qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* hw/ppc/mac_newworld: Problem with Uninorth IRQ lines
@ 2023-02-21 13:44 Philippe Mathieu-Daudé
  2023-02-21 15:17 ` BALATON Zoltan
  2023-03-01 12:24 ` Mark Cave-Ayland
  0 siblings, 2 replies; 3+ messages in thread
From: Philippe Mathieu-Daudé @ 2023-02-21 13:44 UTC (permalink / raw)
  To: Mark Cave-Ayland, QEMU Developers; +Cc: qemu-ppc

Hi Mark,

The mac99 machines use the same PIC input IRQs for the Uninorth
output IRQs:

379     if (PPC_INPUT(env) != PPC_FLAGS_INPUT_970) {
380         /* Uninorth AGP bus */
381         for (i = 0; i < 4; i++) {
382             qdev_connect_gpio_out(uninorth_agp_dev, i,
383                                   qdev_get_gpio_in(pic_dev, 0x1b + i));
384         }
385
386         /* Uninorth internal bus */
387         for (i = 0; i < 4; i++) {
388             qdev_connect_gpio_out(uninorth_internal_dev, i,
389                                   qdev_get_gpio_in(pic_dev, 0x1b + i));
390         }
391     }

Did you mean to use an OR gate here?

Thanks,

Phil.


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: hw/ppc/mac_newworld: Problem with Uninorth IRQ lines
  2023-02-21 13:44 hw/ppc/mac_newworld: Problem with Uninorth IRQ lines Philippe Mathieu-Daudé
@ 2023-02-21 15:17 ` BALATON Zoltan
  2023-03-01 12:24 ` Mark Cave-Ayland
  1 sibling, 0 replies; 3+ messages in thread
From: BALATON Zoltan @ 2023-02-21 15:17 UTC (permalink / raw)
  To: Philippe Mathieu-Daudé; +Cc: Mark Cave-Ayland, QEMU Developers, qemu-ppc

[-- Attachment #1: Type: text/plain, Size: 1201 bytes --]

On Tue, 21 Feb 2023, Philippe Mathieu-Daudé wrote:
> The mac99 machines use the same PIC input IRQs for the Uninorth
> output IRQs:
>
> 379     if (PPC_INPUT(env) != PPC_FLAGS_INPUT_970) {
> 380         /* Uninorth AGP bus */
> 381         for (i = 0; i < 4; i++) {
> 382             qdev_connect_gpio_out(uninorth_agp_dev, i,
> 383                                   qdev_get_gpio_in(pic_dev, 0x1b + i));
> 384         }
> 385
> 386         /* Uninorth internal bus */
> 387         for (i = 0; i < 4; i++) {
> 388             qdev_connect_gpio_out(uninorth_internal_dev, i,
> 389                                   qdev_get_gpio_in(pic_dev, 0x1b + i));
> 390         }
> 391     }
>
> Did you mean to use an OR gate here?

This probably does not matter as these PCI busses are not used at all, 
they are just there to match real hardware but on QEMU everyting is 
connected to the 3rd PCI bus (which is actually the second numbered 
between the two others above, but OpenBIOS only knows about that PCI bus 
and does not care about the AGP and internal PCI buses so in practice 
these IRQs likely never raised so probably does not really matter where 
they are connected now).

Regards,
BALATON Zoltan

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: hw/ppc/mac_newworld: Problem with Uninorth IRQ lines
  2023-02-21 13:44 hw/ppc/mac_newworld: Problem with Uninorth IRQ lines Philippe Mathieu-Daudé
  2023-02-21 15:17 ` BALATON Zoltan
@ 2023-03-01 12:24 ` Mark Cave-Ayland
  1 sibling, 0 replies; 3+ messages in thread
From: Mark Cave-Ayland @ 2023-03-01 12:24 UTC (permalink / raw)
  To: Philippe Mathieu-Daudé, QEMU Developers; +Cc: qemu-ppc

On 21/02/2023 13:44, Philippe Mathieu-Daudé wrote:

> Hi Mark,
> 
> The mac99 machines use the same PIC input IRQs for the Uninorth
> output IRQs:
> 
> 379     if (PPC_INPUT(env) != PPC_FLAGS_INPUT_970) {
> 380         /* Uninorth AGP bus */
> 381         for (i = 0; i < 4; i++) {
> 382             qdev_connect_gpio_out(uninorth_agp_dev, i,
> 383                                   qdev_get_gpio_in(pic_dev, 0x1b + i));
> 384         }
> 385
> 386         /* Uninorth internal bus */
> 387         for (i = 0; i < 4; i++) {
> 388             qdev_connect_gpio_out(uninorth_internal_dev, i,
> 389                                   qdev_get_gpio_in(pic_dev, 0x1b + i));
> 390         }
> 391     }
> 
> Did you mean to use an OR gate here?

Hi Phil,

These are separate PCI buses so I suspect they should be wired up with different 
IRQs, but this wiring has been the same ever since the code was introduced. I think 
it should be possible to dig the values out from a suitable device tree but it's not 
something I've looked at in any detail.


ATB,

Mark.


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2023-03-01 12:25 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-02-21 13:44 hw/ppc/mac_newworld: Problem with Uninorth IRQ lines Philippe Mathieu-Daudé
2023-02-21 15:17 ` BALATON Zoltan
2023-03-01 12:24 ` Mark Cave-Ayland

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).