> On Mar 6, 2019, at 11:41 PM, Yongseok Koh wrote:
>
> RFC:
> https://mails.dpdk.org/archives/dev/2019-March/125517.html
>
> Yongseok Koh (6):
> net/mlx: remove debug messages on datapath
> net/mlx5: fix external memory registration
> net/mlx5: add control of excessive memory pinning by ker
The Memory Region (MR) for DMA memory can't be created from secondary
process due to lib/driver limitation. Whenever it is needed, secondary
process can make a request to primary process through the EAL IPC channel
(rte_mp_msg) which is established on initialization. Once a MR is created
by primary
The Memory Region (MR) for DMA memory can't be created from secondary
process due to lib/driver limitation. Whenever it is needed, secondary
process can make a request to primary process through the EAL IPC channel
(rte_mp_msg) which is established on initialization. Once a MR is created
by primary
RFC:
https://mails.dpdk.org/archives/dev/2019-March/125517.html
Yongseok Koh (6):
net/mlx: remove debug messages on datapath
net/mlx5: fix external memory registration
net/mlx5: add control of excessive memory pinning by kernel
net/mlx5: enable secondary process to register DMA memory
ne
Secondary process is not allowed to register MR due to a restriction of
library and kernel driver.
Fixes: 7e43a32ee060 ("net/mlx5: support externally allocated static memory")
Cc: sta...@dpdk.org
Signed-off-by: Yongseok Koh
---
doc/guides/nics/mlx5.rst | 5 +
drivers/net/mlx5/mlx5_mr.c |
Cc: sta...@dpdk.org
Signed-off-by: Yongseok Koh
---
drivers/net/mlx4/mlx4_mr.c | 4
drivers/net/mlx5/mlx5_mr.c | 6 --
2 files changed, 10 deletions(-)
diff --git a/drivers/net/mlx4/mlx4_mr.c b/drivers/net/mlx4/mlx4_mr.c
index 01894faecf..0ba55fda04 100644
--- a/drivers/net/mlx4/mlx4_m
A new PMD parameter (mr_ext_memseg_en) is added to control extension of
memseg when creating a MR. It is enabled by default.
If enabled, mlx4_mr_create() tries to maximize the range of MR
registration so that the LKey lookup tables on datapath become smaller and
get the best performance. However,
A new PMD parameter (mr_ext_memseg_en) is added to control extension of
memseg when creating a MR. It is enabled by default.
If enabled, mlx5_mr_create() tries to maximize the range of MR
registration so that the LKey lookup tables on datapath become smaller and
get the best performance. However,
RFC:
https://mails.dpdk.org/archives/dev/2019-March/125516.html
Yongseok Koh (3):
net/mlx4: change device reference for secondary process
net/mlx4: add external allocator for Verbs object
net/mlx4: add secondary process support
doc/guides/nics/features/mlx4.ini | 1 +
doc/guides/nics/mlx
To support secondary process, the memory allocated by library such as
completion rings (CQ) and buffer rings (WQ) must be manageable by EAL, in
order to share it with secondary processes. With new changes in rdma-core
and kernel driver, it is possible to provide an external allocator to the
library
rte_eth_devices[] is not shared between primary and secondary process, but
a static array to each process. The reverse pointer of device (priv->dev)
becomes invalid if mlx4 supports secondary process. Instead, priv has the
pointer to shared data of the device,
struct rte_eth_dev_data *dev_data;
In order to support secondary process, a few features are required.
a) rdma-core library should allocate device resources using DPDK's memory
allocator.
b) UAR should be remapped for secondary processes. Currently, in order not
to use different data structure for secondary processes, PMD tr
Rx/Tx burst function pointers are stored in the rte_eth_dev structure,
which is local to a process. Even though primary process replaces the
function pointers, secondary will not run the new ones. With rte_mp APIs,
primary can easily broadcast a request to stop/start the datapath of
secondary proce
There's more need to have PMD global data structure. It should be
initialized once per a process regardless of how many PMD instances are
probed. mlx5_init_once() is called during probing and make sure all the
init functions are called once per a process. The existing shared memory
gets more extens
Socket API is used for IPC in order for secondary process to acquire Verb
command file descriptor. The FD is used to remap UAR address. The new
multi-process APIs (rte_mp) in EAL are newly introduced. mlx5_socket.c is
replaced with mlx5_mp.c, which uses the new APIs.
As it is PMD global infrastruc
The existing socket-based IPC channel is replaced with the new rte_mp APIs of
EAL and extended to request stop/start of dataplane to secondary processes.
Also, initialization of PMD global data including the new IPC channel is
reworked to provide more generic framework for future use.
Yongseok Koh
As the memory event is propagated to secondary processes, the event is
processed redundantly. This should be processed once because the data
structure used for MR and the event is global across the processes.
Fixes: 974f1e7ef146 ("net/mlx5: add new memory region support")
Cc: sta...@dpdk.org
Sign
Hi,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, March 06, 2019 20:31
> To: Xu, Rosen ; dev@dpdk.org
> Cc: Zhang, Tianfei ; Wei, Dan
> ; Pei, Andy ; Yang, Qiming
> ; Wang, Haiyue ; Chen,
> Santos ; Zhang, Zhang ;
> Hemant Agrawal ; Shreyansh Jain
>
> Subject: Re: [PATCH v
Hi ilya,
Thanks for your comments, inline comments and new commit message in V2.
/Gavin
> -Original Message-
> From: Ilya Maximets
> Sent: Wednesday, March 6, 2019 7:49 PM
> To: Gavin Hu (Arm Technology China) ;
> dev@dpdk.org
> Cc: nd ; tho...@monjalon.net; jer...@marvell.com;
> hemant
In weak memory models, like arm64, reading the {prod,cons}.tail may get
reordered after reading or writing the ring slots, which corrupts the ring
and stale data is observed.
This issue was reported by NXP on 8-A72 DPAA2 board. The problem is most
likely caused by missing the acquire semantics whe
Wednesday, March 6, 2019 11:02 PM, Mit Matelske:
> Subject: Re: [dpdk-dev] mlx5 under FreeBSD
>
> > >
> > > dev.mlx5_core.1.hw.board_id: LNR3270110033
> > > dev.mlx5_core.1.hw.fw_version: 12.17.2020
> >
> > Getting back to you on this one.
> >
> > From the FW it looks like you are trying to create
Hi Akhil, Fiona,
Would the usage of strcmp be alright? Please check my comment inline.
Thanks,
Anoob
> -Original Message-
> From: dev On Behalf Of Anoob Joseph
> Sent: Friday, March 1, 2019 11:55 AM
> To: Trahe, Fiona ; Akhil Goyal
> ; Doherty, Declan ; De
> Lara Guarch, Pablo ; Yigit,
On 3/6/2019 3:31 AM, Thomas Monjalon wrote:
06/03/2019 05:16, Anand Rawat:
Added header files to support windows on x86 platforms.
Updated rte_config to include rte_windows.h for windows
build.
[...]
--- a/config/rte_config.h
+++ b/config/rte_config.h
+/* windows specific*/
+#ifdef RTE_EXEC_EN
Hi, David
Thanks for your review.
On 03/05, David Marchand wrote:
>On Fri, Mar 1, 2019 at 9:13 AM Xiaolong Ye wrote:
>
>> diff --git a/doc/guides/rel_notes/release_18_11.rst
>> b/doc/guides/rel_notes/release_18_11.rst
>> index 65bab557d..e0918441a 100644
>> --- a/doc/guides/rel_notes/release_18_
Hi David,
Thanks for you comments.
On 03/05, David Marchand wrote:
>On Fri, Mar 1, 2019 at 9:13 AM Xiaolong Ye wrote:
>
>> This give the option that applicaiton can configure each
>> memory chunk's size precisely. (by MEMPOOL_F_NO_SPREAD).
>>
>> Signed-off-by: Qi Zhang
>> Signed-off-by: Xiaolon
Hi Gabriel,
Thanks for the clarification.
> -Original Message-
> From: Carrillo, Erik G
> Sent: Wednesday, March 6, 2019 8:46 PM
> To: Varghese, Vipin ; rsanf...@akamai.com
> Cc: dev@dpdk.org; techbo...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v3 0/2] Timer library changes
>
> > -Or
On 3/6/2019 3:20 AM, Richardson, Bruce wrote:
On Wed, Mar 06, 2019 at 11:03:24AM +0100, Thomas Monjalon wrote:
# get binutils version for the workaround of Bug 97
-ldver = run_command('ld', '-v').stdout().strip()
-if ldver.contains('2.30')
- if cc.has_argument('-mno-avx512f')
-
On 3/6/2019 3:52 AM, Richardson, Bruce wrote:
-Original Message-
From: Thomas Monjalon [mailto:tho...@monjalon.net]
Sent: Wednesday, March 6, 2019 11:36 AM
To: Richardson, Bruce
Cc: dev@dpdk.org; Rawat, Anand ; Kadam, Pallavi
; Menon, Ranjit ; Shaw,
Jeffrey B
Subject: Re: [dpdk-dev]
> -Original Message-
> From: Yigit, Ferruh
> Sent: Thursday, March 7, 2019 1:55 AM
> To: Zhang, Tianfei ; Xu, Rosen
> ; dev@dpdk.org
> Cc: Wei, Dan ; Pei, Andy ; Yang,
> Qiming ; Wang, Haiyue ;
> Chen, Santos ; Zhang, Zhang
>
> Subject: Re: [PATCH v1 03/11] drivers/raw/ifpga_rawdev: add
06/03/2019 23:45, Anand Rawat:
> > - Please use --in-reply-to for new revisions in order to have
> > all revisions in the same mail thread.
> > The best is to make all revisions as replies of the first cover letter.
> >
> >
>
> I will make sure to follow this from v3 onward.
> Since now that the
On 3/6/2019 12:29 AM, Thomas Monjalon wrote:
Hi,
06/03/2019 05:16, Anand Rawat:
v2 Changes:
Incorporated reviews from v1.
Fixed license period for updated files.
Renamed 'winapp' folder to 'windows'.
Cleaned unneeded meson changes.
Fixed code style warnings.
Added maintainers information for wi
> >
> > dev.mlx5_core.1.hw.board_id: LNR3270110033
> > dev.mlx5_core.1.hw.fw_version: 12.17.2020
>
> Getting back to you on this one.
>
> From the FW it looks like you are trying to create a TIR (QP) which doesn't
> use Toeplitz hash function (IBV_RX_HASH_FUNC_TOEPLITZ) but the rss key is not
>
06/03/2019 19:54, Carrillo, Erik G:
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > 05/03/2019 23:41, Carrillo, Erik G:
> > > Hi all,
> > >
> > > I'd like to bring this patch proposal up again and see if I can get any
> > > more
> > feedback from the maintainer or others.
> > >
> > > I ne
Hi Bernard,
>
> Call create_inline_session() at initialisition in sa.c
> Call rte_ipsec_session_prepare() in fill_ipsec_session() for inline.
Here and in other places - it probably worth to explain what is the purpose
for these changes.
As a side notice, as these series fixes that problem, i
On 3/6/19 4:42 AM, Zhaohui (zhaohui, Polestar) wrote:
Hi Williams：
We have some questions that need your help. About “Segfault when eal
thread executing mlx5 nic‘s lsc event”.
In order to solve this problem(core dump), What should we do?( Looking forward
to your reply）
The link sta
> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Tuesday, March 5, 2019 4:59 PM
> To: Carrillo, Erik G
> Cc: techbo...@dpdk.org; rsanf...@akamai.com; dev@dpdk.org
> Subject: Re: [dpdk-techboard] [dpdk-dev] [PATCH v3 0/2] Timer library
> changes
>
> 05/03/2
<...>
> @@ -11734,13 +11735,14 @@ static const char *get_bnx2x_flags(uint32_t flags)
>
> for (i = 0; i < 5; i++)
> if (flags & (1 << i)) {
> - strcat(flag_str, flag[i]);
> + strlcat(flag_str, flag[i], sizeof(flag_str));
>
06/03/2019 11:46, Gaëtan Rivet:
> On Tue, Mar 05, 2019 at 06:58:04PM +0100, Thomas Monjalon wrote:
> > 05/03/2019 18:38, Gaëtan Rivet:
> > > What happens when a primary process closes a device before a secondary?
> > > Is the secondary unable to stop / close its own then? Isn't there some
> > > mis
On 3/6/2019 1:59 PM, Zhang, Tianfei wrote:
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Wednesday, March 6, 2019 8:28 PM
>> To: Xu, Rosen ; dev@dpdk.org
>> Cc: Zhang, Tianfei ; Wei, Dan
>> ; Pei, Andy ; Yang, Qiming
>> ; Wang, Haiyue ; Chen,
>> Santos ; Zhang, Zhang
>> Subject:
Do not allow creating a ethernet device with a name over the
allowed maximum (or 0 length).
Signed-off-by: Stephen Hemminger
---
lib/librte_ethdev/rte_ethdev.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index 85
On 06/03/2019 16:45, Walker, Benjamin wrote:
> On Wed, 2019-01-30 at 16:27 +, Kevin Traynor wrote:
>> Hi all,
>>
>> Here is a list of patches targeted for stable release 18.08.1. Please
>> help review and test. The tentative date for the final release is 28,
>> February. Before that, please sho
On 2/28/2019 7:30 AM, Yangchao Zhou wrote:
> va2pa depends on the physical address and virtual address offset of
> current mbuf. It may get the wrong physical address of next mbuf which
> allocated in another hugepage segment.
Hi Yangchao,
The problem you described seems valid, when current mbuf
Currently, the timer library uses a per-process table of structures to
manage skiplists of timers presumably because timers contain arbitrary
function pointers whose value may not resolve properly in other
processes.
However, if the same callback is used handle all timers, and that
callback is onl
Add a function to the timer API that allows a caller to traverse a
specified set of timer lists, stopping each timer in each list,
and invoking a callback function.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_timer/rte_timer.c | 39 ++
lib/librte
This patch series modifies the timer library in such a way that
structures that used to be statically allocated in a process's data
segment are now allocated in shared memory. As these structures contain
lists of timers, new APIs are introduced that allow a caller to specify
the particular structu
test-meson-build.sh generates multiple build directories for various
targets. As these follow a few known patterns of reasonable lengths,
and since they don't need to be tracked in git, we can add them to
the gitignore file.
Signed-off-by: Bruce Richardson
---
.gitignore | 6 ++
1 file chang
Generally hidden files are hidden for good reason and we don't want to
track them in git. They can always be manually added to git tracking
individually if needed.
Signed-off-by: Bruce Richardson
---
.gitignore | 3 +++
1 file changed, 3 insertions(+)
diff --git a/.gitignore b/.gitignore
index
Split the ignored file list into section based on logical groups of files,
putting a comment at the top of each section clarifying what it is.
Signed-off-by: Bruce Richardson
---
.gitignore | 3 +++
1 file changed, 3 insertions(+)
diff --git a/.gitignore b/.gitignore
index 9105e26c5..b23e1dbb7
After you run a python script, a .pyc file is often left behind, which we
don't want to track in git.
Signed-off-by: Bruce Richardson
---
.gitignore | 3 +++
1 file changed, 3 insertions(+)
diff --git a/.gitignore b/.gitignore
index b23e1dbb7..eb3c23705 100644
--- a/.gitignore
+++ b/.gitignore
On 2/22/2019 3:36 PM, Hemant Agrawal wrote:
> These duplicate device id is causing incorrect mapping
> for DPAA_SEC for test case execultion on the basis of
> capabilities.
>
> Fixes: e155ca055e84 ("test/crypto: add tests for AMD CCP")
> Cc: sta...@dpdk.org
> Cc: ravi1.ku...@amd.com
>
> Reported-
On Wed, 2019-01-30 at 16:27 +, Kevin Traynor wrote:
> Hi all,
>
> Here is a list of patches targeted for stable release 18.08.1. Please
> help review and test. The tentative date for the final release is 28,
> February. Before that, please shout if anyone has objections with these
> patches be
On 2/12/2019 4:31 PM, Shally Verma wrote:
>
>> -Original Message-
>> From: dev On Behalf Of Arek Kusztal
>> Sent: 07 February 2019 16:25
>> To: dev@dpdk.org
>> Cc: akhil.go...@nxp.com; fiona.tr...@intel.com;
>> shally.ve...@caviumnetworks.com; sunila.s...@caviumnetworks.com;
>> ashish.g
On 2/13/2019 2:52 PM, Anoob Joseph wrote:
> Fixes: 351fbee21986 ("common/cpt: support hash")
>
> Signed-off-by: Anoob Joseph
> Signed-off-by: Tejasree Kondoj
> ---
> drivers/common/cpt/cpt_ucode.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/common/cpt/cpt
On 2/11/2019 5:38 PM, Jozwiak, TomaszX wrote:
>
>> -Original Message-
>> From: Trahe, Fiona
>> Sent: Thursday, February 7, 2019 7:46 PM
>> To: dev@dpdk.org
>> Cc: Trahe, Fiona ; Jozwiak, TomaszX
>> ; Kusztal, ArkadiuszX
>> ; akhil.go...@nxp.com; sta...@dpdk.org
>> Subject: [PATCH] doc/qat
On 2/5/2019 10:24 PM, Trahe, Fiona wrote:
>
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Arek Kusztal
>> Sent: Tuesday, February 5, 2019 9:13 AM
>> To: dev@dpdk.org
>> Cc: sunila.s...@caviumnetworks.com; akhil.go...@nxp.com;
>> shally.ve...@caviumnetworks
On 2/6/2019 4:04 PM, Arek Kusztal wrote:
> This patch changes modular exponentiation and modular multiplicative
> inverse API comments to make it more precise.
>
> Signed-off-by: Arek Kusztal
> ---
> v2:
> - grammar fixes
>
> v3:
> - remove information about positive integer
>
> lib/librte_cry
Rename the cross files for meson compilation from having linuxapp
in the name to just linux in the name.
Signed-off-by: Bruce Richardson
---
.../arm/{arm64_armv8_linuxapp_gcc => arm64_armv8_linux_gcc} | 0
.../arm/{arm64_dpaa2_linuxapp_gcc => arm64_dpaa2_linux_gcc} | 0
config/arm/{arm64_dpa
Rather than using linuxapp and bsdapp everywhere, we can change things to
use the, more readable, terms "linux" and "freebsd" in our build configs.
Rather than renaming the configs we can just duplicate the existing ones
with the new names using symlinks, and use the new names exclusively
internall
Rename the macro and all instances in DPDK code, but keep a copy of
the old macro defined for legacy code linking against DPDK
Signed-off-by: Bruce Richardson
---
app/test-bbdev/test_bbdev_vector.c| 2 +-
.../cperf_test_vector_parsing.c | 2 +-
app/test-pmd/testpmd.c
Rename the macro to make things shorter and more comprehensible. For
both meson and make builds, keep the old macro around for backward
compatibility.
Signed-off-by: Bruce Richardson
---
app/test/process.h| 2 +-
app/test/test_eal_flags.c | 2 +-
The term "linuxapp" is a legacy one, but just calling the subdirectory
"linux" is just clearer for all concerned.
Signed-off-by: Bruce Richardson
---
MAINTAINERS | 8
devtools/build-tags.sh| 2 +-
doc/guid
The term "bsdapp" is a legacy one, but just calling the subdirectory
"freebsd" is just clearer for all concerned.
Signed-off-by: Bruce Richardson
---
MAINTAINERS | 8
devtools/build-tags.sh| 2 +-
doc/guid
The terms linuxapp and bsdapp are legacy names in DPDK that are equivalent
to the more usual names of "linux" and "freebsd". Therefore, we can
replace the instances of the "app" names with the more usual variants. The
only issue in maintaining backward compatibility which can be done by
preserving
Hi Bernard,
On 3/6/2019 9:30 PM, Bernard Iremonger wrote:
> This patchset fixes the issue of the first inbound packet
> being dropped for inline crypto. It also improves the debug
> output in esp.c, sa.c and ipsec-secgw.c
>
> Bernard Iremonger (6):
>examples/ipsec-secgw: fix 1st pkt dropped fo
Improve debug in ipsec-secgw.c
Fixes: 906257e965b7 ("examples/ipsec-secgw: support IPv6")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
examples/ipsec-secgw/ipsec-secgw.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/examples/ipsec-secgw/ipsec-secgw.c
b/ex
Improve debug code in sa.c
Fixes: 0d547ed03717 ("examples/ipsec-secgw: support configuration file")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
examples/ipsec-secgw/sa.c | 21 ++---
1 file changed, 18 insertions(+), 3 deletions(-)
diff --git a/examples/ipsec-secgw/s
Improve debug code in esp.c
Fixes: f159e70b0922 ("examples/ipsec-secgw: support transport mode")
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
examples/ipsec-secgw/esp.c | 5 +++--
1 file changed, 3 insertions(+),
Call create_inline_session() at initialisition in sa.c
Call rte_ipsec_session_prepare() in fill_ipsec_session() for inline.
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
examples/ipsec-secgw/sa.c | 46 +
Refactor cryprodev_init() and main() in ipsec-secgw.c
Add max_session_size().
Start ports before adding flows in main().
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
examples/ipsec-secgw/ipsec-secgw.c | 246 ++
Refactor create_session() into create_inline_session() and
create_lookaside_session() in ipsec.c
Use socket_ctx in create_inline_session()
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Cc: sta...@dpdk.org
Signed-off-by: Bernard Iremonger
---
examples/ipsec-secgw/ipsec.c
This patchset fixes the issue of the first inbound packet
being dropped for inline crypto. It also improves the debug
output in esp.c, sa.c and ipsec-secgw.c
Bernard Iremonger (6):
examples/ipsec-secgw: fix 1st pkt dropped for inline crypto
examples/ipsec-secgw: fix 1st packet dropped patch
Hi Williams：
We have some questions that need your help. About “Segfault when eal
thread executing mlx5 nic‘s lsc event”.
In order to solve this problem(core dump), What should we do?( Looking forward
to your reply）
When the number of slave slave devices exceeds 8, it will cause the array
subscript to cross the boundary.
---
drivers/net/bonding/rte_eth_bond_8023ad.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c
b/drivers/net/bonding/rte_et
Hi:
I think the flow list may be accessed in the same time by two different
threads and may cause some errors. Do it need a lock to protect the flow list?
Thanks
On 3/4/2019 7:12 PM, Stephen Hemminger wrote:
> On Mon, 4 Mar 2019 12:11:20 +0300
> Andrew Rybchenko wrote:
>
>> On 3/1/19 9:42 PM, Stephen Hemminger wrote:
>>> On Fri, 1 Mar 2019 10:48:58 +0300
>>> Andrew Rybchenko wrote:
>>>
On 3/1/19 1:47 AM, Stephen Hemminger wrote:
> Don't need
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Luca Boccassi
> Sent: Wednesday, March 6, 2019 3:28 PM
> To: dpdk-dev
> Cc: Mcnamara, John ; Thomas Monjalon
> ; Yigit, Ferruh
> Subject: Re: [dpdk-dev] DPDK Release Status Meeting 21/2/2019
>
> On Thu, 2019-02
On Thu, 2019-02-21 at 17:43 +, Luca Boccassi wrote:
> On Thu, 2019-02-21 at 16:09 +, Ferruh Yigit wrote:
> > Minutes 21 February 2019
> >
>
> cut
>
> > * Coverity is partially back, but new analyses aren't running
> > * Looking for suggestions from the community
ixgevf vlan strip and extend capabilities were removed when
migrating to the bit flags implementation.
Restoring the capbility to enable these offloads at
configuration time.
Fixes: ec3b1124d14d (\"net/ixgbe: convert to new Rx offloads API\")
Signed-off-by: David Harton
---
drivers/net/ixgbe/ix
> -Original Message-
> From: Varghese, Vipin
> Sent: Tuesday, March 5, 2019 8:39 PM
> To: Carrillo, Erik G ; rsanf...@akamai.com
> Cc: dev@dpdk.org; techbo...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v3 0/2] Timer library changes
>
> Hi Erik,
>
> Apologies if I am reaching out a bit late
These handlers allow an application to create a mempool based on the
lock-free ring, with any combination of single/multi producer/consumer.
Also, add a note to the programmer's guide's "known issues" section.
Signed-off-by: Gage Eads
Acked-by: Andrew Rybchenko
---
doc/guides/prog_guide/env_ab
nb_ring_perf_autotest re-uses the ring_perf_autotest code by wrapping its
top-level function with one that takes a 'flags' argument.
Signed-off-by: Gage Eads
---
app/test/test_ring_perf.c | 19 ---
1 file changed, 16 insertions(+), 3 deletions(-)
diff --git a/app/test/test_ring_
ring_nb_autotest re-uses the ring_autotest code by wrapping its top-level
function with one that takes a 'flags' argument.
Signed-off-by: Gage Eads
---
app/test/test_ring.c | 61
1 file changed, 38 insertions(+), 23 deletions(-)
diff --git a/
This commit adds support for lock-free circular ring enqueue and dequeue
functions. The ring is supported on 32- and 64-bit architectures, however
it uses a 128-bit compare-and-swap instruction when run on a 64-bit
architecture, and thus is currently limited to x86_64.
The algorithm is based on Ol
This marker allows us to replace "&r[1]" with "&r->ring" to locate the
start of the ring.
Signed-off-by: Gage Eads
---
lib/librte_ring/rte_ring.h | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index c78db6916..f16d
For 64-bit systems, at current CPU speeds, 64-bit head and tail indexes
will not wrap-around within the author's lifetime. This is important to
avoiding the ABA problem -- in which a thread mistakes reading the same
tail index in two accesses to mean that the ring was not modified in the
intervenin
For some users, the rte ring's "non-preemptive" constraint is not acceptable;
for example, if the application uses a mixture of pinned high-priority threads
and multiplexed low-priority threads that share a mempool.
This patchset introduces a lock-free ring and a mempool based on it. The
lock-free
Hi Mit,
Monday, March 4, 2019 6:45 PM, Mit Matelske:
> Subject: Re: [dpdk-dev] mlx5 under FreeBSD
>
> dev.mlx5_core.1.hw.board_id: LNR3270110033
> dev.mlx5_core.1.hw.fw_version: 12.17.2020
Getting back to you on this one.
>From the FW it looks like you are trying to create a TIR (QP) which does
This commit adds support for lock-free (linked list based) stack mempool
handler.
In mempool_perf_autotest the lock-based stack outperforms the
lock-free handler for certain lcore/alloc count/free count
combinations*, however:
- For applications with preemptible pthreads, a standard (lock-based)
This commit adds lock-free stack variants of stack_autotest
(stack_lf_autotest) and stack_perf_autotest (stack_lf_perf_autotest), which
differ only in that the lock-free versions pass the RTE_STACK_F_LF flag to
all rte_stack_create() calls.
Signed-off-by: Gage Eads
Reviewed-by: Olivier Matz
---
This commit adds an implementation of the lock-free stack push, pop, and
length functions that use __atomic builtins, for systems that benefit from
the finer-grained memory ordering control.
Signed-off-by: Gage Eads
---
lib/librte_stack/Makefile| 3 +-
lib/librte_stack/meson.build
This commit adds support for a lock-free (linked list based) stack to the
stack API. This behavior is selected through a new rte_stack_create() flag,
RTE_STACK_F_LF.
The stack consists of a linked list of elements, each containing a data
pointer and a next pointer, and an atomic stack depth counte
stack_perf_autotest tests the following with one lcore:
- Cycles to attempt to pop an empty stack
- Cycles to push then pop a single object
- Cycles to push then pop a burst of 32 objects
It also tests the cycles to push then pop a burst of 8 and 32 objects with
the following lcore combinations (i
stack_autotest performs positive and negative testing of the stack API, and
exercises the push and pop datapath functions with all available lcores.
Signed-off-by: Gage Eads
---
MAINTAINERS | 1 +
app/test/Makefile | 2 +
app/test/meson.build | 3 +
app/test/test_stack.c | 4
This patchset introduces a stack library, supporting both lock-based and
lock-free stacks, and a lock-free stack mempool handler.
The lock-based stack code is derived from the existing stack mempool handler,
and that handler is refactored to use the stack library.
The lock-free stack mempool hand
The new rte_stack library is derived from the mempool handler, so this
commit removes duplicated code and simplifies the handler by migrating it
to this new API.
Signed-off-by: Gage Eads
Reviewed-by: Olivier Matz
---
MAINTAINERS | 2 +-
drivers/mempool/stack/Makef
The rte_stack library provides an API for configuration and use of a
bounded stack of pointers. Push and pop operations are MT-safe, allowing
concurrent access, and the interface supports pushing and popping multiple
pointers at a time.
The library's interface is modeled after another DPDK data st
> -Original Message-
> From: Yigit, Ferruh
> Sent: Wednesday, March 6, 2019 8:28 PM
> To: Xu, Rosen ; dev@dpdk.org
> Cc: Zhang, Tianfei ; Wei, Dan
> ; Pei, Andy ; Yang, Qiming
> ; Wang, Haiyue ; Chen,
> Santos ; Zhang, Zhang
> Subject: Re: [PATCH v1 03/11] drivers/raw/ifpga_rawdev: add OP
On Wed, 2019-03-06 at 15:51 +0100, Rastislav Černay wrote:
> >
> > --Why is -lfdt added? I don't see any symbols from libfdt used in
> > the
> > --rest of the patch?
> >
>
> Hi,
> libnfb is dependent on fdt and there is a special case, where some
> functions are defined in header (netcope/rxmac.
>
>
> --Why is -lfdt added? I don't see any symbols from libfdt used in the
> --rest of the patch?
>
Hi,
libnfb is dependent on fdt and there is a special case, where some
functions are defined in header (netcope/rxmac.h, netcope/txmac.h).
So if one wants to use functions, from these headers fdt n
On 2/28/2019 7:13 AM, Rosen Xu wrote:
> Add MAINTAINERS for Intel FPGA Acceleration NIC IPN3KE.
I think this patch shouldn't be separate, can you please merge with the main
patch that adds the pmd.
>
> Signed-off-by: Rosen Xu
> ---
> MAINTAINERS | 7 +++
> 1 file changed, 7 insertions(+)
>
1 - 100 of 120 matches
Mail list logo
