Alex Williamson
---
drivers/vfio/vfio.c | 20
include/linux/vfio.h |4
2 files changed, 20 insertions(+), 4 deletions(-)
dif
Changes since v1 are primarily to patch 3/7 where the commit log is
rewritten, along with option parsing and failure logging based on
upstream discussions. The primary user visible difference is that
option parsing is now much more strict. If a vf_token option is
provided that cannot be used, we
It currently results in messages like:
"vfio-pci :03:00.0: vfio_pci: ..."
Which is quite a bit redundant.
Reviewed-by: Cornelia Huck
Signed-off-by: Alex Williamson
---
drivers/vfio/pci/vfio_pci.c |1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/
The cleanup is getting a tad long.
Signed-off-by: Alex Williamson
---
drivers/vfio/pci/vfio_pci.c | 54 ---
1 file changed, 25 insertions(+), 29 deletions(-)
diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c
index 497ecadef2ba..7d41
With the VF Token interface we can now expect that a vfio userspace
driver must be in collaboration with the PF driver, an unwitting
userspace driver will not be able to get past the GET_DEVICE_FD step
in accessing the device. We can now move on to actually allowing
SR-IOV to be enabled by vfio-pc
Add combined libdpdk.a and libdpdk.so libs for Meson similar to how
it's done for Make builds
Signed-off-by: Ruslan Babayev
---
buildtools/group-libs.sh | 2 ++
buildtools/meson.build | 1 +
meson.build | 17 +
3 files changed, 20 insertions(+)
create mode 10075
This series tries to make the travis configuration a bit easier to read,
then drops jobs that are not essential or functional duplicates in the
hope we can give some oxygen to the robot who seems to lag quite a bit
and can't catch up.
--
David Marchand
David Marchand (4):
ci: remove unnecessar
Maintaining the .travis.yml requires some knowledge of how Travis
computes the jobs list (combination of os: arch: compiler: etc...).
Let's switch to an explicit list to find all jobs at a glance.
To enhance readability, jobs have been sorted per arch/compiler with
comments to isolate blocks.
Set
Following removal of kmod compilation, we don't need to install
linux-headers anymore.
Fixes: ea860973592b ("ci: remove redundant configs disabling kmods")
Signed-off-by: David Marchand
---
.travis.yml | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/.travis.yml b/.travis.yml
Let's prune the jobs list to limit the amount of time spent by the robot
in Travis.
Since meson enables automatically the relevant components, there is not
much gain in testing with extra_packages vs required_packages only.
For a given arch/compiler/env combination, compilation is first tested
in
Reading https://config.travis-ci.com/ and using
https://config.travis-ci.com/explore to check changes, we can cleanup
some warnings reported by the config validation options in Travis.
Signed-off-by: David Marchand
---
.travis.yml | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-
On 2020-02-19 05:46, David Marchand wrote:
On Wed, Feb 19, 2020 at 2:10 AM David Wilder
wrote:
This change follows the example of aarch64 Travis support
by adding support for ppc64le gcc builds. Limitations for
ppc64le are the same as aarch64 as described in commit
31bb45bcfd.
checkpatch di
The quiescent state description refers to an incorrect
thread.
Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
Cc: sta...@dpdk.org
Signed-off-by: Prateek Agarwal
Signed-off-by: Honnappa Nagarahalli
---
V4
1) Seperated the commits for ring and RCU changes
2) Changed the co
1 - 100 of 126 matches
Mail list logo
