From: Dou Liyang <douly.fnst@cn.fujitsu.com>
To: <x86@kernel.org>, <linux-kernel@vger.kernel.org>
Cc: <tglx@linutronix.de>, <mingo@kernel.org>, <hpa@zytor.com>,
<ebiederm@xmission.com>, <bhe@redhat.com>,
<izumi.taku@jp.fujitsu.com>,
Dou Liyang <douly.fnst@cn.fujitsu.com>
Subject: [RFC PATCH v3 05/12] x86/ioapic: Refactor the delay logic in timer_irq_works()
Date: Wed, 10 May 2017 21:44:42 +0800 [thread overview]
Message-ID: <1494423889-25799-6-git-send-email-douly.fnst@cn.fujitsu.com> (raw)
In-Reply-To: <1494423889-25799-1-git-send-email-douly.fnst@cn.fujitsu.com>
Kernel use timer_irq_works() to detects the timer IRQs. It calls
mdelay(10) to delay ten ticks and checks whether the timer IRQs work
or not. The mdelay() depends on the loops_per_jiffy which is set up
in calibrate_delay(). And in "notsc" case, calibrating delay also
should make sure the timer IRQs work well.
Now, initializing interrupt mode behind timer init makes the
timer_irq_works() in advance of calibrate_delay(). the mdelay()
doesn't work well in timer_irq_works().
Refactor the delay logic by waiting for some cycles. At the system
With X86_FEATURE_TSC, Use rdtsc(), others will call __delay() directly.
Signed-off-by: Dou Liyang <douly.fnst@cn.fujitsu.com>
---
v2 --> v3:
Find a new way to for waiting.
Reference to the realization of hpet_clocksource_register() by Thomas.
arch/x86/kernel/apic/io_apic.c | 46 ++++++++++++++++++++++++++++++++++++++++--
1 file changed, 44 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 347bb9f..a63c19e 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1607,6 +1607,44 @@ static int __init notimercheck(char *s)
}
__setup("no_timer_check", notimercheck);
+static void delay_with_tsc(void)
+{
+ unsigned long long start, now;
+ unsigned long ticks = jiffies;
+
+ start = rdtsc();
+
+ /*
+ * We don't know the TSC frequency yet, but waiting for
+ * 40000000000/HZ TSC cycles is safe:
+ * 4 GHz == 10 jiffies
+ * 1 GHz == 40 jiffies
+ */
+ do {
+ rep_nop();
+ now = rdtsc();
+ } while ((now - start) < 40000000000UL / HZ &&
+ time_before_eq(jiffies, ticks + 4));
+}
+
+static void delay_without_tsc(void)
+{
+ int band = 1;
+ unsigned long ticks = jiffies;
+
+ /*
+ * We don't know any frequency yet, but waiting for
+ * 40940000000/HZ cycles is safe:
+ * 4 GHz == 10 jiffies
+ * 1 GHz == 40 jiffies
+ * 1 << 1 + 1 << 2 +...+ 1 << 11 = 4094
+ */
+ do {
+ __delay(((1 << band++) * 10000000UL) / HZ);
+ } while (band < 12 && time_before_eq(jiffies, ticks + 4));
+}
+
+
/*
* There is a nasty bug in some older SMP boards, their mptable lies
* about the timer IRQ. We do the following to work around the situation:
@@ -1625,8 +1663,12 @@ static int __init timer_irq_works(void)
local_save_flags(flags);
local_irq_enable();
- /* Let ten ticks pass... */
- mdelay((10 * 1000) / HZ);
+
+ if (boot_cpu_has(X86_FEATURE_TSC))
+ delay_with_tsc();
+ else
+ delay_without_tsc();
+
local_irq_restore(flags);
/*
--
2.5.5
next prev parent reply other threads:[~2017-05-10 13:45 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-10 13:44 [RFC PATCH v3 00/12] Unify interrupt mode and setup it as soon as possible Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 01/12] x86/apic: Replace init_bsp_APIC() with apic_virtual_wire_mode_setup() Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 02/12] x86/apic: Construct a selector for the interrupt delivery mode Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 03/12] x86/apic: Prepare for unifying the interrupt delivery modes setup Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 04/12] x86/time: Initialize interrupt mode behind timer init Dou Liyang
2017-06-06 6:00 ` [RFC PATCH v4 " Dou Liyang
2017-05-10 13:44 ` Dou Liyang [this message]
2017-05-26 8:10 ` [RFC PATCH v4 05/12] x86/ioapic: Refactor the delay logic in timer_irq_works() Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 06/12] x86/apic: Split local APIC timer setup from the APIC setup Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 07/12] x86/apic: Move the logical APIC ID setup from apic_bsp_setup() Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 08/12] x86/apic: Make interrupt mode setup earlier for SMP-capable system Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 09/12] x86/apic: Setup interrupt mode earlier in case of no SMP motherboard Dou Liyang
2017-05-12 9:55 ` [RFC PATCH v4 " Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 10/12] x86/apic: Make the interrupt mode setup earlier for UP system Dou Liyang
2017-05-11 10:57 ` [RFC PATCH v4 " Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 11/12] x86/apic: Mark the apic_interrupt_mode extern for disable_smp() Dou Liyang
2017-05-10 13:44 ` [RFC PATCH v3 12/12] x86/apic: Remove the apic_virtual_wire_mode_setup() Dou Liyang
2017-05-22 20:23 ` [RFC PATCH v3 00/12] Unify interrupt mode and setup it as soon as possible Thomas Gleixner
2017-05-23 1:29 ` Dou Liyang
2017-06-06 6:50 ` Dou Liyang
2017-06-26 10:08 ` Dou Liyang
2017-06-26 11:50 ` Thomas Gleixner
2017-06-27 1:50 ` Dou Liyang
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=1494423889-25799-6-git-send-email-douly.fnst@cn.fujitsu.com \
--to=douly.fnst@cn.fujitsu.com \
--cc=bhe@redhat.com \
--cc=ebiederm@xmission.com \
--cc=hpa@zytor.com \
--cc=izumi.taku@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
--cc=x86@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
all inboxes | Powered by JetHome®