Re: [PATCH v2 21/21] doc: add table for environment variables used by cnxk
22/02/2022 20:35, Nithin Dabilpuram: > Add table for environment variables used by cnxk drivers. > > Signed-off-by: Nithin Dabilpuram > --- > +.. table:: cnxk environment variables > + > + > +---+-+-+ > + | # | Variable name | Usage > | > + > +===+=+=+ > + | 1 | BPHY_INTR_MLOCK_DISABLE | When defined disables memory locking > in | > + | | | BPHY environment. > | > + > +---+-+-+ > + | 2 | ROC_CN10K_MBOX_TIMEOUT | When set, overrides MBOX timeout by > | > + | | ROC_MBOX_TIMEOUT| value in milli seconds. > | > + > +---+-+-+ > + | 3 | CN10K_ETH_SEC_IV_OVR| When set, overrides outbound inline > SA | > + | | | IV in CN10K. By default IV is > generated | > + | | | by HW. Format of variable is string > | > + | | | of comma separated one byte values as > | > + | | | for ex: "0x0, 0x10, 0x20, ..." > | > + > +---+-+-+ Using tables for such list is a bad idea. The source code is constrained in a small column, and the HTML rendering is constrained by the page width. I recommend switching to the definition list syntax which has a nice rendering for such definitions. Example: ``BPHY_INTR_MLOCK_DISABLE`` Disable memory locking in BPHY environment. See https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html#lists-and-quote-like-blocks
Re: [PATCH v2 21/21] doc: add table for environment variables used by cnxk
On Sat, Feb 26, 2022 at 2:52 PM Thomas Monjalon wrote: > > 22/02/2022 20:35, Nithin Dabilpuram: > > Add table for environment variables used by cnxk drivers. > > > > Signed-off-by: Nithin Dabilpuram > > --- > > +.. table:: cnxk environment variables > > + > > + > > +---+-+-+ > > + | # | Variable name | Usage > > | > > + > > +===+=+=+ > > + | 1 | BPHY_INTR_MLOCK_DISABLE | When defined disables memory > > locking in | > > + | | | BPHY environment. > > | > > + > > +---+-+-+ > > + | 2 | ROC_CN10K_MBOX_TIMEOUT | When set, overrides MBOX timeout by > > | > > + | | ROC_MBOX_TIMEOUT| value in milli seconds. > > | > > + > > +---+-+-+ > > + | 3 | CN10K_ETH_SEC_IV_OVR| When set, overrides outbound inline > > SA | > > + | | | IV in CN10K. By default IV is > > generated | > > + | | | by HW. Format of variable is string > > | > > + | | | of comma separated one byte values > > as | > > + | | | for ex: "0x0, 0x10, 0x20, ..." > > | > > + > > +---+-+-+ > > Using tables for such list is a bad idea. > The source code is constrained in a small column, > and the HTML rendering is constrained by the page width. > I recommend switching to the definition list syntax > which has a nice rendering for such definitions. > > Example: > ``BPHY_INTR_MLOCK_DISABLE`` >Disable memory locking in BPHY environment. Sure, You can remove this doc update while merging to the main tree. We will submit a separate patch for updating this document based on your suggestion. > > See > https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html#lists-and-quote-like-blocks > >
[PATCH] net/ark: support multi-port pkt generation
Added support for packet generation in
multi-port Arkville implementations. The packet
generator is a singleton within the device but is
capable of generating packets for any port within
one device.
Signed-off-by: John Miller
---
drivers/net/ark/ark_ethdev.c | 4 +++-
drivers/net/ark/ark_global.h | 1 +
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ark/ark_ethdev.c b/drivers/net/ark/ark_ethdev.c
index 230a1272e9..980e1a4a3b 100644
--- a/drivers/net/ark/ark_ethdev.c
+++ b/drivers/net/ark/ark_ethdev.c
@@ -441,6 +441,7 @@ ark_config_device(struct rte_eth_dev *dev)
* known state
*/
ark->start_pg = 0;
+ ark->pg_running = 0;
ark->pg = ark_pktgen_init(ark->pktgen.v, 0, 1);
if (ark->pg == NULL)
return -1;
@@ -562,7 +563,7 @@ eth_ark_dev_start(struct rte_eth_dev *dev)
if (ark->start_pg)
ark_pktchkr_run(ark->pc);
- if (ark->start_pg && (dev->data->port_id == 0)) {
+ if (ark->start_pg && !ark->pg_running) {
pthread_t thread;
/* Delay packet generatpr start allow the hardware to be ready
@@ -574,6 +575,7 @@ eth_ark_dev_start(struct rte_eth_dev *dev)
"starter thread\n");
return -1;
}
+ ark->pg_running = 1;
}
if (ark->user_ext.dev_start)
diff --git a/drivers/net/ark/ark_global.h b/drivers/net/ark/ark_global.h
index 49193ac5b3..3c3a712bc8 100644
--- a/drivers/net/ark/ark_global.h
+++ b/drivers/net/ark/ark_global.h
@@ -107,6 +107,7 @@ struct ark_adapter {
/* Pointers to packet generator and checker */
int start_pg;
+ uint16_t pg_running;
ark_pkt_gen_t pg;
ark_pkt_chkr_t pc;
ark_pkt_dir_t pd;
--
2.25.1
Re: [PATCH v2 21/21] doc: add table for environment variables used by cnxk
26/02/2022 10:37, Jerin Jacob: > On Sat, Feb 26, 2022 at 2:52 PM Thomas Monjalon wrote: > > > > 22/02/2022 20:35, Nithin Dabilpuram: > > > Add table for environment variables used by cnxk drivers. > > > > > > Signed-off-by: Nithin Dabilpuram > > > --- > > > +.. table:: cnxk environment variables > > > + > > > + > > > +---+-+-+ > > > + | # | Variable name | Usage > > > | > > > + > > > +===+=+=+ > > > + | 1 | BPHY_INTR_MLOCK_DISABLE | When defined disables memory > > > locking in | > > > + | | | BPHY environment. > > > | > > > + > > > +---+-+-+ > > > + | 2 | ROC_CN10K_MBOX_TIMEOUT | When set, overrides MBOX timeout > > > by | > > > + | | ROC_MBOX_TIMEOUT| value in milli seconds. > > > | > > > + > > > +---+-+-+ > > > + | 3 | CN10K_ETH_SEC_IV_OVR| When set, overrides outbound > > > inline SA | > > > + | | | IV in CN10K. By default IV is > > > generated | > > > + | | | by HW. Format of variable is > > > string | > > > + | | | of comma separated one byte > > > values as | > > > + | | | for ex: "0x0, 0x10, 0x20, ..." > > > | > > > + > > > +---+-+-+ > > > > Using tables for such list is a bad idea. > > The source code is constrained in a small column, > > and the HTML rendering is constrained by the page width. > > I recommend switching to the definition list syntax > > which has a nice rendering for such definitions. > > > > Example: > > ``BPHY_INTR_MLOCK_DISABLE`` > >Disable memory locking in BPHY environment. > > Sure, You can remove this doc update while merging to the main tree. > We will submit a separate patch for updating this document > based on your suggestion. It has been squashed in a commit adding the third variable, so yes it can be dropped and submitted again for the 3 variables.
[PATCH] sched: add parentheses to if clause
Add parentheses to 'if' clause, otherwise will enlarged the
chance of error return.
Fixes: 44c730b0e37971 ("sched: add PIE based congestion management")
Signed-off-by: Weiguo Li
---
lib/sched/rte_pie.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/lib/sched/rte_pie.c b/lib/sched/rte_pie.c
index cdb7bab697..51df403a25 100644
--- a/lib/sched/rte_pie.c
+++ b/lib/sched/rte_pie.c
@@ -18,10 +18,10 @@ rte_pie_rt_data_init(struct rte_pie *pie)
/* Allocate memory to use the PIE data structure */
pie = rte_malloc(NULL, sizeof(struct rte_pie), 0);
- if (pie == NULL)
+ if (pie == NULL) {
RTE_LOG(ERR, SCHED, "%s: Memory allocation fails\n",
__func__);
-
- return -1;
+ return -1;
+ }
}
pie->active = 0;
--
2.25.1
Re: [PATCH v4 3/3] net/ixgbe: Fix SFP detection and linking on hotplug
On 2/25/2022 8:50 PM, Jeff Daly wrote:
Currently the ixgbe driver does not ID any SFP except for the first one
plugged in. This can lead to no-link, or incorrect speed conditions.
For example:
* If link is initially established with a 1G SFP, and later a 1G/10G
multispeed part is later installed, then the MAC link setup functions are
never called to change from 1000BASE-X to 10GBASE-R mode, and the link
stays running at the slower rate.
* If link is initially established with a 1G SFP, and later a 10G only
module is later installed, no link is established, since we are still
trasnsmitting in 1000BASE-X mode to a 10GBASE-R only partner.
Refactor the SFP ID/setup, and link setup code, to more closely match the
flow of the mainline kernel driver which does not have these issues. In
that driver a service task runs periodically to handle these operations
based on bit flags that have been set (usually via interrupt or userspace
request), and then get cleared once the requested subtask has been
completed.
Fixes: af75078fece ("first public release")
Cc:sta...@dpdk.org
Signed-off-by: Stephen Douthit
Signed-off-by: Jeff Daly
Hi Jeff,
Can you please send new version of whole set, instead of only
one of the patches?
It is very hard to manage different versions of individual
patches in a set.
Re: [PATCH] sched: add parentheses to if clause
On Sat, 26 Feb 2022 22:55:30 +0800
Weiguo Li wrote:
> Add parentheses to 'if' clause, otherwise will enlarged the
> chance of error return.
>
> Fixes: 44c730b0e37971 ("sched: add PIE based congestion management")
>
> Signed-off-by: Weiguo Li
> ---
> lib/sched/rte_pie.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/lib/sched/rte_pie.c b/lib/sched/rte_pie.c
> index cdb7bab697..51df403a25 100644
> --- a/lib/sched/rte_pie.c
> +++ b/lib/sched/rte_pie.c
> @@ -18,10 +18,10 @@ rte_pie_rt_data_init(struct rte_pie *pie)
> /* Allocate memory to use the PIE data structure */
> pie = rte_malloc(NULL, sizeof(struct rte_pie), 0);
>
> - if (pie == NULL)
> + if (pie == NULL) {
> RTE_LOG(ERR, SCHED, "%s: Memory allocation fails\n",
> __func__);
> -
> - return -1;
> + return -1;
> + }
> }
>
> pie->active = 0;
This will make the test in test_pie.c fail.
The concept of passing NULL to the routine and expecting allocation
is bad idea because the allocated structure is never initialized.
Since rte_pie_rt_data_init(NULL) always returned -1.
It would make more sense to take out the rte_malloc().
And document it.
P.s: the routing should return a negative rte_errno instead of -1
as well.
[PATCH] ci: remove redundant drivers enabling
No need to explicitly enable drivers bus/vdev and mempool/ring.
bus/vdev is always enabled since
commit 2e33309ebe03 ("config: enable/disable drivers in Arm builds")
mempool/ring is always enabled since
commit 81c2337e044d ("build: make ring mempool driver mandatory")
Signed-off-by: Thomas Monjalon
---
.ci/linux-build.sh| 2 +-
devtools/test-meson-builds.sh | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/.ci/linux-build.sh b/.ci/linux-build.sh
index 67d68535e0..77c0d1976b 100755
--- a/.ci/linux-build.sh
+++ b/.ci/linux-build.sh
@@ -89,7 +89,7 @@ OPTS="$OPTS --default-library=$DEF_LIB"
OPTS="$OPTS --buildtype=debugoptimized"
OPTS="$OPTS -Dcheck_includes=true"
if [ "$MINI" = "true" ]; then
-OPTS="$OPTS -Denable_drivers=bus/vdev,mempool/ring,net/null"
+OPTS="$OPTS -Denable_drivers=net/null"
OPTS="$OPTS -Ddisable_libs=*"
fi
meson build --werror $OPTS
diff --git a/devtools/test-meson-builds.sh b/devtools/test-meson-builds.sh
index c07fd16fdc..a653b253cb 100755
--- a/devtools/test-meson-builds.sh
+++ b/devtools/test-meson-builds.sh
@@ -221,7 +221,7 @@ for c in gcc clang ; do
done
build build-mini cc skipABI $use_shared -Ddisable_libs=* \
- -Denable_drivers=bus/vdev,mempool/ring,net/null
+ -Denable_drivers=net/null
# test compilation with minimal x86 instruction set
# Set the install path for libraries to "lib" explicitly to prevent problems
--
2.34.1
[PATCH] build: try to get kernel version from kernel source
When building the kernel modules, try to get the kernel
version from the kernel sources first. This fixes the
kernel modules installation directory if the target kernel
version differs from the host kernel version, like for
CI build or when packaging for linux distributions.
Signed-off-by: Ferdinand Thiessen
---
kernel/linux/meson.build | 18 --
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/kernel/linux/meson.build b/kernel/linux/meson.build
index d8fb20c1c3..78f28ffb0c 100644
--- a/kernel/linux/meson.build
+++ b/kernel/linux/meson.build
@@ -10,17 +10,23 @@ install = not meson.is_cross_build()
cross_args = []
if not meson.is_cross_build()
-# native build
-kernel_version = run_command('uname', '-r', check: true).stdout().strip()
+# native target build
+kernel_version = run_command('uname', '-r').stdout().strip()
+if kernel_source_dir != ''
+# Try kernel release from sources first
+r = run_command('make', '-s', '-C', kernel_source_dir,
'kernelrelease', check: false)
+if r.returncode() == 0
+kernel_version = r.stdout().strip()
+endif
+else
+# use default path for native builds
+kernel_source_dir = '/lib/modules/' + kernel_version + '/source'
+endif
kernel_install_dir = '/lib/modules/' + kernel_version + '/extra/dpdk'
if kernel_build_dir == ''
# use default path for native builds
kernel_build_dir = '/lib/modules/' + kernel_version + '/build'
endif
-if kernel_source_dir == ''
-# use default path for native builds
-kernel_source_dir = '/lib/modules/' + kernel_version + '/source'
-endif
# test running make in kernel directory, using "make kernelversion"
make_returncode = run_command('make', '-sC', kernel_build_dir,
--
2.35.1
Re: [PATCH] sched: add parentheses to if clause
On Sat, 26 Feb 2022 09:31:37 -0800, Stephen Hemminger wrote:
> > Add parentheses to 'if' clause, otherwise will enlarged the
> > chance of error return.
> >
> > Fixes: 44c730b0e37971 ("sched: add PIE based congestion management")
> >
> > Signed-off-by: Weiguo Li
> > ---
> > lib/sched/rte_pie.c | 6 +++---
> > 1 file changed, 3 insertions(+), 3 deletions(-)
> >
> > diff --git a/lib/sched/rte_pie.c b/lib/sched/rte_pie.c
> > index cdb7bab697..51df403a25 100644
> > --- a/lib/sched/rte_pie.c
> > +++ b/lib/sched/rte_pie.c
> > @@ -18,10 +18,10 @@ rte_pie_rt_data_init(struct rte_pie *pie)
> > /* Allocate memory to use the PIE data structure */
> > pie = rte_malloc(NULL, sizeof(struct rte_pie), 0);
> >
> > - if (pie == NULL)
> > + if (pie == NULL) {
> > RTE_LOG(ERR, SCHED, "%s: Memory allocation fails\n",
> > __func__);
> > -
> > - return -1;
> > + return -1;
> > + }
> > }
> >
> > pie->active = 0;
>
> This will make the test in test_pie.c fail.
>
> The concept of passing NULL to the routine and expecting allocation
> is bad idea because the allocated structure is never initialized.
>
> Since rte_pie_rt_data_init(NULL) always returned -1.
> It would make more sense to take out the rte_malloc().
> And document it.
>
> P.s: the routing should return a negative rte_errno instead of -1
> as well.
>
Hi Stephen,
The 'rte_malloc' and null check is really misleading at the first sight...
Thanks for your suggestion!
-Weiguo
[PATCH v2] sched: remove useless malloc in pie data init
'rte_pie_rt_data_init(NULL)' is not expected, and it's ought to
fail when this happen. The malloc inside this funtion didn't work.
So remove the malloc otherwise will lead to a memory leak.
Fixes: 44c730b0e37971 ("sched: add PIE based congestion management")
Signed-off-by: Weiguo Li
---
v2:
* revise according to Stephen's suggestion.
---
lib/sched/rte_pie.c | 9 ++---
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/lib/sched/rte_pie.c b/lib/sched/rte_pie.c
index cdb7bab697..eed5c12b54 100644
--- a/lib/sched/rte_pie.c
+++ b/lib/sched/rte_pie.c
@@ -15,13 +15,8 @@ int
rte_pie_rt_data_init(struct rte_pie *pie)
{
if (pie == NULL) {
- /* Allocate memory to use the PIE data structure */
- pie = rte_malloc(NULL, sizeof(struct rte_pie), 0);
-
- if (pie == NULL)
- RTE_LOG(ERR, SCHED, "%s: Memory allocation fails\n",
__func__);
-
- return -1;
+ RTE_LOG(ERR, SCHED, "%s: Invalid addr for pie\n", __func__);
+ return -EINVAL;
}
pie->active = 0;
--
2.25.1
RE: [PATCH] net/ice: fix Tx offload path choice
> -Original Message-
> From: Liu, KevinX
> Sent: Friday, December 24, 2021 11:09 PM
> To: dev@dpdk.org
> Cc: Zhang, Qi Z ; Yang, SteveX ;
> Liu, KevinX ; sta...@dpdk.org
> Subject: [PATCH] net/ice: fix Tx offload path choice
>
> Testpmd forwards packets in checksum mode that it needs to calculate the
> checksum of each layer's protocol.
>
> When setting the hardware calculates the outer UDP checksum and the
> software calculates the outer IP checksum, the dev->tx_pkt_burst in
> ice_set_tx_function is set to ice_xmit_pkts_vec_avx2.
> The inner and outer UDP checksum of the tunnel packet after forwarding is
> wrong.The dev->tx_pkt_burst should be set to ice_xmit_pkts.
>
> The patch adds RTE_ETH_TX_OFFLOAD_OUTER_UDP_CKSUM to
> ICE_TX_NO_VECTOR_FLAGS,set dev->tx_pkt_burst to ice_xmit_pkts.After the
> tunnel packet is forwarded, the inner and outer UDP checksum is correct.
>
> At the same time, the patch of "net/ice: fix Tx Checksum offload" will cause
> interrupt errors in a special case that only inner IP and inner UDP checksum
> are
> set for hardware calculation.The patch is updating ICE_TX_NO_VECTOR_FLAGS,
> the problem can be solved, so I will restore the code modification of that
> patch.
>
> Fixes: e6b9d6411e91 ("app/testpmd: add SW L4 checksum in multi-segments")
> Fixes: 28f9002ab67f ("net/ice: add Tx AVX512 offload path")
> Fixes: 295968d17407 ("ethdev: add namespace")
> Fixes: 17c7d0f9d6a4 ("net/ice: support basic Rx/Tx")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Kevin Liu
> ---
> app/test-pmd/csumonly.c | 6 +--
> drivers/net/ice/ice_rxtx.c| 41 ++-
> drivers/net/ice/ice_rxtx_vec_common.h | 59 +--
> 3 files changed, 34 insertions(+), 72 deletions(-)
Please separate testpmd fix and pmd fix into two patches.
[PATCH] common/mlx5: fix missing default devargs initialization
Device arguments list is provided along with its identifier as part of
EAL arguments.
The arguments specified in the list are taken from it, and the rest is
initialized to the default values.
When no list is provided at all, all arguments should have been
initialized to their default values. However, they are mistakenly
initialized to zero which may be a valid value for some.
This patch initializes the default values before checking whether
arguments have been specified.
Fixes: 04054be4331c ("common/mlx5: refactor devargs management")
Signed-off-by: Michael Baum
Acked-by: Matan Azrad
---
drivers/common/mlx5/mlx5_common.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/common/mlx5/mlx5_common.c
b/drivers/common/mlx5/mlx5_common.c
index 94c303ce81..ef1604d223 100644
--- a/drivers/common/mlx5/mlx5_common.c
+++ b/drivers/common/mlx5/mlx5_common.c
@@ -326,8 +326,6 @@ mlx5_common_config_get(struct mlx5_kvargs_ctrl *mkvlist,
};
int ret = 0;
- if (mkvlist == NULL)
- return 0;
/* Set defaults. */
config->mr_ext_memseg_en = 1;
config->mr_mempool_reg_en = 1;
@@ -335,6 +333,8 @@ mlx5_common_config_get(struct mlx5_kvargs_ctrl *mkvlist,
config->dbnc = MLX5_ARG_UNSET;
config->device_fd = MLX5_ARG_UNSET;
config->pd_handle = MLX5_ARG_UNSET;
+ if (mkvlist == NULL)
+ return 0;
/* Process common parameters. */
ret = mlx5_kvargs_process(mkvlist, params,
mlx5_common_args_check_handler, config);
--
2.25.1
