Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1262718
| From | Thomas Gleixner <tglx@linutronix.de> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [patch 01/14] x86/apic: Add a single-target IPI function to the apic |
| Date | 2015-11-05 00:10 +0100 |
| Message-ID | <qrflM-6Yk-15@gated-at.bofh.it> (permalink) |
| References | <qrfc5-6Fb-3@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
From: Linus Torvalds <torvalds@linux-foundation.org>
We still fall back on the "send mask" versions if an apic definition
doesn't have the single-target version, but at least this allows the
(trivial) case for the common clustered x2apic case.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/include/asm/apic.h | 1 +
arch/x86/kernel/smp.c | 16 ++++++++++++++--
2 files changed, 15 insertions(+), 2 deletions(-)
Index: linux/arch/x86/include/asm/apic.h
===================================================================
--- linux.orig/arch/x86/include/asm/apic.h
+++ linux/arch/x86/include/asm/apic.h
@@ -303,6 +303,7 @@ struct apic {
unsigned int *apicid);
/* ipi */
+ void (*send_IPI)(int cpu, int vector);
void (*send_IPI_mask)(const struct cpumask *mask, int vector);
void (*send_IPI_mask_allbutself)(const struct cpumask *mask,
int vector);
Index: linux/arch/x86/kernel/smp.c
===================================================================
--- linux.orig/arch/x86/kernel/smp.c
+++ linux/arch/x86/kernel/smp.c
@@ -115,6 +115,18 @@ static atomic_t stopping_cpu = ATOMIC_IN
static bool smp_no_nmi_ipi = false;
/*
+ * Helper wrapper: not all apic definitions support sending to
+ * a single CPU, so we fall back to sending to a mask.
+ */
+static void send_IPI_cpu(int cpu, int vector)
+{
+ if (apic->send_IPI)
+ apic->send_IPI(cpu, vector);
+ else
+ apic->send_IPI_mask(cpumask_of(cpu), vector);
+}
+
+/*
* this function sends a 'reschedule' IPI to another CPU.
* it goes straight through and wastes no time serializing
* anything. Worst case is that we lose a reschedule ...
@@ -125,12 +137,12 @@ static void native_smp_send_reschedule(i
WARN_ON(1);
return;
}
- apic->send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR);
+ send_IPI_cpu(cpu, RESCHEDULE_VECTOR);
}
void native_send_call_func_single_ipi(int cpu)
{
- apic->send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR);
+ send_IPI_cpu(cpu, CALL_FUNCTION_SINGLE_VECTOR);
}
void native_send_call_func_ipi(const struct cpumask *mask)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[patch 00/14] x86/apic: Implement single target IPI callback Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[patch 06/14] x86/apic: Remove pointless indirections from bigsmp_apic Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[tip:x86/apic] x86/apic: Remove pointless indirections from bigsmp_apic tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2015-11-05 15:50 +0100
[patch 01/14] x86/apic: Add a single-target IPI function to the apic Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[tip:x86/apic] x86/apic: Add a single-target IPI function to the apic tip-bot for Linus Torvalds <tipbot@zytor.com> - 2015-11-05 15:50 +0100
[patch 05/14] x86/apic: Wire up single IPI for apic_physflat Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[tip:x86/apic] x86/apic: Wire up single IPI for apic_physflat tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2015-11-05 15:50 +0100
[patch 04/14] x86/apic: Remove pointless indirections from apic_physflat Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[tip:x86/apic] x86/apic: Remove pointless indirections from apic_physflat tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2015-11-05 15:50 +0100
[patch 08/14] x86/apic: Implement single IPI for x2apic_phys Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[tip:x86/apic] x86/apic: Implement single IPI for x2apic_phys tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2015-11-05 15:50 +0100
[patch 02/14] x86/apic: Implement single target IPI function for x2apic_cluster Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
Re: [patch 02/14] x86/apic: Implement single target IPI function for x2apic_cluster Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-05 08:10 +0100
Re: [patch 02/14] x86/apic: Implement single target IPI function for x2apic_cluster Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 09:10 +0100
Re: [patch 02/14] x86/apic: Implement single target IPI function for x2apic_cluster Ingo Molnar <mingo@kernel.org> - 2015-11-05 08:10 +0100
[tip:x86/apic] x86/apic: Implement single target IPI function for x2apic_cluster tip-bot for Linus Torvalds <tipbot@zytor.com> - 2015-11-05 15:50 +0100
[patch 10/14] x86/apic: Implement single IPI for apic_noop Thomas Gleixner <tglx@linutronix.de> - 2015-11-05 00:10 +0100
[tip:x86/apic] x86/apic: Implement single IPI for apic_noop tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2015-11-05 15:50 +0100
csiph-web