On 04/28/2016 04:30 AM, Roger Quadros wrote: > Hi Joe, > > On 27/04/16 23:41, Joe Lawrence wrote: >> Hello Mathias, Roger, Gabriel >> >> [ ... snip ... ] >> In the meantime I was browsing recent linux-scsi archives and noticed >> Gabriel's [PATCH v2] xhci: Cleanup only when releasing primary hcd: >> >> http://marc.info/?l=linux-usb&m=146134802313366&w=2 >> >> which looks like it modifies xhci_stop() to continue on to >> xhci_cleanup_msix even if XHCI_STATE_HALTED (for primary host >> controller's anyway). >> >> I think this may avoid the issue that I've been seeing and was wondering >> if it would be safe to test this patch on top of 4.6.0-rc3? (An older >> RC because my repro is very timing dependent, so moving the kernel >> forward / backwards or adding excessive trace logging can make it hard >> to debug.) > > Yes, please test it on top of v4.6-rc and I'm hoping it will > be sent for stable as well. Hi Roger, With Gabriel's v2 patch on top of v4.6-rc4, I'm up to ~450 device removals without hitting the IRQ cleanup issue I posted. Previously I was hitting that bug inside of 50 removals. I'll let it continue running, but free to add a: Tested-by: Joe Lawrence <joe.lawrence@xxxxxxxxxxx> # For Stratus hotplug -- Joe -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
