[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-ppc] R128 interrupts
From: |
Jd Lyons |
Subject: |
Re: [Qemu-ppc] R128 interrupts |
Date: |
Fri, 9 Aug 2019 09:41:30 -0400 |
My qemu launch command:
sudo '/home/jam/aty7/bin/qemu-system-ppc' -M mac99 -m 1024 -hda
'/media/jam/60816f11-cba2-4a8f-8ceb-6e406cfdedd7/Tiger.img' -device
ati-vga,guest_hwcursor=true -prom-env 'vga-ndrv?=false' -display gtk
-prom-env 'boot-args=-v' -prom-env 'auto-boot?=false' -bios
'/home/jam/openbios-qemu_30p_pci-map-in.elf' -trace enable="ati*" -trace
enable="pci*" -g 1280x960x32 -netdev tap,id=mynet0 -device
sungem,netdev=mynet0
It takes a while @640x480 while our emulator reads and writes a bunch of ATI*
regs.
The NDRV has to poll for the EDID and set a bunch of properties……….
> On Aug 9, 2019, at 9:22 AM, Jd Lyons <address@hidden> wrote:
>
> I didn’t load the FCode ROM, while it maybe necessary to inject some
> properties the ROM builds into the device tree, at this point it doesn’t
> provide us anything useful and we can’t include it with Qemu because of
> copywrite issues.
>
> I’m not sure we are on the same codebase as far as Qemu, but it takes a while
> to get to the desktop, and you need to remove the ATIRage128GA.plugin.
>
> At this point, what we need is correct handling of the interrupts, I’ve
> looked into how the USB and Network adapters handle interrupts, but it didn’t
> provide me with any code I could hack into the Rage128 support.
>
> It’s going to take a skilled programmer, not a hacker like me.
>
> Here is what we have:
>
> qemu-system-ppc->ati-vga->mac99->openbios->dummy-interrupt->bootx->mach_kernel->’NDRV’->Rage128.kext
>
> The next step would be the GA plugin, but we can’t debug that until we get a
> correctly working interrupt situation.
>
>> On Aug 9, 2019, at 8:42 AM, Howard Spoelstra <address@hidden> wrote:
>>
>>
>>
>> On Fri, Aug 9, 2019 at 2:07 PM Jd Lyons <address@hidden> wrote:
>> First a small change is needed to the ati.c:
>>
>> - assert(bpp != 0);
>> + assert(bpp != 8);
>>
>> I’m sure this isn’t correct, but it works for the purposes of testing there
>> ‘NDRV’.
>>
>> Rebuild qemu-system-ppc, then
>>
>> dev /pci/ATY
>> “ ATY,Rage128Ps” encode-string “ name” property
>> 1 encode-int “ interrupts” property \\ We can’t boot OS 9 here
>> boot
>>
>> No cursor is rendered, even with guest_hwcursor=true, and the screen is an
>> off shade of blue that doesn’t show up in the screenshots.
>>
>> That gives me:
>> 1528@1565354293.016508:ati_mm_read 4 0x54 CRTC_EXT_CNTL -> 0x0
>> 1528@1565354293.016515:ati_mm_write 4 0x54 CRTC_EXT_CNTL <- 0x0
>> ati_mm_write: Display enabled
>> ati_vga_switch_mode: 1 -> 1
>> ati_vga_switch_mode: Switching to 640x480 640 15 @ 8000
>> qemu-system-ppc: warning: CRTC offset is not multiple of pitch
>> ati_vga_switch_mode: VBE offset (128,51), vbe_start_addr=4000
>> 1528@1565354293.034520:ati_mm_read 4 0x284 unknown -> 0x0
>> 1528@1565354293.034544:ati_mm_write 4 0x284 unknown <- 0x2
>> 1528@1565354293.034552:ati_mm_read 4 0x288 unknown -> 0x0
>> 1528@1565354293.034564:ati_mm_write 4 0x288 unknown <- 0x0
>> 1528@1565354293.034705:ati_mm_read 4 0x284 unknown -> 0x0
>> 1528@1565354293.034717:ati_mm_write 4 0x284 unknown <- 0x0
>>
>> and a garbled openbios window at 640x480.
>>
>> Do you pass anything on the command line? Do you load a rom?
>>
>> Best,
>> Howard
>
- Re: [Qemu-ppc] R128 interrupts, Jd Lyons, 2019/08/08
- Re: [Qemu-ppc] R128 interrupts, BALATON Zoltan, 2019/08/08
- Message not available
- Message not available
- Message not available
- Message not available
- Re: [Qemu-ppc] R128 interrupts, Jd Lyons, 2019/08/09
- Re: [Qemu-ppc] R128 interrupts,
Jd Lyons <=
- Re: [Qemu-ppc] R128 interrupts, BALATON Zoltan, 2019/08/10
- Re: [Qemu-ppc] R128 interrupts, BALATON Zoltan, 2019/08/10
- Re: [Qemu-ppc] R128 interrupts, Gerd Hoffmann, 2019/08/12
- Re: [Qemu-ppc] R128 interrupts, BALATON Zoltan, 2019/08/14