I forgot to add the *.txt extension to the above, sorry about that.
On Tue, Feb 25, 2014 at 6:42 PM, Adam Vaughan <[email protected]> wrote: > I don't have a WiFi or Bluetooth adapter with me to try at the moment, but > I just tried booting with my USB flash drive installed prior to power on > and didn't see an issue both with and without the USB FIQ enabled. I can > try with a Bluetooth adapter tomorrow, if needed. > > Your logs are more useful since they show the actual issue, but maybe my > boot logs (attached) will be at least a little helpful in troubleshooting? > Also, the patch I listed above is only needed if you want to boot with the > USB FIQ disabled. > > > On Tue, Feb 25, 2014 at 6:20 PM, Gregory Dymarek <[email protected]>wrote: > >> well, this does not work for me. It boots fine when there is no USB device >> plugged in. >> However, when my WIFI or bluetooth adapter is plugged in, the boot log >> shows: http://pastebin.com/vjLJVDGS >> >> I also tried the patch Adam suggests but it does not seem to affect >> anything. >> >> >> >> On 25 February 2014 22:30, Adam Vaughan <[email protected]> wrote: >> >> > I just tried the above patch and the warning doesn't show up at boot >> > anymore. I tried unplugging and plugging in a flash drive / keyboard >> and >> > saw no issues in dmesg. >> > >> > I mentioned it in a previous email, but just as a friendly reminder this >> > patch is still needed to allow you to boot with a disabled FIQ: >> > >> > diff --git a/drivers/usb/host/dwc_otg/dwc_otg_hcd_intr.c >> > b/drivers/usb/host/dwc_otg/dwc_otg_hcd_intr.c >> > index 19abea0..78172ea 100644 >> > --- a/drivers/usb/host/dwc_otg/dwc_otg_hcd_intr.c >> > +++ b/drivers/usb/host/dwc_otg/dwc_otg_hcd_intr.c >> > @@ -742,8 +742,10 @@ int32_t dwc_otg_hcd_handle_sof_intr(dwc_otg_hcd_t * >> > hcd) >> > } >> > >> > /* Clear interrupt */ >> > - //gintsts.b.sofintr = 1; >> > - //DWC_WRITE_REG32(&hcd->core_if->core_global_regs->gintsts, >> > gintsts.d32); >> > + if (!fiq_fix_enable) { >> > + gintsts.b.sofintr = 1; >> > + DWC_WRITE_REG32(&hcd->core_if->core_global_regs->gintsts, >> > gintsts.d32); >> > + } >> > >> > return 1; >> > } >> > >> > >> > On Tue, Feb 25, 2014 at 4:31 PM, Gilles Chanteperdrix < >> > [email protected]> wrote: >> > >> >> On 02/25/2014 10:09 PM, Gregory Dymarek wrote: >> >> > So the frame freeze I got on my version is on line 145 in here: >> >> > >> https://github.com/raspberrypi/linux/blob/rpi-3.8.y/kernel/irq/handle.c >> >> > >> >> > The dwc_otg_hcd_handle_intr is here: >> >> > >> >> >> https://github.com/raspberrypi/linux/blob/rpi-3.8.y/drivers/usb/host/dwc_otg/dwc_otg_hcd_intr.c >> >> > >> >> > >> >> > Is the line 523 the problem? >> >> > local_fiq_enable(); >> >> >> >> Please try the following patch: >> >> >> >> diff --git a/arch/arm/include/asm/ipipe_hwirq.h >> >> b/arch/arm/include/asm/ipipe_hwirq.h >> >> index 6b864aa..bd8cda1 100644 >> >> --- a/arch/arm/include/asm/ipipe_hwirq.h >> >> +++ b/arch/arm/include/asm/ipipe_hwirq.h >> >> @@ -200,9 +200,9 @@ static inline void hard_local_irq_restore(unsigned >> >> long x) >> >> ipipe_unstall_root(); \ >> >> } while (0) >> >> >> >> -#define local_fiq_enable() ipipe_unstall_root() >> >> +#define local_fiq_enable() hard_local_fiq_enable_notrace() >> >> >> >> -#define local_fiq_disable() ipipe_stall_root() >> >> +#define local_fiq_disable() hard_local_fiq_disable_notrace() >> >> >> >> #define arch_local_irq_restore(flags) \ >> >> do { \ >> >> >> >> >> >> -- >> >> Gilles. >> >> >> >> _______________________________________________ >> >> Xenomai mailing list >> >> [email protected] >> >> http://www.xenomai.org/mailman/listinfo/xenomai >> >> >> > >> > >> _______________________________________________ >> Xenomai mailing list >> [email protected] >> http://www.xenomai.org/mailman/listinfo/xenomai >> > > _______________________________________________ Xenomai mailing list [email protected] http://www.xenomai.org/mailman/listinfo/xenomai
