Bug 48731

Give me some time to get organized this(attachments).

user@user-HP-655-Notebook-PC:/$ lsb_release -rd 
Description:	Ubuntu 12.10
Release:	12.10

user@user-HP-655-Notebook-PC:/$ cat /proc/cpuinfo 
processor	: 0
vendor_id	: AuthenticAMD
cpu family	: 20
model		: 2
model name	: AMD E2-1800 APU with Radeon(tm) HD Graphics
stepping	: 0
microcode	: 0x500010d
cpu MHz		: 850.000
cache size	: 512 KB
physical id	: 0
siblings	: 2
core id		: 0
cpu cores	: 2
apicid		: 0
initial apicid	: 0
fpu		: yes
fpu_exception	: yes
cpuid level	: 6
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni monitor ssse3 cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch ibs skinit wdt arat hw_pstate npt lbrv svm_lock nrip_save pausefilter
bogomips	: 3393.92
TLB size	: 1024 4K pages
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor	: 1
vendor_id	: AuthenticAMD
cpu family	: 20
model		: 2
model name	: AMD E2-1800 APU with Radeon(tm) HD Graphics
stepping	: 0
microcode	: 0x500010d
cpu MHz		: 850.000
cache size	: 512 KB
physical id	: 0
siblings	: 2
core id		: 1
cpu cores	: 2
apicid		: 1
initial apicid	: 1
fpu		: yes
fpu_exception	: yes
cpuid level	: 6
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni monitor ssse3 cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch ibs skinit wdt arat hw_pstate npt lbrv svm_lock nrip_save pausefilter
bogomips	: 3393.92
TLB size	: 1024 4K pages
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

Created attachment 83211 [details]
proc/modules output

Created attachment 83221 [details]
proc/ioports

Created attachment 83231 [details]
proc/iomem

Created attachment 83241 [details]
lspci -vvv

Created attachment 83251 [details]
proc/scsi

Created attachment 83261 [details]
var_log_dmesg

Created attachment 83271 [details]
proc/acpi

Created attachment 83281 [details]
sys/class/backlight

Created attachment 83291 [details]
acpidump

Created attachment 83301 [details]
acpiextract

Created attachment 83311 [details]
DSDT.dat

Created attachment 83321 [details]
SSDT1.dat

Created attachment 83331 [details]
SSDT2.dat

Created attachment 83341 [details]
DSDT.dsl

Created attachment 83351 [details]
fwts results.log

Created attachment 83361 [details]
fwts_method results.log

Created attachment 83371 [details]
dmesg_grep_ACPI_video

Created attachment 83381 [details]
dmicode.log

I have tried to follow this

https://wiki.ubuntu.com/Bugs/Upstream/kernel

and this 

https://wiki.ubuntu.com/Kernel/Debugging/Backlight

guide.

Hope I have all requiered infromation gathered.


greetings

melchiaros

The original report can be found at:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1054647

what if you echo different values, from 0 to /sys/class/backlight/acpi_video0/max_brightness to /sys/class/backlight/acpi_video0/brightness? 
is the backlight actually changed?

No, does not change.

Still present with kernel 3.7rc6

Sorry, I am not verry experienced with bugzilla. Could you please set it back to new(What is the right status here?)

I have the same problem (also in 3.7). The proprietary driver of AMD fglrx seems to fix the problem.

The requested values are:

cat /sys/class/backlight/acpi_video0/max_brightness
10

When pressing those fn-keys for brightness the value of /sys/class/backlight/acpi_video0/brightness changes between 0 and 10. Default is 6. Ubuntu also displays this change, but the brightness do not change. I guess there is something special with the ACPI internals of this display?

Still there with kernel3.8rc3

Is there a known working kernel? Or it always failed?
Thanks.

There is no working kernel.

Created attachment 95671 [details]
Test DSDT table

Please try this hacked DSDT table, see if manually poke /sys/class/backlight/acpi_video0/brightness works, and if hotkey works, thanks.

Hi melchiaros,

Any update?

And here is a document on how to override DSDT table in initrd:

https://github.com/torvalds/linux/blob/master/Documentation/acpi/initrd_table_override.txt

Sorry, my queqe is full in moment. I will not come up with the lest until the end of the next week.

There are 3 people affected by this.

Could anybody do in faster?

Thanks 

melchiaros

(In reply to comment #28)
> I have the same problem (also in 3.7). The proprietary driver of AMD fglrx
> seems to fix the problem.

I'm confused by this, what do you mean by fix? Manually poke work or fn key work? And when it works, please list /sys/class/backlight, thanks.

> 
> The requested values are:
> 
> cat /sys/class/backlight/acpi_video0/max_brightness
> 10
> 
> When pressing those fn-keys for brightness the value of
> /sys/class/backlight/acpi_video0/brightness changes between 0 and 10. Default
> is 6. Ubuntu also displays this change, but the brightness do not change. I
> guess there is something special with the ACPI internals of this display?

This means acpi video driver can't handle backlight, it probably is a bug of the acpi table.

(In reply to comment #35)
> (In reply to comment #28)
> > I have the same problem (also in 3.7). The proprietary driver of AMD fglrx
> > seems to fix the problem.
> 
> I'm confused by this, what do you mean by fix? Manually poke work or fn key
> work? And when it works, please list /sys/class/backlight, thanks.

Both things work. Even if you use the radeon driver of the kernel, hitting fn keys will result in changing backlight value of the kernel. Ubuntu and Mint display these changes visually.

So the difference between radeon and fglrx is, that the lcd display of the laptop won't change brightness, when the kernel value in /sys/class/backlight does change.

I'm sorry, but I am not very experienced in kernel hacking. I think, that I am not able to apply the DSDT to my laptop. I tried to apply the file inspired by the linked site, but after this there are no dmesg lines, that indicate, that the custom DSDT table is used. Also changing the kernel values in /sys/class/backlight won't change brightness of the display.

I think I did something wrong. I will try it again with another tutorial.

(In reply to comment #36)
> (In reply to comment #35)
> > (In reply to comment #28)
> > > I have the same problem (also in 3.7). The proprietary driver of AMD
> fglrx
> > > seems to fix the problem.
> > 
> > I'm confused by this, what do you mean by fix? Manually poke work or fn key
> > work? And when it works, please list /sys/class/backlight, thanks.
> 
> Both things work. Even if you use the radeon driver of the kernel, hitting fn
> keys will result in changing backlight value of the kernel. Ubuntu and Mint
> display these changes visually.
> 
> So the difference between radeon and fglrx is, that the lcd display of the
> laptop won't change brightness, when the kernel value in /sys/class/backlight
> does change.

Please list the output of the following command:
$ ls /sys/class/backlight

I need to know which file exists there, thanks.

> 
> I'm sorry, but I am not very experienced in kernel hacking. I think, that I
> am
> not able to apply the DSDT to my laptop. I tried to apply the file inspired
> by
> the linked site, but after this there are no dmesg lines, that indicate, that
> the custom DSDT table is used. Also changing the kernel values in
> /sys/class/backlight won't change brightness of the display.
> 
> I think I did something wrong. I will try it again with another tutorial.

Don't, unless your laptop is also HP 655. Is it the case?

$ ls /sys/class/backlight
acpi_video0

> Don't, unless your laptop is also HP 655. Is it the case?

That is the reason, why I am posting here!? It is an HP 655. But the product number is more important on HP laptops: It is B6M62EA#ABD.

(In reply to comment #38)
> $ ls /sys/class/backlight
> acpi_video0
> 
> > Don't, unless your laptop is also HP 655. Is it the case?
> 
> That is the reason, why I am posting here!? It is an HP 655. But the product
> number is more important on HP laptops: It is B6M62EA#ABD.

Then please attach your acpidump.

This is the _BCM for this system:

Method (_BCM, 1, NotSerialized)  // _BCM: Brightness Control Method
{
    Store (One, DFBL)
    Store (Divide (Arg0, 0x0A, ), Local0)
    If (LEqual (^^^LPC0.EC0.VRID, Zero))
    {
        Acquire (^^^LPC0.PSMX, 0xFFFF)
        Store (BGTB (PBCM, ShiftLeft (Local0, One)), ^^^LPC0.EC0.LPWN)
        Release (^^^LPC0.PSMX)
    }
    Else
    {
        Store (BGTB (PBCM, ShiftLeft (Local0, One)), Local3)
        AFN7 (Local3)
    }
}

And this is AFN7:
Method (AFN7, 1, Serialized)
{
    CreateBitField (PSBR, 0x07, PBRT)
    Store (One, PBRT)
    CreateByteField (ATIB, 0x0C, BRTL)
    Store (Arg0, BRTL)
    Notify (VGA, 0xD0)
}

My guess is, when fglrx driver is not installed, the VRID variable in EC space will be zero, and EC firmware is expected to change the backlight on _BCM call, but obviously, nothing happened, so EC firmware doesn't do anything. And when fglrx is installed, it will change VRID to 1, and it will install a notify handler for VGA to handle the 0xD0 notification value, and then to change the backlight according to the value AFN7 has saved in PBRT/BRTL.

So quite frankly, I don't see anything we can do in video driver to fix this problem, you will always need the fglrx driver for backlight control here.