From ecb93d1ccd0aac63f03be2db3cac3fa974716f4c Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 28 Jan 2009 14:35:05 -0800
Subject: x86/paravirt: add register-saving thunks to reduce caller register
 pressure

Impact: Optimization

One of the problems with inserting a pile of C calls where previously
there were none is that the register pressure is greatly increased.
The C calling convention says that the caller must expect a certain
set of registers may be trashed by the callee, and that the callee can
use those registers without restriction.  This includes the function
argument registers, and several others.

This patch seeks to alleviate this pressure by introducing wrapper
thunks that will do the register saving/restoring, so that the
callsite doesn't need to worry about it, but the callee function can
be conventional compiler-generated code.  In many cases (particularly
performance-sensitive cases) the callee will be in assembler anyway,
and need not use the compiler's calling convention.

Standard calling convention is:
	 arguments	    return	scratch
x86-32	 eax edx ecx	    eax		?
x86-64	 rdi rsi rdx rcx    rax		r8 r9 r10 r11

The thunk preserves all argument and scratch registers.  The return
register is not preserved, and is available as a scratch register for
unwrapped callee code (and of course the return value).

Wrapped function pointers are themselves wrapped in a struct
paravirt_callee_save structure, in order to get some warning from the
compiler when functions with mismatched calling conventions are used.

The most common paravirt ops, both statically and dynamically, are
interrupt enable/disable/save/restore, so handle them first.  This is
particularly easy since their calls are handled specially anyway.

XXX Deal with VMI.  What's their calling convention?

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/vsmp_64.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

(limited to 'arch/x86/kernel/vsmp_64.c')

diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index a688f3bfaec..c609205df59 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -37,6 +37,7 @@ static unsigned long vsmp_save_fl(void)
 		flags &= ~X86_EFLAGS_IF;
 	return flags;
 }
+PV_CALLEE_SAVE_REGS_THUNK(vsmp_save_fl);
 
 static void vsmp_restore_fl(unsigned long flags)
 {
@@ -46,6 +47,7 @@ static void vsmp_restore_fl(unsigned long flags)
 		flags |= X86_EFLAGS_AC;
 	native_restore_fl(flags);
 }
+PV_CALLEE_SAVE_REGS_THUNK(vsmp_restore_fl);
 
 static void vsmp_irq_disable(void)
 {
@@ -53,6 +55,7 @@ static void vsmp_irq_disable(void)
 
 	native_restore_fl((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC);
 }
+PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_disable);
 
 static void vsmp_irq_enable(void)
 {
@@ -60,6 +63,7 @@ static void vsmp_irq_enable(void)
 
 	native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC));
 }
+PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_enable);
 
 static unsigned __init_or_module vsmp_patch(u8 type, u16 clobbers, void *ibuf,
 				  unsigned long addr, unsigned len)
@@ -90,10 +94,10 @@ static void __init set_vsmp_pv_ops(void)
 	       cap, ctl);
 	if (cap & ctl & (1 << 4)) {
 		/* 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_irq_ops.irq_disable = PV_CALLEE_SAVE(vsmp_irq_disable);
+		pv_irq_ops.irq_enable  = PV_CALLEE_SAVE(vsmp_irq_enable);
+		pv_irq_ops.save_fl  = PV_CALLEE_SAVE(vsmp_save_fl);
+		pv_irq_ops.restore_fl  = PV_CALLEE_SAVE(vsmp_restore_fl);
 		pv_init_ops.patch = vsmp_patch;
 
 		ctl &= ~(1 << 4);
-- 
cgit v1.2.3-70-g09d2


From 129d8bc828e011bda0b7110a097bf3a0167f966e Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 25 Feb 2009 21:20:50 -0800
Subject: x86: don't compile vsmp_64 for 32bit

Impact: cleanup

that is only needed when CONFIG_X86_VSMP is defined with 64bit
also remove dead code about PCI, because CONFIG_X86_VSMP depends on PCI

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/apic.h  |  7 +++++++
 arch/x86/include/asm/setup.h |  4 ++++
 arch/x86/kernel/Makefile     |  2 +-
 arch/x86/kernel/setup.c      |  2 --
 arch/x86/kernel/vsmp_64.c    | 12 +-----------
 5 files changed, 13 insertions(+), 14 deletions(-)

(limited to 'arch/x86/kernel/vsmp_64.c')

diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index 860504178e9..24e21273e30 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -75,7 +75,14 @@ static inline void default_inquire_remote_apic(int apicid)
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
 
+#ifdef CONFIG_X86_VSMP
 extern int is_vsmp_box(void);
+#else
+static inline int is_vsmp_box(void)
+{
+	return 0;
+}
+#endif
 extern void xapic_wait_icr_idle(void);
 extern u32 safe_xapic_wait_icr_idle(void);
 extern void xapic_icr_write(u32, u32);
diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h
index 126877e502e..05c6f6b11fd 100644
--- a/arch/x86/include/asm/setup.h
+++ b/arch/x86/include/asm/setup.h
@@ -64,7 +64,11 @@ extern void x86_quirk_time_init(void);
 #include <asm/bootparam.h>
 
 /* Interrupt control for vSMPowered x86_64 systems */
+#ifdef CONFIG_X86_VSMP
 void vsmp_init(void);
+#else
+static inline void vsmp_init(void) { }
+#endif
 
 void setup_bios_corruption_check(void);
 
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index de5657c039e..95f216bbfaf 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -70,7 +70,7 @@ obj-$(CONFIG_FUNCTION_GRAPH_TRACER)	+= ftrace.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
-obj-y				+= vsmp_64.o
+obj-$(CONFIG_X86_VSMP)		+= vsmp_64.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_MODULES)		+= module_$(BITS).o
 obj-$(CONFIG_EFI) 		+= efi.o efi_$(BITS).o efi_stub_$(BITS).o
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 2280d93c5b9..4c54bc0d8ff 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -863,9 +863,7 @@ void __init setup_arch(char **cmdline_p)
 
 	reserve_initrd();
 
-#ifdef CONFIG_X86_64
 	vsmp_init();
-#endif
 
 	io_delay_init();
 
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index c609205df59..74de562812c 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -22,7 +22,7 @@
 #include <asm/paravirt.h>
 #include <asm/setup.h>
 
-#if defined CONFIG_PCI && defined CONFIG_PARAVIRT
+#ifdef CONFIG_PARAVIRT
 /*
  * Interrupt control on vSMPowered systems:
  * ~AC is a shadow of IF.  If IF is 'on' AC should be 'off'
@@ -114,7 +114,6 @@ static void __init set_vsmp_pv_ops(void)
 }
 #endif
 
-#ifdef CONFIG_PCI
 static int is_vsmp = -1;
 
 static void __init detect_vsmp_box(void)
@@ -139,15 +138,6 @@ int is_vsmp_box(void)
 		return 0;
 	}
 }
-#else
-static void __init detect_vsmp_box(void)
-{
-}
-int is_vsmp_box(void)
-{
-	return 0;
-}
-#endif
 
 void __init vsmp_init(void)
 {
-- 
cgit v1.2.3-70-g09d2