Skip to contentSkip to conversation

[2.5.35] CPUfreq documentation (4/5)

PATCH In review Dominik Brodowski 3 participants 5 messages Sep 17, 2002 Documentation

lore
Share & tools
lore

Dominik Brodowski

author
#0 lore raw

PATCH 1/1 [2.5.35] CPUfreq documentation (4/5)

CPUFreq documentation for 2.5.35:
CREDITS one further CREDIT entry
Documentation/cpufreq documentation of CPU frequency and voltage scaling
support in the Linux kernel.
MAINTAINERS one further MAINTAINERS entry
arch/i386/Config.help Config.help texts for i386 CPUFreq drivers

CREDITS +7 -0
diff -ruN linux-2535original/CREDITS linux/CREDITS
@@ -438,6 +438,13 @@
 S: 411 13  Goteborg
 S: Sweden
 
+N: Dominik Brodowski
+E: [email protected]
+W: http://www.brodo.de/
+P: 1024D/725B37C6  190F 3E77 9C89 3B6D BECD  46EE 67C3 0308 725B 37C6
+D: parts of CPUFreq code, ACPI bugfixes
+S: Tuebingen, Germany
+
 N: Andries Brouwer
 E: [email protected]
 D: random Linux hacker
Documentation/cpufreq +362 -0

362 lines not on this page · raw message

MAINTAINERS +5 -0
diff -ruN linux-2535original/MAINTAINERS linux/MAINTAINERS
@@ -364,6 +364,11 @@
 W:	http://www.fi.muni.cz/~kas/cosa/
 S:	Maintained
 
+CPU FREQUENCY DRIVERS
+L:      [email protected]
+W:      http://www.brodo.de/cpufreq/
+S:      Maintained
+
 CPUID/MSR DRIVER
 P:	H. Peter Anvin
 M:	[email protected]
arch/i386/Config.help +69 -1

76 lines not on this page · raw message

#1 lore raw

Documentation/cpufreq line 81

Quoted 8 lines · Dominik Brodowski
Documentation/cpufreq +8 -0
+P4 CPU Clock Modulation:
+    Intel Pentium 4 Xeon processors
+---------------------------------
+    Note that you can only switch the speed of two logical CPUs at
+    once - but each phyiscal CPU may have different throttling levels.
+    Unfortunately, the cpu_khz value 
+
+

It seems like there is a glitch here. The sentence isn't terminated properly.

#2 lore raw
In reply to Dominik Brodowski

Documentation/cpufreq line 230

Quoted 6 lines · Dominik Brodowski
Documentation/cpufreq +6 -0
+The third argument, a void *pointer, points to a struct cpufreq_freqs
+consisting of five values: cpu, min, max, policy and max_cpu_freq. Min 
+and max are the lower and upper frequencies (in kHz) of the new
+policy, policy the new policy, cpu the number of the affected CPU or
+CPUFREQ_ALL_CPUS for all CPUs; and max_cpu_freq the maximum supported
+CPU frequency. This value is given for informational purposes only.

- Why choosing a void* ? that doesn't validate type ..

- The struct cpufreq_freqs actually consists of only three values (cpu, old, new). The five values you cite here are in the struct cpufreq_policy.

#3 lore raw
Quoted 10 lines · Xavier Bestel

Documentation/cpufreq line 230

Quoted 6 lines · Dominik Brodowski
Documentation/cpufreq +6 -0
+The third argument, a void *pointer, points to a struct cpufreq_freqs
+consisting of five values: cpu, min, max, policy and max_cpu_freq. Min 
+and max are the lower and upper frequencies (in kHz) of the new
+policy, policy the new policy, cpu the number of the affected CPU or
+CPUFREQ_ALL_CPUS for all CPUs; and max_cpu_freq the maximum supported
+CPU frequency. This value is given for informational purposes only.

- Why choosing a void* ? that doesn't validate type ..

That's the type specified by the notifier code. You have two choices:

int notifier_foo(struct notifier_block *nb, int foo, void *bar)

{
	struct my_bar *my = bar;
}

struct notifier_block nb = {
.notifier_call = notifier_foo,
};

OR:

int notifier_foo(struct notifier_block *nb, int foo, struct my_bar *my) {
}

struct notifier_block nb = {
.notifier_call = (int (*)(struct notifier_block *, int, void *))notifier_foo,
};

So, you end up with a cast in one place or the other. I know which one I prefer.

Quoted 3 lines

- The struct cpufreq_freqs actually consists of only three values (cpu, old, new). The five values you cite here are in the struct cpufreq_policy.

Yep, it's a little unclear.

The policy notifiers are called with struct cpufreq_policy, which have five values. The transition notifiers are called with struct cpufreq_freqs, which has three values.

Signature · 2 lines
Russell King ([email protected])                The developer of ARM Linux
             http://www.arm.linux.org.uk/personal/aboutme.html
author
#4 lore raw

[PATCH] Re: CPUfreq documentation (4/5)

In reply to Xavier Bestel

Hi Xavier,

Thanks for the two corrections!

Quoted 8 lines · Xavier Bestel

Review

Quoted 2 lines · Dominik Brodowski
(quoted diff) +2 -0
+The third argument, a void *pointer, points to a struct cpufreq_freqs
+consisting of five values: cpu, min, max, policy and max_cpu_freq. Min

- The struct cpufreq_freqs actually consists of only three values (cpu, old, new). The five values you cite here are in the struct cpufreq_policy.

Dominik

Documentation/cpufreq renamed from Documentation/cpufreq.original +1 -2
@@ -76,7 +76,6 @@
 ---------------------------------
     Note that you can only switch the speed of two logical CPUs at
     once - but each phyiscal CPU may have different throttling levels.
-    Unfortunately, the cpu_khz value 
 
 
 PowerNow! K6:

@@ -222,7 +221,7 @@
 
 The phase is specified in the second argument to the notifier.
 
-The third argument, a void *pointer, points to a struct cpufreq_freqs
+The third argument, a void *pointer, points to a struct cpufreq_policy
 consisting of five values: cpu, min, max, policy and max_cpu_freq. Min 
 and max are the lower and upper frequencies (in kHz) of the new
 policy, policy the new policy, cpu the number of the affected CPU or