Re: [RFC PATCH 0/4] Misc OHCI patches

On Fri, Oct 1, 2021 at 11:16 PM BALATON Zoltan <balaton@eik.bme.hu> wrote:

Hello,

This is a first attempt to make some progress with the problems found
with OHCI especially when trying to pass through a usb sound card on
mac99. This does not go all the way to allow multiple async packets on
different endpoints yet but as a first step just try to fix the
interaction and potential breakage of pending async packet by
isochronous transfers. Plus some small clean ups I've found while
trying to make sense of this device model.

Posted as RFC because it's unfinfished and untested as there seems to
be some regression with mac99 so it does not boot for me for some
reason I haven't debugged yet. Hope Howard can test it and see if it
changes any of the traces seen before.

BALATON Zoltan (4):
usb/ohci: Move cancelling async packet to ohci_stop_endpoints()
usb/ohci: Move USBPortOps related functions together
usb/ohci: Merge ohci_async_cancel_device() into ohci_child_detach()
usb/ohci: Don't use packet from OHCIState for isochronous transfers

hw/usb/hcd-ohci.c | 281 ++++++++++++++++++++++------------------------
1 file changed, 136 insertions(+), 145 deletions(-)

Hi all,

I've booted Fedora12 and MacOS 9.2/OSX10.4 ppc with these patches applied. All boot OK.
(Fedora12 requires -mac99,via=pmu. Here there seems to be some interference with the kbd, due to via=pmu presenting usb mouse and kdb to the guest. So I could not test this further.).
All tests done in Fedora 34 host with current master and patched build.

./qemu-system-ppc-ohcipatch02102021 \
-M mac99 \
-L pc-bios \
-display gtk \
-m 512 \
-boot c \
-hda /home/hsp/Mac-disks/10.4.qcow2 \
-device usb-host,loglevel=4,vendorid=0x046d,productid=0x0a37,pcap=ohci-usb1headset-usb1port-ohcipatchV1-macos104.pcap \
-trace "usb_*" -D ohci-usb1headset-usb1port-ohcipatchV1-macos104.txt

Endpoint 4 is the interrupt-based hid device, endpoint 1 would be the isochronous audio stream if it showed up.

Each test consisted of booting to the desktop, opening the system profiler/system report to check of the presence of the usb device, push the various buttons on the headset, attempt to play a sound, unplug-replug and attempt to play a sound again. Then shut down.

Both succesfully open the usb device:

usb_ohci_init_time usb_bit_time=1000000 usb_frame_time=83
usb_port_claim bus 0, port 1
usb_host_auto_scan_enabled
usb_ohci_reset pci-ohci
usb_ohci_stop pci-ohci: USB Suspended
usb_ohci_stop pci-ohci: USB Suspended
usb_host_open_started dev 3:29
usb_host_detach_kernel dev 3:29, if 0
usb_host_detach_kernel dev 3:29, if 1
usb_host_detach_kernel dev 3:29, if 2
usb_host_detach_kernel dev 3:29, if 3
usb_host_parse_config dev 3:29, value 1, active 1
usb_host_parse_interface dev 3:29, num 0, alt 0, active 1
usb_host_parse_interface dev 3:29, num 1, alt 0, active 1
usb_host_parse_interface dev 3:29, num 2, alt 0, active 1
usb_host_parse_interface dev 3:29, num 3, alt 0, active 1
usb_host_parse_endpoint dev 3:29, ep 4, in, int, active 1
usb_port_attach bus 0, port 1, devspeed full, portspeed full
usb_ohci_port_attach port #0
usb_host_open_success dev 3:29

Master with Mac OS 9.2:

usb_ohci_ed_pkt ED @ 0x00152920 h=0 c=0
head=0x00164000 tailp=0x00164030 next=0x00152020
usb_ohci_ed_pkt_flags fa=37 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x00152900 h=0 c=0
head=0x001645a0 tailp=0x00164660 next=0x001528a0
usb_ohci_ed_pkt_flags fa=37 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x001645a0 8 of 8 bytes setup r=1 cbp=0x01661b90 be=0x01661b97
usb_ohci_td_pkt_full OUT data: 01 0b 00 00 01 00 00 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x00152920 h=0 c=0
head=0x00164000 tailp=0x00164030 next=0x00152020
usb_ohci_ed_pkt_flags fa=37 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x00152900 h=0 c=0
head=0x001645a0 tailp=0x00164660 next=0x001528a0
usb_ohci_ed_pkt_flags fa=37 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x001645a0 8 of 8 bytes setup r=1 cbp=0x01661b90 be=0x01661b97
usb_ohci_td_pkt_full OUT data: 01 0b 00 00 01 00 00 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x00152920 h=0 c=0
head=0x00164000 tailp=0x00164030 next=0x00152020
usb_ohci_ed_pkt_flags fa=37 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x00152900 h=0 c=0
head=0x001645a0 tailp=0x00164660 next=0x001528a0
usb_ohci_ed_pkt_flags fa=37 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x001645a0 8 of 8 bytes setup r=1 cbp=0x01661b90 be=0x01661b97
usb_ohci_td_pkt_full OUT data: 01 0b 00 00 01 00 00 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x00152920 h=0 c=0
head=0x00164000 tailp=0x00164030 next=0x00152020
usb_ohci_ed_pkt_flags fa=37 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x00152900 h=0 c=0
head=0x001645a0 tailp=0x00164660 next=0x001528a0
usb_ohci_ed_pkt_flags fa=37 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x001645a0 8 of 8 bytes setup r=1 cbp=0x01661b90 be=0x01661b97
usb_ohci_td_pkt_full OUT data: 01 0b 00 00 01 00 00 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x00152920 h=0 c=0
head=0x00164000 tailp=0x00164030 next=0x00152020
usb_ohci_ed_pkt_flags fa=37 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async

OHCI patch with MacOS 9.2:

usb_ohci_ed_pkt ED @ 0x001528e0 h=0 c=0
head=0x001609c0 tailp=0x001609f0 next=0x00152020
usb_ohci_ed_pkt_flags fa=41 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x001528c0 h=0 c=0
head=0x00161030 tailp=0x001610c0 next=0x001528a0
usb_ohci_ed_pkt_flags fa=41 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x00161030 8 of 8 bytes setup r=1 cbp=0x01587b08 be=0x01587b0f
usb_ohci_td_pkt_full OUT data: 80 06 00 02 00 00 0a 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x001528e0 h=0 c=0
head=0x001609c0 tailp=0x001609f0 next=0x00152020
usb_ohci_ed_pkt_flags fa=41 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x001528c0 h=0 c=0
head=0x00161030 tailp=0x001610c0 next=0x001528a0
usb_ohci_ed_pkt_flags fa=41 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x00161030 8 of 8 bytes setup r=1 cbp=0x01587b08 be=0x01587b0f
usb_ohci_td_pkt_full OUT data: 80 06 00 02 00 00 0a 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x001528e0 h=0 c=0
head=0x001609c0 tailp=0x001609f0 next=0x00152020
usb_ohci_ed_pkt_flags fa=41 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x001528c0 h=0 c=0
head=0x00161030 tailp=0x001610c0 next=0x001528a0
usb_ohci_ed_pkt_flags fa=41 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x00161030 8 of 8 bytes setup r=1 cbp=0x01587b08 be=0x01587b0f
usb_ohci_td_pkt_full OUT data: 80 06 00 02 00 00 0a 00
usb_ohci_td_too_many_pending
usb_ohci_ed_pkt ED @ 0x001528e0 h=0 c=0
head=0x001609c0 tailp=0x001609f0 next=0x00152020
usb_ohci_ed_pkt_flags fa=41 en=4 d=2 s=0 k=0 f=0 mps=37
usb_ohci_td_skip_async
usb_ohci_ed_pkt ED @ 0x001528c0 h=0 c=0
head=0x00161030 tailp=0x001610c0 next=0x001528a0
usb_ohci_ed_pkt_flags fa=41 en=0 d=0 s=0 k=0 f=0 mps=64
usb_ohci_td_pkt_hdr TD @ 0x00161030 8 of 8 bytes setup r=1 cbp=0x01587b08 be=0x01587b0f
usb_ohci_td_pkt_full OUT data: 80 06 00 02 00 00 0a 00
usb_ohci_td_too_many_pending

Both have issues communicating with endpoint 4 (the hid controls volume up/down and mute).

Endpoint 1 should receive the isochronous audio stream, but never does.

After some experimentation with unplugging/plugging the headset and probing the usb stack (using the usb prober from the mac usb ddk for Mac OS 9.2) at some point endpoint 4 communication works for both guests tested. Only once was I able to get sound out and in working in Mac OS 9.2. For OSX I could only once get audio in working.

Pcap and text logs for both MacOS 9.2 and OSX 10.4 tests included...

Best,

Howard

From:	Howard Spoelstra
Subject:	Re: [RFC PATCH 0/4] Misc OHCI patches
Date:	Sat, 2 Oct 2021 15:50:34 +0200