Slightly modified by trenn@suse.de -> only do this on fam 10h and fam 11h.
Currently powernow-k8 determines CPU frequency from ACPI PSS objects, but
according to AMD family 11h BKDG this frequency is just a rounded value:
"CoreFreq (MHz) = The CPU COF specified by MSRC001_00[6B:64][CpuFid]
rounded to the nearest 100 Mhz."
As a consequnce powernow-k8 reports wrong CPU frequency on some systems,
e.g. on Turion X2 Ultra:
powernow-k8: Found 1 AMD Turion(tm)X2 Ultra DualCore Mobile ZM-82
processors (2 cpu cores) (version 2.20.00)
powernow-k8: 0 : pstate 0 (2200 MHz)
powernow-k8: 1 : pstate 1 (1100 MHz)
powernow-k8: 2 : pstate 2 (600 MHz)
But this is wrong as frequency for Pstate2 is 550 MHz. x86info reports it
correctly:
#x86info -a |grep Pstate
...
Pstate-0: fid=e, did=0, vid=24 (2200MHz)
Pstate-1: fid=e, did=1, vid=30 (1100MHz)
Pstate-2: fid=e, did=2, vid=3c (550MHz) (current)
Solution is to determine the frequency directly from Pstate MSRs instead
of using rounded values from ACPI table.
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Dave Jones <davej@redhat.com>
- Make the message shorter and easier to grep for
- Use printk_once instead of WARN_ONCE (functionality of these was mixed)
Signed-off-by: Thomas Renninger <trenn@suse.de>
Cc: Langsdorf, Mark <mark.langsdorf@amd.com>
Signed-off-by: Dave Jones <davej@redhat.com>
* Rafael J. Wysocki (rjw@sisk.pl) wrote:
> This message has been generated automatically as a part of a report
> of regressions introduced between 2.6.28 and 2.6.29.
>
> The following bug entry is on the current list of known regressions
> introduced between 2.6.28 and 2.6.29. Please verify if it still should
> be listed and let me know (either way).
>
>
> Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13186
> Subject : cpufreq timer teardown problem
> Submitter : Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> Date : 2009-04-23 14:00 (24 days old)
> References : http://marc.info/?l=linux-kernel&m=124049523515036&w=4
> Handled-By : Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> Patch : http://patchwork.kernel.org/patch/19754/
> http://patchwork.kernel.org/patch/19753/
>
(updated changelog)
cpufreq fix timer teardown in ondemand governor
The problem is that dbs_timer_exit() uses cancel_delayed_work() when it should
use cancel_delayed_work_sync(). cancel_delayed_work() does not wait for the
workqueue handler to exit.
The ondemand governor does not seem to be affected because the
"if (!dbs_info->enable)" check at the beginning of the workqueue handler returns
immediately without rescheduling the work. The conservative governor in
2.6.30-rc has the same check as the ondemand governor, which makes things
usually run smoothly. However, if the governor is quickly stopped and then
started, this could lead to the following race :
dbs_enable could be reenabled and multiple do_dbs_timer handlers would run.
This is why a synchronized teardown is required.
The following patch applies to, at least, 2.6.28.x, 2.6.29.1, 2.6.30-rc2.
Depends on patch
cpufreq: remove rwsem lock from CPUFREQ_GOV_STOP call
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: gregkh@suse.de
CC: stable@kernel.org
CC: cpufreq@vger.kernel.org
CC: Ingo Molnar <mingo@elte.hu>
CC: rjw@sisk.pl
CC: Ben Slusky <sluskyb@paranoiacs.org>
Signed-off-by: Dave Jones <davej@redhat.com>
* Rafael J. Wysocki (rjw@sisk.pl) wrote:
> This message has been generated automatically as a part of a report
> of regressions introduced between 2.6.28 and 2.6.29.
>
> The following bug entry is on the current list of known regressions
> introduced between 2.6.28 and 2.6.29. Please verify if it still should
> be listed and let me know (either way).
>
>
> Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13186
> Subject : cpufreq timer teardown problem
> Submitter : Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> Date : 2009-04-23 14:00 (24 days old)
> References : http://marc.info/?l=linux-kernel&m=124049523515036&w=4
> Handled-By : Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> Patch : http://patchwork.kernel.org/patch/19754/
> http://patchwork.kernel.org/patch/19753/
>
(re-send with updated changelog)
cpufreq fix timer teardown in conservative governor
The problem is that dbs_timer_exit() uses cancel_delayed_work() when it should
use cancel_delayed_work_sync(). cancel_delayed_work() does not wait for the
workqueue handler to exit.
The ondemand governor does not seem to be affected because the
"if (!dbs_info->enable)" check at the beginning of the workqueue handler returns
immediately without rescheduling the work. The conservative governor in
2.6.30-rc has the same check as the ondemand governor, which makes things
usually run smoothly. However, if the governor is quickly stopped and then
started, this could lead to the following race :
dbs_enable could be reenabled and multiple do_dbs_timer handlers would run.
This is why a synchronized teardown is required.
Depends on patch
cpufreq: remove rwsem lock from CPUFREQ_GOV_STOP call
The following patch applies to 2.6.30-rc2. Stable kernels have a similar
issue which should also be fixed, but the code changed between 2.6.29
and 2.6.30, so this patch only applies to 2.6.30-rc.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: gregkh@suse.de
CC: stable@kernel.org
CC: cpufreq@vger.kernel.org
CC: Ingo Molnar <mingo@elte.hu>
CC: rjw@sisk.pl
CC: Ben Slusky <sluskyb@paranoiacs.org>
Signed-off-by: Dave Jones <davej@redhat.com>
* Rafael J. Wysocki (rjw@sisk.pl) wrote:
> This message has been generated automatically as a part of a report
> of regressions introduced between 2.6.28 and 2.6.29.
>
> The following bug entry is on the current list of known regressions
> introduced between 2.6.28 and 2.6.29. Please verify if it still should
> be listed and let me know (either way).
>
>
> Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13186
> Subject : cpufreq timer teardown problem
> Submitter : Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> Date : 2009-04-23 14:00 (24 days old)
> References : http://marc.info/?l=linux-kernel&m=124049523515036&w=4
> Handled-By : Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> Patch : http://patchwork.kernel.org/patch/19754/
> http://patchwork.kernel.org/patch/19753/
The patches linked above depend on the following patch to remove
circular locking dependency :
cpufreq: remove rwsem lock from CPUFREQ_GOV_STOP call
(the following issue was faced when using cancel_delayed_work_sync() in the
timer teardown (which fixes a race).
* KOSAKI Motohiro (kosaki.motohiro@jp.fujitsu.com) wrote:
> Hi
>
> my box output following warnings.
> it seems regression by commit 7ccc7608b836e58fbacf65ee4f8eefa288e86fac.
>
> A: work -> do_dbs_timer() -> cpu_policy_rwsem
> B: store() -> cpu_policy_rwsem -> cpufreq_governor_dbs() -> work
>
>
Hrm, I think it must be due to my attempt to fix the timer teardown race
in ondemand governor mixed with new locking behavior in 2.6.30-rc.
The rwlock seems to be taken around the whole call to
cpufreq_governor_dbs(), when it should be only taken around accesses to
the locked data, and especially *not* around the call to
dbs_timer_exit().
Reverting my fix attempt would put the teardown race back in place
(replacing the cancel_delayed_work_sync by cancel_delayed_work).
Instead, a proper fix would imply modifying this critical section :
cpufreq.c: __cpufreq_remove_dev()
...
if (cpufreq_driver->target)
__cpufreq_governor(data, CPUFREQ_GOV_STOP);
unlock_policy_rwsem_write(cpu);
To make sure the __cpufreq_governor() callback is not called with rwsem
held. This would allow execution of cancel_delayed_work_sync() without
being nested within the rwsem.
Applies on top of the 2.6.30-rc5 tree.
Required to remove circular dep in teardown of both conservative and
ondemande governors so they can use cancel_delayed_work_sync().
CPUFREQ_GOV_STOP does not modify the policy, therefore this locking seemed
unneeded.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Greg KH <greg@kroah.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: "Rafael J. Wysocki" <rjw@sisk.pl>
CC: Ben Slusky <sluskyb@paranoiacs.org>
CC: Chris Wright <chrisw@sous-sol.org>
CC: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Jones <davej@redhat.com>
arch/x86/kernel/cpu/cpufreq/powernow-k7.c:172: warning: 'invalidate_entry' defined but not used
Reported-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: Dave Jones <davej@redhat.com>
Some atom procs don't do freq scaling (such as the atom 330 on my own
littlefalls2 board). By adding the atom family here, we at least get
the benefit of passive cooling in a thermal emergency. Not sure how
to see that its actually helping any, but the driver does bind and
claim its functioning on my atom 330.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Dave Jones <davej@redhat.com>
perf: Don't assume /proc/kallsyms is ordered
Since we _are_ ordering it by the symbol start, just traverse the
freshly built rbtree setting the prev->end members to curr->start - 1.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090526152134.GF4424@ghostprotocols.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch adds a debug mode to make the codec communication
synchronous. Define SND_HDA_SUPPORT_SYNC_WRITE in hda_codec.c,
and the call of snd_hda_codec_write*() will become synchronous,
i.e. wait for the reply from the codec at each time issuing a verb.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The watchdog code currently hardcodes the base address
of the timer its using. To support new SoCs, make it
support timers at any address. Use the soc_info structure
to do this.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
The davinci timer code currently hardcodes the timer register
base addresses, the timer irq numbers, and the timers to use
for clock events and clocksource. This won't work for some
a new SoC so put those values into the soc_info structure
and set them up in the SoC-specific files.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Use the SoC infrastructure to hold the interrupt controller
information (i.e., base address, default priorities,
interrupt controller type, and the number of IRQs).
The interrupt controller base, although initially put
in the soc_info structure's intc_base field, is eventually
put in the global 'davinci_intc_base' so the low-level
interrupt code can access it without a dereference.
These changes enable the SoC default irq priorities to be
put in the SoC-specific files, and the interrupt controller
to be at any base address.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
The pinmux register base and setup can be different for different
SoCs so move the pinmux reg base, pinmux table (and its size) to
the SoC infrastructure.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
The current code to support the DaVinci Power and Sleep Controller (PSC)
assumes that there is only one controller. This assumption is no longer
valid so expand the support to allow greater than one PSC.
To accomplish this, put the base addresses for the PSCs in the SoC
infrastructure so it can be referenced by the PSC code. This also
requires adding an extra parameter to davinci_psc_config() to specify
the PSC that is to be enabled/disabled.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
All of the davinci SoCs need to call davinci_clk_init() so
put the call in the common init routine.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
The Davinci cpu_is_davinci_*() macros use the SoC part number
and variant retrieved from the JTAG ID register to determine the
type of cpu that the kernel is running on. Currently, the code to
read the JTAG ID register assumes that the register is always at
the same base address. This isn't true on some newer SoCs.
To solve this, have the SoC-specific code set the JTAG ID register
base address in soc_info structure and add a 'cpu_id' member to it.
'cpu_id' will be used by the cpu_is_davinci_*() macros to match
the cpu id. Also move the info used to identify the cpu type into
the SoC-specific code to keep all SoC-specific code together.
The common code will read the JTAG ID register, search through
an array of davinci_id structures to identify the cpu type.
Once identified, it will set the 'cpu_id' member of the soc_info
structure to the proper value and the cpu_is_davinci_*() macros
will now work.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Create a structure to encapsulate SoC-specific information.
This will assist in generalizing code so it can be used by
different SoCs that have similar hardware but with minor
differences such as having a different base address.
The idea is that the code for each SoC fills out a structure
with the correct information. The board-specific code then
calls the SoC init routine which in turn will call a common
init routine that makes a copy of the structure, maps in I/O
regions, etc.
After initialization, code can get a pointer to the structure
by calling davinci_get_soc_info(). Eventually, the common
init routine will make a copy of all of the data pointed to
by the structure so the original data can be made __init_data.
That way the data for SoC's that aren't being used won't consume
memory for the entire life of the kernel.
The structure will be extended in subsequent patches but
initially, it holds the map_desc structure for any I/O
regions the SoC/board wants statically mapped.
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
This fixes a build error where references to pci_cache_line_size are
undefined, as this ceases to be exported when PCI_DISABLE_MWI is enabled,
as is now the default.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Since we can cat /proc/mounts there is no need to have this
subdirectory in the gfs2 sysfs files. In fact this does not
reflect the full range of possible mount argumenmts, where
as /proc/mounts does.
There was only one userland user of this set of sysfs files
and it will function perfectly well without these files
being present (in fact that subcommand of gfs2_tool is
obsolete anyway).
The tune/* subdirectory is also considered mostly obsolete,
but there are a few uses of this until mount arguments can
be added for the last few functions for which there are no
equivalents currently. However the tune/* directory is still
in my sights and new code should avoid using it. Only the gfs2_quota
and gfs2_tool programs are know to use tune/* at the moment.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
