> Are you sure you want "offload_zxdh" prefix for each log, instead of
> shorter 'zxdh' one?
>> +__func__, __VA_ARGS__)
>> +
>> +extern int zxdh_logtype_driver;
>> +#define RTE_LOGTYPE_ZXDH_DRIVER zxdh_logtype_driver
>> +#define PMD_DRV_LOG(level, ...) \
>> +RTE_LOG_LINE_PREFIX(level,
Snipped
>
> > +struct topology_config {
> > +#ifdef RTE_EAL_HWLOC_TOPOLOGY_PROBE
> > + hwloc_topology_t topology;
> > +#endif
> > +
> > + /* domain count */
> > + uint16_t l1_count;
> > + uint16_t l2_count;
> > + uint8_t l3_count;
> > + uint8_t io_count;
> > +
> > + /*
> From: Vipin Varghese [mailto:vipin.vargh...@amd.com]
> Sent: Wednesday, 30 October 2024 06.42
>
> Introduce topology aware lcore mapping into lcore API.
> With higher core density, more and more cores are categorized
> into various chiplets based on IO (memory and PCIe) and
> Last Level Cache (m
Setting the cpu affinity of the child thread from the parent thread is
racy when using pthread_setaffinity_np, as the child thread may start
running and initialize before affinity is set.
On the other hand, setting the cpu affinity from the child thread itself
may fail, so the parent thread waits
Add support for the following EEPROM function callbacks:
.get_eeprom_len
Get the maximum size of the device EEPROM data.
.get_eeprom
Get the device EEPROM data at a certain offset and length.
.set_eeprom
Set the device EEPROM data at a certain offset and length.
Note that for an nfp NIC, the "d
This patch series fix one problem imported by mistake and add support of
two new APIs.
---
v4:
* Also update the 'nfp.ini' file.
v3:
* Adapt to the multiple PF firmware.
v2:
* Add one missing commit.
---
Chaoyong He (4):
net/nfp: fix port index problem
net/nfp: extract function to check physi
Implement the necessary functions to allow user to visually identify a
physical port associated with a netdev by blinking an LED on that port.
Signed-off-by: James Hershaw
Signed-off-by: Chaoyong He
---
doc/guides/nics/features/nfp.ini | 1 +
.../net/nfp/flower/nfp_flower_represen
Fix one port index problem imported by mistake.
Fixes: 97b6825d9a7b ("net/nfp: extract function to allocate PHY")
Fixes: fd1ec7bc8f0a ("net/nfp: extract function to initialize PF")
Fixes: bb9f9fdcbe00 ("net/nfp: extract function to allocate PF")
Fixes: c8e29c168c20 ("net/nfp: extract function to a
Extract a helper function to check the physical representor port.
Signed-off-by: Chaoyong He
---
.../net/nfp/flower/nfp_flower_representor.c| 18 --
.../net/nfp/flower/nfp_flower_representor.h| 1 +
2 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/drivers/
Snipped
>
> On Wed, 30 Oct 2024 11:11:31 +0530
> Vipin Varghese wrote:
>
> > topo_cnfg.l3[j] = (struct core_domain_mapping *)
> > + malloc(sizeof(struct
> > core_domain_mapping));
>
> Minor nit, no need to cast return value from malloc
Snipped
> > ---
>
>
> Test fails on ARM platform:
>
>
> lcore 19, socket 0, role RTE, cpuset 19
> Control thread running successfully
> INFO: lcore count topology: none (12), io (1), l3 (1), l2 (1), l1 (1).
> INFO: worker lcore count topology: none (11), io (1), l3 (1), l2 (1), l1 (1).
> INFO:
Snipped
>
> On Wed, 30 Oct 2024 11:11:31 +0530
> Vipin Varghese wrote:
>
> > + if (topo_cnfg.io) {
> > + free(topo_cnfg.io);
> > + topo_cnfg.io = NULL;
> > + }
>
> No need to check for NULL before calling free.
Thank you Stepehen on this, the reason why we speci
Snipped
>
> > + /* domain count */
> > + uint16_t l1_count;
> > + uint16_t l2_count;
> > + uint8_t l3_count;
> > + uint8_t io_count;
>
> Make them all uint16_t, the space is there already.
Thank you, fixing this in v4
Hi thomas,
I submitted a patch on October 29 and haven't received response.
Wondering if any modifications are needed? Could you provide an update on the
review?
Thanks. Looking forward to hearing from you.
Best regards,
>> net/zxdh: add zxdh get device backend infos
>> net/zxdh: add configure zxdh intr implementation
>> net/zxdh: add zxdh dev infos get ops
>> net/zxdh: add zxdh dev configure ops
>>
> Hi Junlong,
> I can see not all of the eth_dev_ops implemented, and datapath not
> implemented, so drive
Fix some bugs in txgbe/ngbe, and optimize the Tx flow.
v2 -> v3:
- Reset Tx desc error statistics in .stats_reset().
- Add comment for Tx desc error count.
v1 -> v2:
- Fix build errors on ARM platform.
Jiawen Wu (13):
net/txgbe: fix swfw mbox failure
net/txgbe: fix VF-PF mbox interrupt
net
There was a incorrect bit to define TXGBE_ICRMISC_VFMBX that prevents the
interrupt from being handled correctly.
Fixes: a6712cd029a4 ("net/txgbe: add PF module init and uninit for SRIOV")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/base/txgbe_regs.h | 2 +-
1 file change
Dear Ferruh,
Thank you for your comments. I will make the changes you suggested. Our DPDK
PMD is developed based on our company’s Linux kernel drivers r8125 and r8126,
which are more comprehensive and detailed. To facilitate synchronized
maintenance, I have retained the CamelCase.
Best Regards
Drv_load bit will be reset to default 0 after hardware LAN reset,
reconfigure it to inform firmware that driver is loaded. And set it to 0
when device is closed.
Fixes: b1f596677d8e ("net/txgbe: support device start")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/txgbe_ethd
The hardware does not support outer UDP checksum for tunnel packets.
It's wrong to claim this Tx offload capability, so fix it.
Bugzilla ID: 1529
Fixes: b950203be7f1 ("net/txgbe: support VXLAN-GPE")
Fixes: 295968d17407 ("ethdev: add namespace")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
d
Snipped
> > > >
> > > > I see compilation failure on ARM platforms due to missing header
> > > > include.
> > > >
> > > > ../examples/helloworld/main.c: In function 'parse_topology':
> > > > ../examples/helloworld/main.c:83:13: error: implicit declaration
> > > > of function 'strtoul'; did you me
Enable Tdm_desc_chk then handle the interrupt of TDM desc error. If it's a
non-fatal error, clear the error to reset the queue. If it's a fatal error,
require users to manually restart the port. This flow prevents the hardware
from PCIe pending due to Tx hang, resulting in a reboot to recover.
But
Add checking of the Tx packet length to avoid TDM fatal error as far as
possible. Set the pkt_len=1518 for invalid packet in simple Tx code path,
and drop it directly in featured Tx code path.
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/txgbe_rxtx.c | 33 +
dr
There is a hardware limitation that Rx ring config register is not
writable when Rx ring is enabled, i.e. the NGBE_RXCFG_ENA bit is set.
But disabling the ring when there is traffic will cause ring get stuck.
So restrict the configuration of VLAN strip offload only if device is
started.
Fixes: 59b
When interrupt is legacy or MSI mode, shared interrupt may cause the
interrupt cannot be re-enabled. So fix to read the shared interrupt.
Fixes: b9246b8fa280 ("net/ngbe: support link update")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/ngbe_ethdev.c | 13 +
1 f
Count the number of packets not sent due to Tx descriptor error.
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/txgbe_ethdev.c | 13 +
drivers/net/txgbe/txgbe_rxtx.c | 6 ++
drivers/net/txgbe/txgbe_rxtx.h | 1 +
3 files changed, 20 insertions(+)
diff --git a/drivers/net/tx
Count the number of packets not sent due to Tx descriptor error.
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/ngbe_ethdev.c | 13 +
drivers/net/ngbe/ngbe_rxtx.c | 2 ++
drivers/net/ngbe/ngbe_rxtx.h | 1 +
3 files changed, 16 insertions(+)
diff --git a/drivers/net/ngbe/ngbe_et
Add checking of the Tx packet length to avoid TDM fatal error as far as
possible. Set the pkt_len=1518 for invalid packet in simple Tx code path,
and drop it directly in featured Tx code path. Althrough the hardware
does not support TDM desc check.
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/n
When link status changes, there is a probability that no more packets
can be received on the port, due to hardware defects. These MAC Rx
registers should be reconfigured to fix this problem.
Fixes: b9246b8fa280 ("net/ngbe: support link update")
Fixes: a7c5f95ed9c2 ("net/ngbe: reconfigure MAC Rx wh
Drv_load bit will be reset to default 0 after hardware LAN reset,
reconfigure it to inform firmware that driver is loaded. And set it to 0
when device is closed.
Fixes: 3518df5774c7 ("net/ngbe: support device start/stop")
Fixes: cc63194e89cb ("net/ngbe: support close and reset device")
Cc: sta...@
There is a unknown bug that the register TXGBE_MNGMBX cannot be written
in the loop, when DPDK is built with GCC high version. Access any register
before write TXGBE_MNGMBX can fix it.
Bugzilla ID: 1531
Fixes: 35c90ecccfd4 ("net/txgbe: add EEPROM functions")
Cc: sta...@dpdk.org
Signed-off-by: Jia
>> +uint16_t zxdh_vport_to_vfid(union zxdh_virport_num v)
>> +{
>> +/* epid > 4 is local soft queue. return 1192 */
>> +if (v.epid > 4)
>> +return 1192;
>> +if (v.vf_flag)
>> +return v.epid * 256 + v.vfid;
>> +else
>> +return (v.epid * 8 + v.pfid) + 1152;
>>
>> err_zxdh_init:
>> +zxdh_bar_msg_chan_exit();
>>
> Should 'zxdh_bar_msg_chan_exit()' called during zxdh_eth_dev_uninit()?
Yes, it is. I forget it.
I will put it in the zxdh_eth_dev_uninit().
Thanks.
On Sat, Nov 2, 2024 at 8:58 PM Stephen Hemminger
wrote:
>
> On Sat, 2 Nov 2024 11:08:39 +0100
> David Marchand wrote:
>
> > Setting the cpu affinity of the child thread from the parent thread is
> > racy when using pthread_setaffinity_np, as the child thread may start
> > running and initialize
Acked-by: Chengwen Feng
On 2024/11/3 19:25, David Marchand wrote:
> Setting the cpu affinity of the child thread from the parent thread is
> racy when using pthread_setaffinity_np, as the child thread may start
> running and initialize before affinity is set.
>
> On the other hand, setting the c
> > net_af_packet PMD multi "queue" support relies on Linux FANOUT
> capability.
> > Linux FANOUT is a SW based load balancer that is similar to HW RSS
> which
> > is more common for DPDK PMDs. Instead of multiple HW descriptor queues,
> > AF PACKET uses multiple sockets.
> > HW RSS will typically
36 matches
Mail list logo
