The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
The current location used for __rte_aligned(a) for alignment of types
and variables is not compatible with MSVC. There is only a single
location accepted by both toolchains.
For variables standard C11 offers alignas(a) supported by conformant
compilers i.e. both MSVC and GCC.
For types the standa
On 2/26/24 18:05, David Marchand wrote:
A simple comment in vhost_user_msg_handler() is not that robust.
Add a lock_all_qps property to message handlers so that their
implementation can add a build check and assert a vq is locked.
Signed-off-by: David Marchand
---
Changes since v2:
- droppe
RTE_LOG_LINE cannot be augmented with a prefix format and arguments
without the user of RTE_LOG_LINE using the args... and ## args compiler
extension to conditionally remove trailing comma when the macro receives
only a single argument.
Provide a new/similar macro RTE_LOG_LINE_PREFIX that accepts
Providing a custom prefix when logging is common for components. Lift
ISO C99 compliant helper macros from mlx5_common.h and provide
RTE_LOG_LINE_PREFIX macro that can expand similar to RTE_LOG_LINE with
a custom prefix and argument list.
Signed-off-by: Tyler Retzlaff
---
lib/log/rte_log.h | 97
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/cfgfile/rte_cfgfile.c | 5 +++--
1 file changed, 3 i
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/bpf/bpf_impl.h | 4 ++--
1 file changed, 2 insertion
Remove use of args... and just use __VA_ARGS__. The macros expanding
the argument pack do not require args extension to remove trailing comma.
Signed-off-by: Tyler Retzlaff
---
lib/cmdline/cmdline_parse.c | 2 +-
lib/cmdline/cmdline_parse_num.c | 4 ++--
2 files changed, 3 insertions(+), 3 d
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/compressdev/rte_compressdev_internal.h | 4 ++--
1 f
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/pdump/rte_pdump.c | 4 ++--
1 file changed, 2 insert
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/metrics/rte_metrics_telemetry.c | 12 ++--
1
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/mldev/rte_mldev.h | 4 ++--
1 file changed, 2 insert
Use RTE_LOG_LINE_PREFIX instead of EAL_LOG in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/eal/common/eal_trace.h | 8
1 file changed, 4 in
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/net/rte_net_crc.c | 4 ++--
1 file changed, 2 insert
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/rcu/rte_rcu_qsbr.c | 4 ++--
lib/rcu/rte_rcu_qsbr.h
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/vhost/vhost.h| 8
lib/vhost/vhost_
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/stack/stack_pvt.h | 16
1 file chan
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/power/power_common.h | 6 +++---
1 file changed, 3 i
Use RTE_LOG_LINE_PREFIX instead of RTE_LOG_LINE in macro expansions
which allow a prefix and arguments to be inserted into the log line
without the need to use the ## args variadic argument pack extension.
Signed-off-by: Tyler Retzlaff
---
lib/rawdev/rte_rawdev_pmd.h | 17 +
1 fi
Remove use of args... and just use __VA_ARGS__. The macros expanding
the argument pack do not require args extension to remove trailing comma.
Signed-off-by: Tyler Retzlaff
---
lib/ip_frag/ip_frag_common.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/ip_frag/ip_fra
The dumpcap and pdump commands need to handle more signals.
Stephen Hemminger (2):
app/dumpcap: handle SIGTERM and SIGHUP
app/pdump: handle SIGTERM and SIGHUP
app/dumpcap/main.c | 14 --
app/pdump/main.c | 21 +++--
2 files changed, 27 insertions(+), 8 deletions
If application is killed (SIGTERM), or the console session
ends (SIGHUP) or the write to the output file fails (SIGPIPE)
then do cleanup before exiting.
This also makes DPDK dumpcap behave more like Wireshark dumpcap.
Suggested-by: Isaac Boukris
Signed-off-by: Stephen Hemminger
---
app/dumpcap
Like dumpcap, the pdump process should cleanup if process
terminates due to being killed or hangup.
Signed-off-by: Stephen Hemminger
---
app/pdump/main.c | 21 +++--
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a/app/pdump/main.c b/app/pdump/main.c
index a9205e130
On Mon, 26 Feb 2024 12:19:30 -0800
Tyler Retzlaff wrote:
> RTE_LOG_LINE cannot be augmented with a prefix format and arguments
> without the user of RTE_LOG_LINE using the args... and ## args compiler
> extension to conditionally remove trailing comma when the macro receives
> only a single argum
On Mon, 26 Feb 2024 20:40:57 +0100
Maxime Coquelin wrote:
> > +
> > +/* vhost_user_msg_handler() locks all qps based on a handler's
> > lock_all_qps.
> > + * Later, a handler may need to ensure the vq has been locked (for example,
> > + * when calling lock annotated helpers).
> > + *
> > + * Not
On Fri, 23 Feb 2024 15:00:56 +0100
Morten Brørup wrote:
> Bugfix: The vlan in the bulletin does not contain a VLAN header, only the
> VLAN ID, so only copy 2 byte, not 4. The target structure has padding
> after the field, so copying 2 byte too many is effectively harmless.
> There is no need to
Recheck-request: iol-broadcom-Performance
This patch should not fail a performance test in CI - checking with a rerun
now.
On Mon, Feb 26, 2024 at 5:25 AM Radu Nicolau wrote:
> There are use cases where a SA should be able to use different cryptodevs
> on
> different lcores, for example there c
Recheck-request: iol-unit-arm64-testing
This ran before we had swapped out the old arm v1.2 ipsec container for our
new arm v1.4 ipsec container. We have now made that change, so this retest
should report a pass. Sorry for the inconvenience.
On Fri, Feb 23, 2024 at 7:21 AM Sivaramakrishnan Venka
Hi Patrick,
> I tried mainline earlier today it compiles and links fine. However, build
> failed on
> v1.4. We are working on tagging the mainline, until then please continue
> working on mainline.
>
A new tag SECLIB-IPSEC-2023.10.13 has been created from the tip of arm ipsec-mb
git repo.
Pleas
On Mon, Feb 26, 2024 at 6:24 PM Wathsala Wathawana Vithanage <
wathsala.vithan...@arm.com> wrote:
> Hi Patrick,
> > I tried mainline earlier today it compiles and links fine. However,
> build failed on
> > v1.4. We are working on tagging the mainline, until then please continue
> > working on main
Hi Radu,
Thanks for making the changes. I've one more question. Please see inline.
Thanks,
Anoob
> -Original Message-
> From: Radu Nicolau
> Sent: Monday, February 26, 2024 3:56 PM
> To: dev@dpdk.org
> Cc: Anoob Joseph ; Radu Nicolau
> ; sta...@dpdk.org; Volodymyr Fialko
> ; Ting-Kai Ku
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Provide
inline functions to access compatible type pointer to rearm_data
and rx_descriptor_fields1 which will allow direct references on the
rte marker fields to be removed.
Signed-off-by: Tyler Retzlaff
---
lib/mbuf/rte_mbuf.h |
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. This series
stops referencing and removes the markers.
This series, does the following.
* RTE_MARKER{,816,32,64} fields are removed from all structures.
* Moves duplicated assertions referencing marker fields to
rte_mbuf_core.h and up
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/common/idpf/idpf_common_rxtx.c| 4 +-
d
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/bnxt/bnxt_rxtx_vec_avx2.c | 32 +++
Collect duplicated RTE_BUILD_BUG_ON checks from drivers and place them
at global scope with struct rte_mbuf definition using static_assert.
Signed-off-by: Tyler Retzlaff
---
lib/mbuf/rte_mbuf_core.h | 34 ++
1 file changed, 34 insertions(+)
diff --git a/lib/mbuf/
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/fm10k/fm10k_rxtx_vec.c | 32 +++-
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Use rte_mbuf_prefetch_part2() to prefetch cacheline1 and remove
reference to rte marker fiel
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/ionic/ionic_lif.c | 6 ++
driver
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/iavf/iavf_rxtx_vec_avx2.c | 72 +++
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/ice/ice_rxtx_vec_avx2.c | 36 +++--
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/cnxk/cn10k_rx.h| 36 ++--
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/enic/enic_main.c | 4 +---
drivers
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/hns3/hns3_rxtx_vec.c | 22 +
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/mlx5/mlx5_rxq.c | 2 +-
driver
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/ixgbe/ixgbe_rxtx_vec_common.h | 4 +--
driv
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rx_descriptor_fields1 accessor that provides a compatible
type pointer without using the marker field.
Signed-off-by: Tyler Retzlaff
---
examples/dma/dmafwd.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
Add a check for RTE_MARKER{,8,16,32,64} forbidding new use.
Signed-off-by: Tyler Retzlaff
---
devtools/checkpatches.sh | 8
1 file changed, 8 insertions(+)
diff --git a/devtools/checkpatches.sh b/devtools/checkpatches.sh
index e379700..ea31ba4 100755
--- a/devtools/checkpatches.sh
+++
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/octeon_ep/cnxk_ep_rx.h | 2 +-
drivers/
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/sfc/sfc_ef100_rx.c | 7 ++-
drivers/net
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Remove
RTE_MARKER fields from rte_mbuf struct.
Maintain alignment of fields after removed cacheline1 marker by placing
C11 alignas(RTE_CACHE_LINE_MIN_SIZE).
Signed-off-by: Tyler Retzlaff
---
doc/guides/rel_notes/release_24_03.rst | 3
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Remove
RTE_MARKER fields from rte_mbuf struct.
Maintain alignment of fields after removed cacheline1 marker by placing
C11 alignas(RTE_CACHE_LINE_MIN_SIZE).
Signed-off-by: Tyler Retzlaff
---
doc/guides/rel_notes/release_24_03.rst | 3
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Remove
RTE_MARKER fields from rte_mbuf struct.
Maintain alignment of fields after removed cacheline1 marker by placing
C11 alignas(RTE_CACHE_LINE_MIN_SIZE).
Update implementation of rte_mbuf_prefetch_part1() and
rte_mbuf_prefetch_part2(
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/virtio/virtio_rxtx_packed.h | 4 ++-
RTE_MARKER typedefs are a GCC extension unsupported by MSVC. Use
new rte_mbuf_rearm_data and rte_mbuf_rx_descriptor_fields1 accessors
that provide a compatible type pointer without using the marker fields.
Signed-off-by: Tyler Retzlaff
---
drivers/net/thunderx/nicvf_ethdev.c | 4 +---
drivers/ne
Disable use_va flag for VDPA backend type and fixes the issues
with shadow control command processing, when it is disabled.
This will help to make virtio user driver works in IOVA
as PA mode for vDPA backend.
Signed-off-by: Srujana Challa
---
drivers/net/virtio/virtio_ring.h | 12 ++
The default value for RTE_PKTMBUF_HEADROOM used to be set in
config/rte_config.h. This patch removes it from the file and
instead, a meson option, pkt_mbuf_headroom, is introduced to make
headroom tunable from the build process.
Signed-off-by: Parthakumar Roy
---
config/meson.build | 1 +
confi
This patchset adds SVE support for the signature comparison in the cuckoo
hash lookup and improves the existing NEON implementation. These
optimizations required changes to the data format and signature of the
relevant functions to support dense hitmasks (no padding) and having the
primary and seco
Current hitmask includes padding due to Intel's SIMD
implementation detail. This patch allows non Intel SIMD
implementations to benefit from a dense hitmask.
Signed-off-by: Yoan Picchi
Reviewed-by: Ruifeng Wang
Reviewed-by: Nathan Brown
---
.mailmap | 2 +
lib/hash/rte_cuck
Upon a successful comparison, NEON sets all the bits in the lane to 1
We can skip shifting by simply masking with specific masks.
Signed-off-by: Yoan Picchi
Reviewed-by: Ruifeng Wang
Reviewed-by: Nathan Brown
---
lib/hash/rte_cuckoo_hash.c | 16 +++-
1 file changed, 7 insertions(+)
This patch adds unit test for rte_hash_lookup_bulk().
It also update the test_full_bucket test to the current number of entries
in a hash bucket.
Signed-off-by: Yoan Picchi
Signed-off-by: Harjot Singh
Reviewed-by: Ruifeng Wang
Reviewed-by: Nathan Brown
---
app/test/test_hash.c | 99 ++
- Implemented SVE code for comparing signatures in bulk lookup.
- Added Defines in code for SVE code support.
- Optimise NEON code
- New SVE code is ~5% slower than optimized NEON for N2 processor.
Signed-off-by: Yoan Picchi
Signed-off-by: Harjot Singh
Reviewed-by: Nathan Brown
Reviewed-by: Rui
This patchset adds SVE support for the signature comparison in the cuckoo
hash lookup and improves the existing NEON implementation. These
optimizations required changes to the data format and signature of the
relevant functions to support dense hitmasks (no padding) and having the
primary and seco
Current hitmask includes padding due to Intel's SIMD
implementation detail. This patch allows non Intel SIMD
implementations to benefit from a dense hitmask.
Signed-off-by: Yoan Picchi
Reviewed-by: Ruifeng Wang
Reviewed-by: Nathan Brown
---
.mailmap | 2 +
lib/hash/rte_cuck
Upon a successful comparison, NEON sets all the bits in the lane to 1
We can skip shifting by simply masking with specific masks.
Signed-off-by: Yoan Picchi
Reviewed-by: Ruifeng Wang
Reviewed-by: Nathan Brown
---
lib/hash/rte_cuckoo_hash.c | 16 +++-
1 file changed, 7 insertions(+)
- Implemented SVE code for comparing signatures in bulk lookup.
- Added Defines in code for SVE code support.
- Optimise NEON code
- New SVE code is ~5% slower than optimized NEON for N2 processor.
Signed-off-by: Yoan Picchi
Signed-off-by: Harjot Singh
Reviewed-by: Nathan Brown
Reviewed-by: Rui
This patch adds unit test for rte_hash_lookup_bulk().
It also update the test_full_bucket test to the current number of entries
in a hash bucket.
Signed-off-by: Yoan Picchi
Signed-off-by: Harjot Singh
Reviewed-by: Ruifeng Wang
Reviewed-by: Nathan Brown
---
app/test/test_hash.c | 99 ++
> Subject: RE: [EXT] Re: [PATCH v2 8/8] crypto/ipsec_mb: set and use session ID
>
> Hi Patrick,
> > I tried mainline earlier today it compiles and links fine. However, build
> > failed on
> > v1.4. We are working on tagging the mainline, until then please continue
> > working on mainline.
> >
> A
https://bugs.dpdk.org/show_bug.cgi?id=1311
jiang,yu (yux.ji...@intel.com) changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
101 - 188 of 188 matches
Mail list logo
