From 9260319e7411ff8281700a532caa436f40120ec4 Mon Sep 17 00:00:00 2001 From: Christian Krinitsin Date: Fri, 30 May 2025 16:52:07 +0200 Subject: gitlab scraper: download in toml and text format --- .../issues_text/target_missing/host_missing/accel_missing/2081 | 9 +++++++++ 1 file changed, 9 insertions(+) create mode 100644 gitlab/issues_text/target_missing/host_missing/accel_missing/2081 (limited to 'gitlab/issues_text/target_missing/host_missing/accel_missing/2081') diff --git a/gitlab/issues_text/target_missing/host_missing/accel_missing/2081 b/gitlab/issues_text/target_missing/host_missing/accel_missing/2081 new file mode 100644 index 000000000..538976cb7 --- /dev/null +++ b/gitlab/issues_text/target_missing/host_missing/accel_missing/2081 @@ -0,0 +1,9 @@ +[OHCI] OHCI_CC_DEVICENOTRESPONDING not set when transferring to a disconnected device +Description of problem: +If a USB device is disconnected and is cleaned up by qemu, subsequent transfers to that device address are ignored. On a real OHCI controller `OHCI_CC_DEVICENOTRESPONDING` bit is set and is reported as an error to the host. + +qemu attempts to set it here https://github.com/qemu/qemu/blob/ffd454c67e38cc6df792733ebc5d967eee28ac0d/hw/usb/hcd-ohci.c#L795 which would work fine on a valid device handle. + +However this check https://github.com/qemu/qemu/blob/ffd454c67e38cc6df792733ebc5d967eee28ac0d/hw/usb/hcd-ohci.c#L975 leaves early if no device handle is found so the error code is never set. + +Fix is to set `OHCI_CC_DEVICENOTRESPONDING` if `ohci_find_device` fails before returning. -- cgit 1.4.1