[9fans] usb regression

[9fans] usb regression

[erik quanstrom]

i am now unable to copy more than about 10mb at a time
off the usb sd reader i've been using.  usb/disk isn't just quitting,
the endpoint is dead.  the console says only
	usb/disk... usb/disk: : 062800
	disk: reset: device is detached
it used to work, so maybe something new's come up.
the last copy i made was 20091020.

here's the output with debug on:

usb/disk: : 062800
qhintr: td 0xf004e780 csw 0x10008d40 error 0x40 endpoint stalled
ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
qhintr: td 0xf004e780 csw 0xd8d50 error 0x50 babble detected
qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
usb epctl detach ep7.0
disk: reset: device is detached
usbclose q 0x20 fid 616 ref 2
ehci: epclose ep7.2
ehci: cancelio for qh 0xf004e700 state idle
ehci: doorbell
ehci: qhcoherency: doorbell 1
usb: ep7.2 0xf2c91460 released
usbclose q 0x24 fid 620 ref 2
ehci: epclose ep7.1
ehci: cancelio for qh 0xf004e680 state idle
ehci: doorbell
ehci: qhcoherency: doorbell 1
usb: ep7.1 0xf2c91520 released
usbclose q 0x1c fid 617 ref 2
ehci: epclose ep7.0
ehci: cancelio for qh 0xf004e600 state idle
ehci: doorbell
ehci: qhcoherency: doorbell 1
usb: ep7.0 0xf343dbe0 released

- erik

Re: [9fans] usb regression

[Francisco J Ballesteros]

That's weird, I didn't change anything recently.

Do you know which change in sources broke your device?

On Mon, Mar 15, 2010 at 1:27 AM, erik quanstrom <quanstro@quanstro.net> wrote:
> i am now unable to copy more than about 10mb at a time
> off the usb sd reader i've been using.  usb/disk isn't just quitting,
> the endpoint is dead.  the console says only
>        usb/disk... usb/disk: : 062800
>        disk: reset: device is detached
> it used to work, so maybe something new's come up.
> the last copy i made was 20091020.
>
> here's the output with debug on:
>
> usb/disk: : 062800
> qhintr: td 0xf004e780 csw 0x10008d40 error 0x40 endpoint stalled
> ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
> ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
> ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
> ehci io 0xf2c912e0 qh 0xf004e600 timed out (no intr?)
> qhintr: td 0xf004e780 csw 0xd8d50 error 0x50 babble detected
> qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
> qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
> qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
> qhintr: td 0xf004e780 csw 0x88248 error 0x48 transaction error
> usb epctl detach ep7.0
> disk: reset: device is detached
> usbclose q 0x20 fid 616 ref 2
> ehci: epclose ep7.2
> ehci: cancelio for qh 0xf004e700 state idle
> ehci: doorbell
> ehci: qhcoherency: doorbell 1
> usb: ep7.2 0xf2c91460 released
> usbclose q 0x24 fid 620 ref 2
> ehci: epclose ep7.1
> ehci: cancelio for qh 0xf004e680 state idle
> ehci: doorbell
> ehci: qhcoherency: doorbell 1
> usb: ep7.1 0xf2c91520 released
> usbclose q 0x1c fid 617 ref 2
> ehci: epclose ep7.0
> ehci: cancelio for qh 0xf004e600 state idle
> ehci: doorbell
> ehci: qhcoherency: doorbell 1
> usb: ep7.0 0xf343dbe0 released
>
> - erik
>
>

Re: [9fans] usb regression

[erik quanstrom]

On Mon Mar 15 06:41:08 EDT 2010, nemo@lsub.org wrote:
> That's weird, I didn't change anything recently.
>
> Do you know which change in sources broke your device?
>

no.  in fact, now that i have a bit more experience with
this and reloaded the whole 2gb onto a usb hard drive, it
appears that it's device specific.  which got me thinking.
perhaps this sd card (different mfgr from the rest) causes
problems.  is there a way that usb could recover automaticly
from this sort of situation?

- erik

Re: [9fans] usb regression

[erik quanstrom]

oh, by the way, having the "debug on" option to the ctl file
was great.  i love not rebooting.  :-)

- erik

Re: [9fans] usb regression

[Francisco J Ballesteros]

usually, unplug, then plug.

Otherwise you involve even usbd and the kernel on the process
(you can do a port reset). The port reset code is in, but IIRC, only
the keyboard is using it so far. It's a quite hard measure, so I'd prefer
not to do a port reset unless we are desperated.


On Mon, Mar 15, 2010 at 1:31 PM, erik quanstrom <quanstro@quanstro.net> wrote:
> On Mon Mar 15 06:41:08 EDT 2010, nemo@lsub.org wrote:
>> That's weird, I didn't change anything recently.
>>
>> Do you know which change in sources broke your device?
>>
>
> no.  in fact, now that i have a bit more experience with
> this and reloaded the whole 2gb onto a usb hard drive, it
> appears that it's device specific.  which got me thinking.
> perhaps this sd card (different mfgr from the rest) causes
> problems.  is there a way that usb could recover automaticly
> from this sort of situation?
>
> - erik
>
>

Re: [9fans] usb regression

[erik quanstrom]

On Mon Mar 15 10:45:48 EDT 2010, nemo@lsub.org wrote:
> usually, unplug, then plug.
>
> Otherwise you involve even usbd and the kernel on the process
> (you can do a port reset). The port reset code is in, but IIRC, only
> the keyboard is using it so far. It's a quite hard measure, so I'd prefer
> not to do a port reset unless we are desperated.

i had to repeat this process >20 times to copy
all the data off the sd card.  it took well over an
hour, and it was hard to do other things at the
same time.  i would say the current state of affairs
(for me anyway) is despair.

- erik

Re: [9fans] usb regression

[Francisco J Ballesteros]

Ouch. Ok.
I'll add a port reset upon babling errors for disk drives as soon
as I have a bit of time for that. If it's urgent for you let me know.

On Mon, Mar 15, 2010 at 5:06 PM, erik quanstrom
<quanstro@labs.coraid.com> wrote:
> On Mon Mar 15 10:45:48 EDT 2010, nemo@lsub.org wrote:
>> usually, unplug, then plug.
>>
>> Otherwise you involve even usbd and the kernel on the process
>> (you can do a port reset). The port reset code is in, but IIRC, only
>> the keyboard is using it so far. It's a quite hard measure, so I'd prefer
>> not to do a port reset unless we are desperated.
>
> i had to repeat this process >20 times to copy
> all the data off the sd card.  it took well over an
> hour, and it was hard to do other things at the
> same time.  i would say the current state of affairs
> (for me anyway) is despair.
>
> - erik
>
>

[9fans] partdisk fails in install

[EBo]

I'm having trouble installing on a clean disk with the latest iso (2010/3/12).

I first removed all the partition info (using fdisk on a gparted liveCD) so
that the disk was clear.  I then booted Plan9, and when it asked to install a
MBR, it returned an error that disk/fdisk failed.  This then caused parted to
restart in a loop...  Unfortunately it did not give me much additional info.

I'm hoping to get Plan9 and Inferno installed native and VM on all my
development machines so I can test both versions in various configurations.
Unfortunately, it does not look like Plan9 will install native on any of my 3
machines so far.  One more to try...

  EBo --

Re: [9fans] usb regression

[erik quanstrom]

On Mon Mar 15 13:30:52 EDT 2010, nemo@lsub.org wrote:
> Ouch. Ok.
> I'll add a port reset upon babling errors for disk drives as soon
> as I have a bit of time for that. If it's urgent for you let me know.
>

thanks, now that you've given me a bit of a road map
of what the problem is and where an alternate solution
exists, i'll (try to) take care of it if it becomes an urgent problem.

thanks!

- erik

Re: [9fans] partdisk fails in install

[erik quanstrom]

> I'm having trouble installing on a clean disk with the latest iso (2010/3/12).
>
> I first removed all the partition info (using fdisk on a gparted liveCD) so
> that the disk was clear.  I then booted Plan9, and when it asked to install a
> MBR, it returned an error that disk/fdisk failed.  This then caused parted to
> restart in a loop...  Unfortunately it did not give me much additional info.

if you could send me some information (off list),
i'll take a look at this.
1.  the exact error message that you see.
2.  at the installer prompt, after then fdisk error,
the output of "!cat /dev/kmesg".
3.  the output of lspci -n (or similar) from linux
(or whatever).

- erik

Re: [9fans] partdisk fails in install

[EBo]

It will have to be later this afternoon/night.  I thwacked the disk with
gparted and gave it a couple of partitions, and am installing it on that
machine at the moment.  I at least want to see how far this install gets
before thwacking the disk again, and in a few I have to pick someone up at the
airport...

I will also go through the routine with all my machines and collate the info.

Thanks for the offer.  I look forward to sorting this out a little later.

  EBo --


erik quanstrom <quanstro@quanstro.net> said:

> > I'm having trouble installing on a clean disk with the latest iso (2010/3/12).
> >
> > I first removed all the partition info (using fdisk on a gparted liveCD) so
> > that the disk was clear.  I then booted Plan9, and when it asked to install a
> > MBR, it returned an error that disk/fdisk failed.  This then caused parted to
> > restart in a loop...  Unfortunately it did not give me much additional info.
>
> if you could send me some information (off list),
> i'll take a look at this.
> 1.  the exact error message that you see.
> 2.  at the installer prompt, after then fdisk error,
> the output of "!cat /dev/kmesg".
> 3.  the output of lspci -n (or similar) from linux
> (or whatever).
>
> - erik
>



--

[9fans] install filsystem hangs on bad block

[EBo]

Doing the initial native installation on an hpze1115, the installation hangs
when copying to disk.

Background:
  The old laptop has a disk which is known to have as many as 100 bad blocks,
but has continued to function fine.  While running copydist, the install ran
into the following errors then hung:

------------------------------
fossil: diskWriteRaw failed: /dev/sdC0/fossil: score 0x00000000: date ...
  part=superblock 0: i/o error
fossil: diskWriteRaw failed: /dev/sdC0/fossil: score 0x00000041: date ...
  part=label block 65: i/o error

and it hung at 65% installed.  I'll try a different partition.

Gots to go...

  EBo --

Re: [9fans] install filsystem hangs on bad block

[erik quanstrom]

> fossil: diskWriteRaw failed: /dev/sdC0/fossil: score 0x00000000: date ...
>   part=superblock 0: i/o error
> fossil: diskWriteRaw failed: /dev/sdC0/fossil: score 0x00000041: date ...
>   part=label block 65: i/o error
>
> and it hung at 65% installed.  I'll try a different partition.

those scores look funny.  so also to keep in mind, ...

fossil i/o can fail for lots of reasons that don't have
anything to do with the disk.  for example, if its partitions
are set up incorrectly.  (the installer is not infallable.)

it would be good to double-check that your partition
sizes make sense.

- erik

Re: [9fans] install filsystem hangs on bad block

[EBo]

> > fossil: diskWriteRaw failed: /dev/sdC0/fossil: score 0x00000000: date ...
> >   part=superblock 0: i/o error
> > fossil: diskWriteRaw failed: /dev/sdC0/fossil: score 0x00000041: date ...
> >   part=label block 65: i/o error
>
> those scores look funny.  so also to keep in mind, ...
>
> fossil i/o can fail for lots of reasons that don't have
> anything to do with the disk.  for example, if its partitions
> are set up incorrectly.  (the installer is not infallable.)
>
> it would be good to double-check that your partition
> sizes make sense.

I forget the disk portioning I first set up, but I think that it was 3
consecutive partitions with 20G each, then a forth (logical) with the rest of
the 120G drive.  I had not split up the logical partition yet.  Actually I
think it tried to use the logical one (p4).  I'll try this again again later.

Thanks for taking look at this!

  EBo --

Re: [9fans] install filsystem hangs on bad block

[EBo]

> > it would be good to double-check that your partition
> > sizes make sense.

I was finally able to replicate two related problems.

The first concerns the logical partition that was not further broken into
partitions..  I was able to verify that the install would install directly
onto the logical partition and copying of the os would then cause write errors
later on.

The second problem arose when I repartitioned the disk with partdisk and it
did not appear to have been updated when I continued with the install.  I
wonder if the partition map had not been updated with plan9's partprobe
equivelent.

  EBo --

Re: [9fans] install filsystem hangs on bad block

[erik quanstrom]

> The second problem arose when I repartitioned the disk with partdisk and it
> did not appear to have been updated when I continued with the install.  I
> wonder if the partition map had not been updated with plan9's partprobe
> equivelent.

did you type 'w' then 'q' at the fdisk, the prep prompts, thus accepting
and writing the defaults?

- erik

Re: [9fans] install filsystem hangs on bad block

[EBo]

Yes.

As a note, I kept having write errors pop up, so I ran a low level disk
format.   So far it is running better, but have not tried to replicate various
weird things I tripped over earlier.  I'll try that again later.

  EBo --

erik quanstrom <quanstro@quanstro.net> said:

> > The second problem arose when I repartitioned the disk with partdisk and it
> > did not appear to have been updated when I continued with the install.  I
> > wonder if the partition map had not been updated with plan9's partprobe
> > equivelent.
>
> did you type 'w' then 'q' at the fdisk, the prep prompts, thus accepting
> and writing the defaults?
>
> - erik
>



--