On 4/27/2016 5:41 PM, Or Gerlitz wrote:
On 4/25/2016 4:55 PM, Alex Williamson wrote:
My comment is trying to highlight the problem that users are not likely
to check for updated firmware for every device *before* installing a new
kernel, so there's a reasonable chance that a user will see this change
first as a regression in the kernel. So documentation in the firmware
release notes is really only going to help very meticulous users. For
everyone else it still looks like a kernel regression and a detailed
commit log is an after the fact aid to resolving the problem. So, is
there some way we can detect the firmware from the quirk and make a
determination whether to enable broken intx, with a dmesg entry
indicating to update device firmware? Thanks,
Yes, we can read the firmware version in the quirk. I'll prepare a new
fix that determines whether to mark INTx masking as broken or not
according to the firmware version.
Yep, makes sense to me.
If we can do that, lets go this way, Majd?
Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
