Skip to contentSkip to conversation

perfmon2 patch for review: modified i386 files

PATCH Chuck Ebbert 2 participants 2 messages May 13, 2006

lore
Share & tools
lore
#0 lore raw

In-Reply-To: <[email protected]>

On Fri, 12 May 2006 09:33:48 -0700, Stephane Eranian wrote:

<snip>

Quoted 9 lines
arch/i386/kernel/entry.S 2006-05-12 03:18:52.000000000 -0700 renamed from arch/i386/kernel/entry.S 2006-05-12 03:16:09.000000000 -0700 +3 -0
@@ -436,6 +436,16 @@
 /* The include is where all of the SMP etc. interrupts come from */
 #include "entry_arch.h"
 
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_PERFMON)
+ENTRY(pmu_interrupt)
+     pushl $LOCAL_PERFMON_VECTOR-256

arch/i386/kernel/entry.S 2006-05-12 03:18:52.000000000 -0700 line 451

Quoted 10 lines
arch/i386/kernel/entry.S 2006-05-12 03:18:52.000000000 -0700 renamed from arch/i386/kernel/entry.S 2006-05-12 03:16:09.000000000 -0700 +7 -0
+     SAVE_ALL
+     pushl %esp
+     call pfm_intr_handler
+     addl $4, %esp
+     jmp ret_from_intr
+#endif
+
 ENTRY(divide_error)
      pushl $0                        # no error code
      pushl $do_divide_error

You should rename pfm_intr_handler to smp_pmu_interrupt (yes, it's not really SMP but other functions have that problem too, e.g. smp_error_interrupt) and make it fastcall, then you can do:

BUILD_INTERRUPT(pmu_interrupt, LOCAL_PERFMON_VECTOR)

instead of open-coding it. Then the Xen patch that extends the interrupt vector range won't have to change to accommodate your patch.

You should also probably move the BUILD_INTERRUPT() into entry_arch.h with the rest of them.

Signature · 4 lines
Chuck

"The x86 isn't all that complex -- it just doesn't make a lot of sense."
                                                        -- Mike Johnson
#1 lore raw

Chuck,

Quoted 30 lines · Chuck Ebbert

In-Reply-To: <[email protected]>

On Fri, 12 May 2006 09:33:48 -0700, Stephane Eranian wrote:

<snip>

Quoted 9 lines
arch/i386/kernel/entry.S 2006-05-12 03:18:52.000000000 -0700 renamed from arch/i386/kernel/entry.S 2006-05-12 03:16:09.000000000 -0700 +3 -0
@@ -436,6 +436,16 @@
 /* The include is where all of the SMP etc. interrupts come from */
 #include "entry_arch.h"
 
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_PERFMON)
+ENTRY(pmu_interrupt)
+     pushl $LOCAL_PERFMON_VECTOR-256

arch/i386/kernel/entry.S 2006-05-12 03:18:52.000000000 -0700 line 451

Quoted 10 lines
arch/i386/kernel/entry.S 2006-05-12 03:18:52.000000000 -0700 renamed from arch/i386/kernel/entry.S 2006-05-12 03:16:09.000000000 -0700 +7 -0
+     SAVE_ALL
+     pushl %esp
+     call pfm_intr_handler
+     addl $4, %esp
+     jmp ret_from_intr
+#endif
+
 ENTRY(divide_error)
      pushl $0                        # no error code
      pushl $do_divide_error

You should rename pfm_intr_handler to smp_pmu_interrupt (yes, it's not really SMP but other functions have that problem too, e.g. smp_error_interrupt) and make it fastcall, then you can do:

Quoted 8 lines

BUILD_INTERRUPT(pmu_interrupt, LOCAL_PERFMON_VECTOR)

instead of open-coding it. Then the Xen patch that extends the interrupt vector range won't have to change to accommodate your patch.

You should also probably move the BUILD_INTERRUPT() into entry_arch.h with the rest of them.

Thanks you for the tip. I made the change for i386 and everything is working fine. It will be in my next release.

Signature · 1 line
-Stephane