[Public]
-----Original Message----- From: Bartosz Golaszewski brgl@bgdev.pl Sent: Monday, April 25, 2022 13:43 To: Limonciello, Mario Mario.Limonciello@amd.com Cc: Linus Torvalds torvalds@linux-foundation.org; Natikar, Basavaraj Basavaraj.Natikar@amd.com; Gong, Richard Richard.Gong@amd.com; regressions@lists.linux.dev; Thorsten Leemhuis regressions@leemhuis.info; Greg KH gregkh@linuxfoundation.org; stable stable@vger.kernel.org; Linus Walleij linus.walleij@linaro.org; Shreeya Patel shreeya.patel@collabora.com; Andy Shevchenko andy.shevchenko@gmail.com; open list:GPIO SUBSYSTEM <linux- gpio@vger.kernel.org>; open list linux-kernel@vger.kernel.org Subject: Re: [PATCH v2 0/1] Fix regression in 5.18 for GPIO
On Mon, Apr 25, 2022 at 8:41 PM Bartosz Golaszewski brgl@bgdev.pl wrote:
On Fri, Apr 22, 2022 at 3:15 PM Mario Limonciello mario.limonciello@amd.com wrote:
Linus,
This patch is being sent directly to you because there has been a regression in 5.18 that I identified and sent a fix up that has been reviewed/tested/acked for nearly a week but the current subsystem maintainer (Bartosz) hasn't picked it up to send to you.
Hi Mario!
I don't have any previous submission in my inbox. Are you sure to have used my current address (brgl@bgdev.pl)?
Nevermind, found it in spam. Sorry, this sometimes happens in gmail.
OK glad you found it.
Anyway - it's only been 3 days and I've been travelling. Sometimes reviews take a couple days.
If it was just in the normal development release in an RC I'd agree there wasn't a lot of urgency, but stable picked it up and caused severe regressions. There wasn't an obvious willingness to revert the problematic commit in stable is why Thorsten was making noise about it and suggested me to send it directly to Linus.
Anyway - I'm glad it's sorted now.
Bart
Bart
It's a severe problem; anyone who hits it:
- Power button doesn't work anymore
- Can't resume their laptop from S3 or s2idle
Because the original patch was cc stable@, it landed in stable releases and has been breaking people left and right as distros track the stable channels. The patch is well tested. Would you please consider to pick this up directly to fix that regression?
Thanks,
Mario Limonciello (1): gpio: Request interrupts after IRQ is initialized
drivers/gpio/gpiolib.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
-- 2.34.1