On Tue, Feb 19, 2019 at 09:40:05PM +0800, Jason Wang wrote:
> On 2019/2/19 下午6:59, Tiwei Bie wrote:
> > Below is a quick (unofficial) performance test (macfwd loop, 64B)
> > for the packed ring optimizations in this series on an Intel(R)
> > Xeon(R) Gold 6140 CPU @ 2.30GHz platform:
> >
> > w/o th
On Tue, Feb 19, 2019 at 08:59:51PM +0800, Parthasarathy, JananeeX M wrote:
>
> >-Original Message-
> >From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Maxime Coquelin
> >Sent: Monday, February 04, 2019 3:02 PM
> >To: Poornima, PallantlaX ; dev@dpdk.org
> >Cc: Pattan, Reshma ; Bie, Tiwe
Spawning the ctrl threads on anything that is not part of the eal
coremask is not that polite to the rest of the system, especially
when you took good care to pin your processes on cpu resources with
tools like taskset (linux) / cpuset (freebsd).
Rather than introduce yet another eal options to co
pthread_setaffinity_np returns a >0 value on error.
We could end up letting the ctrl threads on the current process cpu
affinity.
Fixes: d651ee4919cd ("eal: set affinity for control threads")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
Changelog since v2:
- added missing Cc: sta...@dpdk
pthread_getaffinity_np returns a >0 value when failing.
This is mainly for the sake of correctness.
The only case where it could fail is when passing an incorrect cpuset
size wrt to the kernel.
Fixes: 2eba8d21f3c9 ("eal: restrict cores auto detection")
Cc: sta...@dpdk.org
Signed-off-by: David Mar
19/02/2019 12:51, David Marchand:
> On Tue, Feb 19, 2019 at 12:38 PM Burakov, Anatoly
> wrote:
> > On 14-Feb-19 1:30 PM, David Marchand wrote:
> > > --- a/doc/guides/prog_guide/env_abstraction_layer.rst
> > > +++ b/doc/guides/prog_guide/env_abstraction_layer.rst
> > > +Control Thread API
> > > +~~
Add svg images for debug and troubleshoot guide.
Signed-off-by: Vipin Varghese
Acked-by: Marko Kovacevic
---
---
doc/guides/howto/img/dtg_consumer_ring.svg| 24
doc/guides/howto/img/dtg_crypto.svg | 21
.../howto/img/dtg_distributor_worker.svg | 36 ++
doc/gu
Add user guide on debugging and troubleshooting for common
issues and bottleneck found in the sample application model.
Signed-off-by: Vipin Varghese
Acked-by: Marko Kovacevic
---
doc/guides/howto/debug_troubleshoot_guide.rst | 451 ++
doc/guides/howto/index.rst
The patch series adds a how-to guide for debugging and troubleshooting tips.
Motivation
==
DPDK proc-info tool is been enhanced to accommodate the debug information for
the port, traffic manager crypto, ring and mempool contents. With these
additional information, it becomes easy to analy
From: Damian Nowak
This patch enables plain SHA algorithm CAVP test support
in fips_validation sample application.
Signed-off-by: Damian Nowak
Acked-by: Fan Zhang
---
examples/fips_validation/Makefile | 1 +
examples/fips_validation/fips_validation.c | 17
examples/fi
This patch adds the functional test scripts to ipsec-secgw
sample application for both transport and tunnel working
mode.
Signed-off-by: Fan Zhang
---
examples/ipsec-secgw/test/run_test.sh | 8 ++-
.../test/trs_3descbc_sha1_common_defs.sh | 73 ++
exam
This patch adds triple-des CBC mode cipher algorithm to ipsec
library.
Signed-off-by: Fan Zhang
---
lib/librte_ipsec/sa.c | 10 ++
lib/librte_ipsec/sa.h | 6 ++
2 files changed, 16 insertions(+)
diff --git a/lib/librte_ipsec/sa.c b/lib/librte_ipsec/sa.c
index e34dd320a..5c59c4b67 1
This patch adds the functional test scripts to ipsec-secgw
sample application for both transport and tunnel working
mode.
Updated a bit on common_defs to use "mktemp" instead of "tempfile"
as Fedora does not like the command.
Signed-off-by: Fan Zhang
---
examples/ipsec-secgw/test/common_defs.sh
This patch adds AES-CTR cipher algorithm support to ipsec
library.
Signed-off-by: Fan Zhang
---
lib/librte_ipsec/crypto.h | 17 ++
lib/librte_ipsec/sa.c | 133 ++
lib/librte_ipsec/sa.h | 18 +++
3 files changed, 147 insertions(+), 21
This patchset adds the AES-CTR and 3DES-CBC cipher algorithms
support to ipsec library. The test scripts for ipsec-secgw
sample application are added too.
v2:
- removed unsupported tests.
Fan Zhang (4):
ipsec: add AES-CTR algorithm support
ipsec-secgw: add test scripts for aes ctr
ipsec: ad
On Wed, 2018-09-26 at 10:15 +0100, Bruce Richardson wrote:
> Rather than relying on the target machine architecture, use the
> size of a pointer from the compiler to determine if we are 64-bits
> or not. This allows correct behaviour when you pass -m32 as a compile
> option. It also allows us to us
Since all other apps have been moved to the "app" folder, the autotest app
remains alone in the test folder. Rather than having an entire top-level
folder for this, we can move it back to where it all started in early
versions of DPDK - the "app/" folder.
This move has a couple of advantages:
* Th
The bpf folder didn't actual contain a test application, but instead
basic examples of BPF code for use with testpmd. Therefore we can
move it to the `examples` folder. Being different, it also needs
a README with it, explaining what it is and how to use it. References
to the code from the testpmd
Move to the app directory, and add to meson build.
CC: Cristian Dumitrescu
Signed-off-by: Bruce Richardson
---
MAINTAINERS | 2 +-
app/Makefile| 1 +
app/meson.build | 1 +
{test => app}/t
Move app to "app" directory and enable with meson build.
CC: Olivier Matz
Signed-off-by: Bruce Richardson
---
MAINTAINERS | 2 +-
app/Makefile| 1 +
{test => app}/cmdline_test/Makefile | 0
{test => app}/cmdline
Move to "app" directory and enable with meson build.
CC: Konstantin Ananyev
Signed-off-by: Bruce Richardson
---
MAINTAINERS | 2 +-
app/Makefile| 1 +
app/meson.build | 1 +
{test => app}/test-acl/Makefile | 0
{test => app}/test-acl/main.
The apps for testing are split between the "test" and "app" directories,
with a not-very-clear distinction between the two (at least to my mind).
Given how the apps are being built, the easiest path to having
cmdline_test, test-acl and test-pipeline build using meson is to
consolidate all these ap
On 2019/2/19 下午6:59, Tiwei Bie wrote:
Below is a quick (unofficial) performance test (macfwd loop, 64B)
for the packed ring optimizations in this series on an Intel(R)
Xeon(R) Gold 6140 CPU @ 2.30GHz platform:
w/o this series:
packed ring normal/in-order: ~10.4 Mpps
w/ this series:
packed ri
Shifting signed 32-bit values by 31-bits has the potential for
unexpected outcomes as compiler can overwrite a bit.
Specified that values are unsigned.
Errors are observed from running cppcheck.
Bugzilla ID: 58
Fixes: 69e209be5464 ("net/axgbe: add register map and related macros")
Fixes: b5bf771
On Tue, Feb 19, 2019 at 12:40 PM Burakov, Anatoly
wrote:
> On 14-Feb-19 5:41 PM, David Marchand wrote:
> > On Thu, Feb 14, 2019 at 5:44 PM Burakov, Anatoly
> > mailto:anatoly.bura...@intel.com>> wrote:
> > CC: stable?
> >
> > Not really sure about CCing stable for this.
> >
> > I did not get
On Tue, Feb 19, 2019 at 12:38 PM Burakov, Anatoly
wrote:
> On 14-Feb-19 1:30 PM, David Marchand wrote:
> > --- a/doc/guides/prog_guide/env_abstraction_layer.rst
> > +++ b/doc/guides/prog_guide/env_abstraction_layer.rst
> > @@ -498,6 +498,20 @@ Those TLS include *_cpuset* and *_socket_id*:
> > *
On 14-Feb-19 5:41 PM, David Marchand wrote:
On Thu, Feb 14, 2019 at 5:44 PM Burakov, Anatoly
mailto:anatoly.bura...@intel.com>> wrote:
On 14-Feb-19 1:27 PM, David Marchand wrote:
> pthread_getaffinity_np returns a >0 value when failing.
>
> This is mainly for the sake of c
On 14-Feb-19 1:30 PM, David Marchand wrote:
Spawning the ctrl threads on anything that is not part of the eal
coremask is not that polite to the rest of the system, especially
when you took good care to pin your processes on cpu resources with
tools like taskset (linux) / cpuset (freebsd).
Rathe
This patch introduces a helper for clearing the virtio net header
to avoid the code duplication. Macro is used as it shows slightly
better performance.
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_rxtx.c | 46 +---
1 file changed, 18 insertions(+), 28 deleti
Below is a quick (unofficial) performance test (macfwd loop, 64B)
for the packed ring optimizations in this series on an Intel(R)
Xeon(R) Gold 6140 CPU @ 2.30GHz platform:
w/o this series:
packed ring normal/in-order: ~10.4 Mpps
w/ this series:
packed ring normal: ~10.9 Mpps
packed rin
This patch introduces an optimized enqueue function in packed
ring for the case that virtio net header can be prepended to
the unchained mbuf.
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_rxtx.c | 63 +++-
1 file changed, 61 insertions(+), 2 deletions(-)
di
When IN_ORDER feature is negotiated, device may just write out a
single used descriptor for a batch of buffers:
"""
Some devices always use descriptors in the same order in which they
have been made available. These devices can offer the VIRTIO_F_IN_ORDER
feature. If negotiated, this knowledge all
We should try to cleanup at least the 'need' number of descs.
Fixes: 892dc798fa9c ("net/virtio: implement Tx path for packed queues")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
drivers/net/virtio/virtio_rxtx.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/virtio/virtio_rx
When IN_ORDER feature is negotiated, device may just write out a
single used ring entry for a batch of buffers:
"""
Some devices always use descriptors in the same order in which they
have been made available. These devices can offer the VIRTIO_F_IN_ORDER
feature. If negotiated, this knowledge all
v4:
Change some assignment operation in recursive ticket lock to __atomic.
V3:
1.Update ticketlock intrduction(suggested by Honnappa Nagarahalli).
2.Change the implementation of rte_ticketlock_trylock to CAS(suggested by
Honnappa Nagarahalli).
V2:
1.Update commit message(sugg
Add test cases for ticket lock, recursive ticket lock,
and ticket lock performance.
Signed-off-by: Joyce Kong
Reviewed-by: Gavin Hu
Reviewed-by: Phil Yang
---
test/test/Makefile | 1 +
test/test/autotest_data.py | 6 +
test/test/meson.build | 1 +
test/test/test_ticketloc
The spinlock implementation is unfair, some threads may take locks
aggressively while leaving the other threads starving for long time.
This patch introduces ticketlock which gives each waiting thread a
ticket and they can take the lock one by one. First come, first serviced.
This avoids starvatio
37 matches
Mail list logo
