On Tue, Aug 05, 2025 at 02:20:52PM +0100, Marc Zyngier wrote:
On Tue, 05 Aug 2025 14:09:34 +0100,
Sasha Levin <sashal@xxxxxxxxxx> wrote:
From: Marc Zyngier <maz@xxxxxxxxxx>
[ Upstream commit 3cc8f625e4c6a0e9f936da6b94166e62e387fe1d ]
Since changing the affinity of an MSI really is about changing
the target address and that it isn't possible to mask an individual
MSI, it is completely possible for an interrupt to race with itself,
usually resulting in a lost interrupt.
Paper over the design blunder by informing the core code of this
sad state of affairs.
Signed-off-by: Marc Zyngier <maz@xxxxxxxxxx>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@xxxxxxxxxx>
Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
Link: https://lore.kernel.org/r/20250708173404.1278635-11-maz@xxxxxxxxxx
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
LLM Generated explanations, may be completely bogus:
s/may be//. It is an amusing read though, specially when quoting
totally unrelated patches, so thumbs up for the comical value.
Yeah, it's still very much at the "junior engineer" level, but honestly
I think that just the boolean yes/no answers out of it provides a better
noise to signal ratio than the older AUTOSEL.
But I'm not even going to entertain explaining *why* backporting this
patch on its own is nonsense. Reading the original series should be
enlightening enough.
Sadly it doesn't have the context to understand that that specific
conmit is part of a larger series. That information just disappears when
patches are applied into git.
I'll drop it, thanks!
--
Thanks,
Sasha
