From: Bjorn Helgaas <bjorn.helgaas@hp.com>
To: linux-ia64@vger.kernel.org
Subject: Re: [PATCH] fix "timer tick before it's due"
Date: Thu, 13 May 2004 21:25:31 +0000 [thread overview]
Message-ID: <200405131525.31855.bjorn.helgaas@hp.com> (raw)
In-Reply-To: <200405121721.30863.bjorn.helgaas@hp.com>
On Thursday 13 May 2004 2:33 pm, David Mosberger wrote:
> I think it may really be necessary to distinguish between "I'm a new
> CPU, tell me what to set the ITC to" and "my CPU's may haved drifted,
> let me resync". ...
>
> IMHO, it makes sense to limit the syncing code to the former case for
> the time being (there hasn't been a real need for the latter).
I agree. How about this (same as the first patch, plus the BUG_ON()
and removal of the "jumped tick" check):
=== arch/ia64/kernel/smpboot.c 1.49 vs edited ==--- 1.49/arch/ia64/kernel/smpboot.c Thu Mar 25 12:53:03 2004
+++ edited/arch/ia64/kernel/smpboot.c Thu May 13 15:15:57 2004
@@ -202,6 +202,14 @@
} t[NUM_ROUNDS];
#endif
+ /*
+ * Make sure local timer ticks are disabled while we sync. If
+ * they were enabled, we'd have to worry about nasty issues
+ * like setting the ITC ahead of (or a long time before) the
+ * next scheduled tick.
+ */
+ BUG_ON((ia64_get_itv() & (1 << 16)) = 0);
+
go[MASTER] = 1;
if (smp_call_function_single(master, sync_master, NULL, 1, 0) < 0) {
@@ -247,16 +255,6 @@
printk(KERN_INFO "CPU %d: synchronized ITC with CPU %u (last diff %ld cycles, "
"maxerr %lu cycles)\n", smp_processor_id(), master, delta, rt);
-
- /*
- * Check whether we sync'd the itc ahead of the next timer interrupt. If so, just
- * reset it.
- */
- if (time_after(ia64_get_itc(), local_cpu_data->itm_next)) {
- Dprintk("CPU %d: oops, jumped a timer tick; resetting timer.\n",
- smp_processor_id());
- ia64_cpu_local_tick();
- }
}
/*
@@ -289,11 +287,6 @@
smp_setup_percpu_timer();
/*
- * Get our bogomips.
- */
- ia64_init_itm();
-
- /*
* Set I/O port base per CPU
*/
ia64_set_kr(IA64_KR_IO_BASE, __pa(ia64_iobase));
@@ -305,11 +298,6 @@
#endif
local_irq_enable();
- calibrate_delay();
- local_cpu_data->loops_per_jiffy = loops_per_jiffy;
-#ifdef CONFIG_IA32_SUPPORT
- ia32_gdt_init();
-#endif
if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT)) {
/*
@@ -321,6 +309,17 @@
Dprintk("Going to syncup ITC with BP.\n");
ia64_sync_itc(0);
}
+
+ /*
+ * Get our bogomips.
+ */
+ ia64_init_itm();
+ calibrate_delay();
+ local_cpu_data->loops_per_jiffy = loops_per_jiffy;
+
+#ifdef CONFIG_IA32_SUPPORT
+ ia32_gdt_init();
+#endif
/*
* Allow the master to continue.
=== include/asm-ia64/delay.h 1.8 vs edited ==--- 1.8/include/asm-ia64/delay.h Wed Oct 15 15:07:14 2003
+++ edited/include/asm-ia64/delay.h Thu May 13 11:31:29 2004
@@ -44,6 +44,16 @@
ia64_srlz_d();
}
+static __inline__ unsigned long
+ia64_get_itv (void)
+{
+ unsigned long result;
+
+ result = ia64_getreg(_IA64_REG_CR_ITV);
+ ia64_srlz_d();
+ return result;
+}
+
static __inline__ void
ia64_set_itc (unsigned long val)
{
next prev parent reply other threads:[~2004-05-13 21:25 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-05-12 23:21 [PATCH] fix "timer tick before it's due" Bjorn Helgaas
2004-05-12 23:42 ` David Mosberger
2004-05-13 19:02 ` Bjorn Helgaas
2004-05-13 20:33 ` David Mosberger
2004-05-13 21:25 ` Bjorn Helgaas [this message]
2004-05-13 22:36 ` David Mosberger
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=200405131525.31855.bjorn.helgaas@hp.com \
--to=bjorn.helgaas@hp.com \
--cc=linux-ia64@vger.kernel.org \
/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