1. Patchew
  2. linux
  3. View series

[PATCH iwl-net v2] igb: Fix not clearing TimeSync interrupts for 82580

Daiwei Li posted 1 patch 1 year, 10 months ago
There is a newer version of this series
drivers/net/ethernet/intel/igb/igb_main.c | 10 ++++++++++
1 file changed, 10 insertions(+)
[PATCH iwl-net v2] igb: Fix not clearing TimeSync interrupts for 82580
Posted by Daiwei Li 1 year, 10 months ago 
82580 NICs have a hardware bug that makes it
necessary to write into the TSICR (TimeSync Interrupt Cause) register
to clear it:
https://lore.kernel.org/all/CDCB8BE0.1EC2C%25matthew.vick@intel.com/

Add a conditional so only for 82580 we write into the TSICR register,
so we don't risk losing events for other models.

This (partially) reverts commit ee14cc9ea19b ("igb: Fix missing time sync events").

Fixes: ee14cc9ea19b ("igb: Fix missing time sync events")
Closes: https://lore.kernel.org/intel-wired-lan/CAN0jFd1kO0MMtOh8N2Ztxn6f7vvDKp2h507sMryobkBKe=xk=w@mail.gmail.com/
Tested-by: Daiwei Li <daiweili@google.com>
Signed-off-by: Daiwei Li <daiweili@google.com>
---

@Vinicius Gomes, this is my first time submitting a Linux kernel patch,
so apologies if I missed any part of the procedure (e.g. this is
currently on top of 6.7.12, the kernel I am running; should I be
rebasing on inline?). Also, is there any way to annotate the patch
to give you credit for the original change?

 drivers/net/ethernet/intel/igb/igb_main.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index ada42ba63549..1210ddc5d81e 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -6986,6 +6986,16 @@ static void igb_tsync_interrupt(struct igb_adapter *adapter)
 	struct e1000_hw *hw = &adapter->hw;
 	u32 tsicr = rd32(E1000_TSICR);
 	struct ptp_clock_event event;
+	const u32 mask = (TSINTR_SYS_WRAP | E1000_TSICR_TXTS |
+			  TSINTR_TT0 | TSINTR_TT1 |
+			  TSINTR_AUTT0 | TSINTR_AUTT1);
+
+	if (hw->mac.type == e1000_82580) {
+		/* 82580 has a hardware bug that requires a explicit
+		 * write to clear the TimeSync interrupt cause.
+		 */
+		wr32(E1000_TSICR, tsicr & mask);
+	}
 
 	if (tsicr & TSINTR_SYS_WRAP) {
 		event.type = PTP_CLOCK_PPS;
-- 
2.46.0.76.ge559c4bf1a-goog
Re: [PATCH iwl-net v2] igb: Fix not clearing TimeSync interrupts for 82580
Posted by Vinicius Costa Gomes 1 year, 10 months ago 
Daiwei Li <daiweili@google.com> writes:

> 82580 NICs have a hardware bug that makes it
> necessary to write into the TSICR (TimeSync Interrupt Cause) register
> to clear it:
> https://lore.kernel.org/all/CDCB8BE0.1EC2C%25matthew.vick@intel.com/
>
> Add a conditional so only for 82580 we write into the TSICR register,
> so we don't risk losing events for other models.

Please add some information in the commit message about how to reproduce
the issue, as Paul suggested.

>
> This (partially) reverts commit ee14cc9ea19b ("igb: Fix missing time sync events").
>
> Fixes: ee14cc9ea19b ("igb: Fix missing time sync events")
> Closes: https://lore.kernel.org/intel-wired-lan/CAN0jFd1kO0MMtOh8N2Ztxn6f7vvDKp2h507sMryobkBKe=xk=w@mail.gmail.com/
> Tested-by: Daiwei Li <daiweili@google.com>
> Signed-off-by: Daiwei Li <daiweili@google.com>
> ---
>
> @Vinicius Gomes, this is my first time submitting a Linux kernel patch,
> so apologies if I missed any part of the procedure (e.g. this is
> currently on top of 6.7.12, the kernel I am running; should I be
> rebasing on inline?). Also, is there any way to annotate the patch
> to give you credit for the original change?

Your submission format looks fine. Just a couple details:
 - No need for setting in-reply-to (or something like it);
 
 - For this particular patch, you got lucky and it applies cleanly
 against current tip, but for future submissions, for intel-wired-lan
 and patches intended for the stable tree, please rebase against:

 https://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue.git/

For credits, you can add something like:

Suggested-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>

>
>  drivers/net/ethernet/intel/igb/igb_main.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
>
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index ada42ba63549..1210ddc5d81e 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -6986,6 +6986,16 @@ static void igb_tsync_interrupt(struct igb_adapter *adapter)
>  	struct e1000_hw *hw = &adapter->hw;
>  	u32 tsicr = rd32(E1000_TSICR);
>  	struct ptp_clock_event event;
> +	const u32 mask = (TSINTR_SYS_WRAP | E1000_TSICR_TXTS |
> +			  TSINTR_TT0 | TSINTR_TT1 |
> +			  TSINTR_AUTT0 | TSINTR_AUTT1);
> +

Please move the declaration of 'mask' up, to follow the convention, the
"reverse christmas tree" rule. Or separate the attribution from the
declaration.

> +	if (hw->mac.type == e1000_82580) {
> +		/* 82580 has a hardware bug that requires a explicit

And as pointed by Paul, "*an* explicit".

> +		 * write to clear the TimeSync interrupt cause.
> +		 */
> +		wr32(E1000_TSICR, tsicr & mask);
> +	}
>  
>  	if (tsicr & TSINTR_SYS_WRAP) {
>  		event.type = PTP_CLOCK_PPS;
> -- 
> 2.46.0.76.ge559c4bf1a-goog
>

-- 
Vinicius
Re: [PATCH iwl-net v2] igb: Fix not clearing TimeSync interrupts for 82580
Posted by Daiwei Li 1 year, 10 months ago 
Thank you for the review! I've sent out another patch that hopefully
addresses the comments.


On Tue, Aug 13, 2024 at 3:26 PM Vinicius Costa Gomes
<vinicius.gomes@intel.com> wrote:
>
> Daiwei Li <daiweili@google.com> writes:
>
> > 82580 NICs have a hardware bug that makes it
> > necessary to write into the TSICR (TimeSync Interrupt Cause) register
> > to clear it:
> > https://lore.kernel.org/all/CDCB8BE0.1EC2C%25matthew.vick@intel.com/
> >
> > Add a conditional so only for 82580 we write into the TSICR register,
> > so we don't risk losing events for other models.
>
> Please add some information in the commit message about how to reproduce
> the issue, as Paul suggested.
>
> >
> > This (partially) reverts commit ee14cc9ea19b ("igb: Fix missing time sync events").
> >
> > Fixes: ee14cc9ea19b ("igb: Fix missing time sync events")
> > Closes: https://lore.kernel.org/intel-wired-lan/CAN0jFd1kO0MMtOh8N2Ztxn6f7vvDKp2h507sMryobkBKe=xk=w@mail.gmail.com/
> > Tested-by: Daiwei Li <daiweili@google.com>
> > Signed-off-by: Daiwei Li <daiweili@google.com>
> > ---
> >
> > @Vinicius Gomes, this is my first time submitting a Linux kernel patch,
> > so apologies if I missed any part of the procedure (e.g. this is
> > currently on top of 6.7.12, the kernel I am running; should I be
> > rebasing on inline?). Also, is there any way to annotate the patch
> > to give you credit for the original change?
>
> Your submission format looks fine. Just a couple details:
>  - No need for setting in-reply-to (or something like it);
>
>  - For this particular patch, you got lucky and it applies cleanly
>  against current tip, but for future submissions, for intel-wired-lan
>  and patches intended for the stable tree, please rebase against:
>
>  https://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue.git/
>
> For credits, you can add something like:
>
> Suggested-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
>
> >
> >  drivers/net/ethernet/intel/igb/igb_main.c | 10 ++++++++++
> >  1 file changed, 10 insertions(+)
> >
> > diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> > index ada42ba63549..1210ddc5d81e 100644
> > --- a/drivers/net/ethernet/intel/igb/igb_main.c
> > +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> > @@ -6986,6 +6986,16 @@ static void igb_tsync_interrupt(struct igb_adapter *adapter)
> >       struct e1000_hw *hw = &adapter->hw;
> >       u32 tsicr = rd32(E1000_TSICR);
> >       struct ptp_clock_event event;
> > +     const u32 mask = (TSINTR_SYS_WRAP | E1000_TSICR_TXTS |
> > +                       TSINTR_TT0 | TSINTR_TT1 |
> > +                       TSINTR_AUTT0 | TSINTR_AUTT1);
> > +
>
> Please move the declaration of 'mask' up, to follow the convention, the
> "reverse christmas tree" rule. Or separate the attribution from the
> declaration.
>
> > +     if (hw->mac.type == e1000_82580) {
> > +             /* 82580 has a hardware bug that requires a explicit
>
> And as pointed by Paul, "*an* explicit".
>
> > +              * write to clear the TimeSync interrupt cause.
> > +              */
> > +             wr32(E1000_TSICR, tsicr & mask);
> > +     }
> >
> >       if (tsicr & TSINTR_SYS_WRAP) {
> >               event.type = PTP_CLOCK_PPS;
> > --
> > 2.46.0.76.ge559c4bf1a-goog
> >
>
> --
> Vinicius



-- 
Daiwei Li
Software Engineer
Mobile: 415-736-8670
waymo.com

Patchew 2.1-devel-ea86937

© 2016 - 2026 Red Hat, Inc.