* [PATCH 0/2 v3] speed up cpu_up()
@ 2015-05-08 17:48 Len Brown
2015-05-08 17:48 ` [PATCH 1/2] x86: Add cmdline "cpu_init_udelay=N" to specify cpu_up() delay Len Brown
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Len Brown @ 2015-05-08 17:48 UTC (permalink / raw)
To: x86, linux-pm, linux-kernel
Thanks for the feedback, here is a simplifed pair of patches.
I also like the x2apic optimization, since it addresses additional delays.
-- though we may be able to optimize those in the non-x2apic case too.
thanks,
Len Brown, Intel Open Source Technology Center
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH 1/2] x86: Add cmdline "cpu_init_udelay=N" to specify cpu_up() delay
2015-05-08 17:48 [PATCH 0/2 v3] speed up cpu_up() Len Brown
@ 2015-05-08 17:48 ` Len Brown
2015-05-08 17:48 ` [PATCH 2/2] x86: remove 10ms delay from cpu_up() on modern processors Len Brown
2015-05-08 20:43 ` [PATCH 0/2 v3] speed up cpu_up() Rafael J. Wysocki
2015-05-10 10:44 ` Ingo Molnar
2 siblings, 1 reply; 5+ messages in thread
From: Len Brown @ 2015-05-08 17:48 UTC (permalink / raw)
To: x86, linux-pm, linux-kernel; +Cc: Len Brown
From: Len Brown <len.brown@intel.com>
No change to default behavior.
Replace the hard-coded mdelay(10) in cpu_up() with a variable udelay,
that is set to a defined default -- rather than a magic number.
Add a boot-time override, "cpu_init_udelay=N"
Signed-off-by: Len Brown <len.brown@intel.com>
---
Documentation/kernel-parameters.txt | 6 ++++++
arch/x86/kernel/smpboot.c | 23 ++++++++++++++++++++++-
2 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index bfcb1a6..0a16309 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -737,6 +737,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
cpuidle.off=1 [CPU_IDLE]
disable the cpuidle sub-system
+ cpu_init_udelay=N
+ [X86] Delay for N microsec between assert and de-assert
+ of APIC INIT to start processors. This delay occurs
+ on every CPU online, such as boot, and resume from suspend.
+ Default: 10000
+
cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
Format:
<first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index febc6aa..0ddc28b 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -517,6 +517,27 @@ void __inquire_remote_apic(int apicid)
}
/*
+ * The Multiprocessor Specification 1.4 (1997) example code suggests
+ * that there should be a 10ms delay between the BSP asserting INIT
+ * and de-asserting INIT, when starting a remote processor.
+ * But that slows boot and resume on modern processors, which include
+ * many cores and don't require that delay.
+ *
+ * Cmdline "init_cpu_udelay=" is available to over-ride this delay.
+ */
+#define UDELAY_10MS_DEFAULT 10000
+
+static unsigned int init_udelay = UDELAY_10MS_DEFAULT;
+
+static int __init cpu_init_udelay(char *str)
+{
+ get_option(&str, &init_udelay);
+
+ return 0;
+}
+early_param("cpu_init_udelay", cpu_init_udelay);
+
+/*
* Poke the other CPU in the eye via NMI to wake it up. Remember that the normal
* INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
* won't ... remember to clear down the APIC, etc later.
@@ -586,7 +607,7 @@ wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip)
pr_debug("Waiting for send to finish...\n");
send_status = safe_apic_wait_icr_idle();
- mdelay(10);
+ udelay(init_udelay);
pr_debug("Deasserting INIT\n");
--
2.4.0.rc1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH 2/2] x86: remove 10ms delay from cpu_up() on modern processors
2015-05-08 17:48 ` [PATCH 1/2] x86: Add cmdline "cpu_init_udelay=N" to specify cpu_up() delay Len Brown
@ 2015-05-08 17:48 ` Len Brown
0 siblings, 0 replies; 5+ messages in thread
From: Len Brown @ 2015-05-08 17:48 UTC (permalink / raw)
To: x86, linux-pm, linux-kernel; +Cc: Len Brown
From: Len Brown <len.brown@intel.com>
Modern processor familes do not require the 10ms delay
in cpu_up() to de-assert INIT. This speeds up boot
and resume by 10ms per (application) processor.
Signed-off-by: Len Brown <len.brown@intel.com>
---
arch/x86/kernel/smpboot.c | 15 +++++++++++++++
1 file changed, 15 insertions(+)
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 0ddc28b..3ab3df5 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -524,6 +524,7 @@ void __inquire_remote_apic(int apicid)
* many cores and don't require that delay.
*
* Cmdline "init_cpu_udelay=" is available to over-ride this delay.
+ * Modern processor families are quirked to remove the delay entirely.
*/
#define UDELAY_10MS_DEFAULT 10000
@@ -537,6 +538,18 @@ static int __init cpu_init_udelay(char *str)
}
early_param("cpu_init_udelay", cpu_init_udelay);
+static void __init smp_quirk_init_udelay(void)
+{
+ /* if cmdline changed it from default, leave it alone */
+ if (init_udelay != UDELAY_10MS_DEFAULT)
+ return;
+
+ /* if modern processor, use no delay */
+ if (((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 == 6)) ||
+ ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && (boot_cpu_data.x86 >= 0xF)))
+ init_udelay = 0;
+}
+
/*
* Poke the other CPU in the eye via NMI to wake it up. Remember that the normal
* INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
@@ -1191,6 +1204,8 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
uv_system_init();
set_mtrr_aps_delayed_init();
+
+ smp_quirk_init_udelay();
}
void arch_enable_nonboot_cpus_begin(void)
--
2.4.0.rc1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 0/2 v3] speed up cpu_up()
2015-05-08 17:48 [PATCH 0/2 v3] speed up cpu_up() Len Brown
2015-05-08 17:48 ` [PATCH 1/2] x86: Add cmdline "cpu_init_udelay=N" to specify cpu_up() delay Len Brown
@ 2015-05-08 20:43 ` Rafael J. Wysocki
2015-05-10 10:44 ` Ingo Molnar
2 siblings, 0 replies; 5+ messages in thread
From: Rafael J. Wysocki @ 2015-05-08 20:43 UTC (permalink / raw)
To: Len Brown; +Cc: x86, linux-pm, linux-kernel
On Friday, May 08, 2015 01:48:05 PM Len Brown wrote:
> Thanks for the feedback, here is a simplifed pair of patches.
>
> I also like the x2apic optimization, since it addresses additional delays.
> -- though we may be able to optimize those in the non-x2apic case too.
For both [1-2/2]:
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 0/2 v3] speed up cpu_up()
2015-05-08 17:48 [PATCH 0/2 v3] speed up cpu_up() Len Brown
2015-05-08 17:48 ` [PATCH 1/2] x86: Add cmdline "cpu_init_udelay=N" to specify cpu_up() delay Len Brown
2015-05-08 20:43 ` [PATCH 0/2 v3] speed up cpu_up() Rafael J. Wysocki
@ 2015-05-10 10:44 ` Ingo Molnar
2 siblings, 0 replies; 5+ messages in thread
From: Ingo Molnar @ 2015-05-10 10:44 UTC (permalink / raw)
To: Len Brown; +Cc: x86, linux-pm, linux-kernel
* Len Brown <lenb@kernel.org> wrote:
> Thanks for the feedback, here is a simplifed pair of patches.
>
> I also like the x2apic optimization, since it addresses additional delays.
> -- though we may be able to optimize those in the non-x2apic case too.
>
> thanks,
> Len Brown, Intel Open Source Technology Center
Could you please send it against tip:master, so that it applies
cleanly to the x2apic optimization which is already applied?
Thanks,
Ingo
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2015-05-10 10:44 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-05-08 17:48 [PATCH 0/2 v3] speed up cpu_up() Len Brown
2015-05-08 17:48 ` [PATCH 1/2] x86: Add cmdline "cpu_init_udelay=N" to specify cpu_up() delay Len Brown
2015-05-08 17:48 ` [PATCH 2/2] x86: remove 10ms delay from cpu_up() on modern processors Len Brown
2015-05-08 20:43 ` [PATCH 0/2 v3] speed up cpu_up() Rafael J. Wysocki
2015-05-10 10:44 ` Ingo Molnar
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).