From: Wenxuan Wu
Due to the change of struct rte_gtp_psc_generic_hdr, kernel driver can
not handle gtp_psc properly, we introduce a new structure to fix this
gap between kernel driver and struct rte_gtp_psc_generic_hdr.
Fixes: d5eb3e600d9e ("net/iavf: support flow director basic rule")
Cc: simei
From: Wenxuan Wu
Due to the change of struct rte_gtp_psc_generic_hdr, kernel driver can
not handle gtp_psc properly, we introduce a new structure to fix this
gap between kernel driver and struct rte_gtp_psc_generic_hdr.
Fixes: d5eb3e600d9e ("net/iavf: support flow director basic rule")
Cc: simei
From: Wenxuan Wu
Due to the change of struct rte_gtp_psc_generic_hdr, firmware can
not handle gtp_psc properly, we induce a new structure to fix this
gap between firmware and struct rte_gtp_psc_generic_hdr.
Fixes: d5eb3e600d9e ("net/iavf: support flow director basic rule")
Cc: simei...@intel.com
From: Wenxuan Wu
gcc 12 with -O2 flag would raise the following warning:
../drivers/net/ice/base/ice_switch.c:7220:61: error: writing 1 byte into a
region of size 0 [-Werror=stringop-overflow=]
7220 | buf[recps].content.lkup_indx[i + 1] = entry->fv_idx[i];
| ~~~
From: Wenxuan Wu
when n == 1000, and it would overflow the array size of 4 bytes with
nul terminator. Change the logic to avoid this warning.
Fixes: 15c431864000 ("app/flow-perf: add packet forwarding support")
Cc: sta...@dpdk.org
Signed-off-by: Wenxuan Wu
---
app/test-flow-perf/main.c | 2 +-
From: Wenxuan Wu
Gcc with -O2 flag, would retrieve the value in one time.
This patch changed the type of fv_idx in struct ice_recp_grp_entry to
align with its callers which is also u8 type.
When u8 idx[5] = a value u16 index[4], gcc12 would give this warning,
because it is not big enough to stor
From: Wenxuan Wu
when n == 1000, and it would be overflow the array size of 4 bytes with
nul terminator. Change the logic to avoid this warning.
Fixes: 15c431864000 ("app/flow-perf: add packet forwarding support")
Cc: sta...@dpdk.org
Signed-off-by: Wenxuan Wu
---
app/test-flow-perf/main.c | 2
From: Wenxuan Wu
when n == 1000, and it would be overflow the array size of 4 bytes with
nul terminator. Change the logic to avoid this warning.
Fixes: 15c431864000 ("app/flow-perf: add packet forwarding support")
Cc: sta...@dpdk.org
Signed-off-by: Wenxuan Wu
---
app/test-flow-perf/main.c | 2
From: Wenxuan Wu
This patch adds support for protocol based buffer split in normal Rx
data paths. When the Rx queue is configured with specific protocol type,
packets received will be directly split into protocol header and
payload parts limitation of ice pmd. And the two parts will be
put into d
From: Wenxuan Wu
Add command line parameter:
--rxhdrs=mac,[ipv4,udp]
Set the protocol_hdr of segments to scatter packets on receiving if
split feature is engaged. And the queues with BUFFER_SPLIT flag.
Add interactive mode command:
testpmd>set rxhdrs mac,ipv4,l3,tcp,udp,sctp
(protocol sequence
From: Wenxuan Wu
Currently, Rx buffer split supports length based split. With Rx queue
offload RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT enabled and Rx packet segment
configured, PMD will be able to split the received packets into
multiple segments.
However, length based buffer split is not suitable for N
From: Wenxuan Wu
This patch added new ethdev API to retrieve supported protocol header mask
of a PMD, which helps to configure protocol header based buffer split.
Signed-off-by: Wenxuan Wu
---
doc/guides/rel_notes/release_22_07.rst | 2 ++
lib/ethdev/ethdev_driver.h | 18 +
From: Wenxuan Wu
Protocol type based buffer split consists of splitting a received packet into
several separate segments based on the packet content. It is useful in some
scenarios, such as GPU acceleration. The splitting will help to enable
true zero copy and hence improve the performance signif
From: Wenxuan Wu
GCC version greater than 10.0, with compile option -O2, several warnings info
would appear,
this fix omitted these warnings.
Fixes: cfacbcb5a23b ("build: disable gcc 10 zero-length-bounds warning")
Cc: sta...@dpdk.org
Signed-off-by: Wenxuan Wu
---
config/meson.build | 7
From: Wenxuan Wu
Add command line parameter:
--rxhdrs=mac,[ipv4,udp]
Set the protocol_hdr of segments to scatter packets on receiving if
split feature is engaged. And the queues with BUFFER_SPLIT flag.
Add interative mode command:
testpmd>set rxhdrs mac,ipv4,l3,tcp,udp,sctp
(protocol sequence a
From: Wenxuan Wu
This patch adds support for proto based buffer split in normal Rx data
paths. When the Rx queue is configured with specific protocol type,
packets received will be directly splitted into protocol header and
payload parts. And the two parts will be put into different mempools.
Cu
From: Wenxuan Wu
Currently, Rx buffer split supports length based split. With Rx queue
offload RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT enabled and Rx packet segment
configured, PMD will be able to split the received packets into
multiple segments.
However, length based buffer split is not suitable for N
From: Wenxuan Wu
Currently, Rx buffer split supports length based split. With Rx queue
offload RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT enabled and Rx packet segment
configured, PMD will be able to split the received packets into
multiple segments.
However, length based buffer split is not suitable for N
From: Wenxuan Wu
Buffer split consists of splitting a received packet into two separate
regions based on the packet content. It is useful in some scenarios,
such as GPU acceleration. The splitting will help to enable true zero
copy and hence improve the performance significantly.
This patchset e
From: Wenxuan Wu
Add command line parameter:
--rxhdrs=mac,[ipv4,udp]
Set the protocol_hdr of segments to scatter packets on receiving if
split feature is engaged. And the queues with BUFFER_SPLIT flag.
Add interative mode command:
testpmd>set rxhdrs mac,ipv4,l3,tcp,udp,sctp
(protocol sequence a
From: Wenxuan Wu
This patch adds support for proto based buffer split in normal Rx data
paths. When the Rx queue is configured with specific protocol type,
packets received will be directly splitted into protocol header and
payload parts. And the two parts will be put into different mempools.
Cu
From: Wenxuan Wu
Currently, Rx buffer split supports length based split. With Rx queue
offload RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT enabled and Rx packet segment
configured, PMD will be able to split the received packets into
multiple segments.
However, length based buffer split is not suitable for N
From: Wenxuan Wu
Buffer split consists of splitting a received packet into two separate
regions based on the packet content. It is useful in some scenarios,
such as GPU acceleration. The splitting will help to enable true zero
copy and hence improve the performance significantly.
This patchset e
From: Wenxuan Wu
Add command line parameter:
--rxhdrs=mac,[ipv4,udp]
Set the protocol_hdr of segments to scatter packets on receiving if
split feature is engaged. And the queues with BUFFER_SPLIT flag.
Add interative mode command:
testpmd>set rxhdrs mac,ipv4,l3,tcp,udp,sctp
(protocol sequence a
From: Wenxuan Wu
This patch adds support for proto based buffer split in normal Rx data
paths. When the Rx queue is configured with specific protocol type,
packets received will be directly splitted into protocol header and
payload parts. And the two parts will be put into different mempools.
Cu
From: Wenxuan Wu
Currently, Rx buffer split supports length based split. With Rx queue
offload RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT enabled and Rx packet segment
configured, PMD will be able to split the received packets into
multiple segments.
However, length based buffer split is not suitable for N
From: Wenxuan Wu
Buffer split consists of splitting a received packet into two separate
regions based on the packet content. It is useful in some scenarios,
such as GPU acceleration. The splitting will help to enable true zero
copy and hence improve the performance significantly.
This patchset e
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, and the wait
time is 1 sec. Startup time of 10G_BASET longer than 1s would result in
failure.
Acctually, max frame size of media type I40E_MEDIA_TYPE_BASET can be set
regardless of link status.
This patch omitted the l
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, and the wait
time is 1 sec. Startup time of 10G_BASET longer than 1s would result in
failure.
Acctually, max frame size of media type I40E_MEDIA_TYPE_BASET can be set
regardless of link status.
This patch omitted the l
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, and the wait
time is 1 sec. Startup time of 10G_BASET longer than 1s would result in
failure.
Acctually, max frame size of media type I40E_MEDIA_TYPE_BASET can be set
regardless of link status.
This patch omitted the l
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, and the wait
time is 1 sec. For media type of I40E_10G_BASET would consume longer
time which is too short to up would result in error.
Acctually, max frame size of media type I40E_MEDIA_TYPE_BASET can be set
regardless
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, and the wait
time is 1 sec. For media type of I40E_10G_BASET would consume longer
time which is too short to up would result in error.
Acctually, max frame size of media type I40E_MEDIA_TYPE_BASET can be set
regardless
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, and the wait
time is 1 sec. But for nic media type of I40E_10G_BASET would consume longer
time which is too short to up would result in error.
Acctually, max frame size of media type I40E_MEDIA_TYPE_BASET can be set
reg
From: Wenxuan Wu
This fix resolve max frame size can only be set when link is up.
hw.type would result in different behavior:
For X710, max frame size can be set regardless of link status.
For X722, link up should be guaranteed.
Fixes: a4ba77367923
Cc: sta...@dpdk.org
Signed-off-by: Wenxuan Wu
From: Wenxuan Wu
In 810 CA series, reset time would be longer to wait the kernel return
value.
this patch enable reset wait time longer to finish kernel reset
operation.
Signed-off-by: Wenxuan Wu
---
drivers/net/iavf/iavf.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/dr
From: Wenxuan Wu
This patch adds support for proto based buffer split in normal Rx data
paths. When the Rx queue is configured with specific protocol type,
packets received will be directly splitted into protocol header and
payload parts. And the two parts will be put into different mempools.
Cu
From: Wenxuan Wu
This patch adds protocol based buffer split configuration in testpmd.
The protocol split feature is off by default. To enable protocol split,
you need:
1. Start testpmd with two mempools. e.g. --mbuf-size=2048,2048
2. Configure Rx queue with rx_offload buffer split on.
3. Set the
From: Wenxuan Wu
Protocol based buffer split consists of splitting a received packet into two
separate regions based on its content. The split happens after the packet
protocol header and before the packet payload. Splitting is usually between
the packet protocol header that can be posted to a de
From: Wenxuan Wu
Protocol based buffer split consists of splitting a received packet into
two separate regions based on the packet content. It is useful in some
scenarios, such as GPU acceleration. The splitting will help to enable
true zero copy and hence improve the performance significantly.
From: Wenxuan Wu
Previously, max frame size can only be set when link is up, testpmd
configuration
when port is down, so there is a confilict to apply this max frame size before
link up.
Actual results:
Set max frame size at port level not applicable on link down
RX-packets: 0
From: Xuan Ding
This patch adds support for header split in normal Rx data paths.
When the Rx queue is configured with header split for specific
protocol type, packets received will be directly splited into
header and payload parts. And the two parts will be put into
different mempools.
Currentl
From: Xuan Ding
This patch adds header split configuration in testpmd. The header split
feature is off by default. To enable header split, you need:
1. Configure Rx queue with rx_offload header split on.
2. Set the protocol type of header split.
Command for set header split protocol type:
testpm
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. The split happens after the
packet header and before the packet payload. Splitting is usually between
the packet header that can be posted to a dedicated buffer and the pack
From: Xuan Ding
Header split consists of splitting a received packet into two separate
regions based on the packet content. It is useful in some scenarios,
such as GPU acceleration. The splitting will help to enable true zero
copy and hence improve the performance significantly.
This patchset ex
From: wenxuan wu
When testpmd use func get_eth_dev_info() while related resource
had been released.
Change the logic of func port_is_bonding_slave, this func
eth_dev_info_get_print_err while pf is released would result
in this error. Use ports instead to avoid this bug.
Fixes: 0a0821bcf312 ("
From: wenxuan wu
When testpmd start ed with 1 pf and 2 vfs, testpmd quited while vfs
were still alive would result in failure. Root cause is that pf had
been released already but vfs were still accessing by func
rte_eth_dev_info_get, which would result in heap-free-after-use error.
By quitting
From: wenxuan wu
There is a heap-free-after-use bug when quit testpmd
with pf and vfs, stop and close ports in reverse order
is a more reasonable approach.
Cc: sta...@dpdk.org
Signed-off-by: wenxuan wu
---
lib/ethdev/rte_ethdev.h | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(
From: wenxuan wu
Apply our reverse iterator macro into testpmd to
avoid testpmd startups with pf and vfs quit without
error heap-free-before-use
wenxuan wu (2):
lib/ethdev: add reverse macro to quit testpmd
app/testpmd:fix testpmd quit failure
app/test-pmd/testpmd.c | 4 ++--
lib/ethde
From: wenxuan wu
There is a heap-free-after-use bug when quit testpmd
with pf and vfs, stop and close ports in reverse order
is a more reasonable approach.
Cc: sta...@dpdk.org
Signed-off-by: wenxuan wu
---
lib/ethdev/rte_ethdev.h | 14 +-
1 file changed, 13 insertions(+), 1 deleti
From: wenxuan wu
When testpmd startups with pf and vfs,this error occurs when quitting,
results in pf is released before vfs ,so the vf would access an
freed heap memory.
The solution is that release our allocated ports in reverse
order,add two macros RTE_ETH_FOREACH_DEV_REVERSE_OWNED_BY and
RTE
From: wenxuan wu
Apply our reverse iterator macro into testpmd to avoid testpmd startups with pf
and vfs quit without error heap-free-before-use
wenxuan wu (2):
app/testpmd: fix quit testpmd with vfs and pf
lib/ethdev: add reverse macro to quit testpmd
app/test-pmd/testpmd.c | 4 ++--
From: Wenxuan Wu
The ret value in rte_dev_event_monitor_stop stands for whether the
monitor has been successfully closed ,and should not bind with
rte_intr_callback_unregister,so once it goes to the right exit point of
rte_dev_event_monitor, the ret value should be set to 0.
Also ,the refmonitor
From: Wenxuan Wu
The ret value stands for whether the device monitor has been
successfully closed, and has nothing to do with
rte_intr_callback_unregister funcs once it return a value greater than 0
.So if the closure procedure has gone to the right exit point ,we should
set this value to 0.
Als
From: Wenxuan Wu
when testpmd startup with pf and vfs, it is ok when running, while
exiting, it will result in heap-free-after-use which means pf is
released but vf is still accessing.
Change the logic of func port_is_bonding_slave ,this func
eth_dev_info_get_print_err while pf is released would
From: Wenxuan Wu
when testpmd starup with pf and vfs, it is ok when running, while
exiting, it will result in heap-free-after-use which means pf is
released but vf is still accessing.
Change the logic of func port_is_bonding_slave ,this func
eth_dev_info_get_print_err while pf is released would
55 matches
Mail list logo
