net/mlx5: probe LAG representor with PF1 PCI address
PATCH 1/1 net/mlx5: probe LAG representor with PF1 PCI address
In case of bonding, orchestrator wants to use same devargs for LAG and non-LAG scenario, to probe representor on PF1 using PF1 PCI address like "<DBDF_PF1>,representor=pf1vf[0-3]".
This patch changes PCI address check policy to allow PF1 PCI address for representors on PF1.
Note: detaching PF0 device can't remove representors on PF1. It's recommended to use primary(PF0) PCI address to probe representors on both PFs.
--- drivers/net/mlx5/linux/mlx5_os.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/net/mlx5/linux/mlx5_os.c +5 -2
@@ -1875,11 +1875,14 @@ mlx5_device_bond_pci_match(const struct ibv_device *ibv_dev, tmp_str);break ; } -/* Match PCI address. */ +/* Match PCI address, allows BDF0+pfx or BDFx+pfx. */ if (pci_dev->domain == pci_addr.domain && pci_dev->bus == pci_addr.bus && pci_dev->devid == pci_addr.devid && - pci_dev->function + owner == pci_addr.function) + ((pci_dev->function ==0 && + pci_dev->function + owner == pci_addr.function) || + (pci_dev->function == owner && + pci_addr.function == owner))) pf = info.port_name;/* Get ifindex. */ snprintf(tmp_str, sizeof(tmp_str),
Signature · 1 line
2.25.1
Hi, Xuemig
This patch looks like a fix - could you, please add the "Fixes:" tag and modify the headline?
With best regards,
Slava
Quoted 49 lines
-----Original Message-----
From: Xueming Li <[email protected]>
Sent: Thursday, April 22, 2021 10:25
Cc: [email protected]; Xueming(Steven) Li <[email protected]>; Matan Azrad <[email protected]>; Shahaf Shuler <[email protected]>; Slava Ovsiienko <[email protected]>
Subject: [PATCH] net/mlx5: probe LAG representor with PF1 PCI addressIn case of bonding, orchestrator wants to use same devargs for LAG and non- LAG scenario, to probe representor on PF1 using PF1 PCI address like "<DBDF_PF1>,representor=pf1vf[0-3]".
This patch changes PCI address check policy to allow PF1 PCI address for representors on PF1.
Note: detaching PF0 device can't remove representors on PF1. It's recommended to use primary(PF0) PCI address to probe representors on both PFs.
Signed-off-by Xueming Li <[email protected]>--- drivers/net/mlx5/linux/mlx5_os.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)drivers/net/mlx5/linux/mlx5_os.c +5
-2@@ -1875,11 +1875,14 @@ mlx5_device_bond_pci_match(const structibv_device *ibv_dev, tmp_str);break ; } -/* Match PCI address. */ +/* Match PCI address, allows BDF0+pfx or BDFx+pfx. */ if (pci_dev->domain == pci_addr.domain && pci_dev->bus == pci_addr.bus && pci_dev->devid == pci_addr.devid && - pci_dev->function + owner == pci_addr.function) + ((pci_dev->function ==0 && + pci_dev->function + owner == pci_addr.function) || + (pci_dev->function == owner && + pci_addr.function == owner))) pf = info.port_name;/* Get ifindex. */ snprintf(tmp_str,
sizeof(tmp_str),
--
2.25.1
PATCH v2 1/1 net/mlx5: fix LAG representor probe on PF1 PCI
In reply to Xueming LiIn case of bonding, orchestrator wants to use same devargs for LAG and non-LAG scenario to probe representor on PF1 using PF1 PCI address like "<DBDF_PF1>,representor=pf1vf[0-3]".
This patch changes PCI address check policy to allow PF1 PCI address for representors on PF1.
Note: detaching PF0 device can't remove representors on PF1. It's recommended to use primary(PF0) PCI address to probe representors on both PFs.
--- drivers/net/mlx5/linux/mlx5_os.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/net/mlx5/linux/mlx5_os.c +5 -2
@@ -1877,11 +1877,14 @@ mlx5_device_bond_pci_match(const struct ibv_device *ibv_dev, tmp_str);break ; } -/* Match PCI address. */ +/* Match PCI address, allows BDF0+pfx or BDFx+pfx. */ if (pci_dev->domain == pci_addr.domain && pci_dev->bus == pci_addr.bus && pci_dev->devid == pci_addr.devid && - pci_dev->function + owner == pci_addr.function) + ((pci_dev->function ==0 && + pci_dev->function + owner == pci_addr.function) || + (pci_dev->function == owner && + pci_addr.function == owner))) pf = info.port_name;/* Get ifindex. */ snprintf(tmp_str, sizeof(tmp_str),
Signature · 1 line
2.25.1
Re: [dpdk-dev] [PATCH v2] net/mlx5: fix LAG representor probe on PF1 PCI
Quoted 15 lines
-----Original Message-----
From: dev <[email protected]> On Behalf Of Xueming Li
Sent: Monday, May 10, 2021 16:14
Cc: [email protected]; Xueming(Steven) Li <[email protected]>; [email protected]
Subject: [dpdk-dev] [PATCH v2] net/mlx5: fix LAG representor probe on PF1 PCIIn case of bonding, orchestrator wants to use same devargs for LAG and non- LAG scenario to probe representor on PF1 using PF1 PCI address like "<DBDF_PF1>,representor=pf1vf[0-3]".
This patch changes PCI address check policy to allow PF1 PCI address for representors on PF1.
Quoted 8 lines
Note: detaching PF0 device can't remove representors on PF1. It's recommended to use primary(PF0) PCI address to probe representors on both PFs.
Fixes f926cce3fa94 ("net/mlx5: refactor bonding representor probing")Signed-off-by Xueming Li <[email protected]>
Thomas Monjalon
AppliedRe: [dpdk-dev] [PATCH v2] net/mlx5: fix LAG representor probe on PF1 PCI
Quoted 11 lines
Quoted 11 lines
In case of bonding, orchestrator wants to use same devargs for LAG and non- LAG scenario to probe representor on PF1 using PF1 PCI address like "<DBDF_PF1>,representor=pf1vf[0-3]".
This patch changes PCI address check policy to allow PF1 PCI address for representors on PF1.
Note: detaching PF0 device can't remove representors on PF1. It's recommended to use primary(PF0) PCI address to probe representors on both PFs.
Quoted 5 lines
Quoted 4 lines
Fixes f926cce3fa94 ("net/mlx5: refactor bonding representor probing")Signed-off-by Xueming Li <[email protected]>
Applied to next-net-mlx, thanks.