fishy code in arch/x86/kernel/tsc.c:time_cpufreq_notifier()
Just notice the following error from gcc 4.4:
arch/x86/kernel/tsc.c: In function 'time_cpufreq_notifier':arch/x86/kernel/tsc.c:634: warning: 'dummy' may be used uninitialized in this function
dummy is only used in the following way in this function:
lpj = &dummy;
and then dummy might be overriden in the following odd way:
if (!(freq->flags & CPUFREQ_CONST_LOOPS)) #ifdef CONFIG_SMP lpj = &cpu_data(freq->cpu).loops_per_jiffy; #else lpj = &boot_cpu_data.loops_per_jiffy;
#endif
and then is used in
if (!ref_freq) {
ref_freq = freq->old;
loops_per_jiffy_ref = *lpj;
tsc_khz_ref = tsc_khz;
}to me that looks like it can indeed be used unitialized for the case where we do have CONFIG_SMP set, freq->flags & CPUFREQ_CONST_LOOPS is true and ref_freq is false.
Can that case actually happen?
Quoted 9 lines · Christoph Hellwig
Just notice the following error from gcc 4.4:
arch/x86/kernel/tsc.c: In function 'time_cpufreq_notifier':arch/x86/kernel/tsc.c:634: warning: 'dummy' may be used uninitialized in this functionwhere we do have
CONFIG_SMPset,freq->flags&CPUFREQ_CONST_LOOPSis true andref_freqis false.Can that case actually happen?
I think you're right, though the circumstances for hitting it are really low. Nearly all SMP capable cpufreq drivers set CPUFREQ_CONST_LOOPS. powernow-k8 is really the only exception. The older CPUs were typically only ever UP. (powernow-k7 never supported SMP for eg)
It's probably worth fixing.
I think the patch below is closer to the actual intent, with the bonus of being a lot more readable.
Sanity check?
Dave
Fix possible uninitialized use of dummy, by just removing it, and making the setting of lpj more obvious.
arch/x86/kernel/tsc.c +3 -5
@@ -631,17 +631,15 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) {struct cpufreq_freqs *freq = data; -unsigned long *lpj, dummy; +unsigned long *lpj;if (cpu_has (&cpu_data (freq->cpu),X86_FEATURE_CONSTANT_TSC ))return 0 ; - lpj = &dummy; - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + lpj = &boot_cpu_data.loops_per_jiffy;#ifdef CONFIG_SMP +if (!(freq->flags &CPUFREQ_CONST_LOOPS )) lpj = &cpu_data (freq->cpu).loops_per_jiffy; -#else - lpj = &boot_cpu_data.loops_per_jiffy;#endif if (!ref_freq) {
Quoted 13 lines · Dave Jones
arch/x86/kernel/tsc.c +1
-1@@ -631,17 +631,15 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) {struct cpufreq_freqs *freq = data; -unsigned long *lpj, dummy; +unsigned long *lpj;if (cpu_has (&cpu_data (freq->cpu),X86_FEATURE_CONSTANT_TSC ))return 0 ;
arch/x86/kernel/tsc.c static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, · line 643
Quoted 10 lines
arch/x86/kernel/tsc.c +2
-4- lpj = &dummy; - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + lpj = &boot_cpu_data.loops_per_jiffy;#ifdef CONFIG_SMP +if (!(freq->flags &CPUFREQ_CONST_LOOPS )) lpj = &cpu_data (freq->cpu).loops_per_jiffy; -#else - lpj = &boot_cpu_data.loops_per_jiffy;#endif
This makes the code look a lot more sane and should fix the potential issue.
Quoted 15 lines · Christoph Hellwig
Quoted 14 lines · Dave Jones
arch/x86/kernel/tsc.c +1
-1@@ -631,17 +631,15 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) {struct cpufreq_freqs *freq = data; -unsigned long *lpj, dummy; +unsigned long *lpj;if (cpu_has (&cpu_data (freq->cpu),X86_FEATURE_CONSTANT_TSC ))return 0 ;
Quoted 12 lines
arch/x86/kernel/tsc.c static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, · line 643
Quoted 9 lines
arch/x86/kernel/tsc.c +2
-4- lpj = &dummy; - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + lpj = &boot_cpu_data.loops_per_jiffy;#ifdef CONFIG_SMP +if (!(freq->flags &CPUFREQ_CONST_LOOPS )) lpj = &cpu_data (freq->cpu).loops_per_jiffy; -#else - lpj = &boot_cpu_data.loops_per_jiffy;#endif This makes the code look a lot more sane and should fix the potential issue.
Tiny niggle that you wind up setting lpj (loops per jiffy) twice if you're on SMP and have CPUFREQ_CONST_LOOPS.
Jon.
Quoted 29 lines · Jon Masters
Quoted 15 lines · Christoph Hellwig
Quoted 14 lines · Dave Jones
arch/x86/kernel/tsc.c +1
-1@@ -631,17 +631,15 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) {struct cpufreq_freqs *freq = data; -unsigned long *lpj, dummy; +unsigned long *lpj;if (cpu_has (&cpu_data (freq->cpu),X86_FEATURE_CONSTANT_TSC ))return 0 ;Quoted 12 lines
arch/x86/kernel/tsc.c static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, · line 643
Quoted 9 lines
arch/x86/kernel/tsc.c +2
-4- lpj = &dummy; - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + lpj = &boot_cpu_data.loops_per_jiffy;#ifdef CONFIG_SMP +if (!(freq->flags &CPUFREQ_CONST_LOOPS )) lpj = &cpu_data (freq->cpu).loops_per_jiffy; -#else - lpj = &boot_cpu_data.loops_per_jiffy;#endif This makes the code look a lot more sane and should fix the potential issue.
Quoted 3 lines
Tiny niggle that you wind up setting lpj (loops per jiffy) twice if you're on SMP and have
CPUFREQ_CONST_LOOPS.
At least it is consistent. ;)
You set lpj twice if your not on SMP and have CPUFREG_CONST_LOOPS.
Mike
Quoted 10 lines
Jon.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
On Mon June 1 2009, Michael S. Zick wrote:
Just ignore the last - I hit the "send" rather than "discard".
Mike
Quoted 51 lines · Jon Masters
Quoted 29 lines · Christoph Hellwig
Quoted 15 lines · Dave Jones
Quoted 14 lines
arch/x86/kernel/tsc.c +1
-1@@ -631,17 +631,15 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) {struct cpufreq_freqs *freq = data; -unsigned long *lpj, dummy; +unsigned long *lpj;if (cpu_has (&cpu_data (freq->cpu),X86_FEATURE_CONSTANT_TSC ))return 0 ;Quoted 12 lines
arch/x86/kernel/tsc.c static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, · line 643
Quoted 9 lines
arch/x86/kernel/tsc.c +2
-4- lpj = &dummy; - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + lpj = &boot_cpu_data.loops_per_jiffy;#ifdef CONFIG_SMP +if (!(freq->flags &CPUFREQ_CONST_LOOPS )) lpj = &cpu_data (freq->cpu).loops_per_jiffy; -#else - lpj = &boot_cpu_data.loops_per_jiffy;#endif This makes the code look a lot more sane and should fix the potential issue.
Quoted 3 lines
Tiny niggle that you wind up setting lpj (loops per jiffy) twice if you're on SMP and have
CPUFREQ_CONST_LOOPS.At least it is consistent. ;)
You set lpj twice if your not on SMP and haveCPUFREG_CONST_LOOPS.Mike
Quoted 10 lines
Jon.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Quoted 9 lines · Christoph Hellwig
Just notice the following error from gcc 4.4:
arch/x86/kernel/tsc.c: In function 'time_cpufreq_notifier':arch/x86/kernel/tsc.c:634: warning: 'dummy' may be used uninitialized in this functiondummy is only used in the following way in this function:
lpj = &dummy;
Quoted 8 lines
and then dummy might be overriden in the following odd way:
if (!(freq->flags & CPUFREQ_CONST_LOOPS)) #ifdef CONFIG_SMP lpj = &cpu_data(freq->cpu).loops_per_jiffy; #else lpj = &boot_cpu_data.loops_per_jiffy;#endif
This is misindented; the if applies to both CONFIG_SMP and otherwise. For that matter, cpu_data(anything) == boot_cpu_data if !CONFIG_SMP. So the current code is equivalent to:
if (!(freq->flags & CPUFREQ_CONST_LOOPS)) lpj = &cpu_data(freq->cpu).loops_per_jiffy;
Quoted 8 lines
and then is used in
if (!ref_freq) { ref_freq = freq->old; loops_per_jiffy_ref = *lpj; tsc_khz_ref = tsc_khz; }
Quoted 5 lines
to me that looks like it can indeed be used unitialized for the case where we do have
CONFIG_SMPset,freq->flags&CPUFREQ_CONST_LOOPSis true andref_freqis false.Can that case actually happen?
Looks to me like loops_per_jiffy_ref is only used to compute a new value for *lpj. So the case that matters is if this function can be called the first time with freq->flags & CPUFREQ_CONST_LOOPS and then without; otherwise, the uninitialized values only contribute to a dead assignment (and the junk in the static variable).
I'd guess that, if freq->flags & CPUFREQ_CONST_LOOPS, no processor's loops_per_jiffy should get scaled, so the current code is essentially correct, although it's hard to read and far too hard for the compiler to analyze.
Probably the right answer is to move the *lpj = ... in with mark_tsc_unstable and drop the earlier if and dummy.
-Daniel
*This .sig left intentionally blank*