Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
printing test page gets stuck
View unanswered posts
View posts from last 24 hours

Goto page 1, 2  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Wed Nov 16, 2016 3:12 pm    Post subject: printing test page gets stuck Reply with quote

Hi all,

I have a canon pixma MP240 that was working fine on my old computer; unfortunately, on my new computer it gets stuck (extremely slow process) half way through printing. It seems that at a certain point the printer is getting a huge amount of data from the computer for no reason (I may be wrong here).

The printer is connected through USB port, I am using cups and gutenprint. I tried:

net-print/gutenprint-5.2.10 and net-print/gutenprint-5.2.9
net-print/cups-2.1.3-r1 and net-print/cups-2.0.3
app-text/ghostscript-gpl-9.15-r1 and app-text/ghostscript-gpl-9.10-r2

cups doesnt complain about the kernel setting...

could
00:14.0 USB controller: Intel Corporation Sunrise Point-H USB 3.0 xHCI Controller (rev 31)
cause the issue?

I am running out of ideas, any help is appreciated.

Thanks,

Efo
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Wed Nov 16, 2016 3:45 pm    Post subject: Reply with quote

Efo,

Your printer expects raster data that describes the placement of the ink on the paper almost line by line, dot by dot.
On *NIX, applications typically output printed pages using the PostScript page description language.
The idea being that rendering a page onto the printed media is a left up to the output device. This means that the output device needs to understand how to get PostScript onto hard copy.

This is no use at all to your printer, nor to lots of simple devices like it. This is where ghostscript comes in. It converts PostScript into the raster format needed by your printer.
It can be slow. PostScript uses scalable vector graphics where it can. This allows the PostScript interpreter (ghostscript) to scale the image to any size, however, scaling is slow.
Windows users on the other hand, don't get scaling at all.

That it works at all is a good sign. For more diagnostics do the following.

In /etc/cups/cupsd.conf change.
Code:
# Log general information in error_log - change "warn" to "debug"
# for troubleshooting...
LogLevel warn
to
Code:
LogLevel debug
You will get a lot ouf output in the cups log.

Remove the file /var/log/cups/error_log since we only want output from the next test.
Restart cupsd, so it rereads the cupsd.conf file.
Print the test page.
Stop cupsd
Use wgetpaste to post /var/log/cups/error_log
Reset the cupsd Loglevel.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Thu Nov 17, 2016 1:43 am    Post subject: Reply with quote

Hi NeddySeagoon,

Thank you for the reply and the good explanation. I followed your instructions and now for some reason it wont print and the status is pending.

Here is the error_log.

Thanks!
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Thu Nov 17, 2016 10:25 am    Post subject: Reply with quote

Efo,

Here is the interesting bit, from sending the test page to the printer to the broken pipe error.

Code:
D [16/Nov/2016:17:34:33 -0800] [CGI] cgiSetVariable: TITLE=\"Print Test Page\"
D [16/Nov/2016:17:34:33 -0800] [CGI] cgiSetVariable: SERVER_NAME=\"localhost\"
D [16/Nov/2016:17:34:33 -0800] [CGI] cgiSetVariable: CUPS_VERSION=\"CUPS v2.1.3\"
D [16/Nov/2016:17:34:33 -0800] [Client 1] CGI data ready to be sent.
D [16/Nov/2016:17:34:33 -0800] [Client 1] con->http=0x5573227256e0
D [16/Nov/2016:17:34:33 -0800] [Client 1] cupsdWriteClient error=0, used=0, state=HTTP_STATE_POST_SEND, data_encoding=HTTP_ENCODING_LENGTH, data_remaining=2147483647, response=(nil)(), pipe_pid=25307, file=16
D [16/Nov/2016:17:34:33 -0800] [Client 1] Waiting for CGI data.
D [16/Nov/2016:17:34:33 -0800] [Client 1] Script header: Content-Type: text/html;charset=utf-8
D [16/Nov/2016:17:34:33 -0800] [Client 1] Script header:
D [16/Nov/2016:17:34:33 -0800] [Client 1] Sending status 200 for CGI.
D [16/Nov/2016:17:34:33 -0800] [Client 1] cupsdSendHeader: code=200, type="(null)", auth_type=0
D [16/Nov/2016:17:34:33 -0800] [CGI] cgiSetVariable: PRINTER_NAME=\"Canon_MP240_series\"
D [16/Nov/2016:17:34:33 -0800] [Client 1] con->http=0x5573227256e0
D [16/Nov/2016:17:34:33 -0800] [Client 1] cupsdWriteClient error=0, used=0, state=HTTP_STATE_POST_SEND, data_encoding=HTTP_ENCODING_CHUNKED, data_remaining=0, response=(nil)(), pipe_pid=25307, file=16
D [16/Nov/2016:17:34:33 -0800] [Client 1] Waiting for CGI data.
D [16/Nov/2016:17:34:33 -0800] [Client 1] CGI data ready to be sent.
D [16/Nov/2016:17:34:33 -0800] [Client 2] HTTP_STATE_WAITING Closing for error 32 (Broken pipe)


A pipe is a link between two processes. If you ever use the | command like
Code:
dmesg | grep sda
that 'pipes' the output of dmesg to the input of grep.
At
Code:
data_remaining=2147483647
we see that there is about 2G of data in your test page. (That's a lot)

I am not sure if the pipe is also a buffer - that is the output program expects the pip to store output data until the input program is ready for it. The pipe is in /run, whird is in tmpfs in your RAM. By default, its half of RAM.

Repeat the experiment. This time post the cups log, the output of dmesg and the output of
Code:
df -h

dmesg may say something about broken pipes and df-h will show your filesystem sizes.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Thu Nov 17, 2016 2:50 pm    Post subject: Reply with quote

Hi NeddySeagoon,

I repeated the process, this time I also got
Code:
 * Stopping cupsd ...                                                     [ ok ]
/lib64/rc/sh/rc-cgroup.sh: line 98: /sys/fs/cgroup/openrc/cupsd/tasks: No such file or directory
 * Starting cupsd ...

upon cupsd restart. Upon restarting few times this message disappeared, not sure whether it is relevant.

Here is the error_log output
Here is the demsg output
Here is df -h
Code:
Filesystem      Size  Used Avail Use% Mounted on
udev             10M  4.0K   10M   1% /dev
/dev/sda4       413G  105G  287G  27% /
tmpfs           3.2G  652K  3.2G   1% /run
shm              16G  176K   16G   1% /dev/shm
cgroup_root      10M     0   10M   0% /sys/fs/cgroup
/dev/sda2       127M   23M  104M  18% /boot


This may also help
Code:
GNU Parted 3.2
Using /dev/sda
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) print
Model: ATA Samsung SSD 850 (scsi)
Disk /dev/sda: 500GB
Sector size (logical/physical): 512B/512B
Partition Table: gpt
Disk Flags:

Number  Start   End     Size    File system     Name    Flags
 1      1049kB  3146kB  2097kB  fat32           grub    bios_grub
 2      3146kB  137MB   134MB   fat32           boot    boot, esp
 3      137MB   50.5GB  50.3GB  linux-swap(v1)  swap
 4      50.5GB  500GB   450GB   ext4            rootfs


Thanks again for the detailed explanation, it really helps understanding what's going on.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Thu Nov 17, 2016 6:05 pm    Post subject: Reply with quote

Efo,

The easy bit first.
df -h:
tmpfs           3.2G  652K  3.2G   1% /run
That's lots of room for the 2G test page.

dmesg:
[40315.343127] usb 1-1: new high-speed USB device number 4 using xhci_hcd
[40315.509553] usb 1-1: New USB device found, idVendor=04a9, idProduct=1732
[40315.509558] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[40315.509560] usb 1-1: Product: MP240 series
[40315.509562] usb 1-1: Manufacturer: Canon
[40315.509564] usb 1-1: SerialNumber: A32EEA
... mouse messages ..
[41086.701659] usb 1-1: USB disconnect, device number 4

shows the printer being connected and disconnected. That's normal, as long as it didn't disconnect on its own.
That would point to a USB bus issue. There are no printing related errors.

Hmm.
Code:
 [    0.442987] usb: port power management may be unreliable
That's probably harmless as if USB power management was an issue, the printer would probably connect and reconnect on its own. There are so signs of that. Keep it in mind for later.

Do you have anything to build in
Code:
emerge @preserved-rebuild
If so let it run. Its fixing things that need te be rebuilt against updated libraries.

Google suggests that
Code:
emerge -1 cups-filters
is worth trying too.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Fri Nov 18, 2016 1:36 am    Post subject: Reply with quote

Hi NeddySeagoon,

I tried
Code:
emerge @preserved-rebuild
but (as expected) nothing came up.

I also
Code:
emerge -1 cups-filters
and restarted cups; the test page gets stuck half way and is very slow. I have a feeling it has something to do with those 2G.... I doubt the test page is that size; what may induce the generation of such a large file?
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Fri Nov 18, 2016 10:16 am    Post subject: Reply with quote

Efo,

The file is so big as it describes every dot of ink to be placed on the page. That's why its slow too.
I suspect that a page of black text, will be much faster. The amount of data to be transferred depends on the dots per inch and colour depth.
e.g. at 600 dots per linear inch (average laser printer) there are 360,000 dots per square inch. If each dot needs one byte, 256 different shades, that's 360kB per square inch.
For colour, it would be one byte per colour, fer 3 times that. Your printer can do 2400x1200 dots per inch and that will be used on the test page. Allow one byte per cartridge and 2400x1200 dots per inch. You can work out the maximum file size to describe a printed page to the printer.

Test pages normally have some complex images to show of the printer - these can be large and slow.

Does the printer actually go very slowly or does printing fail?
look at the end of the log before you stop it. If there is no error, its still doing something.
Code:
tail -f /path/to/log
will update -forever. Close the window or ctrl-C to stop it.

Doing a little more Googling, The Open Printing page shows a method for the MP250.
Its not you exact model but use it as a guide. You may already have done this without realising it, so consider it a checklist.

The Gutenprint 5.2.11 page lists your exact model. It lists two drivers.
Code:
Canon MP240 series   bjc-MP240-series
Canon PIXMA MP240   bjc-MULTIPASS-MP240   
You should expect better results with the second one. Are you using the bjc-MULTIPASS-MP240 driver?
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Sat Nov 19, 2016 1:45 am    Post subject: Reply with quote

Hi NeddySeagoon,

It prints half page and then it gets stuck; the lights on the printer's dial that usually spin around pretty fast are very slow, at times stop and then spin slowly again. It is definitely slow.

I tried both drivers and currently using
Code:
Driver:   Canon PIXMA MP240 - CUPS+Gutenprint v5.2.10 (color)
not sure they are bjc-MULTIPASS-MP240 or bjc-MP240-series. I tried to select bjc-MP240-series directly but it doesnt work (couldnt find bjc-MULTIPASS-MP240).

The link you provided seems to recommend cnijfilter-mp250series; I tried it in the past and it was very painful, that's why I switched to gutenprint. It looks like the "discussion forum" and "miscellaneous" on that website are not functioning.

The printer worked fine on the old computer, so I looked at the backup of the old computer and it was using gutenprint 5.2.7; I tried to use that ppd but cups complains that is not compatible with 5.2.10. I could try to install gutenprint-v5.2.7 but it is not in the tree anymore (not sure it is a good idea to rely on obsolete packages).

I am running out of ideas; it should work as well as it did on the old computer...

EDIT: the printer test page doesnt seem too complex
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Sat Nov 19, 2016 1:16 pm    Post subject: Reply with quote

Efo,

That's valuable information - that it used to work but doesn't now.

Code:
$ eix gutenprint
* net-print/gutenprint
     Available versions:  5.2.9^t 5.2.10^t (~)5.2.11^t {cups foomaticdb gimp gtk nls ppds readline static-libs}
     Homepage:            http://gutenprint.sourceforge.net
     Description:         Ghostscript and cups printer drivers


Going back to obsolete software is not a long term solution but it bay be useful for testing. There are three versions of gutenprint in the tree.
You have 5.2.10. Its worth trying 5.2.9 and/or 5.2.11.

Code:
emerge -1 =net-print/gutenprint-5.2.9
will install version 5.2.9.
5.2.11 is keyword masked for you, so you will need to unmask it before portage will consider installing it.
See Mixing stable with testing.

I would expect rendering the colour wheel to be slow, but after its got past that part, the rest is easy.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Sat Nov 19, 2016 4:45 pm    Post subject: Reply with quote

Hi NeddySeagoon,

I already tried v9 and v10; I gave it a shot to v11 with the same outcome.

Interestingly, the printer gets stuck (slow) past the color/images part; the top 80% is printed when it gets stuck (basically there are only 2 text lines left and the margin).
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Sun Nov 20, 2016 10:53 am    Post subject: Reply with quote

Efo,

I don't understand that at all. Printing normally works or nothing happens at all.

If you want to try the old gutenprint-5.2.7 you will need to dig the ebuild out of the CVS attic. As you say, its not a long term fix.

Ideally, you put it in your own overlay so your next --sync does not remove it.
Once you have the ebuild in place. run
Code:
ebuild /path/to/ebuild/ manifest
This will download the sources and create the manifest file.
The sources will no longer be mirrored by Gentoo, so you may need to find them and add them to your distfiles by hand before the above command succeeds.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Sun Nov 20, 2016 7:28 pm    Post subject: Reply with quote

Hi NeddySeagoon,

Thank you for your help and patience. I installed 5.2.7 and I still have the same problem, at this point I begin to think that is not a driver issue... maybe there is another process that sends data to the printer?

Can useflags on gutenprint or cups cause this? (I tried to change some of them without success)

Maybe some configuration in cups (localhost:631)?

In the meantime I am back to 5.2.10...
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Sun Nov 20, 2016 9:27 pm    Post subject: Reply with quote

Efo,

There are two ways to set this up. The old way, which uses a kernel driver and the new way which uses USE=usb.
The methods are mutually exclusive. Half of one and half ot the other won't work.

USE=usb is preferred and kernel support for USB printing (if its still there) off.

-- edit --

Post
Code:
emerge --info
and pastebin the dmesg output after a failed print.

Make a test page of your own. Just the letter 'A' in each of the four corners of the page.
Does that fail in the same place or print correctly?
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Sun Nov 20, 2016 11:08 pm    Post subject: Reply with quote

Hi NeddySeagoon,

I have disabled usb print in the kernel.

I have : eix -I cups
Code:
[I] net-print/cups
     Available versions:  2.0.3^t ~2.0.4^t 2.1.3-r1^t 2.1.4^t **9999^t {X acl dbus debug java kerberos lprng-compat pam python selinux +ssl static-libs systemd +threads usb xinetd zeroconf ABI_MIPS="n32 n64 o32" ABI_PPC="32 64" ABI_S390="32 64" ABI_X86="32 64 x32" ELIBC="FreeBSD" LINGUAS="+ca +cs +de +es +fr +it +ja +ru" PYTHON_TARGETS="python2_7"}
     Installed versions:  2.1.4^t(08:40:30 AM 11/20/2016)(X acl dbus pam ssl threads usb -debug -java -kerberos -lprng-compat -python -selinux -static-libs -systemd -xinetd -zeroconf ABI_MIPS="-n32 -n64 -o32" ABI_PPC="-32 -64" ABI_S390="-32 -64" ABI_X86="32 64 -x32" ELIBC="-FreeBSD" LINGUAS="ca cs de es fr it ja ru" PYTHON_TARGETS="python2_7")
     Homepage:            http://www.cups.org/
     Description:         The Common Unix Printing System

[I] net-print/cups-filters
     Available versions:  1.5.0 ~1.8.3 ~1.11.6 **9999 {dbus +foomatic jpeg ldap pdf perl png +postscript static-libs tiff zeroconf}
     Installed versions:  1.5.0(05:09:02 PM 11/17/2016)(dbus foomatic jpeg png tiff -ldap -perl -static-libs -zeroconf)
     Homepage:            https://wiki.linuxfoundation.org/openprinting/pdf_as_standard_print_job_format
     Description:         Cups PDF filters

Found 2 matches


and eix -I gutenprint
Code:
[I] net-print/gutenprint
     Available versions:  5.2.7-r1^t[1] 5.2.9^t 5.2.10^t ~5.2.11^t {cups foomaticdb gimp gtk nls ppds readline static-libs}
     Installed versions:  5.2.10^t(11:19:43 AM 11/20/2016)(cups nls readline -foomaticdb -gimp -gtk -ppds -static-libs)
     Homepage:            http://gutenprint.sourceforge.net
     Description:         Ghostscript and cups printer drivers

[1] "fixes" /var/lib/layman/fixes


Here is emerge --info

Here is dmesg while the printer is stuck.

I am printing one and a half pages of text, and the printer went through smoothly the first page but it got stuck on the second half page. It looks like it may have something to do with reaching the end of data.

EDIT: I forgot to mention that I printed a page with "A" at the four corners without a problem. So my theory about the end of data may be wrong.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Mon Nov 21, 2016 12:09 am    Post subject: Reply with quote

Efo,

You should fix
Code:
[    0.368765] i915 0000:00:02.0: Direct firmware load for i915/skl_dmc_ver1.bin failed with error -2
[    0.368788] [drm:i915_firmware_load_error_print] *ERROR* failed to load firmware i915/skl_dmc_ver1.bin (0)
[    0.368810] [drm:i915_firmware_load_error_print] *ERROR* The driver is built-in, so to load the firmware you need to
               include it either in the kernel (see CONFIG_EXTRA_FIRMWARE) or
               in your initrd/initramfs image.
It will improve your graphics performance. You don't have and 3D hardware acceleration without the firmware.

Code:
[  785.741043] kworker/dying (47) used greatest stack depth: 12624 bytes left
[23593.710378] kworker/dying (24810) used greatest stack depth: 12608 bytes left

That's not supposed to happen, nor is all the
Code:
 [30171.525085] psmouse serio2: TouchPad at isa0060/serio2/input0 lost synchronization, throwing 4 bytes away.


I suspect you are right about the printer reaching the end of data. Big print jobs never get completely converted to the raster image due to the "broken pipe" error we saw in the cups debug log, thus the end of the printer image is not available to send to the printer.

Are you having shutdown issues?
Code:
[    3.253630] EXT4-fs (sda4): INFO: recovery required on readonly filesystem
[    4.871561] FAT-fs (sda2): Volume was not properly unmounted. Some data may be corrupt. Please run fsck.

That's your root and /boot. When you shut down cleanly, that shouldn't happen.

I suspect that the printing issue will go away when other things are fixed.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Mon Nov 21, 2016 2:16 am    Post subject: Reply with quote

Hi NeddySeagoon,

I feel like we just open a can of worms :(

EDIT: I fixed the graphic issue by setting
Code:
CONFIG_EXTRA_FIRMWARE="i915/skl_dmc_ver1.bin i915/skl_guc_ver4.bin"
CONFIG_EXTRA_FIRMWARE_DIR="/lib/firmware"


I thought that "used greatest stack depth" is just a debugging message, not sure how to fix this.

I usually reboot the box with
Code:
/sbin/reboot

and shutdown with
Code:
/sbin/shutdown -h now


Now I will google about touchpad...

I think you may be right that the printer issue may be solved once I get rid of these errors.

EDIT: I dont really have a problem with the touchpad
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Mon Nov 21, 2016 10:29 am    Post subject: Reply with quote

Efo,

The greatest stack depth is indeed for information. The problem is kworker/dying. The number in brackets in the process ID.
That's a kernel process that failed to complete.

The touchpad "lost sync" may not show any user symptoms as the kernel recovers. However, this sort of thing happens when the kernel does not read the touchpad data fast enough because its doing something else. It shouldn't happen and indicates a problem elsewhere.

Yes, it could be a can of worms.

/sbin/reboot is OK. Its the same as
Code:
shutdown -r now

The errors on your root filesystem should have gone away as a result of the recovery in your dmesg post.

The errors on /boot will persist until you run fsck on it. A few words of warning before you do that.
Your filesystem is inconsistent. fsck will make guesses about what it should be to resolve the inconsistencies. When it guesses incorrectly, it will make things worse not better. Be sure you have a backup of all the files on /boot before you run fsck on it.
Lastly, fsck can only be run on as unmounted filesystem.
For completeness, this is just a tidy up, its not related to your printing problem.

Connect your printer - make a note of the USB port you use, it may be important later.
Boot the system. Do not print anything.
Pastebin
Code:
lsusb -vvv
this will show the usb bus topology.
Pastebin
Code:
dmesg
so we can see if the errors are fixed. If the touchpad errors do not appear, maybe they are related to printing?
Pastebin your kernel .config
Post your
Code:
lspci
so we can look at your hardware.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Tue Nov 22, 2016 1:16 am    Post subject: Reply with quote

Hi NeddySeagoon,

Thank you for the explanations, I really enjoy learning abot these things.

So, I remember that I forgot to plug in my laptop and that's probably the mounting error we saw. I forced check (after backing up) and it went away (btw SSHD are fast!).

I like a tidy box.

Here is lsusb
Here is dmesg
Here is .config
Here is lspci
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Tue Nov 22, 2016 10:18 am    Post subject: Reply with quote

Efo,

From lsusb, you have two USB 3 root hubs. No USB 2 or USB 1 at all. That's OK as USB 3 is backwards compatible
They are the root of bus 001 and bus 002 respectively.
This is confirmed by lspci
Quote:
00:14.0 USB controller: Intel Corporation Sunrise Point-H USB 3.0 xHCI Controller (rev 31)
There are no EHCI nor UHCI devices.

Bus 001 has
Code:
Bus 001 Device 001: xHCI Host Controller MaxPower                0mA
Bus 001 Device 002: MP240 series'  MaxPower                2mA
Bus 001 Device 003: Chicony USB 2.0 Camera, MaxPower              500mA
Bus 001 Device 004: Bluetooth, MaxPower              100mA


Bus 002 is not used. A USB 3 root hub can provide 900mA to power all connected devices. You have 602mA connected so that's good. We know another thing that its not.
Your kernel has support for USB2 and USB1 hardware, which you don't need. Its just extra clutter so you can remove that next time you build your kernel if you wish.

Code:
# CONFIG_USB_PRINTER is not set
Good. That's the old way. To get anything at all, you must be using libusb, which means you build your system with USE=usb. A mix gives problems.

dmesg says
Code:
[   35.574113] psmouse serio2: TouchPad at isa0060/serio2/input0 lost synchronization, throwing 5 bytes away.
and we know that its not caused by printing because you didn't print anything. It says nothing about the issue causing this interfering with printing though.

So far, its all good. There is nothing obvious. I have something at the back of my mind for the touchpad issue but I need to check a few things first.
I don't want to recommend something that leaves you with a non operational touch pad until you boot an old kernel.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Tue Nov 22, 2016 2:21 pm    Post subject: Reply with quote

Hi NeddySeagoon,

In my hardware description I should have
Code:
3× USB 3.0 Type-A, 1× USB 2.0 Type-A, SD Card Reader, DVD-RW Drive
I definitely have 4 USB ports.

I modified the kernel with
Code:
< >     OHCI HCD (USB 1.1) support


Not sure about EHCI right now since I am supposed to have 1 USB 2.0.

Btw, should I have
Code:
 <*>     UHCI HCD (most Intel and VIA) support
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Tue Nov 22, 2016 3:34 pm    Post subject: Reply with quote

Efo,

XHCI is the driver for USB 3 root hubs
EHCI is the driver for USB 2 root hubs.

USB 1 was a bit of a mess. There were two competing standards. Intel and VIA used UHCI, the rest of the world used OHCI.
After that, the industry backed single standards.

You have a mix of devices. Your webcam is USB2 but its connected to a USB 3 bus.
EHCI, OHCI and UHCI are not used.

Following up on the the broken pipe is a memory issue, lets try resizing /run.
Execute
Code:
df -h
. It will display some information about mounted file systems, including /run
Code:
Filesystem                Size  Used Avail Use% Mounted on
tmpfs                     1.6G  148K  1.6G   1% /run
thats from my system right now.
The size can be changed.
Code:
sudo mount -o remount,size=50%  /run
The % here is the percentage of RAM to allocate.
There are other ways to give the size. This is a maximum amount. tmpfs will only use what's needed.
Code:
Filesystem                Size  Used Avail Use% Mounted on
tmpfs                     7.9G  148K  7.9G   1% /run
so /run has got bigger.
Make your /run about 3G and try the test page again.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Tue Nov 22, 2016 4:17 pm    Post subject: Reply with quote

Hi NeddySeagoon,

If I understand it correctly I can disable EHCI because my USB 2.0 port is on USB 3.0 bus which is managed by XHCI?
EDIT: basically I have 2 USB 3.0 buses: one with 2 USB 3.0 ports and one with 1 USB 3.0 port and 1 USB 2.0 port?

I will disable UHCI and EHCI as well as checking/resizing /run and retest the printer as soon as I get home.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42860
Location: 56N 3W

PostPosted: Tue Nov 22, 2016 5:45 pm    Post subject: Reply with quote

Efo,

You have two USB 3 ports.
USB 3 is backwards compatible with USB 2 and USB 1. No extra drivers are required.

You do have USB 2 devices connected to one of your USB 3 buses but that's OK.
The backwards compatibility takes care of it.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Efo
Guru
Guru


Joined: 23 Jun 2005
Posts: 420

PostPosted: Wed Nov 23, 2016 1:55 am    Post subject: Reply with quote

Hi NeddySeagoon,

I removed the EHCI and UHCI from the kernel. Rebooted and did the following

Code:
df -h

gives
Code:
Filesystem      Size  Used Avail Use% Mounted on
udev             10M  4.0K   10M   1% /dev
/dev/sda4       413G  106G  287G  27% /
tmpfs           3.2G  656K  3.2G   1% /run
shm              16G     0   16G   0% /dev/shm
cgroup_root      10M     0   10M   0% /sys/fs/cgroup
/dev/sda2       127M   23M  104M  18% /boot

looks like it is already 3.2G so I didnt change it.

I tried the test page and I am having the same problem....
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum