2023-12-21 05:26:27

by Kai-Heng Feng

[permalink] [raw]
Subject: [PATCH v2] r8169: Fix PCI error on system resume

Some r8168 NICs stop working upon system resume:

[ 688.051096] r8169 0000:02:00.1 enp2s0f1: rtl_ep_ocp_read_cond == 0 (loop: 10, delay: 10000).
[ 688.175131] r8169 0000:02:00.1 enp2s0f1: Link is Down
...
[ 691.534611] r8169 0000:02:00.1 enp2s0f1: PCI error (cmd = 0x0407, status_errs = 0x0000)

Not sure if it's related, but those NICs have a BMC device at function
0:
02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. Realtek RealManage BMC [10ec:816e] (rev 1a)

Since increase the loop wait on rtl_ep_ocp_read_cond can eliminate the
issue, so let rtl8168ep_driver_start() to wait a bit longer.

Signed-off-by: Kai-Heng Feng <[email protected]>
---
v2:
- Wording

drivers/net/ethernet/realtek/r8169_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index bb787a52bc75..81fd31f6fac4 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -1211,7 +1211,7 @@ static void rtl8168ep_driver_start(struct rtl8169_private *tp)
{
r8168ep_ocp_write(tp, 0x01, 0x180, OOB_CMD_DRIVER_START);
r8168ep_ocp_write(tp, 0x01, 0x30, r8168ep_ocp_read(tp, 0x30) | 0x01);
- rtl_loop_wait_high(tp, &rtl_ep_ocp_read_cond, 10000, 10);
+ rtl_loop_wait_high(tp, &rtl_ep_ocp_read_cond, 10000, 30);
}

static void rtl8168_driver_start(struct rtl8169_private *tp)
--
2.34.1



2023-12-21 13:00:28

by Heiner Kallweit

[permalink] [raw]
Subject: Re: [PATCH v2] r8169: Fix PCI error on system resume

On 21.12.2023 06:25, Kai-Heng Feng wrote:
> Some r8168 NICs stop working upon system resume:
>
> [ 688.051096] r8169 0000:02:00.1 enp2s0f1: rtl_ep_ocp_read_cond == 0 (loop: 10, delay: 10000).
> [ 688.175131] r8169 0000:02:00.1 enp2s0f1: Link is Down
> ...
> [ 691.534611] r8169 0000:02:00.1 enp2s0f1: PCI error (cmd = 0x0407, status_errs = 0x0000)
>
> Not sure if it's related, but those NICs have a BMC device at function
> 0:
> 02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. Realtek RealManage BMC [10ec:816e] (rev 1a)
>
> Since increase the loop wait on rtl_ep_ocp_read_cond can eliminate the
> issue, so let rtl8168ep_driver_start() to wait a bit longer.
>
As this fixes an actual issue, the patch should target net and have a Fixes tag.

> Signed-off-by: Kai-Heng Feng <[email protected]>
> ---
> v2:
> - Wording
>
> drivers/net/ethernet/realtek/r8169_main.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
> index bb787a52bc75..81fd31f6fac4 100644
> --- a/drivers/net/ethernet/realtek/r8169_main.c
> +++ b/drivers/net/ethernet/realtek/r8169_main.c
> @@ -1211,7 +1211,7 @@ static void rtl8168ep_driver_start(struct rtl8169_private *tp)
> {
> r8168ep_ocp_write(tp, 0x01, 0x180, OOB_CMD_DRIVER_START);
> r8168ep_ocp_write(tp, 0x01, 0x30, r8168ep_ocp_read(tp, 0x30) | 0x01);
> - rtl_loop_wait_high(tp, &rtl_ep_ocp_read_cond, 10000, 10);
> + rtl_loop_wait_high(tp, &rtl_ep_ocp_read_cond, 10000, 30);
> }
>
> static void rtl8168_driver_start(struct rtl8169_private *tp)