Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/30/13 15:53, Lundberg, Johannes wrote:

I added printf at every 100,000th iteration in both locations but I didn't
get any output at all.. Seems it stuck in some other place.

During my test I actually passed that stage a couple of times but still the
xhci_do_command times out and I get the mountroot> prompt With the 9.1
driver I never get either of the errors.


Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


Hi,

If you have a git checkout of the FreeBSD source tree, you can try 
stripping away patches for xhci*.[ch] only, since end of December 2012. 
The 9- and 10- XHCI code should be almost the same, except for the 
streams mode support. Here is a list of patches you can try to revert if 
I get it right:


commit 2969ec61bfd9be3fdcaa50432755d24df345874d
Author: hselasky 
Date:   Fri Jul 26 06:24:33 2013 +

MFC r253532:

Fix an XHCI regression:

The Block Event Interrupts, BEI, feature does not
work like expected with the Renesas XHCI chipsets.
Revert feature.

While at it correct the TD SIZE computation in
case of Zero Length Packet, ZLP, in the end of a
multi frame USB transfer.

PR: usb/180726
Approved by:re, hrs

commit 7dbeb07e555c5e9649fac8079a4b3bcdd57b2cfc
Author: hselasky 
Date:   Tue Jun 11 06:18:51 2013 +

MFC r251249, r251251, r251252, r2512, r251254 and r251515:
Correct XHCI DMA descriptor programming.
Correct maximum IRQ rate.

commit 1941cac2597eea224c9e0e76a789692dfa8dda91
Author: marius 
Date:   Sat Mar 9 02:36:32 2013 +

MFC: r227309 (partial)

Mark all SYSCTL_NODEs static that have no corresponding SYSCTL_DECLs.

The SYSCTL_NODE macro defines a list that stores all child-elements of
that node. If there's no SYSCTL_DECL macro anywhere else, there's no
reason why it shouldn't be static.

commit 9b2efd37b6d3afb84b234348094adf8cc533e2c6
Author: hselasky 
Date:   Wed Feb 6 11:08:11 2013 +

MFC r246113 and r246126:
Add missing NULL pointer check.

Reported by:  Lars Engels

commit 63af294dc7c2db6a120eba9cec16fabec13aeec1
Author: hselasky 
Date:   Mon Jan 21 07:25:38 2013 +

MFC r245132 and r245175:
Optimise the XHCI interrupt handling.
This patch will save CPU time when the XHCI interrupt is
shared with other devices.
Only check event rings when interrupt bits are set.
Otherwise would indicate hiding possible hardware fault(s).

Tested by:  sos @
Submitted by:   sos @

commit fa4f2cf0b90d6c5ce9eca2417e200fb693d6f239
Author: hselasky 
Date:   Mon Jan 21 07:22:45 2013 +

MFC r243780:
- Add support for Etron EJ168 USB 3.0 Host Controllers.
This brand of controllers expects that the number of
contexts specified in the input slot context points
to an active endpoint context, else it refuses to
operate.

- Wrap one or two long lines.

Tested by:  Markus Pfeiffer (DragonFlyBSD)

--HPS
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

I added printf at every 100,000th iteration in both locations but I didn't
get any output at all.. Seems it stuck in some other place.

During my test I actually passed that stage a couple of times but still the
xhci_do_command times out and I get the mountroot> prompt With the 9.1
driver I never get either of the errors.


Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Fri, Aug 30, 2013 at 3:32 PM, Hans Petter Selasky wrote:

> On 08/30/13 15:19, Lundberg, Johannes wrote:
>
>> Hi Hans
>>
>> I tried that too and no change... But, the variable overflow you
>> introduced, it is never increased, right? So, it will never become zero...
>>
>>
> It will become zero when it wraps. Maybe 2**32 is too long. You could add
> a printf while it is running, and see if it goes == 100 for example.
>
> --HPS
>
>  Johannes Lundberg
>> BRILLIANTSERVICE CO., LTD. 
>> 
>> >
>>
>>
>> On Fri, Aug 30, 2013 at 2:26 PM, Hans Petter Selasky > >wrote:
>>
>>  On 08/30/13 13:54, Lundberg, Johannes wrote:
>>>
>>>  Still got the same behaviour after applying the patch...

 Johannes Lundberg
 BRILLIANTSERVICE CO., LTD. 
 >

>>>
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/30/13 15:19, Lundberg, Johannes wrote:

Hi Hans

I tried that too and no change... But, the variable overflow you
introduced, it is never increased, right? So, it will never become zero...



It will become zero when it wraps. Maybe 2**32 is too long. You could 
add a printf while it is running, and see if it goes == 100 for example.


--HPS


Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Fri, Aug 30, 2013 at 2:26 PM, Hans Petter Selasky wrote:


On 08/30/13 13:54, Lundberg, Johannes wrote:


Still got the same behaviour after applying the patch...

Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 



___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Hi Hans

I tried that too and no change... But, the variable overflow you
introduced, it is never increased, right? So, it will never become zero...

Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Fri, Aug 30, 2013 at 2:26 PM, Hans Petter Selasky wrote:

> On 08/30/13 13:54, Lundberg, Johannes wrote:
>
>> Still got the same behaviour after applying the patch...
>>
>> Johannes Lundberg
>> BRILLIANTSERVICE CO., LTD. 
>> 
>> >
>>
>
> I've seen something similar with my mac, that the boot menu counter is not
> always counting stable. I think this problem is inside the following
> function, if you say that the code is stuck inside pause():
>
> sys/x86/isa/clock.c:DELAY()
>
> Maybe you can try this patch instead. Just apply it by hand:
>
> === sys/x86/isa/clock.c
> ==**==**==
> --- sys/x86/isa/clock.c (revision 254832)
> +++ sys/x86/isa/clock.c (local)
> @@ -262,6 +262,7 @@
> timecounter_get_t *func;
> uint64_t end, freq, now;
> u_int last, mask, u;
> +   int overflow = 0;
>
> tc = timecounter;
> freq = atomic_load_acq_64(&tsc_freq);
> @@ -281,6 +282,11 @@
> sched_pin();
> last = func(tc) & mask;
> do {
> +   if (--overflow == 0) {
> +   printf("DELAY overflow\n");
> +   break;
> +   }
> +
> cpu_spinwait();
> u = func(tc) & mask;
> if (u < last)
> @@ -304,6 +310,7 @@
>  DELAY(int n)
>  {
> int delta, prev_tick, tick, ticks_left;
> +   int overflow = 0;
>  #ifdef DELAYDEBUG
> int getit_calls = 1;
> int n1;
> @@ -365,6 +372,10 @@
>  / 100;
>
> while (ticks_left > 0) {
> +   if (--overflow == 0) {
> +   printf("DELAY overflow\n");
> +   break;
> +   }
>  #ifdef KDB
> if (kdb_active) {
>  #ifdef PC98
>
>
> --HPS
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/30/13 13:54, Lundberg, Johannes wrote:

Still got the same behaviour after applying the patch...

Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


I've seen something similar with my mac, that the boot menu counter is 
not always counting stable. I think this problem is inside the following 
function, if you say that the code is stuck inside pause():


sys/x86/isa/clock.c:DELAY()

Maybe you can try this patch instead. Just apply it by hand:

=== sys/x86/isa/clock.c
==
--- sys/x86/isa/clock.c (revision 254832)
+++ sys/x86/isa/clock.c (local)
@@ -262,6 +262,7 @@
timecounter_get_t *func;
uint64_t end, freq, now;
u_int last, mask, u;
+   int overflow = 0;

tc = timecounter;
freq = atomic_load_acq_64(&tsc_freq);
@@ -281,6 +282,11 @@
sched_pin();
last = func(tc) & mask;
do {
+   if (--overflow == 0) {
+   printf("DELAY overflow\n");
+   break;
+   }
+
cpu_spinwait();
u = func(tc) & mask;
if (u < last)
@@ -304,6 +310,7 @@
 DELAY(int n)
 {
int delta, prev_tick, tick, ticks_left;
+   int overflow = 0;
 #ifdef DELAYDEBUG
int getit_calls = 1;
int n1;
@@ -365,6 +372,10 @@
 / 100;

while (ticks_left > 0) {
+   if (--overflow == 0) {
+   printf("DELAY overflow\n");
+   break;
+   }
 #ifdef KDB
if (kdb_active) {
 #ifdef PC98


--HPS
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Still got the same behaviour after applying the patch...

Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Fri, Aug 30, 2013 at 12:13 PM, Hans Petter Selasky wrote:

> On 08/30/13 11:35, Lundberg, Johannes wrote:
>
>> Hi Hans
>>
>> I tried the patch and the result is the same. However, I found the command
>> that causes the freeze. Also, it is not always it freezes but maybe 9/10
>> reboots or more frequently.
>>
>> At the end of the function xhci_start_controller(..) there is a for loop:
>>
>> 487for (i = 0; i != 100; i++) {
>> 488usb_pause_mtx(NULL, hz / 100);
>> 489temp = XREAD4(sc, oper, XHCI_USBSTS) & XHCI_STS_HCH;
>> 490if (!temp)
>> 491break;
>> 492}
>>
>> and it freezes at usb_pause_mtx(...)
>> value of i is 0
>> value of hz is 1000
>>
>
> Hi,
>
> Can you try the attached patch?
>
> I think this is a problem inside DELAY(). Maybe you have to select another
> timing source for DELAY(), mav @ CC'ed?
>
> --HPS
>
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/30/13 11:35, Lundberg, Johannes wrote:

Hi Hans

I tried the patch and the result is the same. However, I found the command
that causes the freeze. Also, it is not always it freezes but maybe 9/10
reboots or more frequently.

At the end of the function xhci_start_controller(..) there is a for loop:

487for (i = 0; i != 100; i++) {
488usb_pause_mtx(NULL, hz / 100);
489temp = XREAD4(sc, oper, XHCI_USBSTS) & XHCI_STS_HCH;
490if (!temp)
491break;
492}

and it freezes at usb_pause_mtx(...)
value of i is 0
value of hz is 1000


Hi,

Can you try the attached patch?

I think this is a problem inside DELAY(). Maybe you have to select 
another timing source for DELAY(), mav @ CC'ed?


--HPS

=== sys/kern/kern_synch.c
==
--- sys/kern/kern_synch.c	(revision 254832)
+++ sys/kern/kern_synch.c	(local)
@@ -356,11 +356,8 @@
 int
 pause_sbt(const char *wmesg, sbintime_t sbt, sbintime_t pr, int flags)
 {
-	int sbt_sec;
+	KASSERT(sbt >= 0, ("pause: timeout must be >= 0"));
 
-	sbt_sec = sbintime_getsec(sbt);
-	KASSERT(sbt_sec >= 0, ("pause: timo must be >= 0"));
-
 	/* silently convert invalid timeouts */
 	if (sbt == 0)
 		sbt = tick_sbt;
@@ -370,11 +367,14 @@
 		 * We delay one second at a time to avoid overflowing the
 		 * system specific DELAY() function(s):
 		 */
-		while (sbt_sec > 0) {
+		while (sbt >= SBT_1S) {
 			DELAY(100);
-			sbt_sec--;
+			sbt -= SBT_1S;
 		}
-		DELAY((sbt & 0x) / SBT_1US);
+		/* Do the delay remainder */
+		sbt /= SBT_1US;
+		if (sbt != 0)
+			DELAY(sbt);
 		return (0);
 	}
 	return (_sleep(&pause_wchan[curcpu], NULL, 0, wmesg, sbt, pr, flags));
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Hi Hans

I tried the patch and the result is the same. However, I found the command
that causes the freeze. Also, it is not always it freezes but maybe 9/10
reboots or more frequently.

At the end of the function xhci_start_controller(..) there is a for loop:

487for (i = 0; i != 100; i++) {
488usb_pause_mtx(NULL, hz / 100);
489temp = XREAD4(sc, oper, XHCI_USBSTS) & XHCI_STS_HCH;
490if (!temp)
491break;
492}

and it freezes at usb_pause_mtx(...)
value of i is 0
value of hz is 1000








Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Fri, Aug 30, 2013 at 8:42 AM, Hans Petter Selasky wrote:

> On 08/30/13 08:09, Lundberg, Johannes wrote:
>
>> What I got so far is this;
>>
>> USB driver from current stops after
>> xhci0: 32 byte context size
>>
>> While driver from 9.1 continues to the next step which is
>> usbus0 on xhci0
>> xhci0: usbpf: Attached
>> ...
>>
>> I can try adding some printf's in the code and see if I get some more..
>>
>>
> Hi,
>
> There are only a few commits to the xhci driver since 9.1 was releases, so
> this should be easy to figure out.
>
> I'm doing a wild guess. Can you try the attached patch. It will ensure
> that any BIOS generated interrupts get cleared before we reset the
> controller.
>
> --HPS
>
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/30/13 08:09, Lundberg, Johannes wrote:

What I got so far is this;

USB driver from current stops after
xhci0: 32 byte context size

While driver from 9.1 continues to the next step which is
usbus0 on xhci0
xhci0: usbpf: Attached
...

I can try adding some printf's in the code and see if I get some more..



Hi,

There are only a few commits to the xhci driver since 9.1 was releases, 
so this should be easy to figure out.


I'm doing a wild guess. Can you try the attached patch. It will ensure 
that any BIOS generated interrupts get cleared before we reset the 
controller.


--HPS

=== sys/dev/usb/controller/xhci.c
==
--- sys/dev/usb/controller/xhci.c	(revision 254832)
+++ sys/dev/usb/controller/xhci.c	(local)
@@ -320,6 +320,12 @@
 		device_printf(sc->sc_bus.parent, "32 byte context size.\n");
 	}
 
+	temp = XREAD4(sc, oper, XHCI_USBSTS);
+	/* clear all pending interrupts */
+	XWRITE4(sc, oper, XHCI_USBSTS, temp);
+	/* clear and disable leftover interrupts */
+	XWRITE4(sc, runt, XHCI_IMAN(0), XHCI_IMAN_INTR_PEND);
+
 	/* Reset controller */
 	XWRITE4(sc, oper, XHCI_USBCMD, XHCI_CMD_HCRST);
 
@@ -385,10 +391,6 @@
 	sc->sc_exit_lat_max = XHCI_HCS3_U1_DEL(temp) +
 	XHCI_HCS3_U2_DEL(temp) + 250 /* us */;
 
-	temp = XREAD4(sc, oper, XHCI_USBSTS);
-
-	/* clear interrupts */
-	XWRITE4(sc, oper, XHCI_USBSTS, temp);
 	/* disable all device notifications */
 	XWRITE4(sc, oper, XHCI_DNCTRL, 0);
 
@@ -462,11 +464,8 @@
 	XWRITE4(sc, runt, XHCI_ERSTBA_HI(0), (uint32_t)(addr >> 32));
 
 	/* Setup interrupter registers */
+	XWRITE4(sc, runt, XHCI_IMAN(0), XHCI_IMAN_INTR_ENA);
 
-	temp = XREAD4(sc, runt, XHCI_IMAN(0));
-	temp |= XHCI_IMAN_INTR_ENA;
-	XWRITE4(sc, runt, XHCI_IMAN(0), temp);
-
 	/* setup command ring control base address */
 	addr = buf_res.physaddr;
 	addr += (uintptr_t)&((struct xhci_hw_root *)0)->hwr_commands[0];
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

What I got so far is this;

USB driver from current stops after
xhci0: 32 byte context size

While driver from 9.1 continues to the next step which is
usbus0 on xhci0
xhci0: usbpf: Attached
...

I can try adding some printf's in the code and see if I get some more..

On Tuesday, August 27, 2013, Lundberg, Johannes wrote:

> Hi Hans
>
> Sure, I will try it out later and mail the results.
>
> Now I'm running current with usb part reverted to 9.1. It gets me pass the
> usb probing to the part that I really wanted to confirm. If FreeBSD
> supports the SSD drive on the new MacBook Air. It seems that it doesn't so
> that one more thing that we need to look into.
>
>
>
> Johannes Lundberg
> BRILLIANTSERVICE CO., LTD. 
>
>
> On Tue, Aug 27, 2013 at 10:06 AM, Hans Petter Selasky 
> 
> > wrote:
>
>> On 08/26/13 21:02, Lundberg, Johannes wrote:
>>
>>> Hi Hans
>>>
>>> Thanks but nothing of that makes any difference. Well, it's gonna be
>>> difficult to find the diff I think... The oldest image I could find was
>>> from May.
>>>
>>> What I'm doing now is compiling a bootonly.iso of current with a xhci.h/c
>>> that's reverted to 9.1 release version to see if that will boot.
>>>
>>>
>> Hi,
>>
>> If it just hangs, can be an IRQ looping issue.
>>
>> You can also try to boot using bootverbose and configure the xhci driver
>> with debugging on in the sys/dev/usb/controller/xhci.c by default to see
>> just what is exactly going on.
>>
>> If you can suggest a patch, that would be great.
>>
>> --HPS
>>
>>
>

-- 
Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

The SSD controller driver is hopefully fixed now with the patch I submitted
to -current mailing list.

I will compile a new system with usb as a loadable module so that it is
easier to test different versions and possible fixes and then do some
testing maybe tomorrow.

Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Tue, Aug 27, 2013 at 10:14 AM, Adrian Chadd  wrote:

> Hm!
>
> Is there a Linux driver for this SSD? Does Linux approximately boot on
> this thing?
>
>
>
> -adrian
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Adrian Chadd]

Hm!

Is there a Linux driver for this SSD? Does Linux approximately boot on this
thing?



-adrian
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Hi Hans

Sure, I will try it out later and mail the results.

Now I'm running current with usb part reverted to 9.1. It gets me pass the
usb probing to the part that I really wanted to confirm. If FreeBSD
supports the SSD drive on the new MacBook Air. It seems that it doesn't so
that one more thing that we need to look into.



Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Tue, Aug 27, 2013 at 10:06 AM, Hans Petter Selasky wrote:

> On 08/26/13 21:02, Lundberg, Johannes wrote:
>
>> Hi Hans
>>
>> Thanks but nothing of that makes any difference. Well, it's gonna be
>> difficult to find the diff I think... The oldest image I could find was
>> from May.
>>
>> What I'm doing now is compiling a bootonly.iso of current with a xhci.h/c
>> that's reverted to 9.1 release version to see if that will boot.
>>
>>
> Hi,
>
> If it just hangs, can be an IRQ looping issue.
>
> You can also try to boot using bootverbose and configure the xhci driver
> with debugging on in the sys/dev/usb/controller/xhci.c by default to see
> just what is exactly going on.
>
> If you can suggest a patch, that would be great.
>
> --HPS
>
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/26/13 21:02, Lundberg, Johannes wrote:

Hi Hans

Thanks but nothing of that makes any difference. Well, it's gonna be
difficult to find the diff I think... The oldest image I could find was
from May.

What I'm doing now is compiling a bootonly.iso of current with a xhci.h/c
that's reverted to 9.1 release version to see if that will boot.



Hi,

If it just hangs, can be an IRQ looping issue.

You can also try to boot using bootverbose and configure the xhci driver 
with debugging on in the sys/dev/usb/controller/xhci.c by default to see 
just what is exactly going on.


If you can suggest a patch, that would be great.

--HPS

___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Hi Hans

Thanks but nothing of that makes any difference. Well, it's gonna be
difficult to find the diff I think... The oldest image I could find was
from May.

What I'm doing now is compiling a bootonly.iso of current with a xhci.h/c
that's reverted to 9.1 release version to see if that will boot.


Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Mon, Aug 26, 2013 at 7:22 PM, Hans Petter Selasky wrote:

> On 08/26/13 17:55, Lundberg, Johannes wrote:
>
>> Just looked through the PR mail that came today and found possible
>> duplicates.
>>
>> o usb/180726   usbXHCI umass support breaks between r248085 and
>> r252560
>> o usb/179342   usbFreebsd 10.0-current USB 3.0 not working
>> (xhci_do_coma
>>
>>
> Hi,
>
> What happens if you load xhci after reaching the boot prompt?
>
> Might also be worth trying:
>
> hw.usb.xhci.msi=0
>
> Are you able to bisect the 10-current code until you find the diff that
> broke it?
>
> --HPS
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Hans Petter Selasky]

On 08/26/13 17:55, Lundberg, Johannes wrote:

Just looked through the PR mail that came today and found possible
duplicates.

o usb/180726   usbXHCI umass support breaks between r248085 and
r252560
o usb/179342   usbFreebsd 10.0-current USB 3.0 not working
(xhci_do_coma



Hi,

What happens if you load xhci after reaching the boot prompt?

Might also be worth trying:

hw.usb.xhci.msi=0

Are you able to bisect the 10-current code until you find the diff that 
broke it?


--HPS
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Just looked through the PR mail that came today and found possible
duplicates.

o usb/180726   usbXHCI umass support breaks between r248085 and
r252560
o usb/179342   usbFreebsd 10.0-current USB 3.0 not working
(xhci_do_coma




Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 


On Mon, Aug 26, 2013 at 4:42 PM, Lundberg, Johannes <
johan...@brilliantservice.co.jp> wrote:

> Hi
>
> I'm trying to install 10-CURRENT on a 2013 MacBook Air and there seem to
> be some problem with the xhci driver.
>
> During boot of the memstick image of current from 20130818 I get either of
> the following errors:
>
> - Boot freezes during probe of xhci devices.
> - Probing times out (immediately) and I get the mountroot> prompt but no
> functional keyboard.
>
> I tried an image of 10-current from May but the problem is the same so
> what ever change that made this stop working was committed before May.
>
> I also tried disabling xhci with the following steps but without success
>
> - Enter boot prompt
> - set hint.xhci.0.disabled=1
> - load ehci / uhci
> - boot
>
> The 9.1 release successfully detects all usb devices during install but
> stop working at the partitioning tool because the PCIe SSD drive isn't
> detected. Using 9.1 as LiveCD I managed to pull out the following
> information.
>
> -
>
> MacBook Air 2013, FreeBSD 9.1 LiveCD
>
> dmesg output
>
> xhci0:  mem 0xb0a0-0xb0a0 at
> device 20.0 on pci0
>
> pciconf -l output
>
> xhci0@pci0:0:20:0: class=0x0c0330 card=0x72708086 chip=0x9c318086
> rev=0x04 hdr=0x00
>
> usbconfig output
>
> ugen0.1:  at usbus0, cfg=0 md=HOST spd=SUPER
> (5.0Gbps) pwr=SAVE
> ugen0.2:  at usbus0, cfg=0 md=HOST spd=FULL
> (12Mbps) pwr=SAVE
> ugen0.3:  at usbus0, cfg=0 md=HOST spd=FULL
> (12Mbps) pwr=ON
> ugen0.4:  at usbus0, cfg=0 md=HOST spd=FULL
> (12Mbps) pwr=ON
> ugen0.5:  at usbus0, cfg=0
> md=HOST spd=FULL (12Mbps) pwr=ON
> ugen0.6:  at usbus0, cfg=0
> md=HOST spd=FULL (12Mbps) pwr=ON
> ugen0.7:  at usbus0, cfg=0 md=HOST spd=SUPER
> (5.0Gbps) pwr=ON
>
> -
>
> Anyone has any clue what's going on?
>
> Best regards
>
> Johannes Lundberg
> BRILLIANTSERVICE CO., LTD. 
>
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

xhci broken on 10-CURRENT and 2013 MacBook Air?

[Lundberg, Johannes]

Hi

I'm trying to install 10-CURRENT on a 2013 MacBook Air and there seem to be
some problem with the xhci driver.

During boot of the memstick image of current from 20130818 I get either of
the following errors:

- Boot freezes during probe of xhci devices.
- Probing times out (immediately) and I get the mountroot> prompt but no
functional keyboard.

I tried an image of 10-current from May but the problem is the same so what
ever change that made this stop working was committed before May.

I also tried disabling xhci with the following steps but without success

- Enter boot prompt
- set hint.xhci.0.disabled=1
- load ehci / uhci
- boot

The 9.1 release successfully detects all usb devices during install but
stop working at the partitioning tool because the PCIe SSD drive isn't
detected. Using 9.1 as LiveCD I managed to pull out the following
information.

-

MacBook Air 2013, FreeBSD 9.1 LiveCD

dmesg output

xhci0:  mem 0xb0a0-0xb0a0 at
device 20.0 on pci0

pciconf -l output

xhci0@pci0:0:20:0: class=0x0c0330 card=0x72708086 chip=0x9c318086 rev=0x04
hdr=0x00

usbconfig output

ugen0.1:  at usbus0, cfg=0 md=HOST spd=SUPER
(5.0Gbps) pwr=SAVE
ugen0.2:  at usbus0, cfg=0 md=HOST spd=FULL
(12Mbps) pwr=SAVE
ugen0.3:  at usbus0, cfg=0 md=HOST spd=FULL
(12Mbps) pwr=ON
ugen0.4:  at usbus0, cfg=0 md=HOST spd=FULL
(12Mbps) pwr=ON
ugen0.5:  at usbus0, cfg=0
md=HOST spd=FULL (12Mbps) pwr=ON
ugen0.6:  at usbus0, cfg=0
md=HOST spd=FULL (12Mbps) pwr=ON
ugen0.7:  at usbus0, cfg=0 md=HOST spd=SUPER
(5.0Gbps) pwr=ON

-

Anyone has any clue what's going on?

Best regards

Johannes Lundberg
BRILLIANTSERVICE CO., LTD. 
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"