On 10/26/21 04:07, Ding, Xuan wrote:
Hi Maxime,
-Original Message-
From: Maxime Coquelin
Sent: Tuesday, October 26, 2021 4:47 AM
To: dev@dpdk.org; david.march...@redhat.com; Xia, Chenbo
; Ding, Xuan
Subject: Re: [PATCH] vhost: fix async DMA map
Hi Xuan,
On 10/25/21 22:33, Maxime
https://bugs.dpdk.org/show_bug.cgi?id=738
Peng,Zhihong (zhihongx.p...@intel.com) changed:
What|Removed |Added
Resolution|--- |FIXED
Status|
On 2021/10/26 14:04, Radha Mohan wrote:
> On Thu, Oct 21, 2021 at 7:04 AM Kevin Laatz wrote:
>>
>> On 21/10/2021 13:59, Chengwen Feng wrote:
>>> From: David Marchand
>>>
>>> No need to expose rte_dma_devices out of the dmadev library.
>>> Existing helpers should be enough, and inlines make use of
A meter policy with RSS/Queue action is not supported
when dv_xmeta_en != 0.
Fixes: 51ec04d ("net/mlx5: connect meter policy to created flows")
Cc: sta...@dpdk.org
Signed-off-by: Li Zhang
Acked-by: Matan Azrad
---
doc/guides/nics/mlx5.rst | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc
On Thu, Oct 21, 2021 at 7:04 AM Kevin Laatz wrote:
>
> On 21/10/2021 13:59, Chengwen Feng wrote:
> > From: David Marchand
> >
> > No need to expose rte_dma_devices out of the dmadev library.
> > Existing helpers should be enough, and inlines make use of
> > rte_dma_fp_objs.
I asked this earlier
The mlx5 compress PMD uses HW acceleration for the compress operations.
The mlx5 HW device has no level style mode, which does a tradeoff between
throughput and compression ratio, unlike SW drivers where the CPU is doing
the compress, and more CPU effort can cause a better compression ratio.
The ml
Currently, the compression block size is 15 by default, which
is the maximum.
Add "log-block-size" devarg to select compression block size manually.
The value provided should be between 4 to 15.
Any out-of-range value will be defaulted to 15.
Signed-off-by: Raja Zidane
Acked-by: Matan Azrad
---
The mlx5 compress PMD uses HW acceleration for the compress operations.
The mlx5 HW device has no level style mode, which does a tradeoff between
throughput and compression ratio, unlike SW drivers where the CPU is doing
the compress, and more CPU effort can cause a better compression ratio.
The ml
From: Kalesh AP
Move the Rx queue state update before bnxt_setup_one_vnic()
is called. For Thor, rxq->rx_started and eth_dev->data->rx_queue_state[]
needs to be set for all queues before bnxt_hwrm_vnic_cfg() or
bnxt_vnic_rss_configure() are called.
Fixes: 0105ea1296c9 ("net/bnxt: support runtime
Add functions for the dmadev vchan setup and DMA operations.
Signed-off-by: Radha Mohan Chintakuntla
---
drivers/dma/cnxk/cnxk_dmadev.c | 322 +
drivers/dma/cnxk/cnxk_dmadev.h | 53 ++
drivers/dma/cnxk/version.map | 3 +
3 files changed, 378 insertions(+)
Add the copy_sg function that will do the multiple DMA transfers of
different sizes and different source/destination as well.
Signed-off-by: Radha Mohan Chintakuntla
---
drivers/dma/cnxk/cnxk_dmadev.c | 80 +-
1 file changed, 79 insertions(+), 1 deletion(-)
diff
Add base support as ROC(Rest of Chip) API which will be used by PMD
dmadev driver.
This patch adds routines to init, fini, configure the DPI DMA device
found in Marvell's CN9k or CN10k SoC familes.
Signed-off-by: Radha Mohan Chintakuntla
---
drivers/common/cnxk/hw/dpi.h | 136
This patch creates and initializes a dmadev device on pci probe.
Signed-off-by: Radha Mohan Chintakuntla
---
MAINTAINERS| 7 +-
doc/guides/dmadevs/cnxk.rst| 53 +++
doc/guides/dmadevs/index.rst | 1 +
drivers/dma/cnxk/cnxk_dmadev.c | 119 +++
- For fast tests and perf tests, add test stubs to skip not supported
ones.
- For driver tests, for now skip on Windows totally to avoid
unnecessary amount of test stubs. For example, there are about 30
cryptodev related tests (even though in the meson for CI it only
listed about half) whic
On Windows, strerror returns just "Unknown error" for errnum greater
than MAX_ERRNO, while linux and freebsd returns "Unknown error ",
which is the current expectation for errno_autotest. Differentiate
the error string on Windows to remove a "duplicate error code" failure.
Signed-off-by: Jie Zhou
DPDK logs_autotest on Windows failed at "dynamic log types" tests.
The failures are on 2 test cases for rte_log_set_level_regexp API,
due to regular expression is not supported on Windows in DPDK yet
and regcomp/regexec are just stubs on Windows (in regex.h).
In app\test\test_logs.c, ifndef these
- Add python script to check if system supports hugepages
- Remove corresponding .sh scripts
- Replace calling of .sh with corresponding .py in meson.build
Signed-off-by: Jie Zhou
---
app/test/has-hugepage.sh | 11 ---
app/test/has_hugepage.py | 25 +
app/test/mes
Even though test_interrupts.c can compile on Windows, skip interrupt
tests for now since majority of eal_interrupt on Windows are stubs.
Will remove the skip after interrupt being fully enabled on Windows.
Signed-off-by: Jie Zhou
---
app/test/test_interrupts.c | 10 ++
1 file changed, 10
cmdline tests pass except one failure at the test_cmdline_socket_fns
test case with error: failed to open /dev/null for reading!
Temporarily "skip" this case while enable all other passing cases.
Issue is tracked internally and will add the corresponding case on
Windows in future.
Signed-off-by:
Remove two alarm_autotest test cases which do bogus range check
on Windows.
Signed-off-by: Jie Zhou
---
app/test/test_alarm.c | 4
1 file changed, 4 insertions(+)
diff --git a/app/test/test_alarm.c b/app/test/test_alarm.c
index b4034339b8..70e97a3109 100644
--- a/app/test/test_alarm.c
+++
- Include rte_os_shim.h
- Replace sleep and usleep with rte_delay_us_sleep
- #ifndef RTE_EXEC_ENV_WINDOWS for POSIX code only
Signed-off-by: Jie Zhou
---
app/test/packet_burst_generator.c | 1 +
app/test/process.h| 4 +++-
app/test/test.c | 4
app/test/test
Check rte_errno to exclude ENOTSUP as failures in test_memory.c
Signed-off-by: Jie Zhou
---
app/test/test_memory.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/app/test/test_memory.c b/app/test/test_memory.c
index dbf6871e71..379b0f99ca 100644
--- a/app/test/test_memory.
Remove unnecessary header inclusion
Signed-off-by: Jie Zhou
---
app/test/commands.c | 2 --
app/test/test.c | 1 -
app/test/test_cmdline_lib.c | 1 -
3 files changed, 4 deletions(-)
diff --git a/app/test/commands.c b/app/test/commands.c
index 76f6ee5d23..82bd8bf474 100644
--
This patchset is to enable a subset of unit tests on windows. It mainly
includes:
- Replace POSIX specific codes
- Fix some lib and tests per failures investigation
- Replace .sh script with .py script for meson.build
- Enable build and run subset of unit tests on Windows
Future work:
- Work with
UT memory_autotest on Windows has 2 failed cases on eal APIs
eal_memalloc_get_seg_fd and eal_memalloc_get_seg_fd_offset. These 2
APIs are not supported on Windows yet. Should return ENOTSUP such that
in test_memory.c these 2 ENOTSUP cases will not be marked as failures,
same as other ENOTSUP cases.
- For fast tests and perf tests, add test stubs to skip not supported
ones.
- For driver tests, for now skip on Windows totally to avoid
unnecessary amount of test stubs. For example, there are about 30
cryptodev related tests (even though in the meson for CI it only
listed about half) whic
DPDK logs_autotest on Windows failed at "dynamic log types" tests.
The failures are on 2 test cases for rte_log_set_level_regexp API,
due to regular expression is not supported on Windows in DPDK yet
and regcomp/regexec are just stubs on Windows (in regex.h).
In app\test\test_logs.c, ifndef these
On Windows, strerror returns just "Unknown error" for errnum greater
than MAX_ERRNO, while linux and freebsd returns "Unknown error ",
which is the current expectation for errno_autotest. Differentiate
the error string on Windows to remove a "duplicate error code" failure.
Signed-off-by: Jie Zhou
Remove two alarm_autotest test cases which do bogus range check
on Windows.
Signed-off-by: Jie Zhou
---
app/test/test_alarm.c | 4
1 file changed, 4 insertions(+)
diff --git a/app/test/test_alarm.c b/app/test/test_alarm.c
index b4034339b8..70e97a3109 100644
--- a/app/test/test_alarm.c
+++
Even though test_interrupts.c can compile on Windows, skip interrupt
tests for now since majority of eal_interrupt on Windows are stubs.
Will remove the skip after interrupt being fully enabled on Windows.
Signed-off-by: Jie Zhou
---
app/test/test_interrupts.c | 10 ++
1 file changed, 10
- Add python script to check if system supports hugepages
- Remove corresponding .sh scripts
- Replace calling of .sh with corresponding .py in meson.build
Signed-off-by: Jie Zhou
---
app/test/has-hugepage.sh | 11 ---
app/test/has_hugepage.py | 25 +
app/test/mes
cmdline tests pass except one failure at the test_cmdline_socket_fns
test case with error: failed to open /dev/null for reading!
Temporarily "skip" this case while enable all other passing cases.
Issue is tracked internally and will add the corresponding case on
Windows in future.
Signed-off-by:
- Include rte_os_shim.h
- Replace sleep and usleep with rte_delay_us_sleep
- #ifndef RTE_EXEC_ENV_WINDOWS for POSIX code only
Signed-off-by: Jie Zhou
---
app/test/packet_burst_generator.c | 1 +
app/test/process.h| 4 +++-
app/test/test.c | 4
app/test/test
This patchset is to enable a subset of unit tests on windows. It mainly
includes:
- Replace POSIX specific codes
- Fix some lib and tests per failures investigation
- Replace .sh script with .py script for meson.build
- Enable build and run subset of unit tests on Windows
Future work:
- Work with
Check rte_errno to exclude ENOTSUP as failures in test_memory.c
Signed-off-by: Jie Zhou
---
app/test/test_memory.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/app/test/test_memory.c b/app/test/test_memory.c
index dbf6871e71..379b0f99ca 100644
--- a/app/test/test_memory.
UT memory_autotest on Windows has 2 failed cases on eal APIs
eal_memalloc_get_seg_fd and eal_memalloc_get_seg_fd_offset. These 2
APIs are not supported on Windows yet. Should return ENOTSUP such that
in test_memory.c these 2 ENOTSUP cases will not be marked as failures,
same as other ENOTSUP cases.
Remove unnecessary header inclusion
Signed-off-by: Jie Zhou
---
app/test/commands.c | 2 --
app/test/test.c | 1 -
app/test/test_cmdline_lib.c | 1 -
3 files changed, 4 deletions(-)
diff --git a/app/test/commands.c b/app/test/commands.c
index 76f6ee5d23..82bd8bf474 100644
--
Hi Maxime,
>-Original Message-
>From: Maxime Coquelin
>Sent: Tuesday, October 26, 2021 4:47 AM
>To: dev@dpdk.org; david.march...@redhat.com; Xia, Chenbo
>; Ding, Xuan
>Subject: Re: [PATCH] vhost: fix async DMA map
>
>Hi Xuan,
>
>On 10/25/21 22:33, Maxime Coquelin wrote:
>> This patch fix
From: Dapeng Yu
This patch uses the index value to call the function, instead of the
function pointer assignment to save the selection of Receive Flex
Descriptor profile ID.
Otherwise the secondary process will run with wrong function address
from primary process.
Fixes: 7a340b0b4e03 ("net/ice:
The GTP, GTP-U, GTP-C header fields can be matched, however NIC does not
support GTP tunneling so no items after the GTP header can be specified.
If a GTP-U or GTP-C item is specified without a preceding UDP item, the
UDP destination port is implicitly matched. For GTP, the destination UDP
port mus
> -邮件原件-
> 发件人: dev 代表 Jerin Jacob
> 发送时间: Monday, October 25, 2021 5:44 PM
> 收件人: Feifei Wang
> 抄送: Ruifeng Wang ; Ananyev, Konstantin
> ; dpdk-dev ; nd
>
> 主题: Re: [dpdk-dev] [PATCH v4 1/5] eal: add new definitions for wait scheme
>
> On Mon, Oct 25, 2021 at 3:01 PM Feifei Wang
> wr
> -邮件原件-
> 发件人: Ananyev, Konstantin
> 发送时间: Monday, October 25, 2021 10:29 PM
> 收件人: Feifei Wang ; Ruifeng Wang
>
> 抄送: dev@dpdk.org; nd ; nd
> 主题: RE: [PATCH v4 1/5] eal: add new definitions for wait scheme
>
>
> > > > Introduce macros as generic interface for address monitoring.
>
Hi Kevin,
We test whole patch set and found it should add one judgement:
the ring_size should be less than or equal to MBUF_RING_SIZE.
If ring_size greater than MBUF_RING_SIZE, the tracking DMA bufs
may be overwrited when the DMA copy is not in time.
Thanks.
On 2021/10/14 17:53, Kevin Laatz wrot
On Mon, Oct 25, 2021 at 05:38:42PM +0200, Thomas Monjalon wrote:
> 14/10/2021 18:21, Jie Zhou:
> > Enable building subset of libraries that tests depend on for Windows
> >
> > Signed-off-by: Jie Zhou
> > ---
> > lib/meson.build | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/li
On 2021/10/20 22:26, Bruce Richardson wrote:
> Since we allow line lengths of up to 100, and the CI checkpatches job
> only check for that amount, the rest of our tooling and docs should
> reflect this reality. Therefore we can:
>
> * adjust the editorconfig to use that value, to save editors (e.g
Probing the availability of Flow Manager API may print the following
error log.
PMD: rte_enic_pmd: Devcmd 88 failed with error code -1
The error indicates a flow manager operation failed and happens when
advanced filtering is disabled on vNIC. It is harmless but confusing
to the user. Since advan
Changing MTU after the device start causes a segfault in the Rx
handler. The MTU handler (enic_set_mtu) performs the following steps.
1. Stop NIC Rx
2. Change Rx handler '(struct rte_eth_dev)->rx_pkt_burst' to
the dummy handler and sleep a while to quiesce
3. Re-allocate/initialize Rx structures
On Sun, Oct 24, 2021 at 9:42 AM Andrew Rybchenko
wrote:
>
> There is no point to do rte_eth_dev_mtu_set() before configure since
> set MTU value is overwritten on configure anyway. So, setting of MTU
> before configure is rejected now on ethdev level.
>
> If testpmd is going to do configure (e.g.
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v21.11-rc1
There are 1171 new patches in this snapshot, big as expected.
Release notes:
https://doc.dpdk.org/guides/rel_notes/release_21_11.html
Highlights of 21.11-rc1:
* General
- more
On Mon, Oct 25, 2021 at 11:30 PM yn h wrote:
>
> I downloaded dpdk version 21.08 from
> http://static.dpdk.org/rel/.
>
> Extracted and ran meson build.
> During ninja build, I ran into
>
> FAILED: drivers/rte_common_cpt.pmd.c
>
> /usr/bin/python3 ../buildtools/gen-pmdinfo-cfile.py
> /home/ybuser/d
2021-10-25 17:11 (UTC-0400), yn h:
> I downloaded dpdk version 21.08 from
> http://static.dpdk.org/rel/.
>
> Extracted and ran meson build.
> During ninja build, I ran into
>
> FAILED: drivers/rte_common_cpt.pmd.c
>
> /usr/bin/python3 ../buildtools/gen-pmdinfo-cfile.py
> /home/ybuser/dpdk-21.05/
I downloaded dpdk version 21.08 from
http://static.dpdk.org/rel/.
Extracted and ran meson build.
During ninja build, I ran into
FAILED: drivers/rte_common_cpt.pmd.c
/usr/bin/python3 ../buildtools/gen-pmdinfo-cfile.py
/home/ybuser/dpdk-21.05/build/buildtools ar
/home/ybuser/dpdk-21.05/build/drive
Hi Xuan,
On 10/25/21 22:33, Maxime Coquelin wrote:
This patch fixes possible NULL-pointer dereferencing
reported by Coverity and also fixes NUMA reallocation
of the async DMA map.
Fixes: 7c61fa08b716 ("vhost: enable IOMMU for async vhost")
Coverity issue: 373655
Signed-off-by: Maxime Coquelin
25/10/2021 22:29, Ananyev, Konstantin:
>
> > > > > > There is a concern about getting efficient log report,
> > > > > > especially when looking at CI issues.
> > > > >
> > > > > +1.
> > > > > The current solution with logs is a real pain.
> > > >
> > > > Are you guys talking about problems with
>
This patch fixes possible NULL-pointer dereferencing
reported by Coverity and also fixes NUMA reallocation
of the async DMA map.
Fixes: 7c61fa08b716 ("vhost: enable IOMMU for async vhost")
Coverity issue: 373655
Signed-off-by: Maxime Coquelin
---
lib/vhost/vhost_user.c | 45 +++
> > > > > There is a concern about getting efficient log report,
> > > > > especially when looking at CI issues.
> > > >
> > > > +1.
> > > > The current solution with logs is a real pain.
> > >
> > > Are you guys talking about problems with
> > > app/test/sample_packet_forward.* David reported?
>
11/10/2021 20:23, lir...@marvell.com:
> From: Liron Himi
>
> Removing Guy Kaneti
> Adding Liron Himi
>
> Signed-off-by: Liron Himi
Applied, thanks.
21/10/2021 09:27, Junfeng Guo:
> Remove Xiaoyun and add Junfeng.
>
> Signed-off-by: Junfeng Guo
> Acked-by: Xiaoyun Li
Applied, thanks.
02/09/2021 18:04, Stephen Hemminger:
> On Thu, 2 Sep 2021 16:50:14 +0100
> Bruce Richardson wrote:
> > On Thu, Sep 02, 2021 at 08:24:58AM -0700, Stephen Hemminger wrote:
> > > On Thu, 2 Sep 2021 09:10:47 +0100
> > > Bruce Richardson wrote:
> > >
> > > > On Wed, Sep 01, 2021 at 02:27:07PM -0700
On 23/10/2021 11:18, David Marchand wrote:
On Sat, Oct 23, 2021 at 8:56 AM David Marchand
wrote:
Windows compilation gives us a splat:
In file included from ../drivers/dma/idxd/idxd_pci.c:10:
In file included from ..\drivers\dma\idxd/idxd_internal.h:11:
..\drivers\dma\idxd/idxd_hw_defs.h:46:21:
On 22/10/2021 20:48, Thomas Monjalon wrote:
14/10/2021 11:53, Kevin Laatz:
This patchset first adds some additional command line options to the
existing ioatfwd application to enhance usability.
Please could you rebase?
I see some errors about max_rx_pkt_len and DEV_RX_OFFLOAD_JUMBO_FRAME.
Cc F
01/07/2021 15:24, Christian Ehrhardt:
> On Thu, Jul 1, 2021 at 10:23 AM Christian Ehrhardt
> wrote:
> > On Thu, Jun 17, 2021 at 10:25 AM Marco Varlese
> > wrote:
> > > On 6/17/21 8:41 AM, Thomas Monjalon wrote:
> > > > 17/06/2021 08:14, Christian Ehrhardt:
> > > >> On Thu, Jun 10, 2021 at 12:30
13/10/2021 10:28, Bruce Richardson:
> On Wed, Oct 13, 2021 at 12:23:03PM +0500, Sarosh Arif wrote:
> > The default behaviour of l2fwd is to exit if we are unable to turn
> > promiscuous mode on. On some aws instances turning promiscuous mode
> > on is not permitted. In such cases there should be a
On Fri, Oct 22, 2021 at 10:35 AM Olivier Matz wrote:
>
> On Thu, Oct 21, 2021 at 04:32:21PM -0500, Honnappa Nagarahalli wrote:
> > Add a testcase to test launching of control threads.
> >
> > Signed-off-by: Honnappa Nagarahalli
>
> Reviewed-by: Olivier Matz
Series applied, thanks.
--
David M
> > Add optional destination ports parameter for port-id action.
> > The parameter is not must, and the value is 1 by default as before if the
> > parameter not provided.
> >
> > For example:
> >
> > $ dpdk-test-flow-perf -w 08:00.0,representor=[0,1] -- --transfer \
> > > --ingress --transfer --e
On Mon, Oct 25, 2021 at 4:27 PM David Marchand
wrote:
>
> Moving struct rte_intr_handle as an internal structure to
> avoid any ABI breakages in future. Since this structure defines
> some static arrays and changing respective macros breaks the ABI.
> Eg:
> Currently RTE_MAX_RXTX_INTR_VEC_ID impos
04/10/2021 14:55, Wisam Jaddo:
> Some options are needed in the runtime many times, so leaving
> it during compilation is not correct. As a result some options
> has been exported into command line options to be used at run
> time.
>
> The options exported are:
> --txq=N
> --rxq=N
> --txd=N
> --rx
15/03/2021 20:34, Tyler Retzlaff:
> The proposal has resulted from request to review [1] the following
> functions where there appeared to be inconsistency in return type
> or parameter type selections for the following inline functions.
>
> rte_bsf32()
> rte_bsf32_safe()
> rte_bsf64()
> rte_bsf64
> > Windows EAL did not detect IOVA mode and worked incorrectly
> > if physical addresses could not be obtained
> > (if virt2phys driver was missing or inaccessible).
> > In this case, rte_mem_virt2iova() reported RTE_BAD_IOVA for any address.
> > Inability to obtain IOVA, be it PA or VA, should ca
> > From a Telemetry usage point of view,
> >
> > Acked-by: Ciara Power
>
> Agree, this patch is much more in keeping with the existing way of working
> than the v1.
>
> Acked-by: Bruce Richardson
Applied, thanks.
On 10/25/2021 5:20 AM, Dmitry Kozlyuk wrote:
Windows EAL did not detect IOVA mode and worked incorrectly
if physical addresses could not be obtained
(if virt2phys driver was missing or inaccessible).
In this case, rte_mem_virt2iova() reported RTE_BAD_IOVA for any address.
Inability to obtain IO
Hi,
> -Original Message-
> From: David Marchand
> Sent: Monday, October 25, 2021 5:27 PM
> To: hka...@marvell.com; dev@dpdk.org
> Cc: dmitry.kozl...@gmail.com; Raslan Darawsheh ;
> NBU-Contact-Thomas Monjalon
> Subject: [PATCH v8 0/9] make rte_intr_handle internal
>
> Moving struct rte_i
On Thu, 21 Oct 2021 19:54:29 +0100
Vladimir Medvedkin wrote:
> +static uint8_t default_rss_key[] = {
Should this be const?
That way you can make sure API isn't modifying it.
06/09/2021 17:55, Vladimir Medvedkin:
> This patch adds a new parameter to the fib configuration to specify
> the size of the extension for internal RIB structure.
It looks to be an announced API change.
What happens if the new field is not initialized in the app?
At least it would deserve a note
06/09/2021 17:54, Vladimir Medvedkin:
> Fixes: 03b8372a9a73 ("rib: fix max depth IPv6 lookup")
> Cc: ohily...@iol.unh.edu
> Cc: sta...@dpdk.org
>
> Signed-off-by: Vladimir Medvedkin
An explanation would have been appreciated.
> - index = (depth & (UINT8_MAX - 1)) / CHAR_BIT;
> + index =
22/10/2021 11:07, Bruce Richardson:
> On Thu, Oct 21, 2021 at 06:15:49PM +0100, Vladimir Medvedkin wrote:
> > This patch fixes buffer overflow reported by ASAN,
> > please reference https://bugs.dpdk.org/show_bug.cgi?id=819
> >
> > The rte_lpm6 keeps routing information for control plane purpose
>
21/10/2021 20:54, Vladimir Medvedkin:
> This patch add a new Toeplitz hash implementation using
> Galios Fields New Instructions (GFNI).
>
> Signed-off-by: Vladimir Medvedkin
> Acked-by: Konstantin Ananyev
> ---
> --- a/lib/hash/version.map
> +++ b/lib/hash/version.map
> @@ -39,10 +39,12 @@ EXPE
Vladimir, your patches are late and not perfect.
You need reviews. Please ask other maintainers to help with reviews.
21/10/2021 20:54, Vladimir Medvedkin:
> This patch adds documentation for the new optimized Toeplitz hash
> implementation using GFNI.
>
> Signed-off-by: Vladimir Medvedkin
> --
Hi,
> -Original Message-
> From: dev On Behalf Of Harman Kalra
> Sent: Friday, October 22, 2021 11:49 PM
> To: dev@dpdk.org
> Cc: david.march...@redhat.com; dmitry.kozl...@gmail.com;
> m...@ashroe.eu; NBU-Contact-Thomas Monjalon ;
> Harman Kalra
> Subject: [dpdk-dev] [PATCH v5 0/6] make
21/10/2021 20:54, Vladimir Medvedkin:
> This patch adds performance tests for different implementations
> of the Toeplitz hash function.
Please name them.
> Signed-off-by: Vladimir Medvedkin
There are some garbage,
> @@ -320,6 +321,7 @@ perf_test_names = [
> 'hash_readwrite_lf_perf_au
10/09/2021 11:46, Mcnamara, John:
> From: Medvedkin, Vladimir
>
> The git diff makes this look like the ifdef is moving but I see that you are
> moving the doc into the right place so Doxygen can pick it up.
>
> Acked-by: John McNamara
title: hash: fix Doxygen comment of Toeplitz file
Applie
25/10/2021 18:17, Thomas Monjalon:
> Options used to be sorted alphabetically.
> It looks easier to read when major options are first,
> then path tuning, libs options, and drivers options.
Even better, we could insert a blank line between each option.
On Mon, Oct 25, 2021 at 6:53 AM Honnappa Nagarahalli
wrote:
>
> v3:
> a) Added Fixes, Cc:stable#dpdk.org in 1/6
> b) Merged 3/6 & 4/6 and moved after the first commit in the series
> c) Merged 2/6 & 5/6 as they need to be in a single commit
> d) Removed use of volatile in 6/6 (Konstantin)
>
> rte_
Options used to be sorted alphabetically.
It looks easier to read when major options are first,
then path tuning, libs options, and drivers options.
Signed-off-by: Thomas Monjalon
---
meson_options.txt | 75 ---
1 file changed, 39 insertions(+), 36 del
14/10/2021 10:20, Bruce Richardson:
> On Thu, Oct 14, 2021 at 04:54:18AM +0800, Kefu Chai wrote:
> > RTE_MBUF_REFCNT_ATOMIC = 0 is not necessary for applications like
> > Seastar, where it's safe to assume that the mbuf refcnt is only
> > updated by a single core only.
> >
> > Signed-off-by: Kefu
14/10/2021 14:26, Aaron Conole:
> Juraj Linkeš writes:
>
> > The way we're building DPDK in CI, with -Dmachine=default, has not been
> > updated when the option got replaced to preserve a backwards-complatible
> > build call to facilitate ABI verification between DPDK versions. Update
> > the cal
14/10/2021 18:21, Jie Zhou:
> Enable building subset of libraries that tests depend on for Windows
>
> Signed-off-by: Jie Zhou
> ---
> lib/meson.build | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/lib/meson.build b/lib/meson.build
> index b2ba7258d8..bd6c27deef 100644
> --- a/lib/m
21/10/2021 10:51, Eli Britstein:
> Functions and macros in x86 rte_memcpy.h may cause cast-align warnings,
> when using strict cast align flag with supporting gcc:
> gcc (Ubuntu 9.3.0-17ubuntu1~20.04) 9.3.0
> CFLAGS="-Wcast-align=strict" make V=1 -C examples/l2fwd clean static
>
> For example:
> I
15/10/2021 14:26, Olivier Matz:
> On Fri, Oct 15, 2021 at 10:39:41AM +0200, David Marchand wrote:
> > This option should be prefixed with -- for consistency with others.
> >
> > Fixes: a103a97e7191 ("eal: allow user to override default mempool driver")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-b
20/10/2021 21:31, Robin Jarry:
> Hello,
>
> 2021-10-19, David Marchand:
> > Migration to argparse was incomplete.
> >
> > $ dpdk-pmdinfo.py -p $(which dpdk-testpmd)
> > Traceback (most recent call last):
> > File "/usr/bin/dpdk-pmdinfo.py", line 626, in
> > main()
> > File "/usr/bin/dpdk
25/10/2021 15:33, Olivier Matz:
> On Sat, Oct 23, 2021 at 12:09:19AM +0300, Dmitry Kozlyuk wrote:
> > When mempool had been created with RTE_MEMPOOL_F_NO_IOVA_CONTIG flag
> > but later populated with valid IOVA, RTE_MEMPOOL_F_NON_IO was unset,
> > while it should be kept. The unit test did not catc
Minutes of Technical Board Meeting, 2021-Oct-20
Members Attending
-
-Bruce
-Ferruh
-Hemant
-Honnappa
-Jerin
-Kevin
-Konstantin (Chair)
-Maxime
-Stephen
-Thomas
NOTE: The technical board meetings every second Wednesday at
https://meet.jit.si/DPDK at 3 pm UTC.
Meetings are public,
From: Harman Kalra
Dynamically allocating the efds and elist array of intr_handle
structure, based on size provided by user. Eg size can be
MSIX interrupts supported by a PCI device.
Signed-off-by: Harman Kalra
Signed-off-by: David Marchand
Acked-by: Dmitry Kozlyuk
---
Changes since v6:
- rem
> > > Introduce macros as generic interface for address monitoring.
> > >
> > > Signed-off-by: Feifei Wang
> > > Reviewed-by: Ruifeng Wang
> > > ---
> > > lib/eal/arm/include/rte_pause_64.h | 126
> > > lib/eal/include/generic/rte_pause.h |
> > > 32 +++
> > >
From: Harman Kalra
VFIO/UIO are mutually exclusive, storing file descriptor in a single
field is enough.
Signed-off-by: Harman Kalra
Signed-off-by: David Marchand
---
Changes since v5:
- split from patch5,
---
lib/eal/common/eal_common_interrupts.c | 6 +++---
lib/eal/common/eal_interrupts.h
From: Harman Kalra
Moving interrupt handle structure definition inside a EAL private
header to make its fields totally opaque to the outside world.
Signed-off-by: Harman Kalra
Signed-off-by: David Marchand
---
Changes since v5:
- let rte_intr_handle fields untouched:
- split vfio / uio fd re
From: Harman Kalra
Removing direct access to interrupt handle structure fields,
rather use respective get set APIs for the same.
Making changes to all the libraries access the interrupt handle fields.
Signed-off-by: Harman Kalra
Signed-off-by: David Marchand
---
Changes since v5:
- split from
From: Harman Kalra
Removing direct access to interrupt handle structure fields,
rather use respective get set APIs for the same.
Making changes to all the libraries access the interrupt handle fields.
Implementing alarm cleanup routine, where the memory allocated
for interrupt instance can be fr
From: Harman Kalra
Updating the interrupt testsuite to make use of interrupt
handle get set APIs.
Signed-off-by: Harman Kalra
Signed-off-by: David Marchand
---
Changes since v5:
- fixed leak on when some interrupt handle can't be allocated,
---
app/test/test_interrupts.c | 164 ++
From: Harman Kalra
Making changes to the interrupt framework to use interrupt handle
APIs to get/set any field.
Signed-off-by: Harman Kalra
Signed-off-by: David Marchand
---
Changes since v6:
- fixed compilation on FreeBSD,
Changes since v5:
- used new helper rte_intr_instance_dup,
---
lib/
1 - 100 of 169 matches
Mail list logo
