[GIT PULL] irqchip fixes for 5.6, take #2

[GIT PULL] irqchip fixes for 5.6, take #2

[Marc Zyngier]

Hi Thomas,

This is hopefully the last irqchip update for 5.6. This time, a single
patch working around a hardware issue on the Cavium ThunderX and its
derivatives.

Please pull.

	M.

The following changes since commit 5186a6cc3ef5a3fa327c258924ef098b0de77006:

  irqchip/gic-v3-its: Rename VPENDBASER/VPROPBASER accessors (2020-02-08 10:01:33 +0000)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms.git tags/irqchip-fixes-5.6-2

for you to fetch changes up to d01fd161e85904064290435f67f4ed59af5daf74:

  irqchip/gic-v3: Workaround Cavium erratum 38539 when reading GICD_TYPER2 (2020-03-14 10:15:19 +0000)

----------------------------------------------------------------
irqchip fixes for 5.6, take #2

- Add workaround for Cavium/Marvell ThunderX unimplemented GIC registers

----------------------------------------------------------------
Marc Zyngier (1):
      irqchip/gic-v3: Workaround Cavium erratum 38539 when reading GICD_TYPER2

 Documentation/arm64/silicon-errata.rst |  2 ++
 drivers/irqchip/irq-gic-v3.c           | 30 +++++++++++++++++++++++++++++-
 2 files changed, 31 insertions(+), 1 deletion(-)

Re: [GIT PULL] irqchip fixes for 5.6, take #2

[Lukas Wunner]

On Sat, Mar 14, 2020 at 10:30:00AM +0000, Marc Zyngier wrote:
> This is hopefully the last irqchip update for 5.6. This time, a single
> patch working around a hardware issue on the Cavium ThunderX and its
> derivatives.

Hm, I was hoping to see the BCM2835 irqchip fix in this pull:

https://lore.kernel.org/lkml/f97868ba4e9b86ddad71f44ec9d8b3b7d8daa1ea.1582618537.git.lukas@wunner.de/

That patch fixes a pretty grave issue so I'd be really grateful
if anyone could pick it up (or provide feedback why it can't be
picked up just yet).

Thanks!

Lukas

Re: [GIT PULL] irqchip fixes for 5.6, take #2

[Marc Zyngier]

Lukas,

On 2020-03-15 08:48, Lukas Wunner wrote:
> On Sat, Mar 14, 2020 at 10:30:00AM +0000, Marc Zyngier wrote:
>> This is hopefully the last irqchip update for 5.6. This time, a single
>> patch working around a hardware issue on the Cavium ThunderX and its
>> derivatives.
> 
> Hm, I was hoping to see the BCM2835 irqchip fix in this pull:
> 
> https://lore.kernel.org/lkml/f97868ba4e9b86ddad71f44ec9d8b3b7d8daa1ea.1582618537.git.lukas@wunner.de/
> 
> That patch fixes a pretty grave issue so I'd be really grateful
> if anyone could pick it up (or provide feedback why it can't be
> picked up just yet).

Whilst I don't dispute that this patch addresses a serious issue,
it is in no way an urgent fix -- the issue is already 7.5 year old,
so a couple of week delay isn't going to change the world. Also,
the system does work without this fix, so I'm quite confident
leaving it for 5.7.

But thanks for this email anyway, as it reminded me that although
I had picked that patch for 5.7, I didn't apply it just yet. This
is now fixed, and the patch should be picked up by -next shortly.

Thanks,

         M.
-- 
Jazz is not dead. It just smells funny...

Re: [GIT PULL] irqchip fixes for 5.6, take #2

[Lukas Wunner]

On Sun, Mar 15, 2020 at 10:13:38AM +0000, Marc Zyngier wrote:
> On 2020-03-15 08:48, Lukas Wunner wrote:
> > Hm, I was hoping to see the BCM2835 irqchip fix in this pull:
> > 
> > https://lore.kernel.org/lkml/f97868ba4e9b86ddad71f44ec9d8b3b7d8daa1ea.1582618537.git.lukas@wunner.de/
> 
> Whilst I don't dispute that this patch addresses a serious issue,
> it is in no way an urgent fix -- the issue is already 7.5 year old,
> so a couple of week delay isn't going to change the world. Also,
> the system does work without this fix, so I'm quite confident
> leaving it for 5.7.
> 
> But thanks for this email anyway, as it reminded me that although
> I had picked that patch for 5.7, I didn't apply it just yet. This
> is now fixed, and the patch should be picked up by -next shortly.

Sure, fair enough, thanks.

Bonne fin de week-end,

Lukas