From: Srivatsa Vaddagiri <vatsa@in.ibm.com>
To: anton@samba.org, benh@kernel.crashing.org, paulus@samba.org
Cc: linuxppc-dev@ozlabs.org, sri_vatsa_v@yahoo.com
Subject: [PATCH 4/4] tickless idle cpus: decrementer statistics - NOT MEANT FOR INCLUSION
Date: Fri, 7 Apr 2006 12:02:56 +0530 [thread overview]
Message-ID: <20060407063256.GD22416@in.ibm.com> (raw)
This patch (a hack) provides some statistics of decrementer exception on
each CPU. I used this purely to verify that tickless feature is working.
Anton, would this be useful statistics to be exposed (maybe in
/proc/interrupts)? If so, I could work out a more cleaner way of
providing this statistics to user.
Signed-off-by: Srivatsa Vaddagiri <vatsa@in.ibm.com>
---
linux-2.6.17-rc1-root/arch/powerpc/kernel/time.c | 4 ++++
linux-2.6.17-rc1-root/fs/proc/proc_misc.c | 17 +++++++++++++++++
2 files changed, 21 insertions(+)
diff -puN arch/powerpc/kernel/time.c~stat arch/powerpc/kernel/time.c
--- linux-2.6.17-rc1/arch/powerpc/kernel/time.c~stat 2006-04-07 11:29:20.000000000 +0530
+++ linux-2.6.17-rc1-root/arch/powerpc/kernel/time.c 2006-04-07 11:29:20.000000000 +0530
@@ -757,6 +757,8 @@ void start_hz_timer(struct pt_regs *regs
static inline int clear_hzless_mask(void) { return 0;}
#endif
+DEFINE_PER_CPU(int, dec_ticks);
+
/*
* For iSeries shared processors, we have to let the hypervisor
* set the hardware decrementer. We set a virtual decrementer
@@ -780,6 +782,8 @@ void timer_interrupt(struct pt_regs * re
irq_enter();
+ __get_cpu_var(dec_ticks) += 1;
+
clear_hzless_mask();
profile_tick(CPU_PROFILING, regs);
diff -puN fs/proc/proc_misc.c~stat fs/proc/proc_misc.c
--- linux-2.6.17-rc1/fs/proc/proc_misc.c~stat 2006-04-07 11:29:20.000000000 +0530
+++ linux-2.6.17-rc1-root/fs/proc/proc_misc.c 2006-04-07 11:29:20.000000000 +0530
@@ -243,6 +243,22 @@ static int version_read_proc(char *page,
return proc_calc_metrics(page, start, off, count, eof, len);
}
+DECLARE_PER_CPU(int, dec_ticks);
+
+static int decrementer_read_proc(char *page, char **start, off_t off,
+ int count, int *eof, void *data)
+{
+ int len, cpu;
+ char *cp = page;
+
+ for_each_cpu(cpu)
+ cp += sprintf (cp, "%d ", per_cpu(dec_ticks, cpu));
+
+ len = cp - page;
+
+ return proc_calc_metrics(page, start, off, count, eof, len);
+}
+
extern struct seq_operations cpuinfo_op;
static int cpuinfo_open(struct inode *inode, struct file *file)
{
@@ -740,6 +756,7 @@ void __init proc_misc_init(void)
{"uptime", uptime_read_proc},
{"meminfo", meminfo_read_proc},
{"version", version_read_proc},
+ {"decrementer", decrementer_read_proc},
#ifdef CONFIG_PROC_HARDWARE
{"hardware", hardware_read_proc},
#endif
_
--
Regards,
vatsa
reply other threads:[~2006-04-07 6:34 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20060407063256.GD22416@in.ibm.com \
--to=vatsa@in.ibm.com \
--cc=anton@samba.org \
--cc=benh@kernel.crashing.org \
--cc=linuxppc-dev@ozlabs.org \
--cc=paulus@samba.org \
--cc=sri_vatsa_v@yahoo.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).