This commit introduces the flexible interface. This
interface allows the core code to operate in PF mode (direct
hardware access) or bifurcated mode (hardware configured via
kernel driver). This driver currently only supports PF modei,
but bifurcated mode will be added in a future patch-set.
Note t
This commit adds probe-time low level hardware
initialization. It also adds probe-time init for both
primary and secondary DPDK processes.
Signed-off-by: Timothy McDaniel
---
drivers/event/dlb/dlb.c | 158 +++-
drivers/event/dlb/meson.build| 3 +-
driv
Load balanced (ldb) queues are setup here.
Directed queues are not set up until link time, at which
point we know the directed port ID. Directed queue setup
will only fail if this queue is already setup or there are
no directed queues left to configure.
Signed-off-by: Timothy McDaniel
Reviewed-by
Add support for getting the queue and port default configuration.
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
drivers/event/dlb/dlb.c | 29 +
1 file changed, 29 insertions(+)
diff --git a/drivers/event/dlb/dlb.c b/drivers/event/dlb/dlb.c
index c038794
Add support for DLB xstats. Perform initialization and add
standard xstats entry points
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
drivers/event/dlb/dlb.c| 23 +
drivers/event/dlb/dlb_xstats.c | 1222
drivers/event/dlb/meson.bu
Configure the load balanded (ldb) or directed (dir) port.
The consumer queue (CQ) and producer port (PP) are also
set up here.
Signed-off-by: Timothy McDaniel
---
doc/guides/eventdevs/dlb.rst | 40 +
drivers/event/dlb/dlb.c | 516 ++-
drivers/event/dlb/dlb
Add supports for the port unlink(s) eventdev entry points.
The unlink operation is an asynchronous operation executed by
a control thread, and the unlinks-in-progress function reads
a counter shared with the control thread. Port QE and memzone
memory is freed here.
Signed-off-by: Timothy McDaniel
Add support for configuring the DLB hardware.
In particular, this patch configures the DLB
hardware's scheduling domain, such that it is provisioned with
the requested number of ports and queues, provided sufficient
resources are available. Individual queues and ports are
configured later in port s
Add port link entry point. Directed queues are identified and created
at this stage. Their setup deferred until link-time, at which
point we know the directed port ID. Directed queue setup
will only fail if this queue is already setup or there are
no directed queues left to configure.
Signed-off-b
Add support for the eventdev start entry point.
DLB delays setting up single link resources until
eventdev start, because it is only then that it can
ascertain which ports have just one linked queue.
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
drivers/event/dlb/dlb.c
Add support for dequeue, dequeue_burst, ...
DLB does not currently support interrupts, but instead uses
umonitor/umwait if supported by the processor. This allows
the software to monitor and wait on writes to a cache-line.
DLB supports normal and sparse cq mode. In normal mode the
hardware will p
These entry points are NO-OPS. DLB does not support
reconfiguring individual queues or ports. The entire device
must be reconfigured.
Signed-off-by: Timothy McDaniel
---
drivers/event/dlb/dlb.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/drivers/event/dlb/dlb
The PMD uses a public interface to allow applications to
control the token pop mode. Supported token pop modes are
as follows, and they impact core scheduling affinity for
ldb ports.
AUTO_POP: Pop the CQ tokens immediately after dequeueing.
DELAYED_POP: Pop CQ tokens after (dequeue_depth - 1) even
Adds the timeout ticks conversion function.
Adds announcement of availabililty of the new driver
for Intel Dynamic Load Balancer 1.0 hardware.
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
doc/guides/rel_notes/release_20_11.rst | 5 +
drivers/event/dlb/dlb.c|
Add a variety of self-tests for both ldb and directed
ports/queues, as well as configure, start, stop, link, etc...
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
app/test/test_eventdev.c |7 +
drivers/event/dlb/dlb.c |1 +
drivers/event/dlb/dlb_selftest.c |
Add support for eventdev stop and close entry points.
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
drivers/event/dlb/dlb.c | 256 +--
drivers/event/dlb/dlb_iface.c| 6 +
drivers/event/dlb/dlb_iface.h| 6 +
dr
Add support for enqueue and its variants.
Signed-off-by: Timothy McDaniel
Reviewed-by: Gage Eads
---
doc/guides/eventdevs/dlb.rst | 161 ++
drivers/event/dlb/dlb.c | 682 +++
2 files changed, 843 insertions(+)
diff --git a/doc/guides/eventde
On 10/22/2020 1:26 AM, Long Li wrote:
From: Long Li
When sending data, netvsc assumes the tx_rndis buffer is contiguous and
calculates physical addresses based on this assumption.
Use memzone to allocate tx_rndis so it's guaranteed that this buffer is
physically contiguous.
Cc:sta...@dpdk.org
This patch series is targeted at the v20.11 release.
It fixes two separate bugs, on the PF with reducing the latency
from 16us to 2us maximum for a single packet, and on the VF with
adding an ITR timeout to both allow descriptor aggregation for
writeback, and minimize latency, setting the value to
The iavf driver was trying to use writeback on ITR, but was
never setting an ITR, so it didn't work. This caused performance
to be limited due to too much PCIe traffic and partial writes
during most benchmarking workloads.
Set the ITR during queue setup, which can be checked at runtime
by reading r
Just like iavf, setting the value to 2us allows for generally good
streaming packet performance while keeping latency down, and
generally keeps the performance of the PF and VF interfaces similar.
The previous value of 0x10 was making latency on a single packet
receive be as much as 16us.
Fixes:
On Fri, Oct 23, 2020 at 3:54 AM Ferruh Yigit wrote:
>
> On 10/22/2020 11:05 PM, Ajit Khaparde wrote:
> > diff --git a/drivers/net/bnxt/tf_core/cfa_resource_types.h
> > b/drivers/net/bnxt/tf_core/cfa_resource_types.h
> > index 19838c393d..412204f7ed 100644
> > --- a/drivers/net/bnxt/tf_core/cfa_re
Hi Yipeng,
I have addressed the comments. Can you please take a look?
Thank you!
> On Oct 21, 2020, at 5:50 PM, Dharmik Thakkar wrote:
>
> Currently, users have to use external RCU mechanisms to free resources
> when using lock free hash algorithm.
>
> Integrate RCU QSBR process to make it ea
From: Stephen Hemminger
The values for Rx and Tx copy break should be tunable rather
than hard coded constants.
The rx_copybreak sets the threshold where the driver uses an
external mbuf to avoid having to copy data. Setting 0 for copybreak
will cause driver to always create an external mbuf. Se
From: Long Li
When receiving packets, netvsp puts data in a buffer mapped through UIO.
Depending on packet size, netvsc may attach the buffer as an external
mbuf. This is not a problem if this mbuf is consumed in the application,
and the application can correctly read data out of an external mbuf
> -Original Message-
> From: Dharmik Thakkar
> Sent: Wednesday, October 21, 2020 3:50 PM
> To: Wang, Yipeng1 ; Gobriel, Sameh
> ; Richardson, Bruce ;
> Ray Kinsella ; Neil Horman
> Cc: dev@dpdk.org; n...@arm.com; Dharmik Thakkar
>
> Subject: [PATCH v6 2/4] lib/hash: integrate RCU QSBR
>
> -Original Message-
> From: Dharmik Thakkar
> Sent: Wednesday, October 21, 2020 3:50 PM
> To: Wang, Yipeng1 ; Gobriel, Sameh
> ; Richardson, Bruce
> Cc: dev@dpdk.org; n...@arm.com; Dharmik Thakkar
>
> Subject: [PATCH v6 4/4] test/hash: add tests for integrated RCU QSBR
>
> Add function
> >
> > Add functional tests for zero copy APIs. Test enqueue/dequeue
> > functions are created using the zero copy APIs to fit into the
> > existing testing method.
> >
> > Signed-off-by: Honnappa Nagarahalli
> > Reviewed-by: Dharmik Thakkar
> > ---
> > app/test/test_ring.c | 196
> >
Add a new CPUID flag indicating processor support for UMONITOR/UMWAIT
and TPAUSE instructions instruction.
Signed-off-by: Liang Ma
Signed-off-by: Anatoly Burakov
Acked-by: Konstantin Ananyev
---
lib/librte_eal/x86/include/rte_cpuflags.h | 1 +
lib/librte_eal/x86/rte_cpuflags.c | 2 ++
Add two new power management intrinsics, and provide an implementation
in eal/x86 based on UMONITOR/UMWAIT instructions. The instructions
are implemented as raw byte opcodes because there is not yet widespread
compiler support for these instructions.
The power management instructions provide an ar
This patchset proposes a simple API for Ethernet drivers
to cause the CPU to enter a power-optimized state while
waiting for packets to arrive, along with a set of
generic intrinsics that facilitate that. This is achieved
through cooperation with the NIC driver that will allow
us to know address of
Add a simple API to allow getting address of next RX descriptor from the
PMD, as well as release notes information.
Signed-off-by: Liang Ma
Signed-off-by: Anatoly Burakov
Acked-by: Konstantin Ananyev
---
Notes:
v8:
- Rename version map file name.
v7:
- Fixed queue ID validatio
Add a simple on/off switch that will enable saving power when no
packets are arriving. It is based on counting the number of empty
polls and, when the number reaches a certain threshold, entering an
architecture-defined optimized power state that will either wait
until a TSC timestamp expires, or w
Currently, it is not possible to check support for intrinsics that
are platform-specific, cannot be abstracted in a generic way, or do not
have support on all architectures. The CPUID flags can be used to some
extent, but they are only defined for their platform, while intrinsics
will be available
Implement support for the power management API by implementing a
`get_wake_addr` function that will return an address of an RX ring's
status bit.
Signed-off-by: Liang Ma
Signed-off-by: Anatoly Burakov
Acked-by: Konstantin Ananyev
Acked-by: Jeff Guo
---
drivers/net/i40e/i40e_ethdev.c | 1 +
d
Implement support for the power management API by implementing a
`get_wake_addr` function that will return an address of an RX ring's
status bit.
Signed-off-by: Anatoly Burakov
Signed-off-by: Liang Ma
Acked-by: Konstantin Ananyev
---
drivers/net/ixgbe/ixgbe_ethdev.c | 1 +
drivers/net/ixgbe/i
Update programmer's guide to document PMD power management usage.
Signed-off-by: Liang Ma
Signed-off-by: Anatoly Burakov
Acked-by: David Hunt
---
doc/guides/prog_guide/power_man.rst | 42 +
1 file changed, 42 insertions(+)
diff --git a/doc/guides/prog_guide/power_m
Implement support for the power management API by implementing a
`get_wake_addr` function that will return an address of an RX ring's
status bit.
Signed-off-by: Liang Ma
Signed-off-by: Anatoly Burakov
Acked-by: Konstantin Ananyev
---
drivers/net/ice/ice_ethdev.c | 1 +
drivers/net/ice/ice_rxt
Add PMD power management feature support to l3fwd-power sample app.
Signed-off-by: Liang Ma
Signed-off-by: Anatoly Burakov
Acked-by: David Hunt
---
Notes:
v8:
- Add return status check for queue enable
v6:
- Fixed typos in documentation
---
.../sample_app_ug/l3_forward_power_
Adding explicit different ut when testing for validation
or latency (early termination enabled or not).
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 92 ++--
1 file changed, 88 insertions
v4: rebased on main latest
v3: apologize again for typo and not double checking with check-git-log
v2: typos missed in commit messages
Serie updating and extending the app running the bbdev-test for the existing
bbdev PMDs.
Nicolas Chautru (7):
app/bbdev: add explicit ut for latency vs validati
Adding explict check in ut that the stats counters
have the expect values. Was missing for coverage.
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 17 +
1 file changed, 17 insertions(+)
diff --git a/app/test
Replacing magic number for default wait time for hw
offload.
Signed-off-by: Nicolas Chautru
Acked-by: Liu Tianjiao
---
app/test-bbdev/test_bbdev_perf.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/app/test-bbdev/test_bbdev_perf.c b/app/test-bbdev/test_bbdev_perf.
Reducing number of repetitions from 1000 to 100
to save time. Results are accurate enough with
100 loops.
Signed-off-by: Nicolas Chautru
Acked-by: Liu Tianjiao
---
app/test-bbdev/test_bbdev_perf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/test-bbdev/test_bbdev_perf
Run preloading explictly for unit tests. Load each code block
by reusing existing input op then restore for the actual test.
Signed-off-by: Nicolas Chautru
Acked-by: Liu Tianjiao
---
app/test-bbdev/main.h| 1 +
app/test-bbdev/test_bbdev_perf.c | 51 +
update offload dequeue to retrieve the full ring to be
agnostic of implementation.
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/app/test-bbd
bler test results are not valid when LLR compression
is used or for loopback scenarios. Skipping these.
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/app
Adding explicit different ut when testing for validation
or latency (early termination enabled or not).
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 92 ++--
1 file changed, 88 insertions
v5: correcting typos again. Quite worrisome spelling disorder.
v4: rebased on main latest
v3: apologize again for typo and not double checking with check-git-log
v2: typos missed in commit messages
Serie updating and extending the app running the bbdev-test for the existing
bbdev PMDs.
Nicolas C
Run preloading explicitly for unit tests. Load each code block
by reusing existing input op then restore for the actual test.
Signed-off-by: Nicolas Chautru
Acked-by: Liu Tianjiao
---
app/test-bbdev/main.h| 1 +
app/test-bbdev/test_bbdev_perf.c | 51 +---
Adding explicit check in ut that the stats counters
have the expect values. Was missing for coverage.
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 17 +
1 file changed, 17 insertions(+)
diff --git a/app/tes
bler test results are not valid when LLR compression
is used or for loopback scenarios. Skipping these.
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/app
Replacing magic number for default wait time for hw
offload.
Signed-off-by: Nicolas Chautru
Acked-by: Liu Tianjiao
---
app/test-bbdev/test_bbdev_perf.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/app/test-bbdev/test_bbdev_perf.c b/app/test-bbdev/test_bbdev_perf.
update offload dequeue to retrieve the full ring to be
agnostic of implementation.
Signed-off-by: Nicolas Chautru
Acked-by: Aidan Goddard
Acked-by: Dave Burley
---
app/test-bbdev/test_bbdev_perf.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/app/test-bbd
Reducing number of repetitions from 1000 to 100
to save time. Results are accurate enough with
100 loops.
Signed-off-by: Nicolas Chautru
Acked-by: Liu Tianjiao
---
app/test-bbdev/test_bbdev_perf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/test-bbdev/test_bbdev_perf
> Subject: Re: [dpdk-dev] [PATCH 1/2] net/netvsc: allow setting rx and tx copy
> break
>
> On 10/22/2020 8:46 PM, Long Li wrote:
> > From: Stephen Hemminger
> >
> > The values for Rx and Tx copy break should be tunable rather than hard
> > coded constants.
> >
> > The rx_copybreak sets the thresh
>
> > Move test_ring_inc_ptr to header file so that it can be used by
> > functions in other files.
> >
> > Signed-off-by: Honnappa Nagarahalli
> > Reviewed-by: Dharmik Thakkar
> > ---
> > app/test/test_ring.c | 11 --- app/test/test_ring.h | 11
> > +++
> > 2 files changed, 1
On Fri, 23 Oct 2020 23:54:22 +
Honnappa Nagarahalli wrote:
>
>
> >
> > > Move test_ring_inc_ptr to header file so that it can be used by
> > > functions in other files.
> > >
> > > Signed-off-by: Honnappa Nagarahalli
> > > Reviewed-by: Dharmik Thakkar
> > > ---
> > > app/test/test_rin
> > static inline void*
> > test_ring_inc_ptr(void *obj, int esize, unsigned int n) {
> > int sz;
> >
> > sz = esize;
> > /* Legacy queue APIs? */
> > if ((esize) == -1)
>
> Extra (paren) doesn't help readability either
+1
The dynamic flag management is broken if rte_mbuf_dynflag_lookup()
is done in a secondary process because the local pointer to
the memzone is not ever initialized.
Fix it by using the same checks as dynfield_register().
I.e if shared memory zone has not been looked up already,
then discover it.
F
Hi,
I am testing a pipeline application with two or more cores using dpdk 19.05.
The application consists of:
Core1: forever get packets from an Ethernet IF (using rte_eth_rx_burst)
Inspect packet header such as EtherType, UDP desp_port, etc;
determine application (say ap
This is a revised version of the earlier RFC patch set for changing
the blacklist/whitelist terms in DPDK.
The first patch is a duplicate from the other patch set about
use of master/slave in API.
Note:
This may cause some warnings in existing programs in the CI
tests using the -w flag.
It also
Use the new terminology blocked to describe when devices
are excluded from being used.
Signed-off-by: Stephen Hemminger
Acked-by: Luca Boccassi
Acked-by: Hemant Agrawal
---
drivers/bus/dpaa/dpaa_bus.c| 7 +++
drivers/bus/fslmc/fslmc_bus.c | 9 -
drivers/bus/fslmc/fsl
This patch renames the enum values in the EAL include files.
As a backward compatible temporary migration tool, define
a replacement mapping for old values.
The old names relating to blacklist and whitelist are replaced
by block list and allow list, but applications may be using the
older compatib
Test the renamed blocklist and allowlist arguments.
Use new terms in test variable names as well.
Signed-off-by: Stephen Hemminger
Acked-by: Luca Boccassi
---
app/test/autotest.py| 16 ++--
app/test/autotest_runner.py | 18 ++---
app/test/test.c | 2 +-
app/
Replace -w / --pci-whitelist with -a / --allow options
and --pci-blacklist with --block.
The -b short option remains unchanged.
Allow the old options for now, but print a nag
warning since old options are deprecated.
Signed-off-by: Stephen Hemminger
Acked-by: Luca Boccassi
---
lib/librte_eal/c
There are two areas where documentation needed update.
The first was use of whitelist when describing address
filtering.
The other is the legacy -w whitelist option for PCI
which is used in many examples
Signed-off-by: Stephen Hemminger
Acked-by: Luca Boccassi
---
doc/guides/cryptodevs/dpaa2_s
> -Original Message-
> From: Xing, Beilei
> Sent: Friday, October 23, 2020 4:15 PM
> To: dev@dpdk.org
> Cc: Guo, Jia ; Xing, Beilei ;
> sta...@dpdk.org
> Subject: [PATCH] net/i40e: fix FDIR issue for ETH + VLAN pattern
>
> From: Beilei Xing
>
> Currently, can't create more than one fo
This patch set contains multiple-thread flow operations support
for the flow objects.
The new added PMD features after RC1 MT support is not done yet.
The sh lock removing patch will be moved to the next series.
Suanming Mou (11):
net/mlx5: use thread safe index pool for flow objects
net/mlx5
As mlx5 PMD is changed to be thread safe, all the flow-related
sub-objects inside the PMD should be thread safe. This commit
changes the index memory pools' lock configuration to be enabled.
That makes the index pool be thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/
From: Xueming Li
As part of multi-thread flow support, this patch moves flow intermediate
data to thread specific, makes them a flow workspace. The workspace is
allocated per thread, destroyed along with thread life-cycle.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/l
From: Xueming Li
The flow ID generation API used an integer pool to save released ID, The
only usage is to generate RSS flow ID. To support multiple flow, it has
to be enhanced to be thread safe.
Indexed pool could be used to generate unique ID by setting size of pool
entry to zero. Since bitmap
From: Xueming Li
Hairpin flow matching required a unique flow ID for matching.
This patch reuses flow ID as hairpin flow ID, this will save some code
to generate a separate hairpin ID, also saves flow memory by removing
hairpin ID.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/n
From: Xueming Li
To make indexed pool to be used as ID generator, this patch allows entry
size to be zero.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5_utils.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_utils.c b/
This commit adds the spinlock for the meter action to make it be thread
safe. Atomic reference counter in all is not enough as the meter action
should be created synchronized with reference counter increment. With
only atomic reference counter, even the counter is increased, the action
may still no
This commit protects the VLAN VM workaround area using a spinlock
in multiple-thread flow insertion to make it thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/common/mlx5/linux/mlx5_nl.h | 1 +
drivers/net/mlx5/linux/mlx5_vlan_os.c | 5 +
2 files changed, 6 inse
From: Xueming Li
To support multi-thread flow operations, this patch introduces list lock
for the rte_flow list manages all the rte_flow handlers.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 1 +
drivers/net/mlx5/mlx5.h | 1 +
drivers/net/
This commit creates the global default miss action instead of maintain
it in flow insertion time. This makes the action to be thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7
drivers/net/mlx5/mlx5.h | 9 +---
drivers/net/ml
This commit changes the jump action in table to be created with table
creation in advanced. In this case, the jump action is safe to be used
in multiple thread. The jump action will be destroyed when table is not
used anymore and released.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
d
From: Xueming Li
In order to support hash list concurrent access, adding next:
1. List level read/write lock.
2. Entry reference counter.
3. Entry create/match/remove callback.
4. Remove insert/lookup/remove function which are not thread safe.
5. Add register/unregister function to support entry
This commit creates the global drop action for flows instead of
maintain it in flow insertion time. The uniqueu global drop action
makes it thread safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 +
drivers/net/mlx5/mlx5.c| 2 +
From: Xueming Li
To support multi-thread flow insertion/removal, this patch uses thread
safe hash list API for flow table cache hash list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5.c | 102 -
drivers/net/mlx5/mlx5.h | 2
From: Xueming Li
To support multi-thread flow insertion, this patch updates flow modify
action list to use thread safe hash list with write-most mode.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 +-
drivers/net/mlx5/mlx5_flow.h | 14 ++-
dri
All table, tag, header modify, header reformat are supported only on DV
mode. For the OFED version doesn't support these, create the related
redundant DV resources waste the memory.
Add the code section in the HAVE_IBV_FLOW_DV_SUPPORT macro to avoid the
redundant resources allocation.
Fixes: 2eb4
From: Xueming Li
To support multi-thread flow insertion, this patch updates flow tag list
to use thread safe hash list with write-most mode.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 ++-
drivers/net/mlx5/mlx5_flow.h | 5 +++
drivers/net/ml
From: Xueming Li
To support multi-thread flow insertion, this patch updates metadata copy
flow list to use thread safe hash list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 +-
drivers/net/mlx5/mlx5_flow.c | 162 ++---
After non-cache mode feature was implemented, the flows can only be
created when port started. No need to check if the mreg flows are
created in port stopped status, and apply the mreg flows after port
start will also never happen.
This commit removed the relevant not used mreg copy code.
Signed-
In previous commits the hash list objects have been converted
to new thread safe hash list. The legacy hash list code can be
removed now.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5_utils.c | 38 -
drivers/net/mlx5/mlx5_utils.h | 66 --
From: Xueming Li
New API of linked list for cache:
- Optimized for small amount cache list.
- Optimized for read-most list.
- Thread safe.
- Since number of entries are limited, entries allocated by API.
- For dynamic entry size, pass 0 as entry size, then the creation
callback allocate the entry
To support multi-thread flow insertion, this patch updates flow header
reformat action list to use thread safe hash list with write-most mode.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 +-
drivers/net/mlx5/mlx5_flow.h | 7 ++
drivers/net/
From: Xueming Li
To support multi-thread flow insertion, this path converts matcher list
to use thread safe cache list API.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/mlx5.h | 3 +
drivers/net/mlx5/mlx5_flow.h| 15 ++-
drivers/net/mlx5/mlx5_flow_dv.c |
This commit applies the cache linked list to Rx queue to make it thread
safe.
Signed-off-by: Suanming Mou
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 5 +
drivers/net/mlx5/mlx5.c| 1 +
drivers/net/mlx5/mlx5.h| 24 +++-
drivers/net/mlx5/mlx5_flow.h
From: Xueming Li
To support multi-thread flow insertion, this patch convert port id
action cache list to thread safe cache list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 ++
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_flow
From: Xueming Li
To support multi-thread flow insertion, this patch converts push VLAN
action cache list to thread safe cache list.
Signed-off-by: Xueming Li
Acked-by: Matan Azrad
---
drivers/net/mlx5/linux/mlx5_os.c | 7 ++
drivers/net/mlx5/mlx5.h | 2 +-
drivers/net/mlx5/mlx5_f
> Subject: [PATCH v2 2/2] drivers: add headers install helper
>
> A lot of drivers export headers, reproduce the same facility than for
> libraries.
>
> Suggested-by: Bruce Richardson
> Signed-off-by: David Marchand
> ---
> doc/guides/contributing/coding_style.rst | 3 +++
> drivers/baseb
On Fri, Oct 23, 2020 at 12:05 AM Thomas Monjalon wrote:
>
> Below patches are cleaning traces of CONFIG_RTE_ after make removal,
> except one occurence in app/test/test_cryptodev.c (left as exercise).
> PS: In reality I don't know what must be done for this case about QAT.
It seems a reintroducti
RTE_LIBRTE_SECURITY_TEST never existed, the variable under this check is
never used.
Fixes: e117c18a1dbe ("crypto/dpaa2_sec: restructure session management")
Cc: sta...@dpdk.org
Signed-off-by: David Marchand
---
drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 6 --
1 file changed, 6 deletions
When merging this series after Bruce changes on build macros, an old macro
usage has been re-introduced.
Fixes: d82d6ac64338 ("app/procinfo: add crypto security context info")
Signed-off-by: David Marchand
---
app/proc-info/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --gi
Following recent increase of an internal array that was limiting CTF event
descriptions, I had a second look at the code.
All of this is slow path, so I see no reason in keeping this limitation
and we can go with dynamic allocations.
While at it, I tweaked the metadata file output.
I consider thi
CTF event description is currently built by appending all fields in a
single string at trace point registration.
When dumping the metadata, this string is split again and inspected to
fixup reserved keywords and special tokens like "." or "->".
Move this fixup per field at trace point registration
1 - 100 of 245 matches
Mail list logo
