Re: Problems with irq mapping in qemu v5.2

[BALATON Zoltan via <qemu-devel@nongnu.org>]

On Wed, 23 Dec 2020, Mark Cave-Ayland wrote:
> On 22/12/2020 22:23, BALATON Zoltan via wrote:
>> I've just remembered that for sam460ex we had this commit: 484ab3dffadc 
>> (sam460ex: Fix PCI interrupts with multiple devices) that changed that 
>> mapping for that machine so I guess you got the exception with the bamboo 
>> board then. I'm not sure though that similar fix is applicable fot that or 
>> even that this fix is correct for sam460ex but appears to work so far.
>
> FWIW you might want to review this commit: as Peter noticed it is possible to 
> lose interrupts here since if one PCI interrupt is already asserted and then 
> another comes along, the second PCI interrupt will unintentionally clear the 
> first which could cause problems.

Yes, thanks for reminding. I think this was also raised back then but for 
some reason we thought this should not be an issue in this case but I'll 
give it a try and see if using an or-gate would be better. The only 
possible IRQ related problem I know about is choppy sound with an emulated 
PCI sound card but that may be because some other audio related problem as 
well.

Regards,
BALATON Zoltan

> You probably want to keep the 4 separate PCI interrupts but feed them into an 
> OR IRQ, the output of which gets fed into the UIC. Have a look at 
> https://lists.gnu.org/archive/html/qemu-devel/2020-12/msg05503.html for the 
> sun4m variant of this based upon Peter's original example.
>
>
> ATB,
>
> Mark.
>
>