On Wed, Jan 6, 2021 at 4:27 AM Ferruh Yigit wrote:
>
> On 12/14/2020 6:53 PM, Lance Richardson wrote:
> > Limit number of representor packets transferred per poll
> > to requested burst size.
> >
>
> Hi Lance,
>
> Can you please describe the impact of the chan
On Thu, Dec 17, 2020 at 9:08 AM Anatoly Burakov
wrote:
>
> From: Liang Ma
>
> Add a simple API to allow getting the monitor conditions for
> power-optimized monitoring of the RX queues from the PMD, as well as
> release notes information.
>
> Signed-off-by: Liang Ma
> Signed-off-by: Anatoly Bura
Fix mask to include all four bits of hardware packet type
field.
Fixes: 97b1db288dd0 ("net/bnxt: use table based packet type translation")
Cc: sta...@dpdk.org
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
di
On Wed, Sep 16, 2020 at 4:21 AM Juraj Linkeš wrote:
>
> From: Ruifeng Wang
>
> Expand vector PMD support to aarch32.
>
> Signed-off-by: Ruifeng Wang
> Acked-by: Ajit Khaparde
> ---
> drivers/net/bnxt/bnxt_rxq.h | 2 +-
> drivers/net/bnxt/bnxt_rxr.h | 2 +-
> drivers/net/bnxt/bnxt_txr.h | 2 +-
> > Hi Juraj,
> >
> > I might be missing something, but I don't
> > believe these changes are sufficient to
> > enable vector mode for ARM32. For one
> > thing, bnxt_receive_function() in bnxt_ethdev.c
> > would need to be changed to enable the
> > selection of the vector receive function.
>
> Hi L
With this change, the bnxt driver fails to initialize under testpmd:
Configuring Port 0 (socket 0)
Port 0 failed to enable Rx offload JUMBO_FRAME
Fail to configure port 0
EAL: Error - exiting with code: 1
It appears that the cause is this bit of code in bnxt_ethdev.c:
if (bp->eth_dev->da
ile (1) {
> cons = RING_CMP(cpr->cp_ring_struct, raw_cons);
> - rte_prefetch0(&cpr->cp_desc_ring[cons]);
> rxcmp = (struct rx_pkt_cmpl *)&cpr->cp_desc_ring[cons];
>
> if (!CMP_VALID(rxcmp, raw_cons, cpr->cp_ring_struct))
> --
> 2.27.0
>
Acked-by: Lance Richardson
> First the code cause problem in the driver looks in another place, following
> in
> 'bnxt_mtu_set_op()':
>
>if (new_mtu > RTE_ETHER_MTU) {
>bp->flags |= BNXT_FLAG_JUMBO;
>bp->eth_dev->data->dev_conf.rxmode.offloads |=
>
Fixes: f6870a7ed6b3 ("app/testpmd: fix max Rx packet length for VLAN packet")
> Cc: sta...@dpdk.org
>
> Reported-by: Andrew Rybchenko
> Signed-off-by: Thomas Monjalon
> ---
Acked-by: Lance Richardson
On Fri, Nov 6, 2020 at 12:54 PM Thomas Monjalon wrote:
> Bruce added the doc and the command in test-meson-builds.sh.
> Are we fine now? Should we mark this patch as rejected?
That's fine with me, thanks for following up.
Lance
emove deprecated coherent IO memory barriers")
Fixes: bfc1d45875e2 ("net/bnxt: fix performance for Arm")
Signed-off-by: Lance Richardson
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ring.h | 37 +++-
1 file changed, 20 insertions(+), 17 deleti
: set padding flags in Rx descriptor")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kumar Khaparde
Reviewed-by: Somnath Kotur
---
drivers/net/bnxt/bnxt_rxr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_r
On Wed, Dec 9, 2020 at 8:00 AM Juraj Linkeš wrote:
>
> From: Ruifeng Wang
>
> Expand vector PMD support to aarch32.
>
> Signed-off-by: Ruifeng Wang
> Acked-by: Ajit Khaparde
> ---
> drivers/net/bnxt/bnxt_rxq.h | 2 +-
> drivers/net/bnxt/bnxt_rxr.h | 2 +-
> drivers/net/bnxt/bnxt_txr.h | 2 +-
>
r == 'arm' and dpdk_conf.get('RTE_ARCH_64')
> sources += files('bnxt_rxtx_vec_neon.c')
> endif
> --
> 2.20.1
>
With this change, there should be no need for
http://patchwork.dpdk.org/patch/84864/
Reviewed-by: Lance Richardson
--
This
On Wed, Dec 9, 2020 at 2:28 PM Ajit Khaparde wrote:
>
> Change the ring logic so that the index increments
> unbounded and mask it only when needed.
>
> Modify the existing macros so that the index is not masked.
> Add a new macro RING_IDX() to mask it only when needed.
>
> Signed-off-by: Ajit Kha
h | 14 +--
> drivers/net/bnxt/bnxt_rxtx_vec_neon.c | 16 ++--
> drivers/net/bnxt/bnxt_rxtx_vec_sse.c| 16 ++--
> drivers/net/bnxt/bnxt_txr.c | 51 ++-
> drivers/net/bnxt/bnxt_txr.h | 6 +-
> 12 files changed, 179 insertions(+), 135 deletions(-)
>
at;
> - uint8_t flow_xstat;
> uint16_tmax_num_kflows;
> uint16_ttx_cfa_action;
> };
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
drivers/net/bnxt/bnxt_rxr.c| 12
> drivers/net/bnxt/bnxt_rxr.h| 2 +-
> drivers/net/bnxt/bnxt_vnic.c | 4 +--
> drivers/net/bnxt/tf_ulp/bnxt_ulp.c | 2 +-
> 9 files changed, 70 insertions(+), 69 deletions(-)
>
Acked-by: Lance Richardson
le to initialize def_cp_lock\n");
> + return err;
> + }
>
> err = pthread_mutex_init(&bp->health_check_lock, NULL);
> if (err)
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
%d, device:%u\n",
> parms->session_id.internal.domain,
> parms->session_id.internal.bus,
> parms->session_id.internal.device);
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
; + rte_mbuf_raw_free(mbuf);
> }
>
> return 0;
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
t bnxt_hwrm_get_dflt_vnic_svif(struct bnxt *bp, uint16_t fid,
> uint16_t *vnic_id, uint16_t *svif);
> int bnxt_hwrm_parent_pf_qcfg(struct bnxt *bp);
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
G(ERR,
> - "bnxt_init_one_rx_ring failed!\n");
> + PMD_DRV_LOG(ERR, "bnxt_init_one_rx_ring failed!\n");
> bnxt_rx_queue_release_op(rxq);
> rc = -ENOMEM;
> goto err_out;
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
_03f0] = 2,
> @@ -227,6 +232,7 @@ uint16_t ulp_class_sig_tbl[BNXT_ULP_CLASS_SIG_TBL_MAX_SZ]
> = {
> [BNXT_ULP_CLASS_HID_01b4] = 216
> };
>
> +/* Array for the proto matcher list */
> struct bnxt_ulp_class_match_info ulp_class_match_list[] = {
> [1] = {
> .class_hid = BNXT_ULP_CLASS_HID_0138,
> --
> 2.21.1 (Apple Git-122.3)
>
Acked-by: Lance Richardson
On Thu, Sep 17, 2020 at 10:05 AM Olivier Matz wrote:
>
> Hi Lance,
>
> On Mon, Aug 24, 2020 at 04:11:45PM -0400, Lance Richardson wrote:
> > I was looking for some clarification regarding how rx checksum
> > flags should be set for tunnel packets having both inner and
t;net/bnxt: optimize Rx processing")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kumar Khaparde
Reviewed-by: Somnath Kotur
Cc: sta...@dpdk.org
---
drivers/net/bnxt/bnxt_rxr.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_rxr.c b/d
Limit number of representor packets transferred per poll
to requested burst size.
Fixes: 6dc83230b43b ("net/bnxt: support port representor data path")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kumar Khaparde
Reviewed-by: Somnath Kotur
Cc: sta...@dpdk.org
---
driver
("net/bnxt: optimize Rx processing")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kumar Khaparde
Reviewed-by: Somnath Kotur
Cc: sta...@dpdk.org
---
v2 - Fixed typo in commit log.
drivers/net/bnxt/bnxt_rxr.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/d
On Mon, Dec 14, 2020 at 12:41 PM Lance Richardson
wrote:
>
> On Thu, Sep 17, 2020 at 10:05 AM Olivier Matz wrote:
> >
> > Hi Lance,
> >
> > On Mon, Aug 24, 2020 at 04:11:45PM -0400, Lance Richardson wrote:
> > > I was looking for some clarification regardin
From: Lance Richardson
Fixes for fallback mbuf allocation logic.
- Preserve raw (unmasked) producer index.
- Iterate over all processed descriptors (representor and
non-representor) when checking allocation status.
- Invoke fallback allocation logic when an allocation
failure
From: Lance Richardson
Rx outer UDP checksum offload has been supported for
some time, but this has not been advertised in offload
capability flags. Fix this, and allow vector mode
receive to be enabled when DEV_RX_OFFLOAD_OUTER_UDP_CKSUM
is requested.
Fixes: 04a681426d49 ("net/bnxt: f
Use boolean AND operator instead of bitwise operator.
Coverity issue: 323488
Fixes: b42c15c83e88 ("net/bnxt: support trusted VF")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
di
Make SIMD initialization code less verbose by using appropriate
intrinsics when all lanes of a vector are initialized to the
same value.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_rxtx_vec_neon.c | 58 +++
drivers/net/bnxt
Mark "BSD nic_uio", "Usage doc", and "Perf doc" as supported
for the bnxt PMD.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kumar Khaparde
---
doc/guides/nics/features/bnxt.ini | 3 +++
1 file changed, 3 insertions(+)
diff --git a/doc/guides/nics/featur
Return correct value for rx_drop_en. Add per-queue field to
track rx_drop_en configuration.
Fixes: 2fc201884be8 ("net/bnxt: support rxq/txq get information")
Cc: sta...@dpdk.org
Signed-off-by: Lance Richardson
Reviewed-by: Somnath Kotur
Reviewed-by: Ajit Khaparde
---
driver
Return current offloads in rxq_info_get()/txq_info_get().
Fixes: 2fc201884be8 ("net/bnxt: support rxq/txq get information")
Cc: sta...@dpdk.org
Signed-off-by: Lance Richardson
Reviewed-by: Somnath Kotur
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ethdev.c | 2 ++
1 file
On Tue, Sep 22, 2020 at 1:30 PM Lance Richardson
wrote:
>
Apologies for not sending a cover letter. I can resend if needed.
Lance
Add meson cross files for building i686 targets using gcc on x86_64
linux hosts.
Uusage example:
meson --cross-file config/x86/cross-i686-linux-gcc build-i686
ninja -C build-i686
Signed-off-by: Lance Richardson
---
Red Hat distros use a different name for the 32-bit pkg-config
command
On Fri, Sep 25, 2020 at 5:13 AM Bruce Richardson
wrote:
>
> On Thu, Sep 24, 2020 at 12:37:42PM -0400, Lance Richardson wrote:
> > Add meson cross files for building i686 targets using gcc on x86_64
> > linux hosts.
> >
> > Uusage example:
> >
> > meso
On Fri, Sep 25, 2020 at 9:44 AM Bruce Richardson
wrote:
>
> On Fri, Sep 25, 2020 at 09:27:25AM -0400, Lance Richardson wrote:
> > On Fri, Sep 25, 2020 at 5:13 AM Bruce Richardson
> > wrote:
> > >
> > > On Thu, Sep 24, 2020 at 12:37:42PM -0400, Lance Richa
Descriptor valid mask should be 64-bit all ones, use
appropriate initializer type (unsigned long long) to
obtain correct value in 32-bit mode.
Fixes: deae85145c64 ("net/bnxt: handle multiple packets per loop in vector Rx")
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kuma
On Tue, Sep 29, 2020 at 5:01 PM Ajit Khaparde
wrote:
>
> Vector mode is disabled when jumbo frames are enabled.
> This usage guide was reporting it incorrectly.
>
> Fixes: 1adaf0e0f2ee ("doc: update bnxt guide")
> Cc:sta...@dpdk.org
>
> Signed-off-by: Ajit Khaparde
> ---
> doc/guides/nics/bnxt.r
this offload is only supported in vector mode at
this time, support in the features list is indicated
as "partial".
Signed-off-by: Lance Richardson
---
doc/guides/nics/features/bnxt.ini | 1 +
drivers/net/bnxt/bnxt_ethdev.c| 8
drivers/net/bnxt/bnxt_rxtx_vec_n
this offload is only supported in vector mode at
this time, support in the features list is indicated
as "partial".
Signed-off-by: Lance Richardson
---
- v2:
Fixed mangling of advertised tx offload capabilities.
doc/guides/nics/features/bnxt.ini | 1 +
drivers/net/bnxt/bnx
On Tue, Oct 6, 2020 at 4:30 AM Thomas Monjalon wrote:
>
> 25/09/2020 15:27, Lance Richardson:
> > Bruce Richardson wrote:
> > > On Thu, Sep 24, 2020 at 12:37:42PM -0400, Lance Richardson wrote:
> > > > Add meson cross files for building i686 targets using
-off-by: Lance Richardson
---
- v2:
Fixed mangling of advertised tx offload capabilities.
- v3:
Reduced duplicated code.
Enable for non-vector transmit path as well.
doc/guides/nics/features/bnxt.ini | 1 +
drivers/net/bnxt/bnxt_ethdev.c | 11 ---
drivers/net/bnxt
Remove unused structure fields and macro definitions.
Signed-off-by: Lance Richardson
Reviewed-by: Ajit Kumar Khaparde
Reviewed-by: Somnath Kotur
---
drivers/net/bnxt/bnxt_rxq.h | 5 --
drivers/net/bnxt/bnxt_rxr.h | 135
drivers/net/bnxt/bnxt_txq.h | 7
The fast mbuf free offload for non-vector mode requires
additional checks in order to handle long tx buffer
descriptors, so dedicated functions are needed for
vector- and non-vector-modes.
Fixes: 103169df2880 ("net/bnxt: support fast mbuf free")
Signed-off-by: Lance Richardson
Reviewe
I was looking for some clarification regarding how rx checksum
flags should be set for tunnel packets having both inner and outer
IP/L4 headers.
Based on comments in rte_mbuf_core.h, it seems to me. that the
inner (encapsulated) IP header checksum status should determine
which of these goes into o
On Mon, Aug 24, 2020 at 4:11 PM Lance Richardson
wrote:
> The IP header checksum status for the outer IP header should determine
> whether this flag is set in ol_flags:
> PKT_RX_EIP_CKSUM_BAD
A simpler question might be simply "how is PKT_RX_EIP_CKSUM_BAD"
intended to b
On Tue, Aug 25, 2020 at 10:39 AM Lance Richardson
wrote:
>
> On Mon, Aug 24, 2020 at 4:11 PM Lance Richardson
> wrote:
>
> > The IP header checksum status for the outer IP header should determine
> > whether this flag is set in ol_flags:
> > PKT_RX_EIP_CKSUM_BA
Fixes and optimizations to improve bnxt vector mode
performance and functionality.
Lance Richardson (12):
net/bnxt: fix burst mode get for Arm
net/bnxt: fix rxq/txq get information
net/bnxt: use appropriate type for Rx mbuf ring
net/bnxt: require async cq for vector mode
net/bnxt
Return correct values for Rx/Tx offloads and for rx_drop_en.
Fixes: 2fc201884be8 ("net/bnxt: support rxq/txq get information")
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
Cc: sta...@dpdk.org
---
drivers/net/bnxt/bnxt_ethdev.c | 4 +++-
1 file changed, 3 insert
Transmit and receive burst mode get operations incorrectly return
"Vector SSE" on ARM64 platforms, change to return "Vector Neon"
instead.
Fixes: 3983583414 ("net/bnxt: support NEON")
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
Cc: sta...@
fixed maximum burst
size to trigger bulk receive buffer allocation.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxq.c | 6 +++--
drivers/net/bnxt/bnxt_rxtx_vec_common.h | 10 +
drivers/net/bnxt/bnxt_rxtx_vec_neon.c | 29
Kotur
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_ethdev.c| 6 ++--
drivers/net/bnxt/bnxt_reps.c | 21 +++---
drivers/net/bnxt/bnxt_ring.c | 4 +--
drivers/net/bnxt/bnxt_rxq.c | 14 -
drivers/net/bnxt/bnxt_rxr.c | 41
Don't allocate extra completion queue entries for aggregation
ring when aggregation ring will not be used.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_ethdev.c | 11 +--
drivers/net/bnxt/bnxt_rxr.c| 21 +++--
2
.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt.h | 19 ++--
drivers/net/bnxt/bnxt_ethdev.c| 2 +-
drivers/net/bnxt/bnxt_rxtx_vec_neon.c | 121 +++---
drivers/net/bnxt/bnxt_rxtx_vec_sse.c | 116
Use table to translate receive descriptor status flags to
rte_mbuf ol_flags values.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxr.c | 166 --
drivers/net/bnxt/bnxt_rxr.h | 6 +
drivers/net/bnxt
Use table-based method for translating receive packet descriptor
flags into rte_mbuf packet type values.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxr.c | 127 --
drivers/net/bnxt/bnxt_rxr.h | 2
Increase the maximum supported burst size for the bnxt vector
mode PMD from 32 to 64.
Reviewed-by: Kalesh Anakkur Purayil
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxtx_vec_common.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions
Process four receive descriptors per inner loop in vector mode
burst receive functions.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_ethdev.c | 2 +-
drivers/net/bnxt/bnxt_rxq.c | 3 +-
drivers/net/bnxt/bnxt_rxq.h
Simplify and optimize receive mbuf allocation function used
by the vector mode PMDs.
Reviewed-by: Ajit Kumar Khaparde
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxtx_vec_common.h | 40 ++
drivers/net/bnxt/bnxt_rxtx_vec_neon.c | 70 -
drivers
Improve performance of vector burst transmit function by processing
multiple packets per inner loop iteration.
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_rxtx_vec_sse.c | 96 +++-
1 file changed, 66 insertions(+), 30 deletions(-)
diff --git a/drivers/net
On Fri, Sep 11, 2020 at 11:03 AM Ferruh Yigit wrote:
>
> On 9/9/2020 4:52 PM, Lance Richardson wrote:
> > Disable support for vector mode when async completions can be placed
> > in a receive completion ring and change the default for all platforms
> > to use a dedicat
On Fri, Sep 11, 2020 at 11:19 AM Ferruh Yigit wrote:
>
> On 9/9/2020 4:53 PM, Lance Richardson wrote:
> > Increase the maximum supported burst size for the bnxt vector
> > mode PMD from 32 to 64.
>
> What is the motivation here? Like does it improve the performance? If
On Mon, Sep 14, 2020 at 6:03 PM Ferruh Yigit wrote:
>
> Hi Lance, Ajit, Somnath,
>
> Raslan reported a build error for PPC architecture, this happens because
> static inline functions in 'bnxt_rxtx_vec_common.h' are using
> 'rxq->rxrearm_nb' & 'rxq->rxrearm_start' which are not defined for PPC.
>
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
On Fri, Sep 11, 2020 at 10:41 AM Ferruh Yigit wrote:
>
> > - qinfo->conf.rx_drop_en = 0;
> > + qinfo->conf.rx_drop_en = 1;
>
> Why 0 is wrong but 1 is correct?
>
> Technically 'rx_drop_en' is a user configuration, which is set via
> 'rte_eth_rx_queue_setup()' API.
>
> bnxt seems not honori
Fix to use correct value offset for PCI function stats.
Fixes: 5f9374de2a3a ("net/bnxt: add PCI function stats to extended stats")
Reviewed-by: Ajit Kumar Khaparde
Cc: sta...@dpdk.org
Cc: ajit.khapa...@broadcom.com
Signed-off-by: Lance Richardson
---
drivers/net/bnxt/bnxt_stats.c
201 - 270 of 270 matches
Mail list logo
