On 04/14, Gavin Hu wrote:
>V3:
>- fix the 'Fixes' line warning.
>V2:
>- drop the 3/3 auto-vectorization patch from this series, it requires
> more rework and will submit in a separate patch.
>
>This series is to optimize the i40e NEON vPMD performance on aarch64.
>
>The patches were benchmarked b
W dniu 15.04.2020 o 08:42, Sarosh Arif pisze:
> v2:
> remove double freeing of mbufs
>
> Signed-off-by: Sarosh Arif
> ---
> app/test/test_distributor.c | 9 +++
> 1 file changed, 7 insertions(+)
>
> diff --git a/app/test/test_distributor.c b/app/test/test_distributor.c
> index ba1f81cf8..
v2:
remove double freeing of mbufs
Signed-off-by: Sarosh Arif
---
app/test/test_distributor.c | 9 +++
1 file changed, 7 insertions(+)
diff --git a/app/test/test_distributor.c b/app/test/test_distributor.c
index ba1f81cf8..5e972bb2e 100644
--- a/app/test/test_distributor.c
+++ b/app/test/te
Allocate metadata object from indexed pool helps rte flow saves the 4
bytes index instead of 8 bytes pointer. For metadata object itself, it
helps save MALLOC_ELEM_OVERHEAD bytes from rte_malloc().
Signed-off-by: Suanming Mou
---
drivers/net/mlx5/mlx5.c | 11 +++
drivers/net/mlx5/ml
Currently, the rte flow structure is not fully aligned and has some
bits wasted. The members can be optimized and reorganized to save
memory.
1. The drv_type uses only limited bits, change the type to 2 bits what
it needs.
2. Align the hairpin_flow_id, drv_type, fdir, copy_applied to 32 bits.
As
This patch allocate the meter object memory from indexed memory pool
which will help to save the MALLOC_ELEM_OVERHEAD memory taken by
rte_malloc().
Signed-off-by: Suanming Mou
---
drivers/net/mlx5/mlx5.c| 11 +++
drivers/net/mlx5/mlx5.h| 1 +
drivers/net/mlx5/mlx
From: Wentao Cui
This commit is for mlx5 fdir flow memory optimization.
Currently for the fdir member in rte_flow structure. It saves the fidr
memory pointer directly. As fidr is fading away, use one bit help to
indicate the function in the flow and add the content to an extra list
save the memo
As only limited bits is used in act_flags for flow destroy, it's a bit
expensive to save the whole 64 bits. Move the act_flags out of flow
handle and save the needed bits for flow destroy to save some bytes for
the flow handle data struct.
The fate action type and mark bits are reserved as they wi
When destroy the flow with RSS, only RSS queue is needed. Currently,
even the RSS key, types, level, are all saved to the rte flow.
Saves only the mlx5 flow RSS queue information to save memory for
rte flow.
As the maximum queue number is UINT16_MAX, the queue number type is
changed to uint16_t.
Currently, the mlx5_flow_handle struct is not fully aligned and has some
bits wasted. The members can be optimized and reorganized to save memory.
1. As metadata and meter is sharing the same flow match id, now the flow
id is limited to 24 bits due to the 8 MSBs are used as for the meter color.
Al
While flow attaches the meter handle, the meter id can be the unique tag
for the flow to get the meter handle. It's no need for flow to save the
pointer of the meter handle.
Save the meter id instead of pointer helps reduce the size for rte flow
structure.
As the supported maximum meter rule is 4
Currently, one flow only has one fate action, the fate actions members
in the flow struct can be reorganized as union to save the memory for
flow struct.
This commit reorganizes the fate actions as union, the act_flags helps
to identify the fate action type when flow destroys.
Signed-off-by: Suan
Currently, rte flow with RSS action may have different queue number. The
indexed memory pool is not suitable for the flow with RSS action.
For flows without RSS action, the size are fixed. Allocate the none RSS
rte flow memory from indexed memory pool helps save MALLOC_ELEM_OVERHEAD
which is more
This patch set is third part of the flow memory consumption optimization.
It mainly optimizes the rte flow and mlx5 flow handle structure.
For mlx5 flow handle structure:
1. As one flow has only one fate action, the fate actions are reorganized
as union.
2. The action flags are optimized to be the
W dniu 10.04.2020 o 19:27, Bruce Richardson pisze:
> The zlib compression driver, as well as the aesni-gcm, aesni-mb and openssl
> crypto drivers all defined the logtype variable in the header file
> directly. This gives errors with gcc 10, due to -fno-common being the
> default, so we need to ap
The four crypto drivers: kasumi, mvsam, snow3g and zuc define logtype
variables in their header file. As the header files are included
in more than one compilation unit, it might cause appearance
of multiple instances of the variable and a linker error.
Such situation can occur, when no common sect
Hi,
> -Original Message-
> From: Chautru, Nicolas
> Sent: Tuesday, April 14, 2020 8:16
> To: Xu, Rosen ; dev@dpdk.org; akhil.go...@nxp.com
> Cc: Richardson, Bruce
> Subject: RE: [dpdk-dev] [PATCH v2 06/13] baseband/fpga_5gnr_fec: add
> queue configuration
>
> Thanks Rosen for your thoro
On 2020-04-15 01:42, Tonghao Zhang wrote:
> On Tue, Apr 14, 2020 at 11:37 PM Mattias Rönnblom
> wrote:
>> On 2020-04-14 15:35, David Marchand wrote:
>>> On Tue, Apr 14, 2020 at 3:20 PM Mattias Rönnblom
>>> wrote:
On 2020-04-14 06:43, Tonghao Zhang wrote:
> On Tue, Apr 14, 2020 at 12:07 P
From: Zhu Tao
When the thread exits normally, pthread_join() is not called, which can
result in a resource leak. Therefore, the thread is set to separation
mode using function pthread_detach(), so that no program call
pthread_join() is required to recycle, and when the thread exits,
the system au
> -Original Message-
> From: Di, ChenxuX
> Sent: Tuesday, April 14, 2020 2:37 PM
> To: dev@dpdk.org
> Cc: Xing, Beilei ; Di, ChenxuX
> Subject: [PATCH v8] net/i40e: enable advanced RSS
>
> This patch supports:
>
> - symmetric hash configuration
> - Input set configuration
>
> Signed
> >> -Original Message-
> >> From: dev On Behalf Of Pablo de Lara
> >> Sent: Tuesday, April 14, 2020 6:30 PM
> >> To: Doherty, Declan
> >> Cc: dev@dpdk.org; akhil.go...@nxp.com; De Lara Guarch, Pablo
> >> ; sta...@dpdk.org
> >> Subject: [dpdk-dev] [PATCH] cryptodev: add missing feature na
From: Zhu Tao
When the thread exits normally, pthread_join() is not called, which can
result in a resource leak. Therefore, the thread is set to separation
mode using function pthread_detach(), so that no program call
pthread_join() is required to recycle, and when the thread exits,
the system au
> -Original Message-
> From: Su, Simei
> Sent: Wednesday, April 15, 2020 10:55 AM
> To: Zhang, Qi Z ; Ye, Xiaolong
> ; Wu, Jingjing
> Cc: dev@dpdk.org; Cao, Yahui ; Su, Simei
>
> Subject: [PATCH v4 0/5] net/iavf: support FDIR capabiltiy
>
> [PATCH v4 1/5] support FDIR common patterns
Ori and qi
On 4/15/2020 10:31 AM, Zhang, Qi Z wrote:
-Original Message-
From: Ori Kam
Sent: Tuesday, April 14, 2020 5:43 PM
To: Guo, Jia ; Iremonger, Bernard
; Ye, Xiaolong ;
Zhang, Qi Z
Cc: dev@dpdk.org; Wu, Jingjing ; Cao, Yahui
; Su, Simei
Subject: RE: [dpdk-dev v4 1/3] ethdev:
This patch enables GTPU with TEID and QFI for flow director filter.
Signed-off-by: Simei Su
---
drivers/net/iavf/iavf_fdir.c | 63
1 file changed, 63 insertions(+)
diff --git a/drivers/net/iavf/iavf_fdir.c b/drivers/net/iavf/iavf_fdir.c
index 9b03d29
This patch enables mark action support and takes mark only case
into consideration.
Signed-off-by: Simei Su
---
drivers/net/iavf/iavf.h | 1 +
drivers/net/iavf/iavf_fdir.c | 45 +++-
2 files changed, 45 insertions(+), 1 deletion(-)
diff --git a/driv
This patch enables PFCP node and sesssion packets with S_FIELD
for flow director filter.
Signed-off-by: Simei Su
---
drivers/net/iavf/iavf_fdir.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/drivers/net/iavf/iavf_fdir.c b/drivers/net/iavf/iavf_fdir.c
index 1e5
This patch enables L2TPv3 with SESSION_ID, ESP/AH with SPI, NAT-T
with SPI and IP src/dst for flow director filter.
Signed-off-by: Simei Su
---
drivers/net/iavf/iavf_fdir.c | 91
1 file changed, 91 insertions(+)
diff --git a/drivers/net/iavf/iavf_fdi
[PATCH v4 1/5] support FDIR common patterns and actions.
[PATCH v4 2/5] support FDIR GTPU pattern.
[PATCH v4 3/5] support FDIR L2TPv3, ESP, AH and NAT-T pattern.
[PATCH v4 4/5] support FDIR PFCP node and session pattern.
[PATCH v4 5/5] support FDIR mark action.
This patchset depend on the followin
This patch adds FDIR create/destroy/validate function in AVF.
Common pattern and queue/qgroup/passthru/drop actions are supported.
Signed-off-by: Simei Su
---
doc/guides/rel_notes/release_20_05.rst | 1 +
drivers/net/iavf/Makefile | 1 +
drivers/net/iavf/iavf.h|
> -Original Message-
> From: Ori Kam
> Sent: Tuesday, April 14, 2020 5:43 PM
> To: Guo, Jia ; Iremonger, Bernard
> ; Ye, Xiaolong ;
> Zhang, Qi Z
> Cc: dev@dpdk.org; Wu, Jingjing ; Cao, Yahui
> ; Su, Simei
> Subject: RE: [dpdk-dev v4 1/3] ethdev: add new RSS offload types
>
> Hi Jeff
> -Original Message-
> From: Guo, Jia
> Sent: Wednesday, April 15, 2020 1:42 AM
> To: or...@mellanox.com; Iremonger, Bernard
> ; Ye, Xiaolong ;
> Zhang, Qi Z
> Cc: dev@dpdk.org; Wu, Jingjing ; Cao, Yahui
> ; Su, Simei ; Guo, Jia
>
> Subject: [dpdk-dev v4 2/3] net/iavf: add RSS configu
> -Original Message-
> From: Iremonger, Bernard
> Sent: Tuesday, March 31, 2020 9:57 PM
> To: dev@dpdk.org; Xing, Beilei ; Zhang, Qi Z
> ; Doherty, Declan
> Cc: Ananyev, Konstantin ; Iremonger,
> Bernard
> Subject: [PATCH v3 1/3] librte_ethdev: add RSS offload types for ESP and AH
>
Hi, Sarosh
I believe commit message title should begin with app/test not the file name
Other comments inline
W dniu 13.04.2020 o 11:19, Sarosh Arif pisze:
> rte_mempool_get_bulk is used to get bufs/many_bufs from the pool,
> but at some locations when test passes/fails the bufs/many_bufs are
> n
1400 series adapters support multiple MARK and FLAG action types.
e.g.: mark id 10 / queue index 2 / mark id 11 / queue index 3
Remove the restriction in the Flow Manager implementation.
Signed-off-by: John Daley
Reviewed-by: Hyong Youb Kim
---
drivers/net/enic/enic_fm_flow.c | 9 +
Support rte_flow RSS action on outer headers (level 0). RSS ranges on
the non-default port is OK.
Restrictions:
- The RETA is ignored. The hash function is simply applied across
the RSS queue range.
- The queues used in the RSS group must be sequential.
- There is a performance hit if the num
Update the VIC Flow Manager API. The extentions will allow support for:
- Decap and strip VLAN
- Remove outer VLAN
- Set Egress port
- Set VLAN when replicating encapped packets
- RSS queue ranges on outer header
Signed-off-by: John Daley
Reviewed-by: Hyong Youb Kim
---
drivers/net/en
Each RTE RQ is represented on enic as a Start Of Packet (SOP) queue and
and overflow queue (DATA). There were arranged SOP0/DATA0, SOP1/DATA1,...
but need to be arranged SOP0, SOP1,..., DATA0, DATA1... so that
rte_flow RSS queue ranges work.
Signed-off-by: John Daley
Reviewed-by: Hyong Youb Kim
Tested-by: Wang, Yinan
> -Original Message-
> From: dev On Behalf Of Marvin Liu
> Sent: 2020年4月14日 20:56
> To: maxime.coque...@redhat.com; Wang, Zhihong ;
> Ye, Xiaolong
> Cc: dev@dpdk.org; Ding, Xuan ; Liu, Yong
>
> Subject: [dpdk-dev] [PATCH] net/virtio: fix crash when device reconne
From: Hyong Youb Kim
The current implementation produces wrong ordering for several cases
like these:
1. mark, decap, steer
Current: steer, mark, decap
Correct: mark, steer, decap
2. decap, steer, steer
Current: steer, steer, decap
Correct: steer, decap, steer
Simplify the logic and swap 1st s
On Tue, Apr 14, 2020 at 11:37 PM Mattias Rönnblom
wrote:
>
> On 2020-04-14 15:35, David Marchand wrote:
> > On Tue, Apr 14, 2020 at 3:20 PM Mattias Rönnblom
> > wrote:
> >> On 2020-04-14 06:43, Tonghao Zhang wrote:
> >>> On Tue, Apr 14, 2020 at 12:07 PM Stephen Hemminger
> >>> wrote:
> On S
On 4/14/2020 12:44 PM, Dmitry Kozlyuk wrote:
Note: no changes in cover letter since v2.
This patchset implements basic MM with the following features:
* Hugepages are dynamically allocated in user-mode.
* Only 2MB hugepages are supported.
* IOVA is always PA, obtained through kernel-mode dri
On 4/14/2020 12:44 PM, Dmitry Kozlyuk wrote:
This driver supports Windows EAL memory management by translating
current process virtual addresses to physical addresses (IOVA).
Standalone virt2phys allows using DPDK without PMD and provides a
reference implementation.
Suggested-by: Ranjit Menon
S
In case of i40evf pmd, when ADD_VLAN is sent down the linux i40e driver,
along with add the vlan the kernel driver also enables the vlan stripping
by default.
This might have issues if the app configured DEV_RX_OFFLOAD_VLAN_STRIP
as off at the port configuration. The app after adding the VLAN will
> > >
> > > > +
> > > > +/**
> > > > + * Start to enqueue several objects on the ring.
> > > > + * Note that no actual objects are put in the queue by this
> > > > +function,
> > > > + * it just reserves for user such ability.
> > > > + * User has to call appropriate enqueue_finish() to copy object
In case of i40evf pmd, when ADD_VLAN is sent down the linux i40e driver,
along with add the vlan the kernel driver also enables the vlan stripping
by default.
This might have issues if the app configured DEV_RX_OFFLOAD_VLAN_STRIP
as off at the port configuration. The app after adding the VLAN will
From: "Dey, Souvik"
In case of i40evf pmd, when ADD_VLAN is sent down the linux i40e driver,
along with add the vlan the kernel driver also enables the vlan stripping
by default.
This might have issues if the app configured DEV_RX_OFFLOAD_VLAN_STRIP
as off at the port configuration. The app after
Hi,
> -Original Message-
> From: Suanming Mou
> Sent: Monday, April 13, 2020 4:29 PM
> Cc: Raslan Darawsheh ; dev@dpdk.org
> Subject: [PATCH 0/2] net/mlx5: fix table leak issue
>
> Currently, the tables got by jump action and flow meter suffix flow are
> not put correctly.
>
> Fix the b
Hi Tonghao, Thanks, your patch works fine with HPET initialization. I will
look into DPDK discussion which got started with your patch.
Thanks,
Ravi
On Mon, Apr 13, 2020 at 5:33 PM Tonghao Zhang
wrote:
> On Tue, Apr 14, 2020 at 7:20 AM Ilya Maximets wrote:
> >
> > On 4/13/20 5:27 PM, Ravi Ker
Basic memory management supports core libraries and PMDs operating in
IOVA as PA mode. It uses a kernel-mode driver, virt2phys, to obtain
IOVAs of hugepages allocated from user-mode.
Signed-off-by: Dmitry Kozlyuk
---
config/meson.build| 2 +-
doc/guides/windows_gsg/
Clang on Windows follows MS ABI where enum values are limited to 2^31-1.
Enum rte_page_size has members valued above this limit, which get
wrapped to zero, resulting in compilation error (duplicate values in
enum). Using MS ABI is mandatory for Windows EAL to call Win32 APIs.
Define these values o
Limited version imported previously lacks at least SLIST macros.
Import a complete file from FreeBSD, since its license exception is
already approved by Technical Board.
Signed-off-by: Dmitry Kozlyuk
---
lib/librte_eal/windows/include/sys/queue.h | 663 +++--
1 file changed, 601
All supported OS create memory segment lists (MSL) and reserve VA space
for them in a nearly identical way. Move common code into EAL private
functions to reduce duplication.
Signed-off-by: Dmitry Kozlyuk
---
lib/librte_eal/common/eal_common_memory.c | 54 ++
lib/librte_eal/commo
EAL common code uses file locking and truncation. Introduce
OS-independent wrappers in order to support both Linux/FreeBSD
and Windows:
* eal_file_lock: lock or unlock an open file.
* eal_file_truncate: enforce a given size for an open file.
Wrappers follow POSIX semantics, but interface is not P
System meory management is implemented differently for POSIX and
Windows. Introduce wrapper functions for operations used across DPDK:
* rte_mem_map()
Create memory mapping for a regular file or a page file (swap).
This supports mapping to a reserved memory region even on Windows.
* rte_mem_u
Add hugepages discovery ("large pages" in Windows terminology)
and update documentation for required privilege setup.
Signed-off-by: Dmitry Kozlyuk
---
config/meson.build | 2 +
doc/guides/windows_gsg/build_dpdk.rst | 20 -
doc/guides/windows_gsg/index.rst | 1
1. Map CPU cores to their respective NUMA nodes as reported by system.
2. Support systems with more than 64 cores (multiple processor groups).
3. Fix magic constants, styling issues, and compiler warnings.
4. Add EAL private function to map DPDK socket ID to NUMA node number.
Fixes: 53ffd9f080fc (
Note: no changes in cover letter since v2.
This patchset implements basic MM with the following features:
* Hugepages are dynamically allocated in user-mode.
* Only 2MB hugepages are supported.
* IOVA is always PA, obtained through kernel-mode driver.
* No 32-bit support (presumably not demanded)
This driver supports Windows EAL memory management by translating
current process virtual addresses to physical addresses (IOVA).
Standalone virt2phys allows using DPDK without PMD and provides a
reference implementation.
Suggested-by: Ranjit Menon
Signed-off-by: Dmitry Kozlyuk
---
windows/READ
The goal of rte_os.h is to mitigate OS differences for EAL users.
In Windows EAL, rte_os.h did excessive things:
1. It included platform SDK headers (windows.h, etc). Those files are
huge, require specific inclusion order, and are generally unused by
the code including rte_os.h. Declarations
W dniu 14.04.2020 o 20:29, Trahe, Fiona pisze:
>
>> -Original Message-
>> From: dev On Behalf Of Pablo de Lara
>> Sent: Tuesday, April 14, 2020 6:30 PM
>> To: Doherty, Declan
>> Cc: dev@dpdk.org; akhil.go...@nxp.com; De Lara Guarch, Pablo
>> ;
>> sta...@dpdk.org
>> Subject: [dpdk-dev]
>
>
> > /**
> > * @internal Enqueue several objects on the RTS ring.
> > diff --git a/lib/librte_ring/rte_ring_rts_c11_mem.h
> > b/lib/librte_ring/rte_ring_rts_c11_mem.h
> > new file mode 100644
> > index 0..b72901497
> > --- /dev/null
> > +++ b/lib/librte_ring/rte_ring_rts_c11_mem.h
> -Original Message-
> From: dev On Behalf Of Pablo de Lara
> Sent: Tuesday, April 14, 2020 6:30 PM
> To: Doherty, Declan
> Cc: dev@dpdk.org; akhil.go...@nxp.com; De Lara Guarch, Pablo
> ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] cryptodev: add missing feature name
>
> String
Hi Ray,
We're going to need hep to understand the numbering.
The examples here
http://doc.dpdk.org/guides/contributing/abi_versioning.html#major-abi-versions
show only major numbers in the .map file.
Whereas the map files all have major.minor.
The example shows that to add a new version of an ex
Remove PMD bitmask, which selects the PMD to be tested
for each test case. Instead, all PMDs are eligible
to run all tests, and capability checking discards
the PMDs which do not support each test case.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev.c | 1 -
app/tes
Before running any sessionless test cases,
check if device supports this mode.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev_blockcipher.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/app/test/test_cryptodev_blockcipher.c
b/app/test/test_cryptodev_block
Add feature flag for symmetric sessionless support,
so it can be checked by applications.
Signed-off-by: Pablo de Lara
---
doc/guides/cryptodevs/features/aesni_gcm.ini | 1 +
doc/guides/cryptodevs/features/aesni_mb.ini | 1 +
doc/guides/cryptodevs/features/armv8.ini | 1 +
doc/guides/crypto
This patchset is the phase two of the crypto test refactoring
effort. It mainly focuses on removing the PMD bitmask used
to select which PMDs are used to run the tests against.
Instead, now all PMDs will run all test cases and the capability
checking will determine which ones are supported and ther
On 2020-04-14 18:15, Jerin Jacob wrote:
> On Tue, Apr 14, 2020 at 9:27 PM Mattias Rönnblom
> wrote:
>> On 2020-04-10 15:00, Jerin Jacob wrote:
>>> On Thu, Apr 9, 2020 at 7:32 PM Mattias Rönnblom
>>> wrote:
On 2020-04-09 15:32, Jerin Jacob wrote:
> On Thu, Apr 9, 2020 at 5:51 PM Mattias R
14/04/2020 19:22, De Lara Guarch, Pablo:
> Hi Thomas,
>
> From: Thomas Monjalon
> > 14/04/2020 12:22, Pablo de Lara:
> > > Now that capabilities are checked to see if an algorithm is supported
> > > by a device, there is no need to check for a specific version of a
> > > library used in a PMD.
>
Now that capabilities are checked to see if an algorithm
is supported by a device, there is no need to check
for a specific version of a library used in a PMD.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev_hash_test_vectors.h | 25 -
1 file changed, 25 deletion
Check if test case is supported by the crypto device,
including algorithm and some of its parameter, such as key length,
IV length, etc, using the capabilities API.
If it is not supported, test case is skipped.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev_blockcipher.c | 49 +
This patchset adds crypto capability checks
in the cryptodev test code, to be able to skip unsupported
test cases for each crypto device, according to their
capabilities.
Thanks to this patchset, there is no more need to check
for internal PMD information in the test code, making it
more "device-a
> >
> > > +
> > > +/**
> > > + * Start to enqueue several objects on the ring.
> > > + * Note that no actual objects are put in the queue by this
> > > +function,
> > > + * it just reserves for user such ability.
> > > + * User has to call appropriate enqueue_finish() to copy objects
> > > +into
String for asymmetric sesionless support was missing.
Fixes: f2b2a4497100 ("cryptodev: add asymmetric session-less")
Cc: sta...@dpdk.org
Signed-off-by: Pablo de Lara
---
lib/librte_cryptodev/rte_cryptodev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/librte_cryptodev/rte_cryptodev
When authenticating after encrypting, if the operation
is out-of-place, the destination buffer is the one
that will get authenticated.
If the cipher offset is higher than the authentication
offset, it means that part of the text to authenticate
will be plaintext, so this needs to get copied to the
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon
> Sent: Tuesday, April 14, 2020 11:34 AM
> To: Doherty, Declan ; akhil.go...@nxp.com; Zhang,
> Roy Fan ; De Lara Guarch, Pablo
>
> Cc: dev@dpdk.org; De Lara Guarch, Pablo
> Subject: Re: [PATCH 2/2] test/crypto: do not check for IMB
> > > diff --git a/lib/librte_ring/rte_ring_hts_generic.h
> > > b/lib/librte_ring/rte_ring_hts_generic.h
> > > new file mode 100644
> > > index 0..da08f1d94
> > > --- /dev/null
> > > +++ b/lib/librte_ring/rte_ring_hts_generic.h
> > > @@ -0,0 +1,198 @@
> > > +
> > > +/**
> > > + * @inte
> >
> > For rings with producer/consumer in RTE_RING_SYNC_ST,
> > RTE_RING_SYNC_MT_HTS mode, provide an ability to split enqueue/dequeue
> > operation into two phases:
> > - enqueue/dequeue start
> > - enqueue/dequeue finish
> > That allows user to inspect objects in the ring without
Hi guys,
> >
> > Hello Honnappa,
> >
> > On Mon, Apr 13, 2020 at 6:42 PM Honnappa Nagarahalli
> > wrote:
> > > > Reviews on those structures must be extra careful, as we are blind
> > > > with those rules in place.
> > > Yes, this is my concern. Why not remove these fixes and ignore the errors
On Tue, Apr 14, 2020 at 9:27 PM Mattias Rönnblom
wrote:
>
> On 2020-04-10 15:00, Jerin Jacob wrote:
> > On Thu, Apr 9, 2020 at 7:32 PM Mattias Rönnblom
> > wrote:
> >> On 2020-04-09 15:32, Jerin Jacob wrote:
> >>> On Thu, Apr 9, 2020 at 5:51 PM Mattias Rönnblom
> >>> wrote:
> On 2020-04-08
Hi Honnappa,
>
> Hi Konstantin,
> Few nits/comments inline.
>
>
>
> > diff --git a/lib/librte_ring/rte_ring_hts.h
> > b/lib/librte_ring/rte_ring_hts.h new
> > file mode 100644 index 0..062d7be6c
> > --- /dev/null
> > +++ b/lib/librte_ring/rte_ring_hts.h
> > @@ -0,0 +1,210 @@
> >
> Subject: RE: [PATCH v3 3/9] ring: introduce RTS ring mode
>
> > > > >
> > > > > +#ifdef ALLOW_EXPERIMENTAL_API
> > > > > +#include
> > > > > +#endif
> > > > > +
> > > > > /**
> > > > > * Enqueue several objects on a ring.
> > > > > *
> > > > > @@ -484,8 +520,21 @@ static __rte_always_inli
>
> Hello Honnappa,
>
> On Mon, Apr 13, 2020 at 6:42 PM Honnappa Nagarahalli
> wrote:
> > > Reviews on those structures must be extra careful, as we are blind
> > > with those rules in place.
> > Yes, this is my concern. Why not remove these fixes and ignore the errors
> manually (i.e. merge t
On 2020-04-10 15:00, Jerin Jacob wrote:
> On Thu, Apr 9, 2020 at 7:32 PM Mattias Rönnblom
> wrote:
>> On 2020-04-09 15:32, Jerin Jacob wrote:
>>> On Thu, Apr 9, 2020 at 5:51 PM Mattias Rönnblom
>>> wrote:
On 2020-04-08 21:36, Jerin Jacob wrote:
> On Wed, Apr 8, 2020 at 11:27 PM Mattias R
On 2020-04-14 15:35, David Marchand wrote:
> On Tue, Apr 14, 2020 at 3:20 PM Mattias Rönnblom
> wrote:
>> On 2020-04-14 06:43, Tonghao Zhang wrote:
>>> On Tue, Apr 14, 2020 at 12:07 PM Stephen Hemminger
>>> wrote:
On Sun, 12 Apr 2020 16:27:53 +0800
xiangxia.m@gmail.com wrote:
>
On Mon, Apr 13, 2020 at 7:15 PM Nithin Dabilpuram wrote:
>
> From: Nithin Dabilpuram
>
> Disable CQ_DISABLED error interrupt in NIX_LF_ERR_INT
> to fix spurious interrupts in event dev mode. Also skip
> configuring RSS when RQ count is '0' because
> RSS table initialization is done incorrectly du
This patch fixes issue that doesn't support mark only case.
Mark only action is equal to mark + passthru action.
Fixes: f5cafa961fae ("net/ice: add flow director create and destroy")
Cc: sta...@dpdk.org
Signed-off-by: Simei Su
---
drivers/net/ice/ice_fdir_filter.c | 14 +-
1 file ch
Hi Chenxu,
> -Original Message-
> From: dev On Behalf Of Chenxu Di
> Sent: Tuesday, April 14, 2020 7:37 AM
> To: dev@dpdk.org
> Cc: Xing, Beilei ; Di, ChenxuX
>
> Subject: [dpdk-dev] [PATCH v8] net/i40e: enable advanced RSS
>
> This patch supports:
>
> - symmetric hash configuration
>
14/04/2020 16:02, Trahe, Fiona:
> Hi Thomas,
>
> From: Thomas Monjalon
> > 14/04/2020 15:04, Trahe, Fiona:
> > > > 14/04/2020 12:21, Ferruh Yigit:
> > > >
> > http://inbox.dpdk.org/dev/mn2pr11mb35507d4b96677a41e66440c5e3...@mn2pr11mb3550.na
> > > > mprd11.prod.outlook.com/
> > > >
> > > > I am no
> -Original Message-
> From: Zhang, Qi Z
> Sent: Tuesday, April 14, 2020 8:51 PM
> To: Su, Simei ; Ye, Xiaolong
> Cc: dev@dpdk.org; Cao, Yahui ; sta...@dpdk.org
> Subject: RE: [PATCH] net/ice: support mark only action for FDIR
>
>
>
> > -Original Message-
> > From: Su, Simei
On Mon, Apr 13, 2020 at 4:56 PM David Marchand
wrote:
>
> From: Pavan Nikhilesh
>
> Remove setting ALLOW_EXPERIMENTAL_API individually for each Makefile and
> meson.build. Instead, enable ALLOW_EXPERIMENTAL_API flag across app, lib
> and drivers.
> This changes reduces the clutter across the proj
Get rid of hardcoded limit of cryptodev sessions.
Fixes: e1143d7dbbf4 ("examples/ipsec-secgw: get rid of maximum SA limitation")
Cc: vladimir.medved...@intel.com
Signed-off-by: Vladimir Medvedkin
---
examples/ipsec-secgw/ipsec-secgw.c | 12 +++-
examples/ipsec-secgw/ipsec.h | 3 +
Hi,
> -Original Message-
> From: dev On Behalf Of Mohsin Shaikh
> Sent: Thursday, April 9, 2020 11:37 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Alexander Kozyrev ;
> Mohsin Shaikh
> Subject: [dpdk-dev] [PATCH] net/mlx5: use open/read/close for reading ib
> stat
>
> fgets(3)/fread(3)
Hi Thomas,
> -Original Message-
> From: Thomas Monjalon
> Sent: Tuesday, April 14, 2020 2:24 PM
> To: Yigit, Ferruh ; Trahe, Fiona
>
> Cc: Coyle, David ; dev@dpdk.org; Doherty, Declan
> ; De Lara Guarch, Pablo
> ; Ryan,
> Brendan ; shreyansh.j...@nxp.com;
> hemant.agra...@nxp.com;
> a
On 18/03/2020 20:41, Arek Kusztal wrote:
> This patch adds versioned function rte_cryptodev_info_get.
> Node 20.05 function works the same way it was working before.
> Node 20.0 function strips capability added in 20.05 release
> to prevent some issues with ABI policy. To do that new capability
Hi Akhil, Thomas,
> -Original Message-
> From: Thomas Monjalon
> Sent: Tuesday, April 14, 2020 2:04 PM
> To: Kusztal, ArkadiuszX
> Cc: Ray Kinsella ; Richardson, Bruce
> ; dev@dpdk.org;
> Trahe, Fiona ; dev@dpdk.org; Akhil Goyal
>
> Subject: Re: [dpdk-dev] [PATCH] cryptodev: version r
Hello Honnappa,
On Mon, Apr 13, 2020 at 6:42 PM Honnappa Nagarahalli
wrote:
> > Reviews on those structures must be extra careful, as we are blind with
> > those
> > rules in place.
> Yes, this is my concern. Why not remove these fixes and ignore the errors
> manually (i.e. merge the patches kn
On Tue, Apr 14, 2020 at 4:39 AM Rasesh Mody wrote:
>
> From: Souvik Dey
>
> In case of bnx2xvf pmd, tx packets can support vland id in 2 ways:
> 1. setting the mbuf ol_flags=PKT_TX_VLAN_PKT and passing the
> vlanid in mbuf->vlan_tci.
> 2. the tx packet itself has the vlan id included in the packe
On Tue, Apr 14, 2020 at 3:20 PM Mattias Rönnblom
wrote:
>
> On 2020-04-14 06:43, Tonghao Zhang wrote:
> > On Tue, Apr 14, 2020 at 12:07 PM Stephen Hemminger
> > wrote:
> >> On Sun, 12 Apr 2020 16:27:53 +0800
> >> xiangxia.m@gmail.com wrote:
> >>
> >>> From: Tonghao Zhang
> >>>
> >>> When rte
1 - 100 of 187 matches
Mail list logo
