Acer 6592 TM, suspend to ram and resume freeze
First of all, sorry for the duplicate on linux-pm, but as Pavel suggested, this topic gains wider audience as acpi and pm are involved.
I have an issue with my laptop with suspend to ram. My laptop is an Acer Travelmate 6592g :
Quoted 6 lines
[rj@velvet wireshark-1.0.5 (master)]$ sudo s2ram -i This machine can be identified by:
sys_vendor = "Acer" sys_product = "TravelMate 6592" sys_version = "PSMBOU-1234567" bios_version = "V1.53 "
The kernel running is the last I have : 2.6.29-rc1, but the problem is the same whatever the kernel version I take (between 2.6.16 and 2.6.29-rc1).
The issue is that even if the laptop enters suspend correctly, it hangs on comming out of it. After some research, I cornered down the module which changes the behaviour : acpi_cpufreq.ko.
When I forbid the laptop to load this module, the suspend to ram works like a charm, in "init" single mode or multi-user mode. When I load the module, the resume always hangs (fan goes high speed like crazy, no LCD back, hard drive led stays on, caps_lock key not responsive, ...)
Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
As this is not the part of the kernel I'm most at ease with, I'd like some help to investigate. I'll gladly make any test or apply patches to my own git tree.
I join to this mail a URL [1] to see the little information I have :
- kernel messages (/var/log/messages), for single user startup
- ACPI table
- my kernel configuration
- kernel message for a working suspend to ram (ie. without acpi_cpufreq).
- I suspend using s2ram -f -v -p -s
Best regards.
--
Robert
Quoted 16 lines · Robert Jarzmik
First of all, sorry for the duplicate on linux-pm, but as Pavel suggested, this topic gains wider audience as acpi and pm are involved.
I have an issue with my laptop with suspend to ram. My laptop is an Acer Travelmate 6592g :
Quoted 6 lines
[rj@velvet wireshark-1.0.5 (master)]$ sudo s2ram -i This machine can be identified by:
sys_vendor = "Acer" sys_product = "TravelMate 6592" sys_version = "PSMBOU-1234567" bios_version = "V1.53 "The kernel running is the last I have : 2.6.29-rc1, but the problem is the same whatever the kernel version I take (between 2.6.16 and 2.6.29-rc1).
Quoted 8 lines
The issue is that even if the laptop enters suspend correctly, it hangs on comming out of it. After some research, I cornered down the module which changes the behaviour : acpi_cpufreq.ko.
When I forbid the laptop to load this module, the suspend to ram works like a charm, in "init" single mode or multi-user mode. When I load the module, the resume always hangs (fan goes high speed like crazy, no LCD back, hard drive led stays on,
caps_lockkey not responsive, ...)
what if you build in the ACPI processor driver (CONFIG_ACPI_PROCESSOR=y) and boot with "processor.max_cstate=1"?
Quoted 4 lines
Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
I don't think these two are related. can you please make a double check?
thanks,
rui
Quoted 2 lines · ZhangRui
what if you build in the ACPI processor driver (
CONFIG_ACPI_PROCESSOR=y) and boot with "processor.max_cstate=1"?
That doesn't help, thanks for the hint.
Quoted 4 lines
Quoted 3 lines
Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
I don't think these two are related. can you please make a double check?
You're right. Loading thermal.ko leads to the same message, yet the suspend to ram works.
Quoted 3 lines · yakui_zhao
Will you please try the latest kernel(2.6.29-rc3)?
If the problem still exists, please attach the output of dmesg,
acpidump.
Yes, great, it works in 2.6.29-rc3!!! It's amazing, because, when I check acpi in [1], I see no change which could have led to the fix.
I provided acpidump in [2] as information.
Well, I' consider the topic solved. I'll make further tests this weekend to test all of my laptop functions, but my hopes are quite high.
Thanks a lot, the three of you.
--
Robert
[1] : Git ACPI check
velvet:~/kernel_src/linux-2.6# git-log --pretty=oneline v2.6.29-rc3 ^v2.6.29-rc1 drivers/acpi/ 88d998c264722cd6e8fafa42aee191f7d9194960 Merge branch 'misc' into release f1f055f1036a20742608c28ee82139f572dfd031 Merge branches 'bugzilla-11884' and 'bugzilla-8544' into release d08ca2ca743f324eceba59e93188f9439e966bce ACPI: rename main.c to sleep.c 0e4240d94628530a912d216cad1e32d2e3827327 thermal fixup for broken BIOS which has invalid trip points. c6cb0e878446c79f42e7833d7bb69ed6bfbb381f ACPI: EC: Don't trust ECDT tables from ASUS 235c4a59278eb07e61d909f1f0c233733034a8b3 ACPI: EC: Limit workaround for ASUS notebooks even more
[2] http://belgarath.hd.free.fr/download/acer6592tm/acpidump.txt.gz
Quoted 2 lines · ZhangRui
what if you build in the ACPI processor driver (
CONFIG_ACPI_PROCESSOR=y) and boot with "processor.max_cstate=1"?
That doesn't help, thanks for the hint.
Quoted 4 lines
Quoted 3 lines
Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
I don't think these two are related. can you please make a double check?
You're right. Loading thermal.ko leads to the same message, yet the suspend to ram works.
Quoted 3 lines · yakui_zhao
Will you please try the latest kernel(2.6.29-rc3)?
If the problem still exists, please attach the output of dmesg,
acpidump.
Yes, great, it works in 2.6.29-rc3!!! It's amazing, because, when I check acpi in [1], I see no change which could have led to the fix.
I provided acpidump in [2] as information.
Well, I' consider the topic solved. I'll make further tests this weekend to test all of my laptop functions, but my hopes are quite high.
Thanks a lot, the three of you.
--
Robert
[1] : Git ACPI check
velvet:~/kernel_src/linux-2.6# git-log --pretty=oneline v2.6.29-rc3 ^v2.6.29-rc1 drivers/acpi/ 88d998c264722cd6e8fafa42aee191f7d9194960 Merge branch 'misc' into release f1f055f1036a20742608c28ee82139f572dfd031 Merge branches 'bugzilla-11884' and 'bugzilla-8544' into release d08ca2ca743f324eceba59e93188f9439e966bce ACPI: rename main.c to sleep.c 0e4240d94628530a912d216cad1e32d2e3827327 thermal fixup for broken BIOS which has invalid trip points. c6cb0e878446c79f42e7833d7bb69ed6bfbb381f ACPI: EC: Don't trust ECDT tables from ASUS 235c4a59278eb07e61d909f1f0c233733034a8b3 ACPI: EC: Limit workaround for ASUS notebooks even more
[2] http://belgarath.hd.free.fr/download/acer6592tm/acpidump.txt.gz
Quoted 16 lines · Robert Jarzmik
First of all, sorry for the duplicate on linux-pm, but as Pavel suggested, this topic gains wider audience as acpi and pm are involved.
I have an issue with my laptop with suspend to ram. My laptop is an Acer Travelmate 6592g :
Quoted 6 lines
[rj@velvet wireshark-1.0.5 (master)]$ sudo s2ram -i This machine can be identified by:
sys_vendor = "Acer" sys_product = "TravelMate 6592" sys_version = "PSMBOU-1234567" bios_version = "V1.53 "The kernel running is the last I have : 2.6.29-rc1, but the problem is the same whatever the kernel version I take (between 2.6.16 and 2.6.29-rc1).
Quoted 8 lines
The issue is that even if the laptop enters suspend correctly, it hangs on comming out of it. After some research, I cornered down the module which changes the behaviour : acpi_cpufreq.ko.
When I forbid the laptop to load this module, the suspend to ram works like a charm, in "init" single mode or multi-user mode. When I load the module, the resume always hangs (fan goes high speed like crazy, no LCD back, hard drive led stays on,
caps_lockkey not responsive, ...)
what if you build in the ACPI processor driver (CONFIG_ACPI_PROCESSOR=y) and boot with "processor.max_cstate=1"?
Quoted 4 lines
Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
I don't think these two are related. can you please make a double check?
thanks,
rui
Quoted 13 lines · Robert Jarzmik
First of all, sorry for the duplicate on linux-pm, but as Pavel suggested, this topic gains wider audience as acpi and pm are involved.
I have an issue with my laptop with suspend to ram. My laptop is an Acer Travelmate 6592g :
Quoted 6 lines
[rj@velvet wireshark-1.0.5 (master)]$ sudo s2ram -i This machine can be identified by:
sys_vendor = "Acer" sys_product = "TravelMate 6592" sys_version = "PSMBOU-1234567" bios_version = "V1.53 "
Quoted 2 lines
The kernel running is the last I have : 2.6.29-rc1, but the problem is the same whatever the kernel version I take (between 2.6.16 and 2.6.29-rc1).
Will you please try the latest kernel(2.6.29-rc3)?
If the problem still exists, please attach the output of dmesg,
acpidump.
Thanks.
Quoted 33 lines
The issue is that even if the laptop enters suspend correctly, it hangs on comming out of it. After some research, I cornered down the module which changes the behaviour : acpi_cpufreq.ko.
When I forbid the laptop to load this module, the suspend to ram works like a charm, in "init" single mode or multi-user mode. When I load the module, the resume always hangs (fan goes high speed like crazy, no LCD back, hard drive led stays on,
caps_lockkey not responsive, ...)Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
As this is not the part of the kernel I'm most at ease with, I'd like some help to investigate. I'll gladly make any test or apply patches to my own git tree.
I join to this mail a URL [1] to see the little information I have :
- kernel messages (/var/log/messages), for single user startup
- ACPI table
- my kernel configuration
- kernel message for a working suspend to ram (ie. withoutacpi_cpufreq).
- I suspend using s2ram -f -v -p -sBest regards.
--
Robert[1] http://belgarath.hd.free.fr/download/acer6592tm/
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 13 lines · Robert Jarzmik
First of all, sorry for the duplicate on linux-pm, but as Pavel suggested, this topic gains wider audience as acpi and pm are involved.
I have an issue with my laptop with suspend to ram. My laptop is an Acer Travelmate 6592g :
Quoted 6 lines
[rj@velvet wireshark-1.0.5 (master)]$ sudo s2ram -i This machine can be identified by:
sys_vendor = "Acer" sys_product = "TravelMate 6592" sys_version = "PSMBOU-1234567" bios_version = "V1.53 "
Quoted 2 lines
The kernel running is the last I have : 2.6.29-rc1, but the problem is the same whatever the kernel version I take (between 2.6.16 and 2.6.29-rc1).
Will you please try the latest kernel(2.6.29-rc3)?
If the problem still exists, please attach the output of dmesg,
acpidump.
Thanks.
Quoted 33 lines
The issue is that even if the laptop enters suspend correctly, it hangs on comming out of it. After some research, I cornered down the module which changes the behaviour : acpi_cpufreq.ko.
When I forbid the laptop to load this module, the suspend to ram works like a charm, in "init" single mode or multi-user mode. When I load the module, the resume always hangs (fan goes high speed like crazy, no LCD back, hard drive led stays on,
caps_lockkey not responsive, ...)Now, I notice that loading acpi-cpufreq triggers the following message :
Quoted 1 line
log: Jan 19 18:23:42 velvet kernel: ACPI: EC: non-query interrupt received, switching to interrupt mode
As this is not the part of the kernel I'm most at ease with, I'd like some help to investigate. I'll gladly make any test or apply patches to my own git tree.
I join to this mail a URL [1] to see the little information I have :
- kernel messages (/var/log/messages), for single user startup
- ACPI table
- my kernel configuration
- kernel message for a working suspend to ram (ie. withoutacpi_cpufreq).
- I suspend using s2ram -f -v -p -sBest regards.
--
Robert[1] http://belgarath.hd.free.fr/download/acer6592tm/
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html