Skip to contentSkip to conversation

ARM: Switch back to L_PTE_WRITE

RFC PATCH New Steve Capper 4 participants 11 messages Feb 14, 2014 – Feb 24, 2014 Arm (32-bit)

lore
Share & tools
lore
#0 lore raw

cover ARM: Switch back to L_PTE_WRITE

Hello,
This RFC switches both the short and long descriptors back to using L_PTE_WRITE to represent a writable pte, rather than !L_PTE_RDONLY.

This allows us, for long descriptors, to distinguish between writable clean ptes and read only ptes; fixing a few problems with huge pages, and possibly other areas of code.
(For short descriptors the distinction was possible already due to the separate Linux/hardware ptes).

I originally attempted to leave the short descriptor code alone and only modify the long descriptors to use L_PTE_WRITE. This turned out to be both overly complicated and confusing. So, I have instead reverted:
36bb94b ARM: pgtable: provide RDONLY page table bit rather than WRITE bit
And introduced a separate L_PTE_WRITE software bit for long descriptors.

This series has been tested on an Arndale board running 3.14-rc2, and both short and long descriptors. The libhugetlbfs tests have passed for LPAE. I am currently running a barrage of ltp tests for both short and long descriptors.

Whilst testing this patch with libhugetlbfs, I came across a horrible bug in the handling of huge_pte_write(). For LPAE, both L_PTE_DIRTY and L_PTE_WRITE are in the upper 32 bits of the ptes. The huge_pte_write() function downcasts the result of pte_write() to 32-bits, meaning that huge_pte_write() always returned false! I've noticed this downcasting to 32-bits taking place in the following functions:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry
(under arm64, the downcasting doesn't take place on the huge_pte_ functions as they use long, but the others use int....)

To fix this issue, I have added a double logical invert to pte_write() and pte_dirty() when LPAE is enabled (in the second patch in this series). This fixes the problem, but it's leaving me feeling rather filthy. Any suggestions on a cleaner way around this would be very welcome.

Cheers,
--
Steve

Steve Capper (2):
arm: mm: Switch back to L_PTE_WRITE
arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

 arch/arm/include/asm/pgtable-2level.h |  4 +++-
 arch/arm/include/asm/pgtable-3level.h | 18 ++++++++++++++++--
 arch/arm/include/asm/pgtable.h        | 36 +++++++++++++++++------------------
 arch/arm/mm/dump.c                    |  8 ++++----
 arch/arm/mm/mmu.c                     | 25 ++++++++++++------------
 arch/arm/mm/proc-macros.S             | 16 ++++++++--------
 arch/arm/mm/proc-v7-2level.S          |  6 +++---
 arch/arm/mm/proc-v7-3level.S          |  6 ++++--
 arch/arm/mm/proc-xscale.S             |  4 ++--
 9 files changed, 70 insertions(+), 53 deletions(-)
Signature · 1 line
1.8.1.4
#1 lore raw

RFC PATCH 1/2 arm: mm: Switch back to L_PTE_WRITE

For LPAE, we have the following means for encoding writable or dirty ptes:

                              L_PTE_DIRTY       L_PTE_RDONLY
    !pte_dirty && !pte_write        0               1
    !pte_dirty && pte_write         0               1
    pte_dirty && !pte_write         1               1
    pte_dirty && pte_write          1               0

So we can't distinguish between writable clean ptes and read only ptes. This can cause problems with ptes being incorrectly flagged as read only when they are writable but not dirty.

This patch re-introduces the L_PTE_WRITE bit for both short descriptors and long descriptors, by reverting:
36bb94b ARM: pgtable: provide RDONLY page table bit rather than WRITE bit

For short descriptors the L_PTE_RDONLY bit is renamed to L_PTE_WRITE and the pertinent logic changed. For long descriptors, L_PTE_WRITE is implemented as a new software bit and L_PTE_RDONLY is renamed to PTE_RDONLY to highlight the fact that it is a hardware bit.

---
 arch/arm/include/asm/pgtable-2level.h |  2 +-
 arch/arm/include/asm/pgtable-3level.h |  4 ++--
 arch/arm/include/asm/pgtable.h        | 36 +++++++++++++++++------------------
 arch/arm/mm/dump.c                    |  8 ++++----
 arch/arm/mm/mmu.c                     | 25 ++++++++++++------------
 arch/arm/mm/proc-macros.S             | 16 ++++++++--------
 arch/arm/mm/proc-v7-2level.S          |  6 +++---
 arch/arm/mm/proc-v7-3level.S          |  6 ++++--
 arch/arm/mm/proc-xscale.S             |  4 ++--
 9 files changed, 55 insertions(+), 52 deletions(-)
arch/arm/include/asm/pgtable-2level.h +1 -1
index dfff709..ca43b84 100644
@@ -120,7 +120,7 @@
 #define L_PTE_YOUNG		(_AT(pteval_t, 1) << 1)
 #define L_PTE_FILE		(_AT(pteval_t, 1) << 2)	/* only when !PRESENT */
 #define L_PTE_DIRTY		(_AT(pteval_t, 1) << 6)
-#define L_PTE_RDONLY		(_AT(pteval_t, 1) << 7)
+#define L_PTE_WRITE		(_AT(pteval_t, 1) << 7)
 #define L_PTE_USER		(_AT(pteval_t, 1) << 8)
 #define L_PTE_XN		(_AT(pteval_t, 1) << 9)
 #define L_PTE_SHARED		(_AT(pteval_t, 1) << 10)	/* shared(v6), coherent(xsc3) */
arch/arm/include/asm/pgtable-3level.h +2 -2
index 03243f7..8a392ef 100644
@@ -79,12 +79,12 @@
 #define L_PTE_PRESENT		(_AT(pteval_t, 3) << 0)		/* Present */
 #define L_PTE_FILE		(_AT(pteval_t, 1) << 2)		/* only when !PRESENT */
 #define L_PTE_USER		(_AT(pteval_t, 1) << 6)		/* AP[1] */
-#define L_PTE_RDONLY		(_AT(pteval_t, 1) << 7)		/* AP[2] */
+#define PTE_RDONLY		(_AT(pteval_t, 1) << 7)		/* AP[2] */
 #define L_PTE_SHARED		(_AT(pteval_t, 3) << 8)		/* SH[1:0], inner shareable */
 #define L_PTE_YOUNG		(_AT(pteval_t, 1) << 10)	/* AF */
 #define L_PTE_XN		(_AT(pteval_t, 1) << 54)	/* XN */
 #define L_PTE_DIRTY		(_AT(pteval_t, 1) << 55)	/* unused */
-#define L_PTE_SPECIAL		(_AT(pteval_t, 1) << 56)	/* unused */
+#define L_PTE_WRITE		(_AT(pteval_t, 1) << 56)
 #define L_PTE_NONE		(_AT(pteval_t, 1) << 57)	/* PROT_NONE */
 
 #define PMD_SECT_VALID		(_AT(pmdval_t, 1) << 0)
arch/arm/include/asm/pgtable.h +18 -18
index 7d59b52..7a17611 100644
@@ -88,13 +88,13 @@ extern pgprot_t		pgprot_s2_device;
 
 #define _MOD_PROT(p, b)	__pgprot(pgprot_val(p) | (b))
 
-#define PAGE_NONE		_MOD_PROT(pgprot_user, L_PTE_XN | L_PTE_RDONLY | L_PTE_NONE)
-#define PAGE_SHARED		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_XN)
-#define PAGE_SHARED_EXEC	_MOD_PROT(pgprot_user, L_PTE_USER)
-#define PAGE_COPY		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_RDONLY | L_PTE_XN)
-#define PAGE_COPY_EXEC		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_RDONLY)
-#define PAGE_READONLY		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_RDONLY | L_PTE_XN)
-#define PAGE_READONLY_EXEC	_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_RDONLY)
+#define PAGE_NONE		_MOD_PROT(pgprot_user, L_PTE_XN | L_PTE_NONE)
+#define PAGE_SHARED		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_WRITE | L_PTE_XN)
+#define PAGE_SHARED_EXEC	_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_WRITE)
+#define PAGE_COPY		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_XN)
+#define PAGE_COPY_EXEC		_MOD_PROT(pgprot_user, L_PTE_USER)
+#define PAGE_READONLY		_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_XN)
+#define PAGE_READONLY_EXEC	_MOD_PROT(pgprot_user, L_PTE_USER)
 #define PAGE_KERNEL		_MOD_PROT(pgprot_kernel, L_PTE_XN)
 #define PAGE_KERNEL_EXEC	pgprot_kernel
 #define PAGE_HYP		_MOD_PROT(pgprot_kernel, L_PTE_HYP)
@@ -102,13 +102,13 @@ extern pgprot_t		pgprot_s2_device;
 #define PAGE_S2			_MOD_PROT(pgprot_s2, L_PTE_S2_RDONLY)
 #define PAGE_S2_DEVICE		_MOD_PROT(pgprot_s2_device, L_PTE_S2_RDWR)
 
-#define __PAGE_NONE		__pgprot(_L_PTE_DEFAULT | L_PTE_RDONLY | L_PTE_XN | L_PTE_NONE)
-#define __PAGE_SHARED		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_XN)
-#define __PAGE_SHARED_EXEC	__pgprot(_L_PTE_DEFAULT | L_PTE_USER)
-#define __PAGE_COPY		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_RDONLY | L_PTE_XN)
-#define __PAGE_COPY_EXEC	__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_RDONLY)
-#define __PAGE_READONLY		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_RDONLY | L_PTE_XN)
-#define __PAGE_READONLY_EXEC	__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_RDONLY)
+#define __PAGE_NONE		__pgprot(_L_PTE_DEFAULT | L_PTE_XN | L_PTE_NONE)
+#define __PAGE_SHARED		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_WRITE | L_PTE_XN)
+#define __PAGE_SHARED_EXEC	__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_WRITE)
+#define __PAGE_COPY		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_XN)
+#define __PAGE_COPY_EXEC	__pgprot(_L_PTE_DEFAULT | L_PTE_USER)
+#define __PAGE_READONLY		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_XN)
+#define __PAGE_READONLY_EXEC	__pgprot(_L_PTE_DEFAULT | L_PTE_USER)
 
 #define __pgprot_modify(prot,mask,bits)		\
 	__pgprot((pgprot_val(prot) & ~(mask)) | (bits))
@@ -216,7 +216,7 @@ static inline pte_t *pmd_page_vaddr(pmd_t pmd)
 
 #define pte_none(pte)		(!pte_val(pte))
 #define pte_present(pte)	(pte_val(pte) & L_PTE_PRESENT)
-#define pte_write(pte)		(!(pte_val(pte) & L_PTE_RDONLY))
+#define pte_write(pte)		(pte_val(pte) & L_PTE_WRITE)
 #define pte_dirty(pte)		(pte_val(pte) & L_PTE_DIRTY)
 #define pte_young(pte)		(pte_val(pte) & L_PTE_YOUNG)
 #define pte_exec(pte)		(!(pte_val(pte) & L_PTE_XN))
@@ -248,8 +248,8 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
 #define PTE_BIT_FUNC(fn,op) \
 static inline pte_t pte_##fn(pte_t pte) { pte_val(pte) op; return pte; }
 
-PTE_BIT_FUNC(wrprotect, |= L_PTE_RDONLY);
-PTE_BIT_FUNC(mkwrite,   &= ~L_PTE_RDONLY);
+PTE_BIT_FUNC(wrprotect, &= ~L_PTE_WRITE);
+PTE_BIT_FUNC(mkwrite,   |= L_PTE_WRITE);
 PTE_BIT_FUNC(mkclean,   &= ~L_PTE_DIRTY);
 PTE_BIT_FUNC(mkdirty,   |= L_PTE_DIRTY);
 PTE_BIT_FUNC(mkold,     &= ~L_PTE_YOUNG);
@@ -261,7 +261,7 @@ static inline pte_t pte_mkspecial(pte_t pte) { return pte; }
 
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 {
-	const pteval_t mask = L_PTE_XN | L_PTE_RDONLY | L_PTE_USER |
+	const pteval_t mask = L_PTE_XN | L_PTE_WRITE | L_PTE_USER |
 		L_PTE_NONE | L_PTE_VALID;
 	pte_val(pte) = (pte_val(pte) & ~mask) | (pgprot_val(newprot) & mask);
 	return pte;
arch/arm/mm/dump.c +4 -4
index 2b3a564..715ca93 100644
@@ -58,10 +58,10 @@ static const struct prot_bits pte_bits[] = {
 		.set	= "USR",
 		.clear	= "   ",
 	}, {
-		.mask	= L_PTE_RDONLY,
-		.val	= L_PTE_RDONLY,
-		.set	= "ro",
-		.clear	= "RW",
+		.mask	= L_PTE_WRITE,
+		.val	= L_PTE_WRITE,
+		.set	= "RW",
+		.clear	= "ro",
 	}, {
 		.mask	= L_PTE_XN,
 		.val	= L_PTE_XN,
arch/arm/mm/mmu.c +13 -12
index 4f08c13..38f40df 100644
@@ -231,7 +231,7 @@ __setup("noalign", noalign_setup);
 
 #endif /* ifdef CONFIG_CPU_CP15 / else */
 
-#define PROT_PTE_DEVICE		L_PTE_PRESENT|L_PTE_YOUNG|L_PTE_DIRTY|L_PTE_XN
+#define PROT_PTE_DEVICE		L_PTE_PRESENT|L_PTE_YOUNG|L_PTE_DIRTY|L_PTE_WRITE|L_PTE_XN
 #define PROT_SECT_DEVICE	PMD_TYPE_SECT|PMD_SECT_AP_WRITE
 
 static struct mem_type mem_types[] = {
@@ -277,26 +277,26 @@ static struct mem_type mem_types[] = {
 	},
 #endif
 	[MT_LOW_VECTORS] = {
-		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_RDONLY,
+		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.domain    = DOMAIN_USER,
 	},
 	[MT_HIGH_VECTORS] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_USER | L_PTE_RDONLY,
+				L_PTE_USER,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.domain    = DOMAIN_USER,
 	},
 	[MT_MEMORY_RWX] = {
-		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY,
+		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
+				L_PTE_WRITE,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
 		.domain    = DOMAIN_KERNEL,
 	},
 	[MT_MEMORY_RW] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-			     L_PTE_XN,
+				L_PTE_WRITE | L_PTE_XN,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
 		.domain    = DOMAIN_KERNEL,
@@ -307,26 +307,27 @@ static struct mem_type mem_types[] = {
 	},
 	[MT_MEMORY_RWX_NONCACHED] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_MT_BUFFERABLE,
+				L_PTE_WRITE | L_PTE_MT_BUFFERABLE,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
 		.domain    = DOMAIN_KERNEL,
 	},
 	[MT_MEMORY_RW_DTCM] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_XN,
+				L_PTE_WRITE | L_PTE_XN,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_XN,
 		.domain    = DOMAIN_KERNEL,
 	},
 	[MT_MEMORY_RWX_ITCM] = {
-		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY,
+		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
+				L_PTE_WRITE,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.domain    = DOMAIN_KERNEL,
 	},
 	[MT_MEMORY_RW_SO] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_MT_UNCACHED | L_PTE_XN,
+				L_PTE_MT_UNCACHED | L_PTE_WRITE | L_PTE_XN,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE | PMD_SECT_S |
 				PMD_SECT_UNCACHED | PMD_SECT_XN,
@@ -334,7 +335,7 @@ static struct mem_type mem_types[] = {
 	},
 	[MT_MEMORY_DMA_READY] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_XN,
+				L_PTE_WRITE | L_PTE_XN,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.domain    = DOMAIN_KERNEL,
 	},
@@ -588,7 +589,7 @@ static void __init build_mem_type_table(void)
 
 	pgprot_user   = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | user_pgprot);
 	pgprot_kernel = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG |
-				 L_PTE_DIRTY | kern_pgprot);
+				 L_PTE_DIRTY | L_PTE_WRITE | kern_pgprot);
 	pgprot_s2  = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | s2_pgprot);
 	pgprot_s2_device  = __pgprot(s2_device_pgprot);
 	pgprot_hyp_device  = __pgprot(hyp_device_pgprot);
arch/arm/mm/proc-macros.S +8 -8
index e3c48a3..c62fd89 100644
@@ -97,7 +97,7 @@
 #error PTE shared bit mismatch
 #endif
 #if !defined (CONFIG_ARM_LPAE) && \
-	(L_PTE_XN+L_PTE_USER+L_PTE_RDONLY+L_PTE_DIRTY+L_PTE_YOUNG+\
+	(L_PTE_XN+L_PTE_USER+L_PTE_WRITE+L_PTE_DIRTY+L_PTE_YOUNG+\
 	 L_PTE_FILE+L_PTE_PRESENT) > L_PTE_SHARED
 #error Invalid Linux PTE bit settings
 #endif
@@ -152,9 +152,9 @@
 	and	r2, r1, #L_PTE_MT_MASK
 	ldr	r2, [ip, r2]
 
-	eor	r1, r1, #L_PTE_DIRTY
-	tst	r1, #L_PTE_DIRTY|L_PTE_RDONLY
-	orrne	r3, r3, #PTE_EXT_APX
+	tst	r1, #L_PTE_WRITE
+	tstne	r1, #L_PTE_DIRTY
+	orreq	r3, r3, #PTE_EXT_APX
 
 	tst	r1, #L_PTE_USER
 	orrne	r3, r3, #PTE_EXT_AP1
@@ -199,7 +199,7 @@
 	.macro	armv3_set_pte_ext wc_disable=1
 	str	r1, [r0], #2048			@ linux version
 
-	eor	r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY
+	eor	r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_WRITE | L_PTE_DIRTY
 
 	bic	r2, r1, #PTE_SMALL_AP_MASK	@ keep C, B bits
 	bic	r2, r2, #PTE_TYPE_MASK
@@ -208,7 +208,7 @@
 	tst	r3, #L_PTE_USER			@ user?
 	orrne	r2, r2, #PTE_SMALL_AP_URO_SRW
 
-	tst	r3, #L_PTE_RDONLY | L_PTE_DIRTY	@ write and dirty?
+	tst	r3, #L_PTE_WRITE | L_PTE_DIRTY	@ write and dirty?
 	orreq	r2, r2, #PTE_SMALL_AP_UNO_SRW
 
 	tst	r3, #L_PTE_PRESENT | L_PTE_YOUNG	@ present and young?
@@ -242,7 +242,7 @@
 	.macro	xscale_set_pte_ext_prologue
 	str	r1, [r0]			@ linux version
 
-	eor	r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY
+	eor	r3, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_WRITE | L_PTE_DIRTY
 
 	bic	r2, r1, #PTE_SMALL_AP_MASK	@ keep C, B bits
 	orr	r2, r2, #PTE_TYPE_EXT		@ extended page
@@ -250,7 +250,7 @@
 	tst	r3, #L_PTE_USER			@ user?
 	orrne	r2, r2, #PTE_EXT_AP_URO_SRW	@ yes -> user r/o, system r/w
 
-	tst	r3, #L_PTE_RDONLY | L_PTE_DIRTY	@ write and dirty?
+	tst	r3, #L_PTE_WRITE | L_PTE_DIRTY	@ write and dirty?
 	orreq	r2, r2, #PTE_EXT_AP_UNO_SRW	@ yes -> user n/a, system r/w
 						@ combined with user -> user r/w
 	.endm
arch/arm/mm/proc-v7-2level.S +3 -3
index bdd3be4..297fccf 100644
@@ -84,9 +84,9 @@ ENTRY(cpu_v7_set_pte_ext)
 	tst	r1, #1 << 4
 	orrne	r3, r3, #PTE_EXT_TEX(1)
 
-	eor	r1, r1, #L_PTE_DIRTY
-	tst	r1, #L_PTE_RDONLY | L_PTE_DIRTY
-	orrne	r3, r3, #PTE_EXT_APX
+	tst	r1, #L_PTE_WRITE
+	tstne	r1, #L_PTE_DIRTY
+	orreq	r3, r3, #PTE_EXT_APX
 
 	tst	r1, #L_PTE_USER
 	orrne	r3, r3, #PTE_EXT_AP1
arch/arm/mm/proc-v7-3level.S +4 -2
index 01a719e..7793b2e 100644
@@ -78,8 +78,10 @@ ENTRY(cpu_v7_set_pte_ext)
 	tst	r3, #1 << (57 - 32)		@ L_PTE_NONE
 	bicne	r2, #L_PTE_VALID
 	bne	1f
-	tst	r3, #1 << (55 - 32)		@ L_PTE_DIRTY
-	orreq	r2, #L_PTE_RDONLY
+	bic	r2, #PTE_RDONLY
+	tst	r3, #1 << (56 - 32)		@ L_PTE_WRITE
+	tstne	r3, #1 << (55 - 32)		@ L_PTE_DIRTY
+	orreq	r2, #PTE_RDONLY
 1:	strd	r2, r3, [r0]
 	ALT_SMP(W(nop))
 	ALT_UP (mcr	p15, 0, r0, c7, c10, 1)		@ flush_pte
arch/arm/mm/proc-xscale.S +2 -2
index d19b1cf..d5b23e8 100644
@@ -510,8 +510,8 @@ ENTRY(cpu_xscale_set_pte_ext)
 	@
 	@ Erratum 40: must set memory to write-through for user read-only pages
 	@
-	and	ip, r1, #(L_PTE_MT_MASK | L_PTE_USER | L_PTE_RDONLY) & ~(4 << 2)
-	teq	ip, #L_PTE_MT_WRITEBACK | L_PTE_USER | L_PTE_RDONLY
+	and	ip, r1, #(L_PTE_MT_MASK | L_PTE_USER | L_PTE_WRITE) & ~(4 << 2)
+	teq	ip, #L_PTE_MT_WRITEBACK | L_PTE_USER
 
 	moveq	r1, #L_PTE_MT_WRITETHROUGH
 	and	r1, r1, #L_PTE_MT_MASK
Signature · 1 line
1.8.1.4
#2 lore raw

RFC PATCH 1/2 arm: mm: Switch back to L_PTE_WRITE

Hi Steve,

Quoted 12 lines · Steve Capper

For LPAE, we have the following means for encoding writable or dirty ptes:

                              L_PTE_DIRTY       L_PTE_RDONLY
    !pte_dirty && !pte_write        0               1
    !pte_dirty && pte_write         0               1
    pte_dirty && !pte_write         1               1
    pte_dirty && pte_write          1               0

So we can't distinguish between writable clean ptes and read only ptes. This can cause problems with ptes being incorrectly flagged as read only when they are writable but not dirty.

Quoted 8 lines

This patch re-introduces the L_PTE_WRITE bit for both short descriptors and long descriptors, by reverting:
36bb94b ARM: pgtable: provide RDONLY page table bit rather than WRITE bit

For short descriptors the L_PTE_RDONLY bit is renamed to L_PTE_WRITE and the pertinent logic changed. For long descriptors, L_PTE_WRITE is implemented as a new software bit and L_PTE_RDONLY is renamed to PTE_RDONLY to highlight the fact that it is a hardware bit.

This would be a lot easier to review if it was a true revert, but I guess that doesn't apply cleanly to mainline?

Quoted 31 lines
---
 arch/arm/include/asm/pgtable-2level.h |  2 +-
 arch/arm/include/asm/pgtable-3level.h |  4 ++--
 arch/arm/include/asm/pgtable.h        | 36 +++++++++++++++++------------------
 arch/arm/mm/dump.c                    |  8 ++++----
 arch/arm/mm/mmu.c                     | 25 ++++++++++++------------
 arch/arm/mm/proc-macros.S             | 16 ++++++++--------
 arch/arm/mm/proc-v7-2level.S          |  6 +++---
 arch/arm/mm/proc-v7-3level.S          |  6 ++++--
 arch/arm/mm/proc-xscale.S             |  4 ++--
 9 files changed, 55 insertions(+), 52 deletions(-)
arch/arm/include/asm/pgtable-2level.h +1 -1
index dfff709..ca43b84 100644
@@ -120,7 +120,7 @@
 #define L_PTE_YOUNG            (_AT(pteval_t, 1) << 1)
 #define L_PTE_FILE             (_AT(pteval_t, 1) << 2) /* only when !PRESENT */
 #define L_PTE_DIRTY            (_AT(pteval_t, 1) << 6)
-#define L_PTE_RDONLY           (_AT(pteval_t, 1) << 7)
+#define L_PTE_WRITE            (_AT(pteval_t, 1) << 7)
 #define L_PTE_USER             (_AT(pteval_t, 1) << 8)
 #define L_PTE_XN               (_AT(pteval_t, 1) << 9)
 #define L_PTE_SHARED           (_AT(pteval_t, 1) << 10)        /* shared(v6), coherent(xsc3) */
arch/arm/include/asm/pgtable-3level.h +0 -0
index 03243f7..8a392ef 100644

arch/arm/include/asm/pgtable-3level.h line 82

Quoted 5 lines
arch/arm/include/asm/pgtable-3level.h +1 -1
@@ -79,12 +79,12 @@
 #define L_PTE_PRESENT          (_AT(pteval_t, 3) << 0)         /* Present */
 #define L_PTE_FILE             (_AT(pteval_t, 1) << 2)         /* only when !PRESENT */
 #define L_PTE_USER             (_AT(pteval_t, 1) << 6)         /* AP[1] */
-#define L_PTE_RDONLY           (_AT(pteval_t, 1) << 7)         /* AP[2] */
+#define PTE_RDONLY             (_AT(pteval_t, 1) << 7)         /* AP[2] */

Why? I think we're just using L_ for consistency here, rather than to distinguish between h/w and Linux bits (e.g. L_PTE_XN).

arch/arm/include/asm/pgtable-3level.h line 87

Quoted 6 lines
arch/arm/include/asm/pgtable-3level.h +1 -1
 #define L_PTE_SHARED           (_AT(pteval_t, 3) << 8)         /* SH[1:0], inner shareable */
 #define L_PTE_YOUNG            (_AT(pteval_t, 1) << 10)        /* AF */
 #define L_PTE_XN               (_AT(pteval_t, 1) << 54)        /* XN */
 #define L_PTE_DIRTY            (_AT(pteval_t, 1) << 55)        /* unused */
-#define L_PTE_SPECIAL          (_AT(pteval_t, 1) << 56)        /* unused */
+#define L_PTE_WRITE            (_AT(pteval_t, 1) << 56)

Why have you killed L_PTE_SPECIAL? We could actually use that for LPAE...

Quoted 17 lines
arch/arm/include/asm/pgtable-3level.h +0 -0
 #define L_PTE_NONE             (_AT(pteval_t, 1) << 57)        /* PROT_NONE */
 
 #define PMD_SECT_VALID         (_AT(pmdval_t, 1) << 0)
arch/arm/mm/proc-macros.S +1 -1
index e3c48a3..c62fd89 100644
@@ -97,7 +97,7 @@
 #error PTE shared bit mismatch
 #endif
 #if !defined (CONFIG_ARM_LPAE) && \
-       (L_PTE_XN+L_PTE_USER+L_PTE_RDONLY+L_PTE_DIRTY+L_PTE_YOUNG+\
+       (L_PTE_XN+L_PTE_USER+L_PTE_WRITE+L_PTE_DIRTY+L_PTE_YOUNG+\
         L_PTE_FILE+L_PTE_PRESENT) > L_PTE_SHARED
 #error Invalid Linux PTE bit settings
 #endif

arch/arm/mm/proc-macros.S line 157

Quoted 9 lines
arch/arm/mm/proc-macros.S +3 -3
@@ -152,9 +152,9 @@
        and     r2, r1, #L_PTE_MT_MASK
        ldr     r2, [ip, r2]
 
-       eor     r1, r1, #L_PTE_DIRTY
-       tst     r1, #L_PTE_DIRTY|L_PTE_RDONLY
-       orrne   r3, r3, #PTE_EXT_APX
+       tst     r1, #L_PTE_WRITE
+       tstne   r1, #L_PTE_DIRTY
+       orreq   r3, r3, #PTE_EXT_APX

Hehe, I have a patch pending in this macro which is sitting in -next. Take a look at b6ccb9803e90 ("ARM: 7954/1: mm: remove remaining domain support from ARMv6"), since this will probably conflict in horrible ways.

Quoted 22 lines
arch/arm/mm/proc-v7-2level.S +3 -3
index bdd3be4..297fccf 100644
@@ -84,9 +84,9 @@ ENTRY(cpu_v7_set_pte_ext)
        tst     r1, #1 << 4
        orrne   r3, r3, #PTE_EXT_TEX(1)
 
-       eor     r1, r1, #L_PTE_DIRTY
-       tst     r1, #L_PTE_RDONLY | L_PTE_DIRTY
-       orrne   r3, r3, #PTE_EXT_APX
+       tst     r1, #L_PTE_WRITE
+       tstne   r1, #L_PTE_DIRTY
+       orreq   r3, r3, #PTE_EXT_APX
 
        tst     r1, #L_PTE_USER
        orrne   r3, r3, #PTE_EXT_AP1
arch/arm/mm/proc-v7-3level.S +0 -0
index 01a719e..7793b2e 100644

arch/arm/mm/proc-v7-3level.S ENTRY(cpu_v7_set_pte_ext) · line 81

Quoted 6 lines
arch/arm/mm/proc-v7-3level.S +1 -2
@@ -78,8 +78,10 @@ ENTRY(cpu_v7_set_pte_ext)
        tst     r3, #1 << (57 - 32)             @ L_PTE_NONE
        bicne   r2, #L_PTE_VALID
        bne     1f
-       tst     r3, #1 << (55 - 32)             @ L_PTE_DIRTY
-       orreq   r2, #L_PTE_RDONLY
+       bic     r2, #PTE_RDONLY

Why do you need this bic?

Will

#3 lore raw

RFC PATCH 1/2 arm: mm: Switch back to L_PTE_WRITE

Quoted 1 line · Will Deacon

Hi Steve,

Hey Will,

Quoted 18 lines · Steve Capper
Quoted 16 lines

For LPAE, we have the following means for encoding writable or dirty ptes:

                              L_PTE_DIRTY       L_PTE_RDONLY
    !pte_dirty && !pte_write        0               1
    !pte_dirty && pte_write         0               1
    pte_dirty && !pte_write         1               1
    pte_dirty && pte_write          1               0

So we can't distinguish between writable clean ptes and read only ptes. This can cause problems with ptes being incorrectly flagged as read only when they are writable but not dirty.

This patch re-introduces the L_PTE_WRITE bit for both short descriptors and long descriptors, by reverting:
36bb94b ARM: pgtable: provide RDONLY page table bit rather than WRITE bit

Quoted 7 lines
Quoted 4 lines

For short descriptors the L_PTE_RDONLY bit is renamed to L_PTE_WRITE and the pertinent logic changed. For long descriptors, L_PTE_WRITE is implemented as a new software bit and L_PTE_RDONLY is renamed to PTE_RDONLY to highlight the fact that it is a hardware bit.

This would be a lot easier to review if it was a true revert, but I guess that doesn't apply cleanly to mainline?

Unfortunately not, we've had the split to 2/3-level pagetables since then. Also there are minor alterations to the kernel pte dumping code.

Quoted 32 lines
Quoted 32 lines
---
 arch/arm/include/asm/pgtable-2level.h |  2 +-
 arch/arm/include/asm/pgtable-3level.h |  4 ++--
 arch/arm/include/asm/pgtable.h        | 36 +++++++++++++++++------------------
 arch/arm/mm/dump.c                    |  8 ++++----
 arch/arm/mm/mmu.c                     | 25 ++++++++++++------------
 arch/arm/mm/proc-macros.S             | 16 ++++++++--------
 arch/arm/mm/proc-v7-2level.S          |  6 +++---
 arch/arm/mm/proc-v7-3level.S          |  6 ++++--
 arch/arm/mm/proc-xscale.S             |  4 ++--
 9 files changed, 55 insertions(+), 52 deletions(-)
arch/arm/include/asm/pgtable-2level.h +1 -1
index dfff709..ca43b84 100644
@@ -120,7 +120,7 @@
 #define L_PTE_YOUNG            (_AT(pteval_t, 1) << 1)
 #define L_PTE_FILE             (_AT(pteval_t, 1) << 2) /* only when !PRESENT */
 #define L_PTE_DIRTY            (_AT(pteval_t, 1) << 6)
-#define L_PTE_RDONLY           (_AT(pteval_t, 1) << 7)
+#define L_PTE_WRITE            (_AT(pteval_t, 1) << 7)
 #define L_PTE_USER             (_AT(pteval_t, 1) << 8)
 #define L_PTE_XN               (_AT(pteval_t, 1) << 9)
 #define L_PTE_SHARED           (_AT(pteval_t, 1) << 10)        /* shared(v6), coherent(xsc3) */
arch/arm/include/asm/pgtable-3level.h +0 -0
index 03243f7..8a392ef 100644
@@ -79,12 +79,12 @@
Quoted 8 lines

arch/arm/include/asm/pgtable-3level.h line 82

Quoted 5 lines
arch/arm/include/asm/pgtable-3level.h +1 -1
 #define L_PTE_PRESENT          (_AT(pteval_t, 3) << 0)         /* Present */
 #define L_PTE_FILE             (_AT(pteval_t, 1) << 2)         /* only when !PRESENT */
 #define L_PTE_USER             (_AT(pteval_t, 1) << 6)         /* AP[1] */
-#define L_PTE_RDONLY           (_AT(pteval_t, 1) << 7)         /* AP[2] */
+#define PTE_RDONLY             (_AT(pteval_t, 1) << 7)         /* AP[2] */

Why? I think we're just using L_ for consistency here, rather than to distinguish between h/w and Linux bits (e.g. L_PTE_XN).

The name was changed to break anything that used L_PTE_RDONLY, i.e. in case another patch slipped through and started behaving strangely. I will change this to something like L_PTE_HW_RDONLY.

Quoted 10 lines

arch/arm/include/asm/pgtable-3level.h line 87

Quoted 7 lines
arch/arm/include/asm/pgtable-3level.h +1 -1
 #define L_PTE_SHARED           (_AT(pteval_t, 3) << 8)         /* SH[1:0], inner shareable */
 #define L_PTE_YOUNG            (_AT(pteval_t, 1) << 10)        /* AF */
 #define L_PTE_XN               (_AT(pteval_t, 1) << 54)        /* XN */
 #define L_PTE_DIRTY            (_AT(pteval_t, 1) << 55)        /* unused */
-#define L_PTE_SPECIAL          (_AT(pteval_t, 1) << 56)        /* unused */
+#define L_PTE_WRITE            (_AT(pteval_t, 1) << 56)

Why have you killed L_PTE_SPECIAL? We could actually use that for LPAE...

I was trying to be efficient, as it was unused.

On the subject of future use of L_PTE_SPECIAL... It was pointed out to me that my fast_gup series had a bug in that it didn't check for special ptes (and it really should). So I would like to introduce L_PTE_SPECIAL usage in another patch ;-).

Quoted 20 lines
Quoted 20 lines
arch/arm/include/asm/pgtable-3level.h +0 -0
 #define L_PTE_NONE             (_AT(pteval_t, 1) << 57)        /* PROT_NONE */
 
 #define PMD_SECT_VALID         (_AT(pmdval_t, 1) << 0)
arch/arm/mm/proc-macros.S +1 -1
index e3c48a3..c62fd89 100644
@@ -97,7 +97,7 @@
 #error PTE shared bit mismatch
 #endif
 #if !defined (CONFIG_ARM_LPAE) && \
-       (L_PTE_XN+L_PTE_USER+L_PTE_RDONLY+L_PTE_DIRTY+L_PTE_YOUNG+\
+       (L_PTE_XN+L_PTE_USER+L_PTE_WRITE+L_PTE_DIRTY+L_PTE_YOUNG+\
         L_PTE_FILE+L_PTE_PRESENT) > L_PTE_SHARED
 #error Invalid Linux PTE bit settings
 #endif
@@ -152,9 +152,9 @@
        and     r2, r1, #L_PTE_MT_MASK
        ldr     r2, [ip, r2]
 
Quoted 10 lines

arch/arm/mm/proc-macros.S line 157

Quoted 6 lines
arch/arm/mm/proc-macros.S +3 -3
-       eor     r1, r1, #L_PTE_DIRTY
-       tst     r1, #L_PTE_DIRTY|L_PTE_RDONLY
-       orrne   r3, r3, #PTE_EXT_APX
+       tst     r1, #L_PTE_WRITE
+       tstne   r1, #L_PTE_DIRTY
+       orreq   r3, r3, #PTE_EXT_APX

Hehe, I have a patch pending in this macro which is sitting in -next. Take a look at b6ccb9803e90 ("ARM: 7954/1: mm: remove remaining domain support from ARMv6"), since this will probably conflict in horrible ways.

Eeeep, okay. I'll rebase this series to follow that patch. (Yes some careful changes to logic will be needed :-) ).

Quoted 16 lines
Quoted 16 lines
arch/arm/mm/proc-v7-2level.S +3 -3
index bdd3be4..297fccf 100644
@@ -84,9 +84,9 @@ ENTRY(cpu_v7_set_pte_ext)
        tst     r1, #1 << 4
        orrne   r3, r3, #PTE_EXT_TEX(1)
 
-       eor     r1, r1, #L_PTE_DIRTY
-       tst     r1, #L_PTE_RDONLY | L_PTE_DIRTY
-       orrne   r3, r3, #PTE_EXT_APX
+       tst     r1, #L_PTE_WRITE
+       tstne   r1, #L_PTE_DIRTY
+       orreq   r3, r3, #PTE_EXT_APX
 
Quoted 15 lines

arch/arm/mm/proc-v7-3level.S ENTRY(cpu_v7_set_pte_ext) · line 81

Quoted 13 lines
arch/arm/mm/proc-v7-2level.S +0 -0
        tst     r1, #L_PTE_USER
        orrne   r3, r3, #PTE_EXT_AP1
arch/arm/mm/proc-v7-3level.S +1 -2
index 01a719e..7793b2e 100644
@@ -78,8 +78,10 @@ ENTRY(cpu_v7_set_pte_ext)
        tst     r3, #1 << (57 - 32)             @ L_PTE_NONE
        bicne   r2, #L_PTE_VALID
        bne     1f
-       tst     r3, #1 << (55 - 32)             @ L_PTE_DIRTY
-       orreq   r2, #L_PTE_RDONLY
+       bic     r2, #PTE_RDONLY

Why do you need this bic?

I want to clear the read only bit if the pte is writable and dirty.

Cheers,

Signature · 1 line
Steve
#4 lore raw

RFC PATCH 1/2 arm: mm: Switch back to L_PTE_WRITE

Quoted 23 lines · Steve Capper
Quoted 22 lines · Will Deacon
Quoted 20 lines · Steve Capper

For LPAE, we have the following means for encoding writable or dirty ptes:

                              L_PTE_DIRTY       L_PTE_RDONLY
    !pte_dirty && !pte_write        0               1
    !pte_dirty && pte_write         0               1
    pte_dirty && !pte_write         1               1
    pte_dirty && pte_write          1               0

So we can't distinguish between writable clean ptes and read only ptes. This can cause problems with ptes being incorrectly flagged as read only when they are writable but not dirty.

This patch re-introduces the L_PTE_WRITE bit for both short descriptors and long descriptors, by reverting:
36bb94b ARM: pgtable: provide RDONLY page table bit rather than WRITE bit

For short descriptors the L_PTE_RDONLY bit is renamed to L_PTE_WRITE and the pertinent logic changed. For long descriptors, L_PTE_WRITE is implemented as a new software bit and L_PTE_RDONLY is renamed to PTE_RDONLY to highlight the fact that it is a hardware bit.

Quoted 5 lines
Quoted 2 lines

This would be a lot easier to review if it was a true revert, but I guess that doesn't apply cleanly to mainline?

Unfortunately not, we've had the split to 2/3-level pagetables since then. Also there are minor alterations to the kernel pte dumping code.

Yeah, I guess as much. Oh well.

Quoted 14 lines
Quoted 5 lines
Quoted 5 lines
arch/arm/include/asm/pgtable-3level.h +0 -0
index 03243f7..8a392ef 100644
@@ -79,12 +79,12 @@
Quoted 8 lines

arch/arm/include/asm/pgtable-3level.h line 82

Quoted 5 lines
arch/arm/include/asm/pgtable-3level.h +1 -1
 #define L_PTE_PRESENT          (_AT(pteval_t, 3) << 0)         /* Present */
 #define L_PTE_FILE             (_AT(pteval_t, 1) << 2)         /* only when !PRESENT */
 #define L_PTE_USER             (_AT(pteval_t, 1) << 6)         /* AP[1] */
-#define L_PTE_RDONLY           (_AT(pteval_t, 1) << 7)         /* AP[2] */
+#define PTE_RDONLY             (_AT(pteval_t, 1) << 7)         /* AP[2] */

Why? I think we're just using L_ for consistency here, rather than to distinguish between h/w and Linux bits (e.g. L_PTE_XN).

Quoted 3 lines

The name was changed to break anything that used L_PTE_RDONLY, i.e. in case another patch slipped through and started behaving strangely. I will change this to something like L_PTE_HW_RDONLY.

I'd personally just stick with L_PTE_RDONLY and have a quick grep around after the merge window which includes this patch, to see if any new users have turned up.

Quoted 11 lines
Quoted 10 lines

arch/arm/include/asm/pgtable-3level.h line 87

Quoted 7 lines
arch/arm/include/asm/pgtable-3level.h +1 -1
 #define L_PTE_SHARED           (_AT(pteval_t, 3) << 8)         /* SH[1:0], inner shareable */
 #define L_PTE_YOUNG            (_AT(pteval_t, 1) << 10)        /* AF */
 #define L_PTE_XN               (_AT(pteval_t, 1) << 54)        /* XN */
 #define L_PTE_DIRTY            (_AT(pteval_t, 1) << 55)        /* unused */
-#define L_PTE_SPECIAL          (_AT(pteval_t, 1) << 56)        /* unused */
+#define L_PTE_WRITE            (_AT(pteval_t, 1) << 56)

Why have you killed L_PTE_SPECIAL? We could actually use that for LPAE...

Quoted 6 lines

I was trying to be efficient, as it was unused.

On the subject of future use of L_PTE_SPECIAL... It was pointed out to me that my fast_gup series had a bug in that it didn't check for special ptes (and it really should). So I would like to introduce L_PTE_SPECIAL usage in another patch ;-).

Leaving the flag intact should be fine, since the pte_mkspecial pte_special macros don't use it yet iirc (although we *do* have them on arch/arm64, so you should check your GUP code there :).

Quoted 31 lines
Quoted 15 lines
Quoted 15 lines
arch/arm/mm/proc-v7-2level.S +3 -3
index bdd3be4..297fccf 100644
@@ -84,9 +84,9 @@ ENTRY(cpu_v7_set_pte_ext)
        tst     r1, #1 << 4
        orrne   r3, r3, #PTE_EXT_TEX(1)
 
-       eor     r1, r1, #L_PTE_DIRTY
-       tst     r1, #L_PTE_RDONLY | L_PTE_DIRTY
-       orrne   r3, r3, #PTE_EXT_APX
+       tst     r1, #L_PTE_WRITE
+       tstne   r1, #L_PTE_DIRTY
+       orreq   r3, r3, #PTE_EXT_APX
 
Quoted 15 lines

arch/arm/mm/proc-v7-3level.S ENTRY(cpu_v7_set_pte_ext) · line 81

Quoted 13 lines
arch/arm/mm/proc-v7-2level.S +0 -0
        tst     r1, #L_PTE_USER
        orrne   r3, r3, #PTE_EXT_AP1
arch/arm/mm/proc-v7-3level.S +1 -2
index 01a719e..7793b2e 100644
@@ -78,8 +78,10 @@ ENTRY(cpu_v7_set_pte_ext)
        tst     r3, #1 << (57 - 32)             @ L_PTE_NONE
        bicne   r2, #L_PTE_VALID
        bne     1f
-       tst     r3, #1 << (55 - 32)             @ L_PTE_DIRTY
-       orreq   r2, #L_PTE_RDONLY
+       bic     r2, #PTE_RDONLY

Why do you need this bic?

Quoted 1 line

I want to clear the read only bit if the pte is writable and dirty.

Ah yes, because that's no longer done by pte_mkwrite.

Will

#5 lore raw

RFC PATCH 2/2 arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

In reply to Steve Capper

On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits. Unfortunately, results from pte_write() and pte_dirty() are downcast to 32-bits by core code:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry

This patch adds a double logical invert to pte_write() and pte_dirty() for LPAE to ensure that the lower 32-bits are set if true.

---
 arch/arm/include/asm/pgtable-2level.h |  2 ++
 arch/arm/include/asm/pgtable-3level.h | 14 ++++++++++++++
 arch/arm/include/asm/pgtable.h        |  2 --
 3 files changed, 16 insertions(+), 2 deletions(-)
arch/arm/include/asm/pgtable-2level.h +2 -0
index ca43b84..7e8ebe7 100644
@@ -162,6 +162,8 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr)
 
 #define pmd_large(pmd)		(pmd_val(pmd) & 2)
 #define pmd_bad(pmd)		(pmd_val(pmd) & 2)
+#define pte_write(pte)		(pte_val(pte) & L_PTE_WRITE)
+#define pte_dirty(pte)		(pte_val(pte) & L_PTE_DIRTY)
 
 #define copy_pmd(pmdpd,pmdps)		\
 	do {				\
arch/arm/include/asm/pgtable-3level.h +14 -0
index 8a392ef..62efcc5 100644
@@ -135,6 +135,20 @@
 
 #ifndef __ASSEMBLY__
 
+/*
+ * On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits.
+ * Unfortunately, results from pte_write() and pte_dirty() are downcast
+ * to 32-bits by core code:
+ *  o gather_stats
+ *  o huge_pte_dirty
+ *  o huge_pte_write
+ *  o make_migration_entry
+ *
+ * Double logical invert to make sure lower 32-bits are set if true.
+ */
+#define pte_write(pte)		(!!(pte_val(pte) & L_PTE_WRITE))
+#define pte_dirty(pte)		(!!(pte_val(pte) & L_PTE_DIRTY))
+
 #define pud_none(pud)		(!pud_val(pud))
 #define pud_bad(pud)		(!(pud_val(pud) & 2))
 #define pud_present(pud)	(pud_val(pud))
arch/arm/include/asm/pgtable.h +0 -2
index 7a17611..e94a616 100644
@@ -216,8 +216,6 @@ static inline pte_t *pmd_page_vaddr(pmd_t pmd)
 
 #define pte_none(pte)		(!pte_val(pte))
 #define pte_present(pte)	(pte_val(pte) & L_PTE_PRESENT)
-#define pte_write(pte)		(pte_val(pte) & L_PTE_WRITE)
-#define pte_dirty(pte)		(pte_val(pte) & L_PTE_DIRTY)
 #define pte_young(pte)		(pte_val(pte) & L_PTE_YOUNG)
 #define pte_exec(pte)		(!(pte_val(pte) & L_PTE_XN))
 #define pte_special(pte)	(0)
Signature · 1 line
1.8.1.4
#6 lore raw

RFC PATCH 2/2 arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

Quoted 10 lines · Steve Capper

On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits. Unfortunately, results from pte_write() and pte_dirty() are downcast to 32-bits by core code:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry

This patch adds a double logical invert to pte_write() and pte_dirty() for LPAE to ensure that the lower 32-bits are set if true.

Yikes, this sounds like something we should put in -stable, no? If so, better make it patch 1 of this series.

Will

#7 lore raw

RFC PATCH 2/2 arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

Quoted 9 lines · Will Deacon
Quoted 8 lines · Steve Capper

On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits. Unfortunately, results from pte_write() and pte_dirty() are downcast to 32-bits by core code:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry

Quoted 5 lines
Quoted 2 lines

This patch adds a double logical invert to pte_write() and pte_dirty() for LPAE to ensure that the lower 32-bits are set if true.

Yikes, this sounds like something we should put in -stable, no? If so, better make it patch 1 of this series.

Yeah, it makes sense to have this as the first patch. It probably should go through stable, I will mark the next version of this patch for stable.

Quoted 2 lines

Thanks Will.

Cheers,
--
Steve

#8 lore raw

RFC PATCH 2/2 arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

In reply to Will Deacon
Quoted 9 lines · Will Deacon
Quoted 8 lines · Steve Capper

On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits. Unfortunately, results from pte_write() and pte_dirty() are downcast to 32-bits by core code:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry

Quoted 5 lines
Quoted 2 lines

This patch adds a double logical invert to pte_write() and pte_dirty() for LPAE to ensure that the lower 32-bits are set if true.

Yikes, this sounds like something we should put in -stable, no? If so, better make it patch 1 of this series.

That's a bug on arm64 as well since functions like gather_stats() take an int as argument.

However, my preference is for a static inline function instead of "!!" on arm64.

Steve, would you send a patch for arm64? If not, I can do it (with your reported-by).

Thanks.

Signature · 1 line
Catalin
#9 lore raw

RFC PATCH 2/2 arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

In reply to Steve Capper
Quoted 10 lines · Steve Capper

On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits. Unfortunately, results from pte_write() and pte_dirty() are downcast to 32-bits by core code:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry

This patch adds a double logical invert to pte_write() and pte_dirty() for LPAE to ensure that the lower 32-bits are set if true.

We should ensure all those functions return something which is compatible with "int" correctly. It didn't matter for non-LPAE as the PTEs fit in 32-bit, but with LPAE, that really needs fixing independently of your other patch.

Signature · 3 lines
FTTC broadband for 0.8mile line: 5.8Mbps down 500kbps up.  Estimation
in database were 13.1 to 19Mbit for a good line, about 7.5+ for a bad.
Estimate before purchase was "up to 13.2Mbit".
#10 lore raw

RFC PATCH 2/2 arm: mm: Double logical invert for LPAE pte_write(), pte_dirty()

Quoted 9 lines · Russell King - ARM Linux
Quoted 8 lines · Steve Capper

On LPAE, L_PTE_WRITE and L_PTE_DIRTY are in the upper 32-bits. Unfortunately, results from pte_write() and pte_dirty() are downcast to 32-bits by core code:
o gather_stats
o huge_pte_dirty
o huge_pte_write
o make_migration_entry

Quoted 7 lines
Quoted 2 lines

This patch adds a double logical invert to pte_write() and pte_dirty() for LPAE to ensure that the lower 32-bits are set if true.

We should ensure all those functions return something which is compatible with "int" correctly. It didn't matter for non-LPAE as the PTEs fit in 32-bit, but with LPAE, that really needs fixing independently of your other patch.

Agreed, I'm getting that written/tested now for arm and arm64.

Cheers,

Signature · 1 line
Steve
Quoted 5 lines
Signature · 4 lines
-- 
FTTC broadband for 0.8mile line: 5.8Mbps down 500kbps up.  Estimation
in database were 13.1 to 19Mbit for a good line, about 7.5+ for a bad.
Estimate before purchase was "up to 13.2Mbit".