mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH] linux-2.5.63_notsc-panic_A1
@ 2003-02-25  5:00 john stultz
  2003-02-25  5:10 ` [PATCH] linux-2.5.63_x86-tsc-cleanup_A0 john stultz
  0 siblings, 1 reply; 2+ messages in thread
From: john stultz @ 2003-02-25  5:00 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: lkml, Dave H, Andrew Morton

[-- Attachment #1: Type: text/plain, Size: 1592 bytes --]

Linus, all

	In my exuberance to remove #ifdef CONFIG_X86_TSC, I accidentally yanked
one that is needed as long as the symbol itself exists. I've actually
been sitting on this patch for a while, thinking I would get around to
killing off the remaining uses of _X86_TSC, but I've just been too
swamped to get around to it.

This patch ensures that the "notsc" boot parameter is ignored (with a
warning) if the option is passed to a kernel that has been compiled with
CONFIG_X86_TSC. Currently the code will unreasonably panic when this
occurs. 

And since this patch does add an #ifdef _X86_TSC, as a show of goodwill,
I'll follow this patch with another that removes one instance of
_X86_TSC. 

Please apply,

thanks
-john

PS: If anyone is looking for a little project, the usage in pkt_sched.h
and profile.h (include/net) could use some attention :)

diff -Nru a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
--- a/arch/i386/kernel/timers/timer_tsc.c	Mon Feb 24 20:57:33 2003
+++ b/arch/i386/kernel/timers/timer_tsc.c	Mon Feb 24 20:57:33 2003
@@ -299,6 +299,7 @@
 	return -ENODEV;
 }
 
+#ifndef CONFIG_X86_TSC
 /* disable flag for tsc.  Takes effect by clearing the TSC cpu flag
  * in cpu/common.c */
 static int __init tsc_setup(char *str)
@@ -306,7 +307,14 @@
 	tsc_disable = 1;
 	return 1;
 }
-
+#else
+static int __init tsc_setup(char *str)
+{
+	printk(KERN_WARNING "notsc: Kernel compiled with CONFIG_X86_TSC, "
+				"cannot disable TSC.\n");
+	return 1;
+}
+#endif
 __setup("notsc", tsc_setup);
 
 



[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 232 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

* [PATCH] linux-2.5.63_x86-tsc-cleanup_A0
  2003-02-25  5:00 [PATCH] linux-2.5.63_notsc-panic_A1 john stultz
@ 2003-02-25  5:10 ` john stultz
  0 siblings, 0 replies; 2+ messages in thread
From: john stultz @ 2003-02-25  5:10 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: lkml, ak, Andrew Morton

[-- Attachment #1: Type: text/plain, Size: 848 bytes --]

Linus, All,
	As promised, here is a minor cleanup that removes the usage of
CONFIG_X86_TSC in get_cycles(), replacing it with a runtime conditional
of if(cpu_has_tsc).

Andi: I've already posted this a few times without complaint, but since
you didn't like my playing with get_cycles() last time, let me know if
have any issues with this.

thanks
-john


diff -Nru a/include/asm-i386/timex.h b/include/asm-i386/timex.h
--- a/include/asm-i386/timex.h	Mon Feb 24 21:09:32 2003
+++ b/include/asm-i386/timex.h	Mon Feb 24 21:09:32 2003
@@ -40,14 +40,10 @@
 
 static inline cycles_t get_cycles (void)
 {
-#ifndef CONFIG_X86_TSC
-	return 0;
-#else
-	unsigned long long ret;
-
-	rdtscll(ret);
+	unsigned long long ret = 0;
+	if(cpu_has_tsc)
+		rdtscll(ret);
 	return ret;
-#endif
 }
 
 extern unsigned long cpu_khz;




[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 232 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2003-02-25  5:06 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-02-25  5:00 [PATCH] linux-2.5.63_notsc-panic_A1 john stultz
2003-02-25  5:10 ` [PATCH] linux-2.5.63_x86-tsc-cleanup_A0 john stultz

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®