> -Original Message-
> From: Rong, Leyi
> Sent: Wednesday, September 16, 2020 2:30 PM
> To: Jiang, JunyuX ; dev@dpdk.org
> Cc: Zhang, Qi Z ; Yang, Qiming
> ; Sun, GuinanX
> Subject: RE: [PATCH v3 0/5] supports RxDID #22 and FDID
>
>
> > -Original Message-
> > From: Jiang, Jun
> -Original Message-
> From: Jiang, JunyuX
> Sent: Wednesday, September 16, 2020 11:10 AM
> To: dev@dpdk.org
> Cc: Rong, Leyi ; Zhang, Qi Z ;
> Yang, Qiming ; Sun, GuinanX
> ; Jiang, JunyuX
> Subject: [PATCH v3 0/5] supports RxDID #22 and FDID
>
> This patchset supports flex Rx descrip
For RSS attribute don't have an associated RSS type, we need
to reject it.
Signed-off-by: Qi Zhang
---
v4:
- refine macros
v3:
- add missing floating l2 rss attribute check
v2:
- fix checkpatch warning.
drivers/net/iavf/iavf_hash.c | 70
1 file changed, 4
For RSS attribute don't have an associated RSS type, we need
to reject it.
Signed-off-by: Qi Zhang
---
v4:
- refine macros
v3:
- add missing floating l2 rss attribute check
v2:
- fix checkpatch warning.
drivers/net/iavf/iavf_hash.c | 70
1 file changed, 4
For RSS attribute don't have an associated RSS type, we need
to reject it.
Signed-off-by: Qi Zhang
---
v4:
- refine macros
v3:
- add missing floating l2 rss attribute check
v2:
- fix checkpatch warning.
drivers/net/iavf/iavf_hash.c | 75 +---
1 file changed, 53
Inside Macro ASSIGN_UNLESS_EQUAL(var, val), assignment to var is always
failing as assignment done using var_ having local scope only.
This leads to TX packets not going out and found broken due to cleanup
malfunctioning. This patch fixes the wrong variable assignment.
Fixes: 57f90f894588 ("net/vi
testpmd will initialize default max packet length to 1518 which does't
include vlan tag size in ether overheader. Once, send the max mtu length
packet with vlan tag, the max packet length will exceed 1518 that will
cause packets dropped directly from NIC hw side. But for i40e/i40evf,
they should su
testpmd will initialize default max packet length to 1518 which does't
include vlan tag size in ether overheader. Once, send the max mtu length
packet with vlan tag, the max packet length will exceed 1518 that will
cause packets dropped directly from NIC hw side.
configure the correct default max
testpmd will initialize default max packet length to 1518 which does't
include vlan tag size in ether overheader. Once, send the max mtu length
packet with vlan tag, the max packet length will exceed 1518 that will
cause packets dropped directly from NIC hw side.
ice can support dual vlan tags tha
testpmd will initialize default max packet length to 1518 which does't
include vlan tag size in ether overheader. Once, send the max mtu length
packet with vlan tag, the max packet length will exceed 1518 that will
cause packets dropped directly from NIC hw side.
iavf can support dual vlan tags th
testpmd will initialize default max packet length to 1518 which does't
include vlan tag size in ether overheader. Once, send the max mtu length
packet with vlan tag, the max packet length will exceed 1518 that will
cause packets dropped directly from NIC hw side.
e1000 can support single vlan tags
testpmd will initialize default max packet length to 1518 which does't
include vlan tag size in ether overheader. Once, send the max mtu length
packet with vlan tag, the max packet length will exceed 1518 that will
cause packets dropped directly from NIC hw side.
igc can support single vlan tag th
> -Original Message-
> From: dev On Behalf Of Guo, Jia
> Sent: Wednesday, September 16, 2020 1:22 PM
> To: Jiang, JunyuX ; dev@dpdk.org
> Cc: Xing, Beilei ; sta...@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2] net/i40e: fix incorrect byte counters
>
> Acked-by: Jeff Guo
>
> > -Or
Acked-by: Jeff Guo
> -Original Message-
> From: Jiang, JunyuX
> Sent: Wednesday, September 16, 2020 9:51 AM
> To: dev@dpdk.org
> Cc: Guo, Jia ; Xing, Beilei ; Jiang,
> JunyuX ; sta...@dpdk.org
> Subject: [PATCH v2] net/i40e: fix incorrect byte counters
>
> This patch fixed the issue tha
Tested-by: Zhou, Jun
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Chenxu Di
Sent: Tuesday, September 15, 2020 10:41 AM
To: dev@dpdk.org
Cc: hemant.agra...@nxp.com; sachin.sax...@nxp.com; Di, ChenxuX
; sta...@dpdk.org
Subject: [dpdk-dev] [PATCH] net/dpaa2: fix
Although currently only the gtpu inner hash be enabled while not the
gtpu outer hash, but the outer protocol still needed to co-exist with
inner protocol when configure the gtpu inner hash rule, that would
allow the gtpu innner hash support for the different outer protocols.
Signed-off-by: Jeff Gu
Hi Anatoly,
We just dump the memory contents when it fails to allocate the memory.
rte_malloc_dump_stats();
rte_dump_physmem_layout();
rte_memzone_dump();
We could see the fragmented memory
--- MALLOC_STATS ---
Heap id:0
Heap name:socket_0
Heap_size:134217728,
Free_size:13360409
Although currently only the gtpu inner hash be enabled while not the
gtpu outer hash, but the outer protocol still needed to co-exist with
inner protocol when configure the gtpu inner hash rule, that would
allow the gtpu innner hash support for the different outer protocols.
Signed-off-by: Jeff Gu
On Wed, 16 Sep 2020 11:05:38 +0800 (CST)
yang_y_yi wrote:
> No, next_proto_id of inner IP header can clearly identify it is a UDP packet
> even if it is a no-udp-header ip fragment.
Yes GRO will know its UDP, but it will have no port information (in this
packet).
GRO can only safely combine p
From: Guinan Sun
Support flow director mark ID parsing from flexible
Rx descriptor in SSE path.
Signed-off-by: Guinan Sun
---
drivers/net/ice/ice_rxtx_vec_sse.c | 49 ++
1 file changed, 49 insertions(+)
diff --git a/drivers/net/ice/ice_rxtx_vec_sse.c
b/drivers/net
From: Guinan Sun
Support flow director mark ID parsing from flexible
Rx descriptor in AVX path.
Signed-off-by: Guinan Sun
---
drivers/net/ice/ice_rxtx_vec_avx2.c | 64 -
1 file changed, 63 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ice/ice_rxtx_vec_avx2.
From: Guinan Sun
Currently, all data paths already support flow mark, so remove devargs
"flow-mark-support". FDIR matched ID will display in verbose
when packets match the created rule.
Signed-off-by: Guinan Sun
---
doc/guides/nics/ice.rst | 12
drivers/net/ice/ice_e
This patch supports RxDID #22 by the following changes:
-add structure and macro definition for RxDID #22.
-support RxDID #22 format in normal path.
-change RSS hash parsing from RxDID #22 in AVX/SSE data path.
Signed-off-by: Junyu Jiang
---
drivers/net/ice/ice_ethdev.c| 20 ++
drive
From: Guinan Sun
The patch adds fdir_enabled flag to identify if parse flow director mark ID
from flexible Rx descriptor.
Signed-off-by: Guinan Sun
---
drivers/net/ice/ice_ethdev.h | 2 ++
drivers/net/ice/ice_fdir_filter.c | 9 -
drivers/net/ice/ice_rxtx.h| 30 ++
This patchset supports flex Rx descriptor RxDID #22 and
FDID offload in vPMD. Remove devargs "flow-mark-support".
---
v3:
* Check if package does support RXDID.
* Modify commit message.
* Rebase the patchset.
v2:
* Reorder patches.
Guinan Sun (4):
net/ice: add flow director enabled switch valu
> -Original Message-
> From: Medvedkin, Vladimir
> Sent: Wednesday, September 16, 2020 12:28 AM
> To: Bruce Richardson ; Ruifeng Wang
>
> Cc: dev@dpdk.org; Honnappa Nagarahalli
> ; nd
> Subject: Re: [PATCH 2/2] lpm: hide internal data
>
> Hi Ruifeng,
>
> On 15/09/2020 17:02, Bruce Ric
No, next_proto_id of inner IP header can clearly identify it is a UDP packet
even if it is a no-udp-header ip fragment.
At 2020-09-16 10:54:24, "Jiayu Hu" wrote:
>On Mon, Sep 14, 2020 at 05:14:59PM +0800, yang_y_yi wrote:
>>
>> Jiayu, VM to VM case for big UDP packet (say ud
The ptype mask for flexible descriptor in Rx function ice_recv_pkts_vec
has a reversed order, which leads to an incorrect value of the final
ptype. This patch fix the mask to parse the correct ptype of RX packets.
Fixes: c68a52b8b38c ("net/ice: support vector SSE in Rx")
Cc: sta...@dpdk.org
Signe
Acked-by: Liu Tianjiao
> -Original Message-
> From: Chautru, Nicolas
> Sent: Thursday, September 10, 2020 5:16 AM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Liu, Tianjiao ; Chautru, Nicolas
>
> Subject: [PATCH v1] app/bbdev: fix test vector symlink
> 5G DL default symlink was pointing
On Mon, Sep 14, 2020 at 05:14:59PM +0800, yang_y_yi wrote:
>
> Jiayu, VM to VM case for big UDP packet (say udp parload size is 8192K, but
> mtu
> 1500). VM will segment it as UDP framgments, but when ovs dpdk uses VxLan to
> encapsulate them, it doesn't know they are same flow, OVS DPDK has no k
Tested-by: Yingya Han
-Original Message-
From: stable On Behalf Of Junyu Jiang
Sent: Wednesday, September 16, 2020 9:51 AM
To: dev@dpdk.org
Cc: Guo, Jia ; Xing, Beilei ; Jiang,
JunyuX ; sta...@dpdk.org
Subject: [dpdk-stable] [PATCH v2] net/i40e: fix incorrect byte counters
This patch f
Acked-by: Liu Tianjiao
> -Original Message-
> From: Chautru, Nicolas
> Sent: Tuesday, August 18, 2020 5:06 PM
> To: dev@dpdk.org; akhil.go...@nxp.com
> Cc: Richardson, Bruce ; Chautru, Nicolas
>
> Subject: [PATCH v3 6/7] app/bbdev: reduce duration of throughput test
>
> Reducing numbe
This patch fixed the issue that rx/tx bytes overflowed
on 48 bit limitation by enlarging the limitation.
Fixes: 4861cde46116 ("i40e: new poll mode driver")
Cc: sta...@dpdk.org
Signed-off-by: Junyu Jiang
---
drivers/net/i40e/i40e_ethdev.c | 47 ++
drivers/net/i40e
On Tue, 15 Sep 2020 16:28:03 -0700
Pallavi Kadam wrote:
> + if (IsEqualGUID((const void *)&(device_info_data->ClassGuid),
> + (const void *)&GUID_DEVCLASS_NETUIO))
Is there anyway to get rid of casts and have the right types?
Casts are evil
This patch adds implementations to probe PCI devices bound to netuio
with the help of "netuio" class device changes.
Now Windows will support both "netuio" and "net" device class and
can set kernel driver type based on the device class selection.
Note: Few definitions and structures have been copi
> From: Ivan Dyukov [mailto:i.dyu...@samsung.com]
> Sent: Tuesday, September 15, 2020 9:07 PM
>
> There is new link_speed value introduced. It's INT_MAX value which
> means that speed is unknown. To simplify processing of the value
> in application, new function is added which convert link_speed t
Hi, Tal
On 9/13/2020 11:42 AM, Tal Shnaiderman wrote:
Subject: [PATCH] bus/pci: netuio interface for windows
This patch adds implementations to probe PCI devices bound to netuio with
the help of "netuio" class device changes.
Now Windows will support both "netuio" and "net" device class and can
On Tue, Sep 15, 2020 at 4:51 AM Andrey Vesnovaty wrote:
> Hi Ajit, PSB.
>
> Thanks,
> Andrey
>
> > -Original Message-
> > From: Ajit Khaparde
> > Sent: Saturday, September 12, 2020 5:18 AM
> > To: Andrey Vesnovaty
> > Cc: dpdk-dev ; j...@marvell.com; Jerin Jacob
> > ; NBU-Contact-Thomas
On Tue, Sep 15, 2020 at 7:07 AM Lance Richardson <
lance.richard...@broadcom.com> wrote:
> A previous commit broke compilation for builds other than x86 and
> aarch64. Fix by moving macro definitions to more appropriate header
> files.
>
> Fixes: 479387565605 ("net/bnxt: improve small ring sizes s
Statistics of handled packets are cleared and read on main lcore,
while they are increased in workers handlers on different lcores.
Without synchronization occasionally showed invalid values.
This patch uses atomic acquire/release mechanisms to synchronize.
Fixes: c3eabff124e6 ("distributor: add
During quit_workers function distributor's main core processes
some packets to wake up pending worker cores so they can quit.
As quit_workers acts also as a cleanup procedure for next test
case it should also collect these packages returned by workers'
handlers, so the cyclic buffer with returned p
rte_distributor_return_pkt function which is run on worker cores
must wait for distributor core to clear handshake on retptr64
before using those buffers. While the handshake is set distributor
core controls buffers and any operations on worker side might overwrite
buffers which are unread yet.
Fi
Sanity tests with mbuf alloc and shutdown tests assume that
mbufs passed to worker cores are freed in handlers.
Such packets should not be returned to the distributor's main
core. The only packets that should be returned are the packets
send after completion of the tests in quit_workers function.
Synchronization of data exchange between distributor and worker cores
is based on 2 handshakes: retptr64 for returning mbufs from workers
to distributor and bufptr64 for passing mbufs to workers.
Without proper order of verifying those 2 handshakes a deadlock may
occur. This can happen when worker
The sanity test with worker shutdown delegates all bufs
to be processed by a single lcore worker, then it freezes
one of the lcore workers and continues to send more bufs.
Problem occurred if freezed lcore is the same as the one
that is processing the mbufs. The lcore processing mbufs
might be dif
During review and verification of the patch created by Sarosh Arif:
"test_distributor: prevent memory leakages from the pool" I found out
that running distributor unit tests multiple times in a row causes fails.
So I investigated all the issues I found.
There are few synchronization issues that mi
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/server_node_efd/server/init.c | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/examples/server_node_efd/server/init.c
b/examples/server_node_efd/server/i
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/performance-thread/l3fwd-thread/main.c | 13 +
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/examples/performance-thread/l3fwd-thread/main.c
b/examples/performan
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/vm_power_manager/main.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/examples/vm_power_manager/main.c b/examples/vm_power_manager/main.c
index 273bfec29
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/l2fwd/main.c | 13 +
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/examples/l2fwd/main.c b/examples/l2fwd/main.c
index e04c601b5..28bede684 100644
--- a/examples/
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/qos_sched/init.c | 12
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/examples/qos_sched/init.c b/examples/qos_sched/init.c
index 9626c15b8..06328ddb2 100644
---
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/l3fwd-power/main.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
index 9db94ce04..bcd737994 100
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
.../client_server_mp/mp_server/init.c | 15 ++-
examples/multi_process/symmetric_mp/main.c| 13 +
2 files changed, 11 insertions(+), 17 deletions(-)
diff
Add usage of rte_eth_link_strf function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/ntb/ntb_fwd.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/examples/ntb/ntb_fwd.c b/examples/ntb/ntb_fwd.c
index eba8ebf9f..6d9b3b45c 100644
--- a/examples/n
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/l3fwd-graph/main.c | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/examples/l3fwd-graph/main.c b/examples/l3fwd-graph/main.c
index c70270c4d..0f3f4088c 1
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/l2fwd-event/main.c | 13 +
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/examples/l2fwd-event/main.c b/examples/l2fwd-event/main.c
index 4fe500333..d90c71698 1006
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/l2fwd-crypto/main.c | 13 +
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
index 827da9b3e..820470db5 1
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/bbdev_app/main.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/examples/bbdev_app/main.c b/examples/bbdev_app/main.c
index 68a46050c..fe9d8e0a5 100644
--- a/examp
Add usage of rte_eth_link_to_str function to example
applications:
* ip_fragmentation
* ip_reassembly
* l3fwd-acl
Signed-off-by: Ivan Dyukov
---
examples/ip_fragmentation/main.c | 13 +
examples/ip_reassembly/main.c| 13 +
examples/l3fwd-acl/ma
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/ipsec-secgw/ipsec-secgw.c | 13 +
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/examples/ipsec-secgw/ipsec-secgw.c
b/examples/ipsec-secgw/ipsec-secgw.c
index f77
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/ioat/ioatfwd.c | 14 ++
1 file changed, 6 insertions(+), 8 deletions(-)
diff --git a/examples/ioat/ioatfwd.c b/examples/ioat/ioatfwd.c
index b66ee73bc..f77012eb9 100644
--- a/ex
Add usage of rte_eth_link_speed_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/ip_pipeline/cli.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/ip_pipeline/cli.c b/examples/ip_pipeline/cli.c
index d79699e2e..cf3d7d462 100644
---
Add usage of rte_eth_link_to_str function to example
applications
Signed-off-by: Ivan Dyukov
---
examples/kni/main.c | 27 ++-
1 file changed, 10 insertions(+), 17 deletions(-)
diff --git a/examples/kni/main.c b/examples/kni/main.c
index f5d12a5b8..49d888064 100644
--- a
rte_ethdev has declared new NUM_UNKNOWN speed which
could be used in case when no speed information is available and
link is up. NUM_NONE should be returned, if link is down.
Signed-off-by: Ivan Dyukov
Acked-by: Jeff Guo
---
drivers/net/i40e/i40e_ethdev.c| 5 -
drivers/net/i40e/i40e_et
rte_ethdev has declared new NUM_UNKNOWN speed which
could be used in case when no speed information is available
Signed-off-by: Ivan Dyukov
Reviewed-by: Wei Zhao
---
drivers/net/ixgbe/ixgbe_ethdev.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe
Add usage of rte_eth_link_to_str function to example
applications:
* ipv4_multicast
* l2fwd-jobstats
* l2fwd-keepalive
* l3fwd
* link_status_interrupt
Signed-off-by: Ivan Dyukov
---
examples/ipv4_multicast/main.c| 14 ++---
examples/l2fwd-j
rte_ethdev has declared new NUM_UNKNOWN speed which
could be used in case when no speed information is available and
link is up. NUM_NONE should be returned, if link is down.
Signed-off-by: Ivan Dyukov
Reviewed-by: Ferruh Yigit
---
drivers/net/ice/ice_ethdev.c | 5 -
1 file changed, 4 inser
There is new link_speed value introduced. It's INT_MAX value which
means that speed is unknown. To simplify processing of the value
in application, new function is added which convert link_speed to
string. Also dpdk examples have many duplicated code which format
entire link status structure to tex
MAINTAINERS | 1 +
app/proc-info/main.c | 11 ++---
app/test-pipeline/init.c | 11 ++---
app/test-pmd/config.c| 9 ++--
app/test-pmd/testpmd.c
Add usage of rte_eth_link_to_str function to example
applications and docs
Signed-off-by: Ivan Dyukov
---
app/proc-info/main.c | 11 +--
app/test-pipeline/init.c | 11 +--
app/test-pmd/config.c | 9 +
From: Thomas Monjalon
When querying the link information, the link status is
a mandatory major information.
Other boolean values are supposed to be accurate:
- duplex mode (half/full)
- negotiation (auto/fixed)
This API update is making explicit that the link speed information
is
From: Pavan Nikhilesh
Improve single flow performance by moving the point of coherence
to the end of transmit sequence.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/octeontx2/otx2_worker.h | 35 +--
drivers/net/octeontx2/otx2_tx.h | 18 ++
2 files
Since we are not holding the mbufs or creating any references
in the app, hence mbuf fast free offload can be enabled.
Signed-off-by: Harman Kalra
---
app/test-eventdev/test_pipeline_common.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/app/test-eventdev/test_pipeline_common.c
b/app
Issue has been observed in case of multi segments where mbuf
data gets corrupted due to missing barriers. Changes made to
mbuf just before LMTST by one core gets updatded when the
same mbuf is in use by another core, leading to corruption.
It should be ensured that all changes made to mbuf should b
From: Pavan Nikhilesh
Add SWTAG flush operation at the end of transmit sequence to
immediately release the tag held by the core.
Reuse Tag address to check SWTAG completion status.
Signed-off-by: Pavan Nikhilesh
---
drivers/event/octeontx2/otx2_evdev.c | 2 +-
drivers/event/octeontx2/otx2_ev
> -Original Message-
> From: McDaniel, Timothy
> Sent: Friday, September 11, 2020 2:18 PM
> Cc: dev@dpdk.org; Carrillo, Erik G ; Eads, Gage
> ; Van Haaren, Harry ;
> jer...@marvell.com
> Subject: [PATCH v4 04/22] event/dlb: add definitions shared with LKM or
> shared code
>
> Add heade
A previous commit broke compilation for builds other than x86 and
aarch64. Fix by moving macro definitions to more appropriate header
files.
Fixes: 479387565605 ("net/bnxt: improve small ring sizes support")
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxq.c | 1 -
drive
Add module EEPROM/EEPROM dump command
"show port (module_eeprom|eeprom)"
Commands will dump the content of the EEPROM/module
EEPROM for the selected port.
Signed-off-by: David Liu
---
app/test-pmd/cmdline.c | 49
app/test-pmd/config.c | 11
Signed-off-by: David Liu
---
app/test-pmd/cmdline.c | 49
app/test-pmd/config.c | 118
app/test-pmd/testpmd.h | 2 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 7 ++
4 files changed, 176 insertio
Replace '--scalar' command-line option with new one: --alg=
to allow user explicitly select desired classify method.
This is an optional parameter, if not specified default classify
algorithm will be used.
Signed-off-by: Konstantin Ananyev
---
doc/guides/rel_notes/release_20_11.rst| 5
Improve l3fwd-acl usability.
Allow to specify destination MAC and explicitly select classify algorithm.
hese patch series depends on:
https://patches.dpdk.org/project/dpdk/list/?series=12241
to be applied first.
Konstantin Ananyev (3):
examples/l3fwd-acl: add source and destination MAC update
Add ability to select AVX512 classify algorithm.
Signed-off-by: Konstantin Ananyev
---
doc/guides/sample_app_ug/l3_forward_access_ctrl.rst | 2 +-
examples/l3fwd-acl/main.c | 4
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/doc/guides/sample_app_ug/
Introduces two changes into l3fwd-acl behaviour to make
it behave in the same way as l3fwd:
- Add a command-line parameter to allow the user to specify the
destination mac address for each ethernet port used.
- While forwarding the packet update source and destination mac
addresses.
Signed
On 9/15/2020 5:08 PM, Ferruh Yigit wrote:
> On 9/15/2020 4:38 PM, Ajit Khaparde wrote:
>>
>>
>> On Tue, Sep 15, 2020 at 7:07 AM Lance Richardson
>> mailto:lance.richard...@broadcom.com>>
>> wrote:
>>
>> A previous commit broke compilation for builds other than x86 and
>> aarch64. Fix by mov
Add AVX512 classify to the test coverage.
Signed-off-by: Konstantin Ananyev
---
app/test/test_acl.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/app/test/test_acl.c b/app/test/test_acl.c
index 333b34757..11d69d2d5 100644
--- a/app/test/test_acl.c
+++ b/app/test/test_a
With current ACL implementation first field in the rule definition
has always to be one byte long. Though for optimising classify
implementation it might be useful to be able to use 4B reads
(as we do for rest of the fields).
So at build phase, check user provided field definitions to determine
is
Add necessary changes to support new AVX512 specific ACL classify
algorithm:
- changes in meson.build to check that build tools
(compiler, assembler, etc.) do properly support AVX512.
- run-time checks to make sure target platform does support AVX512.
- dummy rte_acl_classify_avx512() for tar
Add ability to use AVX512 classify method.
Signed-off-by: Konstantin Ananyev
---
app/test-acl/main.c | 4
1 file changed, 4 insertions(+)
diff --git a/app/test-acl/main.c b/app/test-acl/main.c
index d9b65517c..19b714335 100644
--- a/app/test-acl/main.c
+++ b/app/test-acl/main.c
@@ -81,6 +8
Add search_avx512x16x2() which uses mostly 512-bit width
registers/instructions and is able to process up to 32 flows in
parallel. That allows to futher speedup rte_acl_classify_avx512()
for bursts with 32+ requests.
Run-time code-path selection is done internally based
on input burst size and is t
Introduce classify implementation that uses AVX512 specific ISA.
Current approach uses mix of 256i/512-bit width registers/instructions
and is able to process up to 16 flows in parallel.
Note that for now only 64-bit version of rte_acl_classify_avx512()
is available.
Signed-off-by: Konstantin Anan
Removal of unused enum value (RTE_ACL_CLASSIFY_NUM).
This enum value is not used inside DPDK, while it prevents
to add new classify algorithms without causing an ABI breakage.
Note that this change introduce a formal ABI incompatibility
with previous versions of ACL library.
Signed-off-by: Konsta
Make classify test to run for all supported methods.
Signed-off-by: Konstantin Ananyev
---
app/test/test_acl.c | 103 ++--
1 file changed, 51 insertions(+), 52 deletions(-)
diff --git a/app/test/test_acl.c b/app/test/test_acl.c
index 316bf4d06..333b34757
- enhance output to print extra stats
- use rte_rdtsc_precise() for cycle measurements
Signed-off-by: Konstantin Ananyev
---
app/test-acl/main.c | 15 ++-
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/app/test-acl/main.c b/app/test-acl/main.c
index 0a5dfb621..d9b6551
Right now ACL library determines best possible (default) classify method
on a given platform with specilal constructor function rte_acl_init().
This patch makes the following changes:
- Move selection of default classify method into a separate private
function and call it for each ACL context c
Add brief description for missing ACL classify algorithms:
RTE_ACL_CLASSIFY_NEON and RTE_ACL_CLASSIFY_ALTIVEC.
Fixes: 34fa6c27c156 ("acl: add NEON optimization for ARMv8")
Fixes: 1d73135f9f1c ("acl: add AltiVec for ppc64")
Cc: sta...@dpdk.org
Signed-off-by: Konstantin Ananyev
---
doc/guides/pro
These patch series introduce support of AVX512 specific classify
implementation for ACL library.
Inside it contains two code-paths –
one uses mostly 256 bit instruction/registers and can
process up to 16 flows in parallel.
second uses 512 bit instruction/registers over majority of
places and can pr
Right now we define dummy version of rte_acl_classify_avx2()
when both X86 and AVX2 are not detected, though it should be
for non-AVX2 case only.
Fixes: e53ce4e41379 ("acl: remove use of weak functions")
Cc: sta...@dpdk.org
Signed-off-by: Konstantin Ananyev
---
lib/librte_acl/rte_acl.c | 2 +-
15/09/2020 18:39, Chautru, Nicolas:
> Hi,
>
> > From: Thomas Monjalon
> > Sent: Tuesday, September 15, 2020 9:11 AM
> > To: David Marchand ; Chautru, Nicolas
> > ; Power, Ciara ;
> > Richardson, Bruce
> > Cc: dev
> > Subject: Re: [PATCH v4 00/31] remove make support in DPDK
> >
> > 15/09/2020
Hi,
> From: Thomas Monjalon
> Sent: Tuesday, September 15, 2020 9:11 AM
> To: David Marchand ; Chautru, Nicolas
> ; Power, Ciara ;
> Richardson, Bruce
> Cc: dev
> Subject: Re: [PATCH v4 00/31] remove make support in DPDK
>
> 15/09/2020 17:57, Richardson, Bruce:
> >
> > > -Original Message
Hi Ruifeng,
On 15/09/2020 17:02, Bruce Richardson wrote:
On Mon, Sep 07, 2020 at 04:15:17PM +0800, Ruifeng Wang wrote:
Fields except tbl24 and tbl8 in rte_lpm structure have no
need to be exposed to the user.
Hide the unneeded exposure of structure fields for better
ABI maintainability.
Sugges
1 - 100 of 227 matches
Mail list logo
