Re: [patch 1/3] x86/msi: Make irq_retrigger() functional for posted MSI

On Tue, Nov 25, 2025 at 11:20 AM Thomas Gleixner tglx@linutronix.de wrote:

Luigi reported that retriggering a posted MSI interrupt does not work correctly. [...]

So instead of playing games with the PIR, this can be actually solved for both cases by:

1. Keeping track of the posted interrupt vector handler state

Tangential comment, but I see that this patch uses this_cpu_read()/write() whereas the rest of the file uses __this_cpu_read()/write()

Given where they are used and the operand size, do we care about preemption/interrupt protection, or the (possibly marginal) extra cost ?

cheers luigi