Re: [PATCH v2 0/1] gpiolib: acpi: Add quirk for ASUS ROG Strix G16 G614 series
From: Mario Limonciello
Date: Sun Jun 21 2026 - 03:58:39 EST
On 6/20/26 00:20, Marco Scardovi wrote:
In data sabato 20 giugno 2026 08:52:27 Ora legale dell’Europa centrale, Andy Shevchenko ha scritto:
On Fri, Jun 19, 2026 at 12:28:49PM +0200, Marco Scardovi wrote:Uhm, right. Not everyone are crazy over updating everything.
In data venerdì 19 giugno 2026 12:17:17 Ora legale dell’Europa centrale, Basavaraj Natikar ha scritto:
On 6/18/2026 11:37 PM, Marco Scardovi wrote:
In data giovedì 18 giugno 2026 20:01:40 Ora legale dell’Europa centrale, Andy Shevchenko ha scritto:
On Thu, Jun 18, 2026 at 06:59:15PM +0200, Marco Scardovi wrote:
On Thu, Jun 18, 2026 at 16:35:37 CEST, Andy Shevchenko wrote:
On Thu, Jun 18, 2026 at 06:46:28PM +0530, Basavaraj Natikar wrote:
On 6/18/2026 4:44 AM, Marco Scardovi wrote:
On Wed, Jun 17, 2026 at 10:33 PM, Armin Wolf wrote:
...
I also checked the BIOS 316 ACPI dump — the stalling path is byte‑for‑byte identical
to 315, so the AML is unchanged and it'll still stall if pin 21 boots low.
On the OEM side, I'm connecting internally with our AMD contact for ASUS to report
this behavior and follow up on a firmware fix.
Thank you. Does that mean that my patch will not be required anymore?
If that's the case you can consider it as null. I don't mind having the boot
flag added until it's completely fixed on ASUS' side. BTW I have the suspect
that my model is not the only one with the long boot bug on ASUS [1] [2]
but actually the only one who came up with an actual patch on the kernel.
I believe your patch is still needed. Not every model might get a BIOS update,
and no guarantee that everybody who possesses the problematic platforms
actually *will* update the BIOS. But it's AMD's call. I will happily accept
the changes acknowledged by AMD.
As soon as I have a complete response from AMD and Asus I'll proceed
with a proper patch.
For now, thank you all for checking it out and giving your time for feedback.
Maybe it's clear to everyone else but me; but if this is ASL used for the dGPU shouldn't we get someone from NVIDIA to add their 2¢?
Maybe *they* had intended this to be both edge triggered and that's how it got here.
I can't imagine this is going to be the only case we see of this. So before we go down that quirk path can we get them to weigh in? Maybe send to dri-devel and CC the nvidia folks that work on the kernel.