From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) Description of problem: Throttling is continually taking CPU down to 87% The HP NX6125 goes very slow - to open a terminal sessions takes at least 10 seconds. This is an HP laptop with AMD Turion processor. It's OK under 32 bit version of FC4 but under AMD 64 bit this happens. the messages file shows Jul 21 18:19:46 eult129 kernel: shpchp: acpi_shpchprm:get_device PCI ROOT HID fail=0x5 Jul 21 18:19:46 eult129 kernel: shpchp: acpi_shpchprm:\_SB_.C046 evaluate _BBN fail=0x5 Jul 21 18:19:46 eult129 kernel: shpchp: acpi_shpchprm:get_device PCI ROOT HID fail=0x5 Jul 21 18:19:46 eult129 kernel: shpchp: acpi_shpchprm:\_SB_.C046 evaluate _BBN fail=0x5 Jul 21 18:19:46 eult129 kernel: shpchp: acpi_shpchprm:get_device PCI ROOT HID fail=0x5 Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Install FC4 on any NX6125 2. Try and open a terminal session from Gnome or KDE 3. Actual Results: It takes at leats 10 seconds to open up Expected Results: Should take less than 1 second Additional info:
I think I've fixed it with the help of a colleague. The thermal trip settings were all set to 16 degrees C so throttling always occurred. I have set the values to a higher level by modifiying the /etc/init.d /set_thermal file to contain these lines echo "95:75:75:65:60:55:45">/proc/acpi/thermal_zone/TZ1/trip_points echo "100:85:85:60:55:45">/proc/acpi/thermal_zone/TZ2/trip_points echo "100:55:55:60:55:45">/proc/acpi/thermal_zone/TZ3/trip_points before the line touch /var/lock/subsys/set_thermal Seems to be OK Can you fix the kernel so that it has the values suggested for this machine by the manufacturer ? I have just put in the unknown values taken from a completely different model. Here are the trip_points beofre I made the mods TZ1 critical (S5): 95 C passive: 16 C: tc1=1 tc2=2 tsp=100 devices=0xffff810057fc9d80 active[0]: 16 C: devices=0xffff810037e16680 active[1]: 16 C: devices=0xffff810037e16540 active[2]: 16 C: devices=0xffff810037e16480 active[3]: 16 C: devices=0xffff810037e16380 TZ2 critical (S5): 100 C passive: 16 C: tc1=1 tc2=2 tsp=300 devices=0xffff810057fc9d80 TZ3 critical (S5): 100 C passive: 16 C: tc1=1 tc2=2 tsp=300 devices=0xffff810057fc9d80
*** Bug 163981 has been marked as a duplicate of this bug. ***
That's a BIOS bug that HP need to fix. As we've no idea about the validity of the values you've entered, we can't rely on this workaround. I suggest looking to see if theres a BIOS update available already, and if not, somehow contact HP about this issue.
I tried a BIOS upgrade which didn't help. In fact the best cure is not the set_thermal above, but to set the kernel parameter no_timer_check=0 which fixes the fast clock problem on and this one. Should I still report this to HP as an issue and if so what do they need to fix ?