Linux Virtualization - Apr 2007 - [RFC/PATCH PV_OPS X86_64 08/17] paravirt_ops - memory managment

plain text document attachment (xx-paravirt-mm.patch)
Memory management for paravirt_ops.

Signed-off-by: Steven Rostedt srostedt@redhat.com
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>


Index: clean-start/arch/x86_64/mm/fault.c
==================================================================---
clean-start.orig/arch/x86_64/mm/fault.c
+++ clean-start/arch/x86_64/mm/fault.c
@@ -180,7 +180,7 @@ void dump_pagetable(unsigned long addres
 	pmd_t *pmd;
 	pte_t *pte;
 
-	asm("movq %%cr3,%0" : "=r" (pgd));
+	pgd = (pgd_t *)read_cr3();
 
 	pgd = __va((unsigned long)pgd & PHYSICAL_PAGE_MASK); 
 	pgd += pgd_index(address);
@@ -347,7 +347,7 @@ asmlinkage void __kprobes do_page_fault(
 	prefetchw(&mm->mmap_sem);
 
 	/* get the address */
-	__asm__("movq %%cr2,%0":"=r" (address));
+	address = read_cr2();
 
 	info.si_code = SEGV_MAPERR;
 
Index: clean-start/arch/x86_64/mm/init.c
==================================================================---
clean-start.orig/arch/x86_64/mm/init.c
+++ clean-start/arch/x86_64/mm/init.c
@@ -384,7 +384,7 @@ void __meminit init_memory_mapping(unsig
 	} 
 
 	if (!after_bootmem)
-		asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features));
+		mmu_cr4_features = read_cr4();
 	__flush_tlb_all();
 }
 
@@ -398,7 +398,7 @@ void __cpuinit zap_low_mappings(int cpu)
 		 * For AP's, zap the low identity mappings by changing the cr3
 		 * to init_level4_pgt and doing local flush tlb all
 		 */
-		asm volatile("movq %0,%%cr3" :: "r"
(__pa_symbol(&init_level4_pgt)));
+		write_cr3(__pa_symbol(&init_level4_pgt));
 	}
 	__flush_tlb_all();
 }
Index: clean-start/include/asm-x86_64/mmu_context.h
==================================================================---
clean-start.orig/include/asm-x86_64/mmu_context.h
+++ clean-start/include/asm-x86_64/mmu_context.h
@@ -42,7 +42,7 @@ static inline void switch_mm(struct mm_s
 		load_cr3(next->pgd);
 
 		if (unlikely(next->context.ldt != prev->context.ldt)) 
-			load_LDT_nolock(&next->context, cpu);
+			load_LDT_nolock(&next->context);
 	}
 #ifdef CONFIG_SMP
 	else {
@@ -55,7 +55,7 @@ static inline void switch_mm(struct mm_s
 			 * to make sure to use no freed page tables.
 			 */
 			load_cr3(next->pgd);
-			load_LDT_nolock(&next->context, cpu);
+			load_LDT_nolock(&next->context);
 		}
 	}
 #endif
Index: clean-start/include/asm-x86_64/page.h
==================================================================---
clean-start.orig/include/asm-x86_64/page.h
+++ clean-start/include/asm-x86_64/page.h
@@ -64,16 +64,44 @@ typedef struct { unsigned long pgd; } pg
 
 typedef struct { unsigned long pgprot; } pgprot_t;
 
+static inline unsigned long native_pte_val(pte_t pte)
+{
+	return pte.pte;
+}
+
+static inline unsigned long native_pud_val(pud_t pud)
+{
+	return pud.pud;
+}
+
+
+static inline unsigned long native_pmd_val(pmd_t pmd)
+{
+	return pmd.pmd;
+}
+
+static inline unsigned long native_pgd_val(pgd_t pgd)
+{
+	return pgd.pgd;
+}
+
+
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+
 #define pte_val(x)	((x).pte)
 #define pmd_val(x)	((x).pmd)
 #define pud_val(x)	((x).pud)
 #define pgd_val(x)	((x).pgd)
-#define pgprot_val(x)	((x).pgprot)
 
 #define __pte(x) ((pte_t) { (x) } )
 #define __pmd(x) ((pmd_t) { (x) } )
 #define __pud(x) ((pud_t) { (x) } )
 #define __pgd(x) ((pgd_t) { (x) } )
+#endif /* CONFIG_PARAVIRT */
+
+#define pgprot_val(x)	((x).pgprot)
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
 #define __PHYSICAL_START	((unsigned long)CONFIG_PHYSICAL_START)
Index: clean-start/include/asm-x86_64/pgtable.h
==================================================================---
clean-start.orig/include/asm-x86_64/pgtable.h
+++ clean-start/include/asm-x86_64/pgtable.h
@@ -55,50 +55,62 @@ extern unsigned long empty_zero_page[PAG
  */
 #define PTRS_PER_PTE	512
 
-#define pte_ERROR(e) \
-	printk("%s:%d: bad pte %p(%016lx).\n", __FILE__, __LINE__, &(e),
pte_val(e))
-#define pmd_ERROR(e) \
-	printk("%s:%d: bad pmd %p(%016lx).\n", __FILE__, __LINE__, &(e),
pmd_val(e))
-#define pud_ERROR(e) \
-	printk("%s:%d: bad pud %p(%016lx).\n", __FILE__, __LINE__, &(e),
pud_val(e))
-#define pgd_ERROR(e) \
-	printk("%s:%d: bad pgd %p(%016lx).\n", __FILE__, __LINE__, &(e),
pgd_val(e))
-
-#define pgd_none(x)	(!pgd_val(x))
-#define pud_none(x)	(!pud_val(x))
+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define set_pte native_set_pte
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+#define set_pmd native_set_pmd
+#define set_pud native_set_pud
+#define set_pgd native_set_pgd
+#define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while
(0)
+#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
+#define pud_clear native_pud_clear
+#define pgd_clear native_pgd_clear
+#define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte, 0))
+#endif
 
-static inline void set_pte(pte_t *dst, pte_t val)
+static inline void native_set_pte(pte_t *dst, pte_t val)
 {
-	pte_val(*dst) = pte_val(val);
+	dst->pte = pte_val(val);
 } 
-#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
 
-static inline void set_pmd(pmd_t *dst, pmd_t val)
+
+static inline void native_set_pmd(pmd_t *dst, pmd_t val)
 {
-        pmd_val(*dst) = pmd_val(val); 
+	dst->pmd = pmd_val(val);
 } 
 
-static inline void set_pud(pud_t *dst, pud_t val)
+static inline void native_set_pud(pud_t *dst, pud_t val)
 {
-	pud_val(*dst) = pud_val(val);
+	dst->pud = pud_val(val);
 }
 
-static inline void pud_clear (pud_t *pud)
+static inline void native_set_pgd(pgd_t *dst, pgd_t val)
 {
-	set_pud(pud, __pud(0));
+	dst->pgd = pgd_val(val);
 }
-
-static inline void set_pgd(pgd_t *dst, pgd_t val)
+static inline void native_pud_clear (pud_t *pud)
 {
-	pgd_val(*dst) = pgd_val(val); 
-} 
+	set_pud(pud, __pud(0));
+}
 
-static inline void pgd_clear (pgd_t * pgd)
+static inline void native_pgd_clear (pgd_t * pgd)
 {
 	set_pgd(pgd, __pgd(0));
 }
 
-#define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte, 0))
+#define pgd_none(x)	(!pgd_val(x))
+#define pud_none(x)	(!pud_val(x))
+
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %p(%016lx).\n", __FILE__, __LINE__, &(e),
pte_val(e))
+#define pmd_ERROR(e) \
+	printk("%s:%d: bad pmd %p(%016lx).\n", __FILE__, __LINE__, &(e),
pmd_val(e))
+#define pud_ERROR(e) \
+	printk("%s:%d: bad pud %p(%016lx).\n", __FILE__, __LINE__, &(e),
pud_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %p(%016lx).\n", __FILE__, __LINE__, &(e),
pgd_val(e))
 
 struct mm_struct;
 
@@ -238,7 +250,6 @@ static inline unsigned long pmd_bad(pmd_
 
 #define pte_none(x)	(!pte_val(x))
 #define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
-#define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while
(0)
 
 #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))	/* FIXME: is this
 						   right? */
@@ -247,11 +258,11 @@ static inline unsigned long pmd_bad(pmd_
 
 static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
 {
-	pte_t pte;
-	pte_val(pte) = (page_nr << PAGE_SHIFT);
-	pte_val(pte) |= pgprot_val(pgprot);
-	pte_val(pte) &= __supported_pte_mask;
-	return pte;
+	unsigned long pte;
+	pte = (page_nr << PAGE_SHIFT);
+	pte |= pgprot_val(pgprot);
+	pte &= __supported_pte_mask;
+	return __pte(pte);
 }
 
 /*
@@ -345,7 +356,6 @@ static inline int pmd_large(pmd_t pte) {
 			pmd_index(address))
 #define pmd_none(x)	(!pmd_val(x))
 #define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
-#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
 #define pfn_pmd(nr,prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val(prot)))
 #define pmd_pfn(x)  ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
 
@@ -362,19 +372,20 @@ static inline int pmd_large(pmd_t pte) {
 /* physical address -> PTE */
 static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
 { 
-	pte_t pte;
-	pte_val(pte) = physpage | pgprot_val(pgprot); 
-	pte_val(pte) &= __supported_pte_mask;
-	return pte; 
+	unsigned long pte;
+	pte = physpage | pgprot_val(pgprot);
+	pte &= __supported_pte_mask;
+	return __pte(pte);
 }
  
 /* Change flags of a PTE */
-static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+static inline pte_t pte_modify(pte_t pte_old, pgprot_t newprot)
 { 
-	pte_val(pte) &= _PAGE_CHG_MASK;
-	pte_val(pte) |= pgprot_val(newprot);
-	pte_val(pte) &= __supported_pte_mask;
-       return pte; 
+	unsigned long pte = pte_val(pte_old);
+	pte &= _PAGE_CHG_MASK;
+	pte |= pgprot_val(newprot);
+	pte &= __supported_pte_mask;
+       return __pte(pte);
 }
 
 #define pte_index(address) \
Index: clean-start/include/asm-x86_64/tlbflush.h
==================================================================---
clean-start.orig/include/asm-x86_64/tlbflush.h
+++ clean-start/include/asm-x86_64/tlbflush.h
@@ -4,6 +4,7 @@
 #include <linux/mm.h>
 #include <asm/processor.h>
 
+
 static inline unsigned long get_cr3(void)
 {
 	unsigned long cr3;
@@ -16,7 +17,7 @@ static inline void set_cr3(unsigned long
 	asm volatile("mov %0,%%cr3" :: "r" (cr3) :
"memory");
 }
 
-static inline void __flush_tlb(void)
+static inline void __native_flush_tlb(void)
 {
 	set_cr3(get_cr3());
 }
@@ -33,17 +34,24 @@ static inline void set_cr4(unsigned long
 	asm volatile("mov %0,%%cr4" :: "r" (cr4) :
"memory");
 }
 
-static inline void __flush_tlb_all(void)
+static inline void __native_flush_tlb_all(void)
 {
 	unsigned long cr4 = get_cr4();
 	set_cr4(cr4 & ~X86_CR4_PGE);	/* clear PGE */
 	set_cr4(cr4);			/* write old PGE again and flush TLBs */
 }
 
-#define __flush_tlb_one(addr) \
+#define __native_flush_tlb_one(addr) \
 	__asm__ __volatile__("invlpg (%0)" :: "r" (addr) :
"memory")
 

+#ifdef CONFIG_PARAVIRT
+#include <asm/paravirt.h>
+#else
+#define __flush_tlb __native_flush_tlb
+#define __flush_tlb_one __native_flush_tlb_one
+#define __flush_tlb_all __native_flush_tlb_all
+#endif /* CONFIG_PARAVIRT */
 /*
  * TLB flushing:
  *

--