Re: [PATCH v5 5/7] hw/isa/vt82c686: Work around missing level sensitive irq in i8259 model

[Bernhard Beschow]

Am 1. März 2023 14:05:31 UTC schrieb David Woodhouse <dwmw2@infradead.org>:
>On Wed, 2023-03-01 at 14:18 +0100, BALATON Zoltan wrote:
>> > Are you sure the PIC ELCR is actually set for the lines you're having
>> > trouble with? Is that something the Pegasos SmartFirmware would have
>> > done, and MorphOS is expecting to inherit but isn't actually setting up
>> > for itself?
>> 
>> No, it works with other guests like Linux and AmigaOS that use PIC as set 
>> up by the firmware but MorphOS tries to use it in level sensitive mode and 
>> likely has an IRQ handler which expects this to work. This is where I've 
>> debugged it and came to this workaround:
>> 
>> https://lists.nongnu.org/archive/html/qemu-ppc/2023-02/msg00403.html
>> 
>> When booting MorphOS with -d unimp I see these logs:
>> 
>> i8259: level sensitive irq not supported
>> i8259: level sensitive irq not supported
>> 
>> which is I guess when it tries to set it for both PICs. (If you want to 
>> try this MorphOS iso is downloadable and instructions how to boot it is 
>> here: http://zero.eik.bme.hu/~balaton/qemu/amiga/#morphos
>
>
>Wow. Even looking at the PIIX3 datasheet from 1996, That 'Edge/Level
>Bank Select (LTIM)' bit was documented as 'This bit is disabled. Its
>function is replaced by the Edge/Level Triggerede Control (ELCR)
>Registers.
>
>We've been able to set the edge/level on a per-pin basis ever since the
>ELCR was introduced with the IBM PS/2, I think.
>
>It isn't a *correct* fix without a little bit more typing, but does
>this make it work?
>
>diff --git a/hw/intc/i8259.c b/hw/intc/i8259.c
>index 17910f3bcb..36ebcff025 100644
>--- a/hw/intc/i8259.c
>+++ b/hw/intc/i8259.c
>@@ -246,6 +246,7 @@ static void pic_ioport_write(void *opaque, hwaddr addr64,
>             if (val & 0x08) {
>                 qemu_log_mask(LOG_UNIMP,
>                               "i8259: level sensitive irq not supported\n");
>+                s->elcr = 0xff;

Thanks so much, David! You're a genious...

Best regards,

Bernhard

>             }
>         } else if (val & 0x08) {
>             if (val & 0x04) {
>
>
>