[patch 3/4] x86: vSMP: Use pvops only if platform has the capability to support it

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Andrew Morton <akpm@...>, Ingo Molnar <mingo@...>
Cc: <linux-kernel@...>, Glauber de Oliveira Costa <gcosta@...>, Andi Kleen <ak@...>, <shai@...>
Date: Thursday, March 20, 2008 - 3:43 am

Re-arrange set_vsmp_pv_ops so that pv_ops are set only if
the platform has capability to support paravirtualized irq ops

Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>

Index: linux.git.trees/arch/x86/kernel/vsmp_64.c
===================================================================
--- linux.git.trees.orig/arch/x86/kernel/vsmp_64.c	2008-03-19 20:53:34.267707163 -0700
+++ linux.git.trees/arch/x86/kernel/vsmp_64.c	2008-03-19 21:21:47.853254516 -0700
@@ -78,12 +78,6 @@ static void __init set_vsmp_pv_ops(void)
 	void *address;
 	unsigned int cap, ctl, cfg;
 
-	pv_irq_ops.irq_disable = vsmp_irq_disable;
-	pv_irq_ops.irq_enable  = vsmp_irq_enable;
-	pv_irq_ops.save_fl  = vsmp_save_fl;
-	pv_irq_ops.restore_fl  = vsmp_restore_fl;
-	pv_init_ops.patch = vsmp_patch;
-
 	/* set vSMP magic bits to indicate vSMP capable kernel */
 	cfg = read_pci_config(0, 0x1f, 0, PCI_BASE_ADDRESS_0);
 	address = early_ioremap(cfg, 8);
@@ -92,7 +86,13 @@ static void __init set_vsmp_pv_ops(void)
 	printk(KERN_INFO "vSMP CTL: capabilities:0x%08x  control:0x%08x\n",
 	       cap, ctl);
 	if (cap & ctl & (1 << 4)) {
-		/* Turn on vSMP IRQ fastpath handling (see system.h) */
+		/* Setup irq ops and turn on vSMP  IRQ fastpath handling */
+		pv_irq_ops.irq_disable = vsmp_irq_disable;
+		pv_irq_ops.irq_enable  = vsmp_irq_enable;
+		pv_irq_ops.save_fl  = vsmp_save_fl;
+		pv_irq_ops.restore_fl  = vsmp_restore_fl;
+		pv_init_ops.patch = vsmp_patch;
+
 		ctl &= ~(1 << 4);
 		writel(ctl, address + 4);
 		ctl = readl(address + 4);
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[patch 0/4] x86: vSMP updates, Ravikiran G Thirumalai, (Thu Mar 20, 3:37 am)
Re: [patch 0/4] x86: vSMP updates, Ingo Molnar, (Fri Mar 21, 4:54 am)
[patch 4/4] x86: apic_is_clustered_box to indicate unsynched..., Ravikiran G Thirumalai, (Thu Mar 20, 3:45 am)
Re: [patch 4/4] x86: apic_is_clustered_box to indicate unsyn..., Ravikiran G Thirumalai, (Thu Mar 20, 3:02 pm)
Re: [patch 4/4] x86: apic_is_clustered_box to indicate unsyn..., Ravikiran G Thirumalai, (Fri Mar 21, 2:52 pm)
Re: [patch 4/4] x86: apic_is_clustered_box to indicate unsyn..., Ravikiran G Thirumalai, (Sat Mar 22, 2:59 pm)
Re: [patch 4/4] x86: apic_is_clustered_box to indicate unsyn..., Ravikiran G Thirumalai, (Mon Mar 24, 5:48 pm)
[patch 3/4] x86: vSMP: Use pvops only if platform has the ca..., Ravikiran G Thirumalai, (Thu Mar 20, 3:43 am)
[patch 2/4] x86: Fix build breakage when PCI is define and P..., Ravikiran G Thirumalai, (Thu Mar 20, 3:41 am)
Re: [patch 2/4] x86: Fix build breakage when PCI is define a..., Jeremy Fitzhardinge, (Fri Mar 21, 12:30 am)
Re: [patch 2/4] x86: Fix build breakage when PCI is define a..., Ravikiran G Thirumalai, (Fri Mar 21, 2:54 pm)
Re: [patch 2/4] x86: Fix build breakage when PCI is define a..., Jeremy Fitzhardinge, (Fri Mar 21, 10:19 pm)
[patch 1/4] x86: vSMP: Fix is_vsmp_box(), Ravikiran G Thirumalai, (Thu Mar 20, 3:39 am)
Re: [patch 1/4] x86: vSMP: Fix is_vsmp_box(), Yinghai Lu, (Thu Mar 20, 3:44 am)
Re: [patch 1/4] x86: vSMP: Fix is_vsmp_box(), Ravikiran G Thirumalai, (Thu Mar 20, 2:40 pm)
Re: [patch 1/4] x86: vSMP: Fix is_vsmp_box(), Ingo Molnar, (Fri Mar 21, 5:11 am)
Re: [patch 1/4] x86: vSMP: Fix is_vsmp_box(), Ravikiran G Thirumalai, (Fri Mar 21, 1:59 pm)
Re: [patch 1/4] x86: vSMP: Fix is_vsmp_box(), Yinghai Lu, (Fri Mar 21, 5:17 am)