| Message ID | 20130807110901.6617.23985.malonedeb@soybean.canonical.com |
|---|---|
| State | New |
| Headers | show |
Your observation is correct, but this is already fixed upstream:
commit 628e54857a82a3cb65ef96c12640c30d6307a064
Author: Gerd Hoffmann <kraxel@redhat.com>
Date: Wed Jun 12 13:17:02 2013 +0200
usb-host-libusb: set USB_DEV_FLAG_IS_HOST
** Changed in: qemu
Status: New => Fix Committed
Thanks, this works as expected. Can this get backported to 1.5.x?
** Changed in: qemu
Status: Fix Committed => Fix Released
diff --git a/hw/usb/core.c b/hw/usb/core.c index 05948ca..20753cc 100644 --- a/hw/usb/core.c +++ b/hw/usb/core.c @@ -409,6 +409,8 @@ void usb_handle_packet(USBDevice *dev, USBPacket *p) /* hcd drivers cannot handle async for isoc */ assert(p->ep->type != USB_ENDPOINT_XFER_ISOC); /* using async for interrupt packets breaks migration */ + printf("type=%x %x %x %p %p\n", p->ep->type, + dev->flags, p->ep->dev->flags, dev, p->ep->dev); assert(p->ep->type != USB_ENDPOINT_XFER_INT || (dev->flags & (1 << USB_DEV_FLAG_IS_HOST))); usb_packet_set_state(p, USB_PACKET_ASYNC);
Public bug reported: After the patch at http://lists.nongnu.org/archive/html/qemu- devel/2013-05/msg01567.html, I cannot get my Logitech Unifying USB receiver passed to the guest anymore. Minimal reproduction command: qemu-system-x86_64 -usbdevice host:046d:c52b Debug patch: yields: type=0 1 1 0x5555565de4c0 0x5555565de4c0 type=0 1 1 0x5555565de4c0 0x5555565de4c0 type=0 1 1 0x5555565de4c0 0x5555565de4c0 type=0 1 1 0x5555565de4c0 0x5555565de4c0 type=0 1 1 0x5555565de4c0 0x5555565de4c0 type=3 1 1 0x5555565de4c0 0x5555565de4c0 qemu-system-x86_64: hw/usb/core.c:415: usb_handle_packet: Assertion `p->ep->type != 3 || (dev->flags & (1 << USB_DEV_FLAG_IS_HOST))' failed. Program received signal SIGABRT, Aborted. 0x00007ffff10c61c9 in raise () from /usr/lib/libc.so.6 (gdb) bt #0 0x00007ffff10c61c9 in raise () from /usr/lib/libc.so.6 #1 0x00007ffff10c75c8 in abort () from /usr/lib/libc.so.6 #2 0x00007ffff10bf356 in __assert_fail_base () from /usr/lib/libc.so.6 #3 0x00007ffff10bf402 in __assert_fail () from /usr/lib/libc.so.6 #4 0x00005555556e4b38 in usb_handle_packet (dev=<optimized out>, p=p@entry=0x5555565da610) at hw/usb/core.c:414 #5 0x00005555556f93ef in uhci_handle_td (s=s@entry=0x5555565ce5d0, q=0x5555566a9dc0, q@entry=0x0, qh_addr=qh_addr@entry=980834, td=td@entry=0x7fffffffdd30, td_addr=<optimized out>, int_mask=int_mask@entry=0x7fffffffdd1c) at hw/usb/hcd-uhci.c:904 #6 0x00005555556f9949 in uhci_process_frame (s=s@entry=0x5555565ce5d0) at hw/usb/hcd-uhci.c:1084 #7 0x00005555556f9c35 in uhci_frame_timer (opaque=0x5555565ce5d0) at hw/usb/hcd-uhci.c:1183 #8 0x000055555573e086 in qemu_run_timers (clock=0x55555654e200) at qemu-timer.c:394 #9 0x000055555573e325 in qemu_run_timers (clock=<optimized out>) at qemu-timer.c:459 #10 qemu_run_all_timers () at qemu-timer.c:452 #11 0x0000555555711bee in main_loop_wait (nonblocking=<optimized out>) at main-loop.c:470 #12 0x00005555555e0161 in main_loop () at vl.c:2029 #13 main (argc=<optimized out>, argv=<optimized out>, envp=<optimized out>) at vl.c:4419 Looking around, I see that host-bsd.c and host-linux.c both set USB_DEV_FLAG_IS_HOST, but host-libusbx.c doesn't do that. Affected QEMU versions: 1.5.1, 1.5.2 Not affected: 1.5.0 libusb version: 1.0.16 ** Affects: qemu Importance: Undecided Status: New