Linux Virtualization - Apr 2007 - [PATCH 1/4] Pte drop ptep_get_and_clear paravirt op.patch

In shadow mode hypervisors, ptep_get_and_clear achieves the desired
purpose of keeping the shadows in sync by issuing a native_get_and_clear,
followed by a call to pte_update, which indicates the PTE has been
modified.

Direct mode hypervisors (Xen) have no need for this anyway, and will trap
the update using writable pagetables.

This means no hypervisor makes use of ptep_get_and_clear; there is no
reason to have it in the paravirt-ops structure.  Change confusing
terminology about raw vs. native functions into consistent use of
native_pte_xxx for operations which do not invoke paravirt-ops.

Signed-off-by: Zachary Amsden <zach@vmware.com>

diff -r c02c6f5e882c arch/i386/kernel/paravirt.c
--- a/arch/i386/kernel/paravirt.c	Wed Apr 11 16:25:09 2007 -0700
+++ b/arch/i386/kernel/paravirt.c	Wed Apr 11 17:09:55 2007 -0700
@@ -315,8 +315,6 @@ struct paravirt_ops paravirt_ops = {
 	.pte_update = paravirt_nop,
 	.pte_update_defer = paravirt_nop,
 
-	.ptep_get_and_clear = native_ptep_get_and_clear,
-
 #ifdef CONFIG_HIGHPTE
 	.kmap_atomic_pte = kmap_atomic,
 #endif
diff -r c02c6f5e882c include/asm-i386/paravirt.h
--- a/include/asm-i386/paravirt.h	Wed Apr 11 16:25:09 2007 -0700
+++ b/include/asm-i386/paravirt.h	Wed Apr 11 17:12:03 2007 -0700
@@ -187,8 +187,6 @@ struct paravirt_ops
 	void (*pte_update)(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
 	void (*pte_update_defer)(struct mm_struct *mm,
 				 unsigned long addr, pte_t *ptep);
-
- 	pte_t (*ptep_get_and_clear)(pte_t *ptep);
 
 #ifdef CONFIG_HIGHPTE
 	void *(*kmap_atomic_pte)(struct page *page, enum km_type type);
@@ -859,12 +857,8 @@ static inline void pmd_clear(pmd_t *pmdp
 	PVOP_VCALL1(pmd_clear, pmdp);
 }
 
-static inline pte_t raw_ptep_get_and_clear(pte_t *p)
-{
-	unsigned long long val = PVOP_CALL1(unsigned long long, ptep_get_and_clear,
p);
-	return (pte_t) { val, val >> 32 };
-}
 #else  /* !CONFIG_X86_PAE */
+
 static inline pte_t __pte(unsigned long val)
 {
 	return (pte_t) { PVOP_CALL1(unsigned long, make_pte, val) };
@@ -899,11 +893,6 @@ static inline void set_pmd(pmd_t *pmdp, 
 static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 	PVOP_VCALL2(set_pmd, pmdp, pmdval.pud.pgd.pgd);
-}
-
-static inline pte_t raw_ptep_get_and_clear(pte_t *p)
-{
-	return (pte_t) { PVOP_CALL1(unsigned long, ptep_get_and_clear, p) };
 }
 #endif	/* CONFIG_X86_PAE */
 
diff -r c02c6f5e882c include/asm-i386/pgtable.h
--- a/include/asm-i386/pgtable.h	Wed Apr 11 16:25:09 2007 -0700
+++ b/include/asm-i386/pgtable.h	Wed Apr 11 17:11:22 2007 -0700
@@ -265,8 +265,6 @@ static inline pte_t pte_mkhuge(pte_t pte
  */
 #define pte_update(mm, addr, ptep)		do { } while (0)
 #define pte_update_defer(mm, addr, ptep)	do { } while (0)
-
-#define raw_ptep_get_and_clear(xp)     native_ptep_get_and_clear(xp)
 #endif
 
 /*
@@ -340,7 +338,7 @@ do {									\
 #define __HAVE_ARCH_PTEP_GET_AND_CLEAR
 static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long
addr, pte_t *ptep)
 {
-	pte_t pte = raw_ptep_get_and_clear(ptep);
+	pte_t pte = native_ptep_get_and_clear(ptep);
 	pte_update(mm, addr, ptep);
 	return pte;
 }

Zachary Amsden wrote:
> In shadow mode hypervisors, ptep_get_and_clear achieves the desired
> purpose of keeping the shadows in sync by issuing a native_get_and_clear,
> followed by a call to pte_update, which indicates the PTE has been
> modified.
>
> Direct mode hypervisors (Xen) have no need for this anyway, and will trap
> the update using writable pagetables.
>
> This means no hypervisor makes use of ptep_get_and_clear; there is no
> reason to have it in the paravirt-ops structure.  Change confusing
> terminology about raw vs. native functions into consistent use of
> native_pte_xxx for operations which do not invoke paravirt-ops.
>   
Looks OK to me.  Are you going to put this into the pvops queue?

Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>

    J

Zachary Amsden wrote:
> In shadow mode hypervisors, ptep_get_and_clear achieves the desired
> purpose of keeping the shadows in sync by issuing a native_get_and_clear,
> followed by a call to pte_update, which indicates the PTE has been
> modified.
>
> Direct mode hypervisors (Xen) have no need for this anyway, and will trap
> the update using writable pagetables.
>   
Jan Beulich just posted a patch for Xen/linux which specialzes
ptep_get_and_clear, which apparently improves kernel-compile performance
by 25-30% on some configurations.  I think we'll want to keep this.

    J