From: Glauber de Oliveira Costa <gcosta@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: akpm@linux-foundation.org, glommer@gmail.com, tglx@linutronix.de,
mingo@elte.hu, rusty@rustcorp.com.au, ak@suse.de,
chrisw@sous-sol.org, avi@qumranet.com, anthony@codemonkey.ws,
virtualization@lists.linux-foundation.org, lguest@ozlabs.org,
kvm-devel@lists.sourceforge.net, zach@vmware.com,
jun.nakajima@intel.com,
Glauber de Oliveira Costa <gcosta@redhat.com>,
Steven Rostedt <rostedt@goodmis.org>
Subject: [PATCH 15/24] native versions for set pagetables
Date: Fri, 9 Nov 2007 16:42:56 -0200 [thread overview]
Message-ID: <11946338683305-git-send-email-gcosta@redhat.com> (raw)
In-Reply-To: <11946338621966-git-send-email-gcosta@redhat.com>
This patch turns the set_p{te,md,ud,gd} functions into their
native_ versions. There is no need to patch any caller.
Also, it adds pte_update() and pte_update_defer() calls whenever
we modify a page table entry. This last part was coded to match
i386 as close as possible.
Pieces of the header are moved to below the #ifdef CONFIG_PARAVIRT
site, as they are users of the newly defined set_* macros.
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
---
include/asm-x86/pgtable_64.h | 192 ++++++++++++++++++++++++++++--------------
1 files changed, 128 insertions(+), 64 deletions(-)
diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h
index 9b0ff47..592d613 100644
--- a/include/asm-x86/pgtable_64.h
+++ b/include/asm-x86/pgtable_64.h
@@ -57,56 +57,107 @@ extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
*/
#define PTRS_PER_PTE 512
-#ifndef __ASSEMBLY__
+#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 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 pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
+#define pud_clear native_pud_clear
+#define pgd_clear native_pgd_clear
+#define pte_update(mm, addr, ptep) do { } while (0)
+#define pte_update_defer(mm, addr, ptep) do { } while (0)
-#define pgd_none(x) (!pgd_val(x))
-#define pud_none(x) (!pud_val(x))
+#endif
-static inline void set_pte(pte_t *dst, pte_t val)
+#ifndef __ASSEMBLY__
+
+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))
+static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
+ pte_t *ptep, pte_t pteval)
+{
+ native_set_pte(ptep, pteval);
+}
+
+static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr,
+ pte_t *ptep)
+{
+ native_set_pte_at(mm, addr, ptep, __pte(0));
+}
+
+static inline void native_pmd_clear(pmd_t *pmd)
+{
+ native_set_pmd(pmd, __pmd(0));
+}
+
+
+#define pte_ERROR(e) \
+ printk("%s:%d: bad pte %p(%016llx).\n", \
+ __FILE__, __LINE__, &(e), (u64)pte_val(e))
+#define pmd_ERROR(e) \
+ printk("%s:%d: bad pmd %p(%016llx).\n", \
+ __FILE__, __LINE__, &(e), (u64)pmd_val(e))
+#define pud_ERROR(e) \
+ printk("%s:%d: bad pud %p(%016llx).\n", \
+ __FILE__, __LINE__, &(e), (u64)pud_val(e))
+#define pgd_ERROR(e) \
+ printk("%s:%d: bad pgd %p(%016llx).\n", \
+ __FILE__, __LINE__, &(e), (u64)pgd_val(e))
+
+#define pgd_none(x) (!pgd_val(x))
+#define pud_none(x) (!pud_val(x))
struct mm_struct;
-static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
+ unsigned long addr, pte_t *ptep)
+{
+ pte_t pte = __pte(xchg(&ptep->pte, 0));
+ pte_update(mm, addr, ptep);
+ return pte;
+}
+
+static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
+ unsigned long addr, pte_t *ptep,
+ int full)
{
pte_t pte;
if (full) {
@@ -246,7 +297,6 @@ static inline unsigned long pmd_bad(pmd_t 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? */
@@ -255,11 +305,11 @@ static inline unsigned long pmd_bad(pmd_t 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);
}
/*
@@ -283,30 +333,6 @@ static inline pte_t pte_mkwrite(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) |
static inline pte_t pte_mkhuge(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) | _PAGE_PSE)); return pte; }
static inline pte_t pte_clrhuge(pte_t pte) { set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_PSE)); return pte; }
-struct vm_area_struct;
-
-static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
-{
- if (!pte_young(*ptep))
- return 0;
- return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte);
-}
-
-static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
- clear_bit(_PAGE_BIT_RW, &ptep->pte);
-}
-
-/*
- * Macro to mark a page protection value as "uncacheable".
- */
-#define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
-
-static inline int pmd_large(pmd_t pte) {
- return (pmd_val(pte) & __LARGE_PTE) == __LARGE_PTE;
-}
-
-
/*
* Conversion functions: convert a page and protection to a page entry,
* and a page entry and page directory to the page they refer to.
@@ -340,7 +366,6 @@ 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)
@@ -352,15 +377,53 @@ static inline int pmd_large(pmd_t pte) {
/* page, protection -> pte */
#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
-#define mk_pte_huge(entry) (pte_val(entry) |= _PAGE_PRESENT | _PAGE_PSE)
-
+
+static inline pte_t __mk_pte_huge(pte_t entry)
+{
+ unsigned long pte;
+ pte = pte_val(entry);
+ pte |= _PAGE_PRESENT | _PAGE_PSE;
+ return __pte(pte);
+}
+#define mk_pte_huge(entry) ((entry) = __mk_pte_huge(entry))
+
+#include <linux/mm_types.h>
+static inline int ptep_test_and_clear_young(struct vm_area_struct *vma,
+ unsigned long addr, pte_t *ptep)
+{
+ int ret = 0;
+ if (!pte_young(*ptep))
+ return 0;
+ ret = test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte);
+ pte_update(vma->vm_mm, addr, ptep);
+ return ret;
+}
+
+static inline void ptep_set_wrprotect(struct mm_struct *mm,
+ unsigned long addr, pte_t *ptep)
+{
+ clear_bit(_PAGE_BIT_RW, &ptep->pte);
+ pte_update(mm, addr, ptep);
+}
+
+/*
+ * Macro to mark a page protection value as "uncacheable".
+ */
+#define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
+
+static inline int pmd_large(pmd_t pte)
+{
+ return (pmd_val(pte) & __LARGE_PTE) == __LARGE_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) \
@@ -387,6 +450,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
int __changed = !pte_same(*(__ptep), __entry); \
if (__changed && __dirty) { \
set_pte(__ptep, __entry); \
+ pte_update_defer((__vma)->vm_mm, (__address), (__ptep)); \
flush_tlb_page(__vma, __address); \
} \
__changed; \
--
1.4.4.2
next prev parent reply other threads:[~2007-11-09 21:32 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-11-09 18:42 [PATCH 0/24] paravirt_ops for unified x86 - that's me again! Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 1/24] mm/sparse-vmemmap.c: make sure init_mm is included Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 2/24] irqflags consolidation Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 3/24] consolidate spinlock.h Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 4/24] tlb functions consolidation Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 5/24] smp x86 consolidation Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 6/24] Add debugreg/load_rsp native hooks Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 7/24] consolidate msr.h Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 8/24] consolidate system.h Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 9/24] Wipe out traditional opt from x86_64 Makefile Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 10/24] paravirt hooks at entry functions Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 11/24] read/write_crX, clts and wbinvd for 64-bit paravirt Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 12/24] provide native irq initialization function Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 13/24] report ring kernel is running without paravirt Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 14/24] export math_state_restore Glauber de Oliveira Costa
2007-11-09 18:42 ` Glauber de Oliveira Costa [this message]
2007-11-09 18:42 ` [PATCH 16/24] add native functions for descriptors handling Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 17/24] This patch add provisions for time related functions so they Glauber de Oliveira Costa
2007-11-09 18:42 ` [PATCH 18/24] export cpu_gdt_descr Glauber de Oliveira Costa
2007-11-09 18:43 ` [PATCH 19/24] turn priviled operation into a macro in head_64.S Glauber de Oliveira Costa
2007-11-09 18:43 ` [PATCH 20/24] tweak io_64.h for paravirt Glauber de Oliveira Costa
2007-11-09 18:43 ` [PATCH 21/24] native versions for page table entries values Glauber de Oliveira Costa
2007-11-09 18:43 ` [PATCH 22/24] prepare x86_64 architecture initialization for paravirt Glauber de Oliveira Costa
2007-11-09 18:43 ` [PATCH 23/24] consolidation of paravirt for 32 and 64 bits Glauber de Oliveira Costa
2007-11-09 18:43 ` [PATCH 24/24] make vsmp a paravirt client Glauber de Oliveira Costa
2007-11-11 18:07 ` Andi Kleen
2007-11-13 11:36 ` Glauber de Oliveira Costa
2007-11-13 12:33 ` Andi Kleen
2007-11-13 12:51 ` Glauber de Oliveira Costa
2007-11-13 13:09 ` Andi Kleen
2007-11-13 18:49 ` Ravikiran Thirumalai
2007-11-13 17:21 ` Jeremy Fitzhardinge
2007-11-13 18:00 ` Ravikiran Thirumalai
2007-11-19 22:21 ` [PATCH 7/24] consolidate msr.h Bastian Blank
2007-11-19 22:36 ` Steven Rostedt
2007-11-20 5:51 ` Ingo Molnar
2007-11-20 6:14 ` Steven Rostedt
2007-11-20 6:16 ` Steven Rostedt
2007-11-20 10:43 ` Rusty Russell
2007-11-10 0:29 ` [PATCH 5/24] smp x86 consolidation Jeremy Fitzhardinge
2007-11-10 1:29 ` Glauber de Oliveira Costa
2007-11-10 2:08 ` Jeremy Fitzhardinge
2007-11-17 9:32 ` Ingo Molnar
2007-11-09 23:00 ` [PATCH 1/24] mm/sparse-vmemmap.c: make sure init_mm is included Jeremy Fitzhardinge
2007-11-10 1:26 ` Glauber de Oliveira Costa
2007-11-12 8:17 ` [kvm-devel] [PATCH 0/24] paravirt_ops for unified x86 - that's me again! Amit Shah
2007-11-13 11:43 ` Glauber de Oliveira Costa
2007-11-13 20:26 ` Jeremy Fitzhardinge
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=11946338683305-git-send-email-gcosta@redhat.com \
--to=gcosta@redhat.com \
--cc=ak@suse.de \
--cc=akpm@linux-foundation.org \
--cc=anthony@codemonkey.ws \
--cc=avi@qumranet.com \
--cc=chrisw@sous-sol.org \
--cc=glommer@gmail.com \
--cc=jun.nakajima@intel.com \
--cc=kvm-devel@lists.sourceforge.net \
--cc=lguest@ozlabs.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rostedt@goodmis.org \
--cc=rusty@rustcorp.com.au \
--cc=tglx@linutronix.de \
--cc=virtualization@lists.linux-foundation.org \
--cc=zach@vmware.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox