On Thu, Jun 3, 2021 at 9:18 PM Thomas Monjalon wrote:
>
> 03/06/2021 20:15, Morten Brørup:
> > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
> > > Sent: Thursday, 3 June 2021 18.11
> > >
> > > 03/06/2021 17:18, Morten Brørup:
> > > > Hi all,
> > > >
> > > > Bugfix releases
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of David Marchand
> Sent: Friday, 4 June 2021 09.01
>
> On Thu, Jun 3, 2021 at 9:18 PM Thomas Monjalon
> wrote:
> >
> > 03/06/2021 20:15, Morten Brørup:
> > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas
> Monjalon
> > > > Sent:
On 5/28/21 10:11 AM, Jiayu Hu wrote:
> Lock protection is needed during the vhost notifies the application of
> device readiness, so the first patch is to add lock protection. After
> performing locking, existed async vhost registration functions will cause
> deadlock, as they acquire lock too.
> -Original Message-
> From: Wang, Yixue
> Sent: Friday, June 4, 2021 2:52 PM
> To: Zhang, Qi Z ; Yang, Qiming
>
> Cc: Zhang, Liheng ; Dong, Yao
> ; dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH v2] net/ice: fix data path corrupt on secondary process
>
> Hi Qi,
>
> Patch v2 has
Sorry, for previous blank reply.
On 5/28/21 10:11 AM, Jiayu Hu wrote:
> Lock protection is needed during the vhost notifies the application of
> device readiness, so the first patch is to add lock protection. After
> performing locking, existed async vhost registration functions will cause
> deadl
On Fri, Jun 4, 2021 at 12:29 AM Dmitry Kozlyuk wrote:
>
> 2021-06-03 18:55 (UTC+0200), David Marchand:
> [...]
> > diff --git a/config/meson.build b/config/meson.build
> > index 017bb2efbb..c6985139b4 100644
> > --- a/config/meson.build
> > +++ b/config/meson.build
> > @@ -172,6 +172,13 @@ if libe
Add the logic to determine how many DD bits have been set
for contiguous packets, for removing the SMP barrier while
reading descs.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
drivers/net/i40e/i40e_rxtx.c | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --gi
03/06/2021 21:36, Morten Brørup:
> > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > Sent: Thursday, 3 June 2021 21.18
> k-dev] code.dpdk.org bugfix releases
> >
> > 03/06/2021 20:15, Morten Brørup:
> > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas
> > Monjalon
> > > > Sen
04/06/2021 07:51, Wang, Haiyue:
> > From: Elena Agostini
> >
> > The new library gpudev is for dealing with GPU from a DPDK application
> > in a vendor-agnostic way.
> >
> > As a first step, the features are focused on memory management.
> > A function allows to allocate memory inside the GPU,
>
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Friday, 4 June 2021 10.12
>
> 03/06/2021 21:36, Morten Brørup:
> > > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > > Sent: Thursday, 3 June 2021 21.18
> > k-dev] code.dpdk.org bugfix releases
> > >
> > > 03/06/
04/06/2021 10:19, Morten Brørup:
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
> > > Now, I have another question: What does code.dpdk.org show for "dpdk"
> > that it does not show for "dpdk-stable"?
> >
> > stable releases are only in dpdk-stable but everything is in dp
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Friday, 4 June 2021 10.48
>
> 04/06/2021 10:19, Morten Brørup:
> > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas
> Monjalon
> > > > Now, I have another question: What does code.dpdk.org show for
> "dpdk"
Since C11 memory model is adopted in DPDK now[1], use GCC's
atomic builtins in test cases.
[1]https://www.dpdk.org/blog/2021/03/26/dpdk-adopts-the-c11-memory-model/
Joyce Kong (8):
test/ticketlock: use GCC atomic builtins for lcores sync
test/spinlock: use GCC atomic builtins for lcores sync
Convert rte_atomic usages to GCC atomic builtins for lcores sync
in ticketlock testcases.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
Reviewed-by: Honnappa Nagarahalli
---
app/test/test_ticketlock.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/app/test/t
Convert rte_atomic usages to GCC atomic builtins for lcores sync
in spinlock testcases.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_spinlock.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/app/test/test_spinlock.c b/app/test/test_spinlo
Convert rte_atomic usages to GCC atomic builtins for lcores sync
in rwlock testcases.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_rwlock.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/app/test/test_rwlock.c b/app/test/test_rwlock.c
index
Convert rte_atomic usages to GCC atomic builtins for lcores sync
in mcslock testcases.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_mcslock.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/app/test/test_mcslock.c b/app/test/test_mcsloc
Remove the unused synchro variable as there is no lcores
sync in mempool function test.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_mempool.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/app/test/test_mempool.c b/app/test/test_mempool.c
index 3adadd673..7675
Convert rte_atomic usages to GCC atomic builtins for lcores sync
in mempool_perf testcases. Meanwhile, remove unnecessary synchro
init as it would be set to 0 when launching cores.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_mempool_perf.c | 12 +---
1 file cha
Convert rte_atomic usages to GCC atomic builtins for lock sync
in service core testcases.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_service_cores.c | 36 +++
1 file changed, 20 insertions(+), 16 deletions(-)
diff --git a/app/test/test
Covert rte_atomic usages to GCC atomic builtins in rcu_perf
testcases.
Signed-off-by: Joyce Kong
Reviewed-by: Ruifeng Wang
---
app/test/test_rcu_qsbr_perf.c | 98 +--
1 file changed, 49 insertions(+), 49 deletions(-)
diff --git a/app/test/test_rcu_qsbr_perf.c b/
Actions VXLAN_DECAP and VXLAN_ENCAP need to be listed
among actions supported for transfer flows.
Fixes: 6ab6c40d1e83 ("net/sfc: support action VXLAN decap in transfer rules")
Fixes: 1bbd1ec2348a ("net/sfc: support action VXLAN encap in MAE backend")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Malov
03/06/2021 11:33, Ferruh Yigit:
> On 6/3/2021 8:47 AM, Jerin Jacob wrote:
> > On Thu, Jun 3, 2021 at 2:05 AM Thomas Monjalon wrote:
> >> + [gpudev] (@ref rte_gpudev.h),
> >
> > Since this device does not have a queue etc? Shouldn't make it a
> > library like mempool with vendor-defin
> -Original Message-
> From: dev On Behalf Of Thomas Monjalon
> Sent: Thursday, June 3, 2021 04:36
> To: dev@dpdk.org
> Cc: Elena Agostini
> Subject: [dpdk-dev] [PATCH] gpudev: introduce memory API
>
> From: Elena Agostini
>
> The new library gpudev is for dealing with GPU from a DPDK
On Fri, Jun 4, 2021 at 3:58 PM Thomas Monjalon wrote:
>
> 03/06/2021 11:33, Ferruh Yigit:
> > On 6/3/2021 8:47 AM, Jerin Jacob wrote:
> > > On Thu, Jun 3, 2021 at 2:05 AM Thomas Monjalon
> > > wrote:
> > >> + [gpudev] (@ref rte_gpudev.h),
> > >
> > > Since this device does not have
Hi Balazs,
On 5/28/21 12:26 PM, Balazs Nemeth wrote:
> Same idea as commit a287ac28919d ("vhost: allocate and free packets
> in bulk in Tx packed"), allocate and free packets in bulk. Also remove
> the unused function virtio_dev_pktmbuf_alloc.
>
> Signed-off-by: Balazs Nemeth
> ---
> lib/vhost/
Including rte_byteorder.h may fail for ARM builds with 'Platform must
be built with RTE_FORCE_INTRINSICS' if rte_config.h is not included
before. Include rte_config.h from rte_byteorder.h to solve the issue.
Signed-off-by: Michael Pfeiffer
---
lib/eal/arm/include/rte_byteorder.h | 9 +
1
Same idea as commit a287ac28919d ("vhost: allocate and free packets
in bulk in Tx packed"), allocate and free packets in bulk. Also remove
the unused function virtio_dev_pktmbuf_alloc.
Signed-off-by: Balazs Nemeth
---
lib/vhost/virtio_net.c | 37 -
1 file chan
04/06/2021 13:07, Wang, Haiyue:
> > From: Elena Agostini
> > +typedef int (*gpu_malloc_t)(struct rte_gpu_dev *dev, size_t size, void
> > **ptr);
> > +typedef int (*gpu_free_t)(struct rte_gpu_dev *dev, void *ptr);
> > +
[...]
> > + /* FUNCTION: Allocate memory on the GPU. */
> > + gpu_malloc_t
04/06/2021 13:09, Jerin Jacob:
> On Fri, Jun 4, 2021 at 3:58 PM Thomas Monjalon wrote:
> > 03/06/2021 11:33, Ferruh Yigit:
> > > On 6/3/2021 8:47 AM, Jerin Jacob wrote:
> > > > On Thu, Jun 3, 2021 at 2:05 AM Thomas Monjalon
> > > > wrote:
> > > >> + [gpudev] (@ref rte_gpudev.h),
> >
03/06/2021 13:38, Jerin Jacob:
> On Thu, Jun 3, 2021 at 4:00 PM Thomas Monjalon wrote:
> > 03/06/2021 12:04, Jerin Jacob:
> > > On Thu, Jun 3, 2021 at 3:06 PM Thomas Monjalon
> > > wrote:
> > > > 03/06/2021 11:20, Jerin Jacob:
> > > > > The device needs have a queue kind of structure
> > > > > a
On Wed, 2021-06-02 at 16:33 +0200, Christian Ehrhardt wrote:
> Like what was done for mainline kernel in commit 38ad54f3bc76 ("kni: fix
> build with Linux 5.6"), a new parameter 'txqueue' has to be added to
> 'ndo_tx_timeout' ndo on SLES 15-SP3 kernel.
>
> Caused by:
> commit c3bf155c40e9db722fe
On 6/4/21 3:46 PM, Thomas Monjalon wrote:
> 04/06/2021 13:09, Jerin Jacob:
>> On Fri, Jun 4, 2021 at 3:58 PM Thomas Monjalon wrote:
>>> 03/06/2021 11:33, Ferruh Yigit:
On 6/3/2021 8:47 AM, Jerin Jacob wrote:
> On Thu, Jun 3, 2021 at 2:05 AM Thomas Monjalon
> wrote:
>> + [gpudev
04/06/2021 15:05, Andrew Rybchenko:
> On 6/4/21 3:46 PM, Thomas Monjalon wrote:
> > 04/06/2021 13:09, Jerin Jacob:
> >> On Fri, Jun 4, 2021 at 3:58 PM Thomas Monjalon wrote:
> >>> 03/06/2021 11:33, Ferruh Yigit:
> On 6/3/2021 8:47 AM, Jerin Jacob wrote:
> > On Thu, Jun 3, 2021 at 2:05 AM
> -Original Message-
> From: Thomas Monjalon
> Sent: Friday, June 4, 2021 20:44
> To: Wang, Haiyue
> Cc: dev@dpdk.org; Elena Agostini
> Subject: Re: [dpdk-dev] [PATCH] gpudev: introduce memory API
>
> 04/06/2021 13:07, Wang, Haiyue:
> > > From: Elena Agostini
> > > +typedef int (*gpu_m
On 6/4/21 4:18 PM, Thomas Monjalon wrote:
> 04/06/2021 15:05, Andrew Rybchenko:
>> On 6/4/21 3:46 PM, Thomas Monjalon wrote:
>>> 04/06/2021 13:09, Jerin Jacob:
On Fri, Jun 4, 2021 at 3:58 PM Thomas Monjalon wrote:
> 03/06/2021 11:33, Ferruh Yigit:
>> On 6/3/2021 8:47 AM, Jerin Jacob w
04/06/2021 15:25, Wang, Haiyue:
> From: Thomas Monjalon
> > Another question is about the function rte_gpu_free().
> > How do we recognize that a memory chunk is from the CPU and GPU visible,
> > or just from GPU?
> >
>
> I didn't find the rte_gpu_free_visible definition, and the rte_gpu_free's
04/06/2021 15:59, Andrew Rybchenko:
> On 6/4/21 4:18 PM, Thomas Monjalon wrote:
> > 04/06/2021 15:05, Andrew Rybchenko:
> >> On 6/4/21 3:46 PM, Thomas Monjalon wrote:
> >>> 04/06/2021 13:09, Jerin Jacob:
> On Fri, Jun 4, 2021 at 3:58 PM Thomas Monjalon
> wrote:
> > 03/06/2021 11:33,
> -Original Message-
> From: Richardson, Bruce
> Sent: Thursday, May 27, 2021 7:58 PM
> To: Laatz, Kevin
> Cc: dev@dpdk.org; sta...@dpdk.org; Pai G, Sunil
> Subject: Re: [PATCH] raw/ioat: fix missing device name in idxd bus scan
>
> On Thu, May 27, 2021 at 02:36:09PM +0100, Kevin Laat
While working on a RISC-V port, using a HiFive Unmatched (FU740) which
does not have IOMMU (hence only RTE_IOVA_PA is available), I've noticed
that some of the EAL tests are failing because of a totally different
reason than the test itself.
Namely the --no-huge flag and --iova-mode=pa can't be use
On Linux systems without IOMMU support available (be it lack of
supported IOMMU or lack of IOMMU support in kernel or explicit --no-huge
EAL parameter), the IOVA mapping will default to DMA with physical
addresses. This implicitly requires hugepage support to work (checked in
rte_eal_using_phys_add
In tests where no-shconf flag is used, no-huge is also passed due to
compatibility with FreeBSD system, as described in: b5d878e6d.
However on Linux systems with RTE_IOVA_PA (lack of or an incompatible
IOMMU) this causes issues since hugepages are required by EAL.
Therefore replace all occurrences
When -n argument became optional, the test logic was fixed (by
1e0b51fd4) but the comment indicating why --no-huge and --no-shconf are
used was not changed.
Today those flags are used for compatibility with FreeBSD (see
b5d878e6d), so change the comment to reflect that.
Signed-off-by: Stanislaw Ka
From: Igor Romanov
Make software index of an Rx queue and ethdev index separate.
When an ethdev RxQ is accessed in ethdev callbacks, an explicit ethdev
queue index is used.
This is a preparation to introducing non-ethdev Rx queues.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
R
Update base driver and support COUNT action in transfer flow rules.
v2:
- add release notes
- add missing documentaion
- fix spelling
- handle query in stopped gracefully
Andrew Rybchenko (6):
net/sfc: do not enable interrupts on internal Rx queues
common/sfc_efx/base: separate target EvQ
rxq_intr flag requests support for interrupt mode for ethdev Rx queues.
There is no internal Rx queues yet.
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc_ev.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/sfc/sfc_ev.c b/drivers/net/sfc/sfc_ev.c
inde
Target EvQ and IRQ number are specified in the same location
in MCDI request. The value is treated as IRQ number if the
event queue is interrupting (corresponding flag is set) and
as target event queue otherwise.
However it is better to separate it on helper API level to
make it more clear.
Signe
Custom mapping is actually supported for EF10 and EF100 families only.
A driver (e.g. DPDK PMD) may require to customize mapping of EvQ
to interrupts if, for example, extra EvQ are used for house-keeping
in polling or wake up (via another EvQ) mode.
Signed-off-by: Andrew Rybchenko
Reviewed-by: A
From: Igor Romanov
Make software index of a Tx queue and ethdev index separate.
When an ethdev TxQ is accessed in ethdev callbacks, an explicit ethdev
queue index is used.
This is a preparation to introducing non-ethdev Tx queues.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Re
Interrupts support has assumptions on interrupt numbers used
for LSC and Rx queues. The first interrupt is used for LSC,
subsequent interrupts are used for Rx queues.
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_ev.c | 32
1
From: Igor Romanov
Add a flag to request support for user mark field on an RxQ.
The field is required to retrieve generation count value from
counter RxQ.
Implement it only for Riverhead and EF10 ESSB since they support
the field in the Rx prefix.
Signed-off-by: Igor Romanov
Signed-off-by: And
From: Igor Romanov
Add extra EFX flags to RxQ info initialization API to support
choosing different RxQ types and make the API public to use
it in for counter queues.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_rx.c | 10 +
From: Igor Romanov
Add a flag to request support for ingress m-port on an RxQ.
Implement it only for Riverhead, other families will return an error
if the flag is set.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/c
From: Igor Romanov
Add a function returning management event queue software index.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_ev.c | 2 +-
drivers/net/sfc/sfc_ev.h | 6 ++
2 files changed, 7 insertions(+), 1 deletion(-)
The driver requires service cores for housekeeping. Share these
cores for many adapters and various purposes to avoid extra CPU
overhead.
Since housekeeping services will talk to NIC, it should be possible
to choose logical core on matching NUMA node.
Signed-off-by: Andrew Rybchenko
Reviewed-by:
From: Igor Romanov
User will be able to create and free MAE counters. Support for
associating counters with action set will be added in upcoming
patches.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/common/sfc_efx/
From: Igor Romanov
The MCDIs will be used to control counter Rx queue packet flow.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/common/sfc_efx/base/efx.h | 32 ++
drivers/common/sfc_efx/base/efx_mae.c | 13
From: Igor Romanov
MAE delivers counters data as special packets via dedicated Rx queue.
Reserve an RxQ so that it does not interfere with ethdev Rx queues.
A routine will be added later to handle these packets.
There is no point to reserve the queue if no service cores are
available and counter
From: Igor Romanov
User will be able to associate counter with MAE action set to
collect counter packets and bytes for a specific action set.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/common/sfc_efx/base/efx.h
From: Igor Romanov
The information about the number of pushed Rx buffers is required
for counter Rx queue to know when to give credits to counter
stream.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/net/sfc/sfc_dp_
From: Igor Romanov
The information about the maximum number of MAE counters is
crucial to the counter support in the driver.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/common/sfc_efx/base/efx.h | 1 +
drivers
Packetiser composes packets with MAE counters update.
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
.../base/efx_regs_counters_pkt_format.h | 87 +++
1 file changed, 87 insertions(+)
create mode 100644 drivers/common/sfc_efx/base/efx_regs_counters_pkt_form
From: Igor Romanov
The query reports the number of hits for a counter associated
with a flow rule.
Signed-off-by: Igor Romanov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
Reviewed-by: Ivan Malov
---
drivers/net/sfc/sfc_flow.c| 48 ++-
drivers/net/sf
From: Igor Romanov
For now, a rule may have only one dedicated counter, shared counters
are not supported.
HW delivers (or "streams") counter readings using special packets.
The driver creates a dedicated Rx queue to receive such packets
and requests that HW start "streaming" the readings to it.
Rx/Tx doorbells stats are essential for performance investigation.
On the way fix ethdev documenation to refine requirements on
driver callback. It allows to make these callbacks a bit simpler.
Add testpmd option to show specified xstats periodically or upon
request, for example:
* --display-xs
From: Ivan Ilchenko
Add MAC stats lock in get xstats by id callback before reading
number of supported MAC stats.
Fixes: 73280c1e4ff ("net/sfc: support xstats retrieval by ID")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
dri
From: Ivan Ilchenko
Update MAC stats function reads adapter state with MAC stats locking
but without adapter locking. Add adapter locking before calling this
function and remove MAC stats locking since there's no point to have
it together with adapter locking. The second place MAC stats locking
i
From: Ivan Ilchenko
Document valid combinations of input arguments in accordance with
current implementation in ethdev.
Fixes: 79c913a42f0 ("ethdev: retrieve xstats by ID")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
lib/eth
From: Ivan Ilchenko
Update xstats by IDs callbacks documentation in accordance with
ethdev usage of these callbacks. Document valid combinations of
input arguments to make driver implementation simpler.
Fixes: 79c913a42f0 ("ethdev: retrieve xstats by ID")
Cc: sta...@dpdk.org
Signed-off-by: Ivan
From: Ivan Ilchenko
Fix xstats by ID callbacks according to ethdev usage.
Handle combinations of input arguments that are required by ethdev
and sanity check and reject other combinations on callback entry.
Fixes: 73280c1e4ff ("net/sfc: support xstats retrieval by ID")
Cc: sta...@dpdk.org
Signe
From: Ivan Ilchenko
Fixes: 1caab2f1e68 ("net/sfc: add basic statistics")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc.h| 2 +-
drivers/net/sfc/sfc_ethdev.c | 6 +++---
drivers/net/sfc/sfc_port.c
From: Ivan Ilchenko
Device may support only some MAC stats. Add mapping from ids to subset
of supported MAC stats for each port.
Fixes: 73280c1e4ff ("net/sfc: support xstats retrieval by ID")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Mo
From: Ivan Ilchenko
There is no point to recalculate number of available xstats on
each request. The number is calculated once on device start
and may be returned on subsequent calls.
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/s
From: Ivan Ilchenko
Move getting MAC stats code that involves locking to separate functions
to simplify addition of new xstats.
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc.h| 4 ++
drivers/net/sfc/sfc_ethdev.c | 73 +
From: Ivan Ilchenko
Rx/Tx doorbells statistics are collected in software and
available per queue. These stats are useful for performance
investigation.
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/meson.build| 1 +
drive
From: Ivan Ilchenko
Add 'display-xstats' option for using in accompanying with Rx/Tx statistics
(i.e. 'stats-period' option or 'show port stats' interactive command) to
display specified list of extended statistics.
Signed-off-by: Ivan Ilchenko
Signed-off-by: Andrew Rybchenko
---
app/test-pmd
> -Original Message-
> From: Min Hu (Connor)
> + ret = rte_eth_dev_get_reg_info(i, ®_info);
> + if (ret) {
> + printf("Error getting device reg info: %d\n", ret);
> + continue;
> + }
> +
> + buf_
On Fri, Jun 4, 2021 at 6:25 PM Thomas Monjalon wrote:
>
> 03/06/2021 13:38, Jerin Jacob:
> > On Thu, Jun 3, 2021 at 4:00 PM Thomas Monjalon wrote:
> > > 03/06/2021 12:04, Jerin Jacob:
> > > > On Thu, Jun 3, 2021 at 3:06 PM Thomas Monjalon
> > > > wrote:
> > > > > 03/06/2021 11:20, Jerin Jacob:
On Fri, Jun 4, 2021 at 7:39 PM Thomas Monjalon wrote:
>
> 04/06/2021 15:59, Andrew Rybchenko:
> > On 6/4/21 4:18 PM, Thomas Monjalon wrote:
> > > 04/06/2021 15:05, Andrew Rybchenko:
> > >> On 6/4/21 3:46 PM, Thomas Monjalon wrote:
> > >>> 04/06/2021 13:09, Jerin Jacob:
> > On Fri, Jun 4, 2021
04/06/2021 17:20, Jerin Jacob:
> On Fri, Jun 4, 2021 at 7:39 PM Thomas Monjalon wrote:
> > 04/06/2021 15:59, Andrew Rybchenko:
> > > On 6/4/21 4:18 PM, Thomas Monjalon wrote:
> > > > 04/06/2021 15:05, Andrew Rybchenko:
> > > >> On 6/4/21 3:46 PM, Thomas Monjalon wrote:
> > > >>> 04/06/2021 13:09,
>
> Add the logic to determine how many DD bits have been set for contiguous
> packets, for removing the SMP barrier while reading descs.
Are there any performance numbers with this change?
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
> ---
> drivers/net/i40e/i40e_rxtx.c | 13 ++
> -Original Message-
> From: Thomas Monjalon
> Sent: Friday, June 4, 2021 22:06
> To: Wang, Haiyue
> Cc: dev@dpdk.org; Elena Agostini ;
> andrew.rybche...@oktetlabs.ru; Yigit, Ferruh
> ; jer...@marvell.com
> Subject: Re: [dpdk-dev] [PATCH] gpudev: introduce memory API
>
> 04/06/2021 15:
> -Original Message-
> From: dev On Behalf Of Thomas Monjalon
> Sent: Friday, June 4, 2021 23:51
> To: Jerin Jacob
> Cc: Honnappa Nagarahalli ; Andrew Rybchenko
> ; Yigit, Ferruh ;
> dpdk-dev ;
> Elena Agostini ; David Marchand
>
> Subject: Re: [dpdk-dev] [PATCH] gpudev: introduce memo
On Fri, 4 Jun 2021 04:46:16 -0500
Joyce Kong wrote:
> Since C11 memory model is adopted in DPDK now[1], use GCC's
> atomic builtins in test cases.
>
> [1]https://www.dpdk.org/blog/2021/03/26/dpdk-adopts-the-c11-memory-model/
>
> Joyce Kong (8):
> test/ticketlock: use GCC atomic builtins for
2021-06-04 09:27 (UTC+0200), David Marchand:
> On Fri, Jun 4, 2021 at 12:29 AM Dmitry Kozlyuk
> wrote:
> >
> > 2021-06-03 18:55 (UTC+0200), David Marchand:
> > [...]
> > > diff --git a/config/meson.build b/config/meson.build
> > > index 017bb2efbb..c6985139b4 100644
> > > --- a/config/meson.bui
From: Narcisa Vasile
EAL thread API
**Problem Statement**
DPDK currently uses the pthread interface to create and manage threads.
Windows does not support the POSIX thread programming model, so it currently
relies on a header file that hides the Windows calls under
pthread matched interfaces. Gi
From: Narcisa Vasile
Use a portable, type-safe representation for the thread identifier.
Add functions for comparing thread ids and obtaining the thread id
for the current thread.
---
lib/eal/common/rte_thread.c | 105 ++
lib/eal/include/rte_thread.h
From: Narcisa Vasile
Implement thread attributes for:
* thread affinity
* thread priority
Implement functions for managing thread attributes.
Priority is represented through an enum that allows for two levels:
- RTE_THREAD_PRIORITY_NORMAL
- RTE_THREAD_PRIORITY_REALTIME_CRITICAL
From: Narcisa Vasile
Add function to translate Windows error codes to
errno-style error codes. The possible return values are chosen
so that we have as much semantical compatibility between platforms as
possible.
---
lib/eal/include/rte_thread.h | 5 +-
lib/eal/windows/rte_thread.c | 90 +++
From: Narcisa Vasile
Add function for setting the priority for a thread.
Priorities on multiple platforms are similarly determined by
a priority value and a priority class/policy.
On Linux, the following mapping is created:
RTE_THREAD_PRIORITY_NORMAL corresponds to
* policy SCHED_OTHER
* priorit
From: Narcisa Vasile
Implement functions for getting/setting thread affinity.
Threads can be pinned to specific cores by setting their
affinity attribute.
---
lib/eal/common/rte_thread.c | 14 +++
lib/eal/include/rte_thread.h | 36
lib/eal/windows/eal_lcore.c | 169 ++
From: Narcisa Vasile
Add function for thread creation, join, canceling, detaching.
The *rte_thread_create()* function can optionally receive an rte_thread_attr_t
object that will cause the thread to be created with the affinity and priority
described by the attributes object. If no rte_thread_at
From: Narcisa Vasile
Add functions for mutex init, destroy, lock, unlock.
On Linux, static initialization of a mutex is possible
through PTHREAD_MUTEX_INITIALIZER.
Windows does not have a static initializer.
Initialization is only done through InitializeCriticalSection().
To simulate static in
From: Narcisa Vasile
Add functions for barrier init, destroy, wait.
A portable type is used to represent a barrier identifier.
The rte_thread_barrier_wait() function returns the same value
on all platforms.
---
lib/eal/common/rte_thread.c | 61
lib/eal/incl
From: Narcisa Vasile
Allow the user to choose the thread priority through an EAL
command line argument.
The user can choose thread priority through an EAL parameter,
when starting an application. If EAL parameter is not used,
the per-platform default value for thread priority is used.
Otherwise
From: Narcisa Vasile
EAL thread API
**Problem Statement**
DPDK currently uses the pthread interface to create and manage threads.
Windows does not support the POSIX thread programming model, so it currently
relies on a header file that hides the Windows calls under
pthread matched interfaces. Gi
From: Narcisa Vasile
Use a portable, type-safe representation for the thread identifier.
Add functions for comparing thread ids and obtaining the thread id
for the current thread.
Signed-off-by: Narcisa Vasile
---
lib/eal/common/rte_thread.c | 105 ++
lib/eal/
From: Narcisa Vasile
Implement thread attributes for:
* thread affinity
* thread priority
Implement functions for managing thread attributes.
Priority is represented through an enum that allows for two levels:
- RTE_THREAD_PRIORITY_NORMAL
- RTE_THREAD_PRIORITY_REALTIME_CRITICAL
From: Narcisa Vasile
Add function to translate Windows error codes to
errno-style error codes. The possible return values are chosen
so that we have as much semantical compatibility between platforms as
possible.
Signed-off-by: Narcisa Vasile
---
lib/eal/include/rte_thread.h | 5 +-
lib/eal/w
From: Narcisa Vasile
Implement functions for getting/setting thread affinity.
Threads can be pinned to specific cores by setting their
affinity attribute.
Signed-off-by: Narcisa Vasile
Signed-off-by: Dmitry Malloy
---
lib/eal/common/rte_thread.c | 14 +++
lib/eal/include/rte_thread.h | 3
1 - 100 of 108 matches
Mail list logo
