[yast-devel] Request for testing a new hwinfo
Hi all, I have reimplemented monitor detection in hwinfo to read the data from /sys instead of running BIOS emulation and reading the data from the VESA BIOS. This should work better and should be more reliable. The new functionality requires kernel mode setting (KMS) to be active (which by default is), when KMS is not active the old BIOS emulation is still used. I have tested it on few machines and for me it works better than the old implementation (e.g. in a dual head setup with two monitors the old code detected just one monitor, with the new code it correctly detects both monitors). However, it would be nice to test it on a broader range of hardware to find possible regressions. The change has been implemented in hwinfo-21.13, it should be available soon in the Factory. To make testing easier I also have built 13.2 and SLE12 packages. How to test it in 13.2 or SLE12 (run as root): # store the original output to see the difference hwinfo --monitor > monitor1.log # add the 13.2 repo zypper ar -f http://download.opensuse.org/repositories/home:/lslezak/openSUSE_13.2_Update... hwinfo_test # or the SLE12 repo zypper ar -f http://download.opensuse.org/repositories/home:/lslezak/SLE12/ hwinfo_test # install the packages zypper dup --from hwinfo_test # run the new version hwinfo --monitor > monitor2.log # check the difference diff -u monitor1.log monitor2.log If you see some regressions (e.g. a monitor not detected) or some strange differences then let me know (either reply here or report it in bugzilla, do not forget to mention the manually updated hwinfo package so it's clear that it's not about the released package). The new version prints some additional properties (monitor manufacture date), so there should be at least a little change in the output... Of course, reporting success or improvements here is also appreciated ;-) Thank you a lot for testing!! -- Best Regards Ladislav Slezák Yast Developer ------------------------------------------------------------------------ SUSE LINUX, s.r.o. e-mail: lslezak@suse.cz Lihovarská 1060/12 tel: +420 284 028 960 190 00 Prague 9 fax: +420 284 028 951 Czech Republic http://www.suse.cz/ -- To unsubscribe, e-mail: yast-devel+unsubscribe@opensuse.org To contact the owner, e-mail: yast-devel+owner@opensuse.org
El Miércoles, 3 de junio de 2015 17:00:35 Ladislav Slezak escribió:
Hi all,
Hi Ladislav,
I have reimplemented monitor detection in hwinfo to read the data from /sys instead of running BIOS emulation and reading the data from the VESA BIOS. This should work better and should be more reliable.
Cool.
The new functionality requires kernel mode setting (KMS) to be active (which by default is), when KMS is not active the old BIOS emulation is still used.
I have tested it on few machines and for me it works better than the old implementation (e.g. in a dual head setup with two monitors the old code detected just one monitor, with the new code it correctly detects both monitors). However, it would be nice to test it on a broader range of hardware to find possible regressions.
Well done! I've tried in my laptop and it worked just fine. Anyway, I'm attaching the monitor1.log, monitor2.log and also a file containing diferences for further reference. [..]
Of course, reporting success or improvements here is also appreciated ;-)
Thank you a lot for testing!!
You are welcome! -- Imobach González Sosa YaST Team at SUSE LINUX GmbH
On 6/3/2015 at 09:00 AM, Ladislav Slezak <lslezak@suse.cz> wrote: I have reimplemented monitor detection in hwinfo to read the data from /sys instead of running BIOS emulation and reading the data from the VESA BIOS. This should work better and should be more reliable.
I have tested it on few machines and for me it works better than the old implementation (e.g. in a dual head setup with two monitors the old code detected just one monitor, with the new code it correctly detects both monitors).
For my dual head setup it now correctly reports two monitors where the old SLE12 version reported three. output attached.
How to test it in 13.2 or SLE12 (run as root):
# store the original output to see the difference hwinfo --monitor > monitor1.log
# add the 13.2 repo zypper ar -f http://download.opensuse.org/repositories/home:/lslezak/openSUSE_13.2_Update / hwinfo_test # or the SLE12 repo zypper ar -f http://download.opensuse.org/repositories/home:/lslezak/SLE12/ hwinfo_test
btw - typo in the URL ^^^
Hello,
Am Mittwoch, 3. Juni 2015 schrieb Ladislav Slezak:
> I have reimplemented monitor detection in hwinfo to read the data from
> /sys instead of running BIOS emulation and reading the data from the
> VESA BIOS. This should work better and should be more reliable.
> The new version prints some additional properties (monitor manufacture
> date), so there should be at least a little change in the output...
>
> Of course, reporting success or improvements here is also appreciated
> ;-)
I installed the 13.2 package on Tumbleweed, because hwinfo 21.13 didn't
make it into Tumbleweed yet.
Oh, and the 21.14 package in system:install:head/hwinfo has a 0 byte
tarball :-/
On my laptop (with an additional external monitor attached), I get
--- monitor1.log 2015-06-06 15:53:24.036726348 +0200
+++ monitor2.log 2015-06-06 15:59:25.362474177 +0200
@@ -1,6 +1,7 @@
-33: None 00.0: 10000 Monitor
- [Created at monitor.95]
+26: None 00.0: 10000 Monitor
+ [Created at monitor.125]
Unique ID: rdCR.+Gf2yBtn0F1
+ Parent ID: VCu0._YOiuWv5Ly7
Hardware Class: monitor
Model: "BenQ GL2450H"
Vendor: BNQ
@@ -20,6 +21,8 @@
Resolution: 1600x900@60Hz
Resolution: 1920x1080@60Hz
Size: 531x298 mm
+ Year of Manufacture: 2014
+ Week of Manufacture: 33
Detailed Timings #0:
Resolution: 1920x1080
Horizontal: 1920 2008 2052 2200 (+88 +132 +280) +hsync
@@ -31,19 +34,24 @@
Hor. Sync Range: 30-83 kHz
Bandwidth: 148 MHz
Config Status: cfg=new, avail=yes, need=no, active=unknown
+ Attached to: #20 (VGA compatible controller)
-34: None 00.2: 10002 LCD Monitor
- [Created at monitor.95]
- Unique ID: aHB6.TXuwCXVLBM9
+27: None 01.0: 10002 LCD Monitor
+ [Created at monitor.125]
+ Unique ID: wkFv.TXuwCXVLBM9
+ Parent ID: VCu0._YOiuWv5Ly7
Hardware Class: monitor
Model: "X176G LCD Monitor"
Vendor: LPL "X176G"
Device: eisa 0x0000
Resolution: 1440x900@60Hz
Size: 367x230 mm
+ Year of Manufacture: 2007
+ Week of Manufacture: 0
Detailed Timings #0:
Resolution: 1440x900
Horizontal: 1440 1520 1568 1968 (+80 +128 +528) -hsync
Vertical: 900 901 904 914 (+1 +4 +14) -vsync
Frequencies: 108.00 MHz, 54.88 kHz, 60.04 Hz
Config Status: cfg=new, avail=yes, need=no, active=unknown
+ Attached to: #20 (VGA compatible controller)
I'm slightly surprised that I get a different Unique ID for the laptop
display - you should know better if this is intentional.
The full hwinfo output has some more differences. Just as an example:
>> misc.1.2: open parallel
------ exec: "/sbin/modprobe parport " -----
+----- exec: "/sbin/rmmod lp" -----
+----- return code: ? -----
+----- exec: "/sbin/rmmod parport_pc" -----
+----- return code: ? -----
+----- exec: "/sbin/rmmod parport" -----
+ rmmod: ERROR: Module parport is in use by: ppdev
----- return code: ? -----
Maybe you need to rmmod ppdev first?
Needless to say that my laptop doesn't have a parport ;-)
If you are interested in the full hwinfo output (from 21.12. and 21.13),
take a look at www.cboltz.de/tmp/dl/hwinfo-logs.tar.gz
Regards,
Christian Boltz
--
Also, monolithisch im Sinne von "das macht einfach alles und streichelt
auch noch die Katze", nicht im Sinne von einem großen Code Blob.
[Andreas Scherbaum über systemd auf
https://plus.google.com/117024231055768477646/posts/7tJUsZidkr6]
--
To unsubscribe, e-mail: yast-devel+unsubscribe@opensuse.org
To contact the owner, e-mail: yast-devel+owner@opensuse.org
participants (4)
-
Christian Boltz -
Imobach González Sosa -
Ladislav Slezak -
Scott Reeves