x86 PAT: track pfnmap mappings with remap_pfn_range vm_insert_pfn - v3

---
 include/linux/mm.h |    3 +++
 mm/memory.c        |   43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 46 insertions(+)

include/linux/mm.h +3 -0

Index: linux-2.6/include/linux/mm.h
===================================================================
@@ -1223,6 +1223,9 @@ struct page *follow_page(struct vm_area_
 #define FOLL_GET	0x04	/* do get_page on page */
 #define FOLL_ANON	0x08	/* give ZERO_PAGE if no pgtable */
 
+int follow_pfnmap_pte(struct vm_area_struct *vma,
+				unsigned long address, pte_t *ret_ptep);
+
 typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
 			void *data);
 extern int apply_to_page_range(struct mm_struct *mm, unsigned long address,

mm/memory.c +43 -0

49 lines not on this page · raw message

[email protected]

author

2008-12-18 20:02 UTC #11 lore raw

PATCH 4/7 x86 PAT: Implement track/untrack of pfnmap regions for x86 - v3

Hookup remap_pfn_range and vm_insert_pfn and corresponding copy and free routines with reserve and free tracking.

reserve and free here only takes care of non RAM region mapping. For RAM region, driver should use set_memory_[uc|wc|wb] to set the cache type and then setup the mapping for user pte. We can bypass below reserve/free in that case.

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 arch/x86/include/asm/pgtable.h |   10 +
 arch/x86/mm/pat.c              |  236 +++++++++++++++++++++++++++++++++++++++++
 2 files changed, 246 insertions(+)

arch/x86/mm/pat.c +236 -0

Index: linux-2.6/arch/x86/mm/pat.c
===================================================================
@@ -596,6 +596,242 @@ void unmap_devmem(unsigned long pfn, uns
 	free_memtype(addr, addr + size);
 }
 
+/*
+ * Internal interface to reserve a range of physical memory with prot.
+ * Reserved non RAM regions only and after successful reserve_memtype,
+ * this func also keeps identity mapping (if any) in sync with this new prot.
+ */
+static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t vma_prot)
+{
+	int is_ram = 0;
+	int id_sz, ret;
+	unsigned long flags;
+	unsigned long want_flags = (pgprot_val(vma_prot) & _PAGE_CACHE_MASK);
+
+	is_ram = pagerange_is_ram(paddr, paddr + size);
+
+	if (is_ram != 0) {
+		/*
+		 * For mapping RAM pages, drivers need to call
+		 * set_memory_[uc|wc|wb] directly, for reserve and free, before
+		 * setting up the PTE.
+		 */
+		WARN_ON_ONCE(1);
+		return 0;
+	}
+
+	ret = reserve_memtype(paddr, paddr + size, want_flags, &flags);
+	if (ret)
+		return ret;
+
+	if (flags != want_flags) {
+		free_memtype(paddr, paddr + size);
+		printk(KERN_ERR
+		"%s:%d map pfn expected mapping type %s for %Lx-%Lx, got %s\n",
+			current->comm, current->pid,
+			cattr_name(want_flags),
+			(unsigned long long)paddr,
+			(unsigned long long)(paddr + size),
+			cattr_name(flags));
+		return -EINVAL;
+	}
+
+	/* Need to keep identity mapping in sync */
+	if (paddr >= __pa(high_memory))
+		return 0;
+
+	id_sz = (__pa(high_memory) < paddr + size) ?
+				__pa(high_memory) - paddr :
+				size;
+
+	if (ioremap_change_attr((unsigned long)__va(paddr), id_sz, flags) < 0) {
+		free_memtype(paddr, paddr + size);
+		printk(KERN_ERR
+			"%s:%d reserve_pfn_range ioremap_change_attr failed %s "
+			"for %Lx-%Lx\n",
+			current->comm, current->pid,
+			cattr_name(flags),
+			(unsigned long long)paddr,
+			(unsigned long long)(paddr + size));
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/*
+ * Internal interface to free a range of physical memory.
+ * Frees non RAM regions only.
+ */
+static void free_pfn_range(u64 paddr, unsigned long size)
+{
+	int is_ram;
+
+	is_ram = pagerange_is_ram(paddr, paddr + size);
+	if (is_ram == 0)
+		free_memtype(paddr, paddr + size);
+}
+
+/*
+ * track_pfn_vma_copy is called when vma that is covering the pfnmap gets
+ * copied through copy_page_range().
+ *
+ * If the vma has a linear pfn mapping for the entire range, we get the prot
+ * from pte and reserve the entire vma range with single reserve_pfn_range call.
+ * Otherwise, we reserve the entire vma range, my ging through the PTEs page
+ * by page to get physical address and protection.
+ */
+int track_pfn_vma_copy(struct vm_area_struct *vma)
+{
+	int retval = 0;
+	unsigned long i, j;
+	u64 paddr;
+	pgprot_t prot;
+	pte_t pte;
+	unsigned long vma_start = vma->vm_start;
+	unsigned long vma_end = vma->vm_end;
+	unsigned long vma_size = vma_end - vma_start;
+
+	if (!pat_enabled)
+		return 0;
+
+	if (is_linear_pfn_mapping(vma)) {
+		/*
+		 * reserve the whole chunk starting from vm_pgoff,
+		 * But, we have to get the protection from pte.
+		 */
+		if (follow_pfnmap_pte(vma, vma_start, &pte)) {
+			WARN_ON_ONCE(1);
+			return -1;
+		}
+		prot = pte_pgprot(pte);
+		paddr = (u64)vma->vm_pgoff << PAGE_SHIFT;
+		return reserve_pfn_range(paddr, vma_size, prot);
+	}
+
+	/* reserve entire vma page by page, using pfn and prot from pte */
+	for (i = 0; i < vma_size; i += PAGE_SIZE) {
+		if (follow_pfnmap_pte(vma, vma_start + i, &pte))
+			continue;
+
+		paddr = pte_pa(pte);
+		prot = pte_pgprot(pte);
+		retval = reserve_pfn_range(paddr, PAGE_SIZE, prot);
+		if (retval)
+			goto cleanup_ret;
+	}
+	return 0;
+
+cleanup_ret:
+	/* Reserve error: Cleanup partial reservation and return error */
+	for (j = 0; j < i; j += PAGE_SIZE) {
+		if (follow_pfnmap_pte(vma, vma_start + j, &pte))
+			continue;
+
+		paddr = pte_pa(pte);
+		free_pfn_range(paddr, PAGE_SIZE);
+	}
+
+	return retval;
+}
+
+/*
+ * track_pfn_vma_new is called when a _new_ pfn mapping is being established
+ * for physical range indicated by pfn and size.
+ *
+ * prot is passed in as a parameter for the new mapping. If the vma has a
+ * linear pfn mapping for the entire range reserve the entire vma range with
+ * single reserve_pfn_range call.
+ * Otherwise, we look t the pfn and size and reserve only the specified range
+ * page by page.
+ *
+ * Note that this function can be called with caller trying to map only a
+ * subrange/page inside the vma.
+ */
+int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot,
+			unsigned long pfn, unsigned long size)
+{
+	int retval = 0;
+	unsigned long i, j;
+	u64 base_paddr;
+	u64 paddr;
+	unsigned long vma_start = vma->vm_start;
+	unsigned long vma_end = vma->vm_end;
+	unsigned long vma_size = vma_end - vma_start;
+
+	if (!pat_enabled)
+		return 0;
+
+	if (is_linear_pfn_mapping(vma)) {
+		/* reserve the whole chunk starting from vm_pgoff */
+		paddr = (u64)vma->vm_pgoff << PAGE_SHIFT;
+		return reserve_pfn_range(paddr, vma_size, prot);
+	}
+
+	/* reserve page by page using pfn and size */
+	base_paddr = (u64)pfn << PAGE_SHIFT;
+	for (i = 0; i < size; i += PAGE_SIZE) {
+		paddr = base_paddr + i;
+		retval = reserve_pfn_range(paddr, PAGE_SIZE, prot);
+		if (retval)
+			goto cleanup_ret;
+	}
+	return 0;
+
+cleanup_ret:
+	/* Reserve error: Cleanup partial reservation and return error */
+	for (j = 0; j < i; j += PAGE_SIZE) {
+		paddr = base_paddr + j;
+		free_pfn_range(paddr, PAGE_SIZE);
+	}
+
+	return retval;
+}
+
+/*
+ * untrack_pfn_vma is called while unmapping a pfnmap for a region.
+ * untrack can be called for a specific region indicated by pfn and size or
+ * can be for the entire vma (in which case size can be zero).
+ */
+void untrack_pfn_vma(struct vm_area_struct *vma, unsigned long pfn,
+			unsigned long size)
+{
+	unsigned long i;
+	u64 paddr;
+	unsigned long vma_start = vma->vm_start;
+	unsigned long vma_end = vma->vm_end;
+	unsigned long vma_size = vma_end - vma_start;
+
+	if (!pat_enabled)
+		return;
+
+	if (is_linear_pfn_mapping(vma)) {
+		/* free the whole chunk starting from vm_pgoff */
+		paddr = (u64)vma->vm_pgoff << PAGE_SHIFT;
+		free_pfn_range(paddr, vma_size);
+		return;
+	}
+
+	if (size != 0 && size != vma_size) {
+		/* free page by page, using pfn and size */
+		paddr = (u64)pfn << PAGE_SHIFT;
+		for (i = 0; i < size; i += PAGE_SIZE) {
+			paddr = paddr + i;
+			free_pfn_range(paddr, PAGE_SIZE);
+		}
+	} else {
+		/* free entire vma, page by page, using the pfn from pte */
+		for (i = 0; i < vma_size; i += PAGE_SIZE) {
+			pte_t pte;
+
+			if (follow_pfnmap_pte(vma, vma_start + i, &pte))
+				continue;
+
+			paddr = pte_pa(pte);
+			free_pfn_range(paddr, PAGE_SIZE);
+		}
+	}
+}
+
 #if defined(CONFIG_DEBUG_FS) && defined(CONFIG_X86_PAT)
 
 /* get Nth element of the linked list */

arch/x86/include/asm/pgtable.h +10 -0

Index: linux-2.6/arch/x86/include/asm/pgtable.h
===================================================================
@@ -219,6 +219,11 @@ static inline unsigned long pte_pfn(pte_
 	return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT;
 }
 
+static inline u64 pte_pa(pte_t pte)
+{
+	return pte_val(pte) & PTE_PFN_MASK;
+}
+
 #define pte_page(pte)	pfn_to_page(pte_pfn(pte))
 
 static inline int pmd_large(pmd_t pte)

@@ -328,6 +333,11 @@ static inline pgprot_t pgprot_modify(pgp
 
 #define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask)
 
+/* Indicate that x86 has its own track and untrack pfn vma functions */
+#define track_pfn_vma_new track_pfn_vma_new
+#define track_pfn_vma_copy track_pfn_vma_copy
+#define untrack_pfn_vma untrack_pfn_vma
+
 #ifndef __ASSEMBLY__
 #define __HAVE_PHYS_MEM_ACCESS_PROT
 struct file;

[email protected]

author

2008-12-18 20:02 UTC #8 lore raw

PATCH 3/7 x86 PAT: hooks in generic vm code to help archs to track pfnmap regions - v3

Introduce generic hooks in remap_pfn_range and vm_insert_pfn and corresponding copy and free routines with reserve and free tracking.

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 include/linux/mm.h |    6 ++++
 mm/memory.c        |   76 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 81 insertions(+), 1 deletion(-)

mm/memory.c +75 -1

Index: linux-2.6/mm/memory.c
===================================================================
@@ -99,6 +99,50 @@ int randomize_va_space __read_mostly =
 					2;
 #endif
 
+#ifndef track_pfn_vma_new
+/*
+ * Interface that can be used by architecture code to keep track of
+ * memory type of pfn mappings (remap_pfn_range, vm_insert_pfn)
+ *
+ * track_pfn_vma_new is called when a _new_ pfn mapping is being established
+ * for physical range indicated by pfn and size.
+ */
+int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot,
+			unsigned long pfn, unsigned long size)
+{
+	return 0;
+}
+#endif
+
+#ifndef track_pfn_vma_copy
+/*
+ * Interface that can be used by architecture code to keep track of
+ * memory type of pfn mappings (remap_pfn_range, vm_insert_pfn)
+ *
+ * track_pfn_vma_copy is called when vma that is covering the pfnmap gets
+ * copied through copy_page_range().
+ */
+int track_pfn_vma_copy(struct vm_area_struct *vma)
+{
+	return 0;
+}
+#endif
+
+#ifndef untrack_pfn_vma
+/*
+ * Interface that can be used by architecture code to keep track of
+ * memory type of pfn mappings (remap_pfn_range, vm_insert_pfn)
+ *
+ * untrack_pfn_vma is called while unmapping a pfnmap for a region.
+ * untrack can be called for a specific region indicated by pfn and size or
+ * can be for the entire vma (in which case size can be zero).
+ */
+void untrack_pfn_vma(struct vm_area_struct *vma, unsigned long pfn,
+			unsigned long size)
+{
+}
+#endif
+
 static int __init disable_randmaps(char *s)
 {
 	randomize_va_space = 0;

@@ -669,6 +713,16 @@ int copy_page_range(struct mm_struct *ds
 	if (is_vm_hugetlb_page(vma))
 		return copy_hugetlb_page_range(dst_mm, src_mm, vma);
 
+	if (is_pfn_mapping(vma)) {
+		/*
+		 * We do not free on error cases below as remove_vma
+		 * gets called on error from higher level routine
+		 */
+		ret = track_pfn_vma_copy(vma);
+		if (ret)
+			return ret;
+	}
+
 	/*
 	 * We need to invalidate the secondary MMU mappings only when
 	 * there could be a permission downgrade on the ptes of the

@@ -915,6 +969,9 @@ unsigned long unmap_vmas(struct mmu_gath
 		if (vma->vm_flags & VM_ACCOUNT)
 			*nr_accounted += (end - start) >> PAGE_SHIFT;
 
+		if (is_pfn_mapping(vma))
+			untrack_pfn_vma(vma, 0, 0);
+
 		while (start != end) {
 			if (!tlb_start_valid) {
 				tlb_start = start;

@@ -1473,6 +1530,7 @@ out:
 int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
 			unsigned long pfn)
 {
+	int ret;
 	/*
 	 * Technically, architectures with pte_special can avoid all these
 	 * restrictions (same for remap_pfn_range).  However we would like

@@ -1487,7 +1545,15 @@ int vm_insert_pfn(struct vm_area_struct 
 
 	if (addr < vma->vm_start || addr >= vma->vm_end)
 		return -EFAULT;
-	return insert_pfn(vma, addr, pfn, vma->vm_page_prot);
+	if (track_pfn_vma_new(vma, vma->vm_page_prot, pfn, PAGE_SIZE))
+		return -EINVAL;
+
+  	ret = insert_pfn(vma, addr, pfn, vma->vm_page_prot);
+
+	if (ret)
+		untrack_pfn_vma(vma, pfn, PAGE_SIZE);
+
+	return ret;
 }
 EXPORT_SYMBOL(vm_insert_pfn);

@@ -1625,6 +1691,10 @@ int remap_pfn_range(struct vm_area_struc
 
 	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
 
+	err = track_pfn_vma_new(vma, prot, pfn, PAGE_ALIGN(size));
+	if (err)
+		return -EINVAL;
+
 	BUG_ON(addr >= end);
 	pfn -= addr >> PAGE_SHIFT;
 	pgd = pgd_offset(mm, addr);

@@ -1636,6 +1706,10 @@ int remap_pfn_range(struct vm_area_struc
 		if (err)
 			break;
 	} while (pgd++, addr = next, addr != end);
+
+	if (err)
+		untrack_pfn_vma(vma, pfn, PAGE_ALIGN(size));
+
 	return err;
 }
 EXPORT_SYMBOL(remap_pfn_range);

include/linux/mm.h +6 -0

Index: linux-2.6/include/linux/mm.h
===================================================================
@@ -155,6 +155,12 @@ static inline int is_pfn_mapping(struct 
 	return (vma->vm_flags & VM_PFNMAP);
 }
 
+extern int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot,
+				unsigned long pfn, unsigned long size);
+extern int track_pfn_vma_copy(struct vm_area_struct *vma);
+extern void untrack_pfn_vma(struct vm_area_struct *vma, unsigned long pfn,
+				unsigned long size);
+
 /*
  * vm_fault is filled by the the pagefault handler and passed to the vma's
  * ->fault function. The vma's ->fault is responsible for returning a bitmask

[email protected]

author

2008-12-18 20:03 UTC #18 lore raw

PATCH 7/7 x86 PAT: update documentation to cover pgprot and remap_pfn related changes - v3

Add documentation related to pgprot_* change.

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 Documentation/x86/pat.txt |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

Documentation/x86/pat.txt +24 -0

Index: linux-2.6/Documentation/x86/pat.txt
===================================================================
@@ -80,6 +80,30 @@ pci proc               |    --    |    -
                        |          |            |                  |
 -------------------------------------------------------------------
 
+Advanced APIs for drivers
+-------------------------
+A. Exporting pages to user with remap_pfn_range, io_remap_pfn_range,
+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace, do it by using mmap
+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being added. So, driver can
+continue to use the above sequence, with either pgprot_noncached() or
+pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or WC in memtype
+list in order to ensure no conflicting mapping.
+
+Note that this set of APIs only work with IO (non RAM) regions. If driver
+wants to export RAM region, it has to do set_memory_uc() or set_memory_wc()
+as step 0 above and also track the usage of those pages and use set_memory_wb()
+before the page is freed to free pool.
+
+
+
 Notes:
 
 -- in the above table mean "Not suggested usage for the API". Some of the --'s

[email protected]

author

2008-12-18 20:03 UTC #17 lore raw

PATCH 6/7 x86 PAT: add pgprot_writecombine() interface for drivers - v3

Add pgprot_writecombine. pgprot_writecombine will be aliased to pgprot_noncached when not supported by the architecture.

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 arch/x86/include/asm/pgtable.h |    3 +++
 arch/x86/mm/pat.c              |    8 ++++++++
 include/asm-generic/pgtable.h  |    4 ++++
 3 files changed, 15 insertions(+)

include/asm-generic/pgtable.h +4 -0

Index: linux-2.6/include/asm-generic/pgtable.h
===================================================================
@@ -129,6 +129,10 @@ static inline void ptep_set_wrprotect(st
 #define move_pte(pte, prot, old_addr, new_addr)	(pte)
 #endif
 
+#ifndef pgprot_writecombine
+#define pgprot_writecombine pgprot_noncached
+#endif
+
 /*
  * When walking page tables, get the address of the next boundary,
  * or the end address of the range if that comes earlier.  Although no

arch/x86/include/asm/pgtable.h +3 -0

Index: linux-2.6/arch/x86/include/asm/pgtable.h
===================================================================
@@ -168,6 +168,9 @@
 
 #ifndef __ASSEMBLY__
 
+#define pgprot_writecombine	pgprot_writecombine
+extern pgprot_t pgprot_writecombine(pgprot_t prot);
+
 /*
  * ZERO_PAGE is a global shared page that is always zero: used
  * for zero-mapped memory areas etc..

arch/x86/mm/pat.c +8 -0

Index: linux-2.6/arch/x86/mm/pat.c
===================================================================
@@ -832,6 +832,14 @@ void untrack_pfn_vma(struct vm_area_stru
 	}
 }
 
+pgprot_t pgprot_writecombine(pgprot_t prot)
+{
+	if (pat_enabled)
+		return __pgprot(pgprot_val(prot) | _PAGE_CACHE_WC);
+	else
+		return pgprot_noncached(prot);
+}
+
 #if defined(CONFIG_DEBUG_FS) && defined(CONFIG_X86_PAT)
 
 /* get Nth element of the linked list */

[email protected]

author

2008-12-18 20:03 UTC #1 lore raw

PATCH 1/7 x86 PAT: store vm_pgoff for all linear_over_vma_region mappings - v3

The following patchset adds x86 PAT attribute tracking and untracking for pfnmap related APIs.

This patch:

While working on x86 PAT, we faced some hurdles with trackking remap_pfn_range() regions, as we do not have any information to say whether that PFNMAP mapping is linear for the entire vma range or it is smaller granularity regions within the vma.

A simple solution to this is to use vm_pgoff as an indicator for linear mapping over the vma region. Currently, remap_pfn_range only sets vm_pgoff for COW mappings. Below patch changes the logic and sets the vm_pgoff irrespective of COW. This will still not be enough for the case where pfn is zero (vma region mapped to physical address zero). But, for all the other cases, we can look at pfnmap VMAs and say whether the mappng is for the entire vma region or not.

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 include/linux/mm.h |    9 +++++++++
 mm/memory.c        |    7 +++----
 2 files changed, 12 insertions(+), 4 deletions(-)

mm/memory.c +3 -4

Index: linux-2.6/mm/memory.c
===================================================================
@@ -1575,11 +1575,10 @@ int remap_pfn_range(struct vm_area_struc
 	 * behaviour that some programs depend on. We mark the "original"
 	 * un-COW'ed pages by matching them up with "vma->vm_pgoff".
 	 */
-	if (is_cow_mapping(vma->vm_flags)) {
-		if (addr != vma->vm_start || end != vma->vm_end)
-			return -EINVAL;
+	if (addr == vma->vm_start && end == vma->vm_end)
 		vma->vm_pgoff = pfn;
-	}
+	else if (is_cow_mapping(vma->vm_flags))
+		return -EINVAL;
 
 	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;

include/linux/mm.h +9 -0

Index: linux-2.6/include/linux/mm.h
===================================================================
@@ -145,6 +145,15 @@ extern pgprot_t protection_map[16];
 #define FAULT_FLAG_WRITE	0x01	/* Fault was a write access */
 #define FAULT_FLAG_NONLINEAR	0x02	/* Fault was via a nonlinear mapping */
 
+static inline int is_linear_pfn_mapping(struct vm_area_struct *vma)
+{
+	return ((vma->vm_flags & VM_PFNMAP) && vma->vm_pgoff);
+}
+
+static inline int is_pfn_mapping(struct vm_area_struct *vma)
+{
+	return (vma->vm_flags & VM_PFNMAP);
+}
 
 /*
  * vm_fault is filled by the the pagefault handler and passed to the vma's

[email protected]

author

2008-12-18 20:04 UTC #16 lore raw

PATCH 5/7 x86 PAT: change pgprot_noncached to uc_minus instead of strong uc - v3

Make pgprot_noncached uc_minus instead of strong UC. This will make pgprot_noncached to be in line with ioremap_nocache() and all the other APIs that map page uc_minus on uc request.

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 arch/x86/include/asm/pgtable.h    |    8 ++++++++
 arch/x86/include/asm/pgtable_32.h |    9 ---------
 arch/x86/include/asm/pgtable_64.h |    6 ------
 3 files changed, 8 insertions(+), 15 deletions(-)

arch/x86/include/asm/pgtable.h +8 -0

Index: linux-2.6/arch/x86/include/asm/pgtable.h
===================================================================
@@ -158,6 +158,14 @@
 #define PGD_IDENT_ATTR	 0x001		/* PRESENT (no other attributes) */
 #endif
 
+/*
+ * Macro to mark a page protection value as UC-
+ */
+#define pgprot_noncached(prot)					\
+	((boot_cpu_data.x86 > 3)				\
+	 ? (__pgprot(pgprot_val(prot) | _PAGE_CACHE_UC_MINUS))	\
+	 : (prot))
+
 #ifndef __ASSEMBLY__
 
 /*

arch/x86/include/asm/pgtable_32.h +0 -9

Index: linux-2.6/arch/x86/include/asm/pgtable_32.h
===================================================================
@@ -101,15 +101,6 @@ extern unsigned long pg0[];
 #endif
 
 /*
- * Macro to mark a page protection value as "uncacheable".
- * On processors which do not support it, this is a no-op.
- */
-#define pgprot_noncached(prot)					\
-	((boot_cpu_data.x86 > 3)				\
-	 ? (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))	\
-	 : (prot))
-
-/*
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.
  */

arch/x86/include/asm/pgtable_64.h +0 -6

Index: linux-2.6/arch/x86/include/asm/pgtable_64.h
===================================================================
@@ -177,12 +177,6 @@ static inline int pmd_bad(pmd_t pmd)
 #define pages_to_mb(x)	((x) >> (20 - PAGE_SHIFT))   /* FIXME: is this right? */
 
 /*
- * Macro to mark a page protection value as "uncacheable".
- */
-#define pgprot_noncached(prot)					\
-	(__pgprot(pgprot_val((prot)) | _PAGE_PCD | _PAGE_PWT))
-
-/*
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.
  */

Randy Dunlap

2008-12-18 21:16 UTC #19 lore raw

Re: [patch 7/7] x86 PAT: update documentation to cover pgprot and remap_pfn related changes - v3

[email protected] wrote:

+Advanced APIs for drivers
+-------------------------
+A. Exporting pages to user with remap_pfn_range, io_remap_pfn_range,

to users
or to userspace

+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace, do it by using mmap

Drop comma.

+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being added. So, driver can

s/pat/PAT/
s/driver/drivers/ or s/driver/a driver/

+continue to use the above sequence, with either pgprot_noncached() or +pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or WC in memtype +list in order to ensure no conflicting mapping.
+
+Note that this set of APIs only work with IO (non RAM) regions. If driver

works with IO (non-RAM) regions. If a driver

+wants to export RAM region, it has to do set_memory_uc() or set_memory_wc()

export a RAM region,

+as step 0 above and also track the usage of those pages and use set_memory_wb() +before the page is freed to free pool.

Signature · 1 line

~Randy

H. Peter Anvin

reviewer

2008-12-18 21:25 UTC #23 lore raw

Quoted 7 lines · [email protected] wrote:

v3: Patches updated based on Andrew's comments on the earlier version.
Drivers use mmap followed by pgprot_* and remap_pfn_range or vm_insert_pfn, in order to export reserved memory to userspace. Currently, such mappings are not tracked and hence not kept consistent with other mappings (/dev/mem, pci resource, ioremap) for the sme memory, that may exist in the system.

Quoted 8 lines

The following patchset adds x86 PAT attribute tracking and untracking for pfnmap related APIs.
First three patches in the patchset are changing the generic mm code to fit in this tracking. Last four patches are x86 specific to make things work with x86 PAT code. The patchset aso introduces pgprot_writecombine interface, which gives writecombine mapping when enabled, falling back to pgprot_noncached otherwise.

Series applied to tip:x86/pat2, thanks!

-hpa

Signature · 2 lines

H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.

Nick Piggin

2008-12-18 21:27 UTC #2 lore raw

Re: [patch 1/7] x86 PAT: store vm_pgoff for all linear_over_vma_region mappings - v3

Quoted 67 lines · On Thu, Dec 18, 2008 at 11:41:27AM -0800, [email protected] wrote:

Drivers use mmap followed by pgprot_* and remap_pfn_range or vm_insert_pfn, in order to export reserved memory to userspace. Currently, such mappings are not tracked and hence not kept consistent with other mappings (/dev/mem, pci resource, ioremap) for the sme memory, that may exist in the system.
The following patchset adds x86 PAT attribute tracking and untracking for pfnmap related APIs.
First three patches in the patchset are changing the generic mm code to fit in this tracking. Last four patches are x86 specific to make things work with x86 PAT code. The patchset aso introduces pgprot_writecombine interface, which gives writecombine mapping when enabled, falling back to pgprot_noncached otherwise.
This patch:
While working on x86 PAT, we faced some hurdles with trackking remap_pfn_range() regions, as we do not have any information to say whether that PFNMAP mapping is linear for the entire vma range or it is smaller granularity regions within the vma.
A simple solution to this is to use vm_pgoff as an indicator for linear mapping over the vma region. Currently, remap_pfn_range only sets vm_pgoff for COW mappings. Below patch changes the logic and sets the vm_pgoff irrespective of COW. This will still not be enough for the case where pfn is zero (vma region mapped to physical address zero). But, for all the other cases, we can look at pfnmap VMAs and say whether the mappng is for the entire vma region or not.
Signed-off-by Venkatesh Pallipadi <[email protected]>
Signed-off-by Suresh Siddha <[email protected]>
---
 include/linux/mm.h |    9 +++++++++
 mm/memory.c        |    7 +++----
 2 files changed, 12 insertions(+), 4 deletions(-)
Index: linux-2.6/mm/memory.c
===================================================================
mm/memory.c +3 -4
@@ -1575,11 +1575,10 @@ int remap_pfn_range(struct vm_area_struc
 	 * behaviour that some programs depend on. We mark the "original"
 	 * un-COW'ed pages by matching them up with "vma->vm_pgoff".
 	 */
-	if (is_cow_mapping(vma->vm_flags)) {
-		if (addr != vma->vm_start || end != vma->vm_end)
-			return -EINVAL;
+	if (addr == vma->vm_start && end == vma->vm_end)
 		vma->vm_pgoff = pfn;
-	}
+	else if (is_cow_mapping(vma->vm_flags))
+		return -EINVAL;
 
 	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
 
Index: linux-2.6/include/linux/mm.h
===================================================================
include/linux/mm.h +2 -0
@@ -145,6 +145,15 @@ extern pgprot_t protection_map[16];
 #define FAULT_FLAG_WRITE	0x01	/* Fault was a write access */
 #define FAULT_FLAG_NONLINEAR	0x02	/* Fault was via a nonlinear mapping */
 
+static inline int is_linear_pfn_mapping(struct vm_area_struct *vma)
+{

include/linux/mm.h extern pgprot_t protection_map[16]; · line 159

Quoted 10 lines

include/linux/mm.h +7 -0

+	return ((vma->vm_flags & VM_PFNMAP) && vma->vm_pgoff);
+}
+
+static inline int is_pfn_mapping(struct vm_area_struct *vma)
+{
+	return (vma->vm_flags & VM_PFNMAP);
+}
 
 /*
  * vm_fault is filled by the the pagefault handler and passed to the vma's

This is fine by me, however:
1. Can you add some comments to say "this is not for core vm but for pat, oh and a pgoff of zero is not going to work".
2. Can you please justify to me (or the changelog) roughly why PAT wants to know if the mapping is linear or not? Presumably it has to handle both types? If performance wasn't an issue, then you could manually scan the ptes to verify (which would solve your zero-offset bug). etc.

Nick Piggin

2008-12-18 21:31 UTC #6 lore raw

Re: [patch 2/7] x86 PAT: Add follow_pfnmp_pte routine to help tracking pfnmap pages - v3

Quoted 6 lines · On Thu, Dec 18, 2008 at 11:41:28AM -0800, [email protected] wrote:

Add a generic interface to follow pfn in a pfnmap vma range. This is used by one of the subsequent x86 PAT related patch to keep track of memory types for vma regions across vma copy and free.
Signed-off-by Venkatesh Pallipadi <[email protected]>
Signed-off-by Suresh Siddha <[email protected]>

Can you please reuse follow_phys for this? (preferably use the same API even if it requires some modification, otherwise if not possible, then at least can you implement a common core for both APIs).

Quoted 76 lines

---
 include/linux/mm.h |    3 +++
 mm/memory.c        |   43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 46 insertions(+)

Index: linux-2.6/include/linux/mm.h
===================================================================

include/linux/mm.h +3 -0

@@ -1223,6 +1223,9 @@ struct page *follow_page(struct vm_area_
 #define FOLL_GET	0x04	/* do get_page on page */
 #define FOLL_ANON	0x08	/* give ZERO_PAGE if no pgtable */
 
+int follow_pfnmap_pte(struct vm_area_struct *vma,
+				unsigned long address, pte_t *ret_ptep);
+
 typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
 			void *data);
 extern int apply_to_page_range(struct mm_struct *mm, unsigned long address,

Index: linux-2.6/mm/memory.c
===================================================================

mm/memory.c +43 -0

@@ -1111,6 +1111,49 @@ no_page_table:
 	return page;
 }
 
+int follow_pfnmap_pte(struct vm_area_struct *vma, unsigned long address,
+			pte_t *ret_ptep)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *ptep, pte;
+	spinlock_t *ptl;
+	struct page *page;
+	struct mm_struct *mm = vma->vm_mm;
+
+	if (!is_pfn_mapping(vma))
+		goto err;
+
+	page = NULL;
+	pgd = pgd_offset(mm, address);
+	if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd)))
+		goto err;
+
+	pud = pud_offset(pgd, address);
+	if (pud_none(*pud) || unlikely(pud_bad(*pud)))
+		goto err;
+
+	pmd = pmd_offset(pud, address);
+	if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
+		goto err;
+
+	ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
+
+	pte = *ptep;
+	if (!pte_present(pte))
+		goto err_unlock;
+
+	*ret_ptep = pte;
+	pte_unmap_unlock(ptep, ptl);
+	return 0;
+
+err_unlock:
+	pte_unmap_unlock(ptep, ptl);
+err:
+	return -EINVAL;
+}
+
 /* Can we do the FOLL_ANON optimization? */
 static inline int use_zero_page(struct vm_area_struct *vma)
 {

Signature · 1 line

--

Nick Piggin

2008-12-18 21:36 UTC #9 lore raw

Re: [patch 3/7] x86 PAT: hooks in generic vm code to help archs to track pfnmap regions - v3

Quoted 2 lines · On Thu, Dec 18, 2008 at 11:41:29AM -0800, [email protected] wrote:

Introduce generic hooks in remap_pfn_range and vm_insert_pfn and corresponding copy and free routines with reserve and free tracking.

These should be inline so that they can be folded out (I'm sure gcc with -Os and "optimize" inlining will do something stupid here). Also, the normal way to add such arch hooks is to put the default into asm-generic and have other archs include it... that would be nicer than sticking it into mm/memory.c wouldn't it?

Sigh, fork/exit paths slow down yet again. But oh well. Maybe can you add some branch hints?

Quoted 156 lines

Signed-off-by Venkatesh Pallipadi <[email protected]>

---
 include/linux/mm.h |    6 ++++
 mm/memory.c        |   76 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 81 insertions(+), 1 deletion(-)

Index: linux-2.6/mm/memory.c
===================================================================

mm/memory.c +75 -1

@@ -99,6 +99,50 @@ int randomize_va_space __read_mostly =
 					2;
 #endif
 
+#ifndef track_pfn_vma_new
+/*
+ * Interface that can be used by architecture code to keep track of
+ * memory type of pfn mappings (remap_pfn_range, vm_insert_pfn)
+ *
+ * track_pfn_vma_new is called when a _new_ pfn mapping is being established
+ * for physical range indicated by pfn and size.
+ */
+int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot,
+			unsigned long pfn, unsigned long size)
+{
+	return 0;
+}
+#endif
+
+#ifndef track_pfn_vma_copy
+/*
+ * Interface that can be used by architecture code to keep track of
+ * memory type of pfn mappings (remap_pfn_range, vm_insert_pfn)
+ *
+ * track_pfn_vma_copy is called when vma that is covering the pfnmap gets
+ * copied through copy_page_range().
+ */
+int track_pfn_vma_copy(struct vm_area_struct *vma)
+{
+	return 0;
+}
+#endif
+
+#ifndef untrack_pfn_vma
+/*
+ * Interface that can be used by architecture code to keep track of
+ * memory type of pfn mappings (remap_pfn_range, vm_insert_pfn)
+ *
+ * untrack_pfn_vma is called while unmapping a pfnmap for a region.
+ * untrack can be called for a specific region indicated by pfn and size or
+ * can be for the entire vma (in which case size can be zero).
+ */
+void untrack_pfn_vma(struct vm_area_struct *vma, unsigned long pfn,
+			unsigned long size)
+{
+}
+#endif
+
 static int __init disable_randmaps(char *s)
 {
 	randomize_va_space = 0;

@@ -669,6 +713,16 @@ int copy_page_range(struct mm_struct *ds
 	if (is_vm_hugetlb_page(vma))
 		return copy_hugetlb_page_range(dst_mm, src_mm, vma);
 
+	if (is_pfn_mapping(vma)) {
+		/*
+		 * We do not free on error cases below as remove_vma
+		 * gets called on error from higher level routine
+		 */
+		ret = track_pfn_vma_copy(vma);
+		if (ret)
+			return ret;
+	}
+
 	/*
 	 * We need to invalidate the secondary MMU mappings only when
 	 * there could be a permission downgrade on the ptes of the

@@ -915,6 +969,9 @@ unsigned long unmap_vmas(struct mmu_gath
 		if (vma->vm_flags & VM_ACCOUNT)
 			*nr_accounted += (end - start) >> PAGE_SHIFT;
 
+		if (is_pfn_mapping(vma))
+			untrack_pfn_vma(vma, 0, 0);
+
 		while (start != end) {
 			if (!tlb_start_valid) {
 				tlb_start = start;

@@ -1473,6 +1530,7 @@ out:
 int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
 			unsigned long pfn)
 {
+	int ret;
 	/*
 	 * Technically, architectures with pte_special can avoid all these
 	 * restrictions (same for remap_pfn_range).  However we would like

@@ -1487,7 +1545,15 @@ int vm_insert_pfn(struct vm_area_struct 
 
 	if (addr < vma->vm_start || addr >= vma->vm_end)
 		return -EFAULT;
-	return insert_pfn(vma, addr, pfn, vma->vm_page_prot);
+	if (track_pfn_vma_new(vma, vma->vm_page_prot, pfn, PAGE_SIZE))
+		return -EINVAL;
+
+  	ret = insert_pfn(vma, addr, pfn, vma->vm_page_prot);
+
+	if (ret)
+		untrack_pfn_vma(vma, pfn, PAGE_SIZE);
+
+	return ret;
 }
 EXPORT_SYMBOL(vm_insert_pfn);

@@ -1625,6 +1691,10 @@ int remap_pfn_range(struct vm_area_struc
 
 	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
 
+	err = track_pfn_vma_new(vma, prot, pfn, PAGE_ALIGN(size));
+	if (err)
+		return -EINVAL;
+
 	BUG_ON(addr >= end);
 	pfn -= addr >> PAGE_SHIFT;
 	pgd = pgd_offset(mm, addr);

@@ -1636,6 +1706,10 @@ int remap_pfn_range(struct vm_area_struc
 		if (err)
 			break;
 	} while (pgd++, addr = next, addr != end);
+
+	if (err)
+		untrack_pfn_vma(vma, pfn, PAGE_ALIGN(size));
+
 	return err;
 }
 EXPORT_SYMBOL(remap_pfn_range);

Index: linux-2.6/include/linux/mm.h
===================================================================

include/linux/mm.h +6 -0

@@ -155,6 +155,12 @@ static inline int is_pfn_mapping(struct 
 	return (vma->vm_flags & VM_PFNMAP);
 }
 
+extern int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot,
+				unsigned long pfn, unsigned long size);
+extern int track_pfn_vma_copy(struct vm_area_struct *vma);
+extern void untrack_pfn_vma(struct vm_area_struct *vma, unsigned long pfn,
+				unsigned long size);
+
 /*
  * vm_fault is filled by the the pagefault handler and passed to the vma's
  * ->fault function. The vma's ->fault is responsible for returning a bitmask

Signature · 1 line

--

Nick Piggin

2008-12-18 21:38 UTC #12 lore raw

Re: [patch 4/7] x86 PAT: Implement track/untrack of pfnmap regions for x86 - v3

In reply to H. Peter Anvin

And that's the end of the mm/ code... yeah the mm/ parts seem OK to me, with my minor comments addressed I would ack them all.

Quoted 2 lines · On Thu, Dec 18, 2008 at 11:41:30AM -0800, [email protected] wrote:

Hookup remap_pfn_range and vm_insert_pfn and corresponding copy and free routines with reserve and free tracking.

Ingo Molnar

2008-12-18 21:47 UTC #14 lore raw

Re: [patch 4/7] x86 PAT: Implement track/untrack of pfnmap regions for x86 - v3

Quoted 8 lines · H. Peter Anvin

Quoted 2 lines · Nick Piggin
And that's the end of the mm/ code... yeah the mm/ parts seem OK to me, with my minor comments addressed I would ack them all.
Great!
Venki, can this be done with incremental patches, or would that be too messy?

would be nice to have them as deltas. The patches are pushed out into tip/master already - and we'll push it towards linux-next once Nick's concerns have been addressed.

Ingo

H. Peter Anvin

reviewer

2008-12-18 21:48 UTC #13 lore raw

Re: [patch 4/7] x86 PAT: Implement track/untrack of pfnmap regions for x86 - v3

Quoted 2 lines · Nick Piggin

And that's the end of the mm/ code... yeah the mm/ parts seem OK to me, with my minor comments addressed I would ack them all.

Great!

Venki, can this be done with incremental patches, or would that be too messy?

-hpa

Signature · 2 lines

H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.

Pallipadi, Venkatesh

author

2008-12-18 21:49 UTC #20 lore raw

Re: [patch 7/7] x86 PAT: update documentation to cover pgprot and remap_pfn related changes - v3

In reply to Randy Dunlap

Quoted 32 lines · Randy Dunlap

[email protected] wrote:

(quoted diff) +3 -0

+Advanced APIs for drivers
+-------------------------
+A. Exporting pages to user with remap_pfn_range, io_remap_pfn_range,
 
                    to users

or to userspace

(quoted diff) +3 -0

+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace, do it by using mmap

Drop comma.

(quoted diff) +5 -0

+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being added. So, driver can

s/pat/PAT/
s/driver/drivers/ or s/driver/a driver/

(quoted diff) +6 -0

+continue to use the above sequence, with either pgprot_noncached() or
+pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or WC in memtype
+list in order to ensure no conflicting mapping.
+

Review

Quoted 10 lines

(quoted diff) +4 -0

+Note that this set of APIs only work with IO (non RAM) regions. If driver
 
                                 works with IO (non-RAM) regions. If a driver
 
+wants to export RAM region, it has to do set_memory_uc() or set_memory_wc()
 
          export a RAM region,
 
+as step 0 above and also track the usage of those pages and use set_memory_wb()
+before the page is freed to free pool.

refreshed patch below

Thanks,
Venki

Add documentation related to pgprot_* change.

Signed-off-by Venkatesh Pallipadi <[email protected]>

In reply to Pallipadi, Venkatesh

---
 Documentation/x86/pat.txt |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

Documentation/x86/pat.txt +24 -0

Index: linux-2.6/Documentation/x86/pat.txt
===================================================================
@@ -80,6 +80,30 @@ pci proc               |    --    |    -
                        |          |            |                  |
 -------------------------------------------------------------------
 
+Advanced APIs for drivers
+-------------------------
+A. Exporting pages to users with remap_pfn_range, io_remap_pfn_range,
+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace do it by using mmap
+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being added. So, drivers can
+continue to use the above sequence, with either pgprot_noncached() or
+pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or WC in memtype
+list in order to ensure no conflicting mapping.
+
+Note that this set of APIs only work with IO (non RAM) regions. If a driver
+wants to export a RAM region, it has to do set_memory_uc() or set_memory_wc()
+as step 0 above and also track the usage of those pages and use set_memory_wb()
+before the page is freed to free pool.
+
+
+
 Notes:
 
 -- in the above table mean "Not suggested usage for the API". Some of the --'s

Pallipadi, Venkatesh

author

2008-12-18 21:53 UTC #15 lore raw

Re: [patch 4/7] x86 PAT: Implement track/untrack of pfnmap regions for x86 - v3

In reply to Ingo Molnar

Quoted 7 lines · Ingo Molnar

Quoted 4 lines · H. Peter Anvin
Quoted 2 lines · Nick Piggin
And that's the end of the mm/ code... yeah the mm/ parts seem OK to me, with my minor comments addressed I would ack them all.

Quoted 9 lines

Quoted 4 lines
Great!
Venki, can this be done with incremental patches, or would that be too messy?
would be nice to have them as deltas. The patches are pushed out into tip/master already - and we'll push it towards linux-next once Nick's concerns have been addressed.

Yes. I will send incremental patches over v3 addressing Nick's comments.

Thanks,
Venki

Randy Dunlap

2008-12-18 21:54 UTC #21 lore raw

Re: [patch 7/7] x86 PAT: update documentation to cover pgprot and remap_pfn related changes - v3

Quoted 37 lines · Venkatesh

Quoted 36 lines · Randy Dunlap

[email protected] wrote:

Documentation/x86/pat.txt +3 -0

+Advanced APIs for drivers
+-------------------------
+A. Exporting pages to user with remap_pfn_range, io_remap_pfn_range,
 
                    to users

or to userspace

Documentation/x86/pat.txt +3 -0

+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace, do it by using mmap

Drop comma.

Documentation/x86/pat.txt +5 -0

+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being added. So, driver can

s/pat/PAT/
s/driver/drivers/ or s/driver/a driver/

Documentation/x86/pat.txt +8 -0

+continue to use the above sequence, with either pgprot_noncached() or
+pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or WC in memtype
+list in order to ensure no conflicting mapping.
+
+Note that this set of APIs only work with IO (non RAM) regions. If driver
 
                                 works with IO (non-RAM) regions. If a driver
 
+wants to export RAM region, it has to do set_memory_uc() or set_memory_wc()

Quoted 8 lines

Documentation/x86/pat.txt pci proc | -- | - · line 103
Quoted 6 lines
export a RAM region,
Documentation/x86/pat.txt +2 -0
+as step 0 above and also track the usage of those pages and use set_memory_wb()
+before the page is freed to free pool.
refreshed patch below

Some of them were changed, others not changed, with no explanation... More changes below:

Quoted 21 lines

Thanks,
Venki
Add documentation related to pgprot_* change.
Signed-off-by Venkatesh Pallipadi <[email protected]>
Signed-off-by Suresh Siddha <[email protected]>
---
 Documentation/x86/pat.txt |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
Index: linux-2.6/Documentation/x86/pat.txt
===================================================================
Documentation/x86/pat.txt +1 -0
@@ -80,6 +80,30 @@ pci proc               |    --    |    -
                        |          |            |                  |
 -------------------------------------------------------------------
 
+Advanced APIs for drivers

Documentation/x86/pat.txt pci proc | -- | - · line 93

Quoted 10 lines

Documentation/x86/pat.txt +10 -0

+-------------------------
+A. Exporting pages to users with remap_pfn_range, io_remap_pfn_range,
+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace do it by using mmap
+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being added. So, drivers can

PAT (please)

Documentation/x86/pat.txt pci proc | -- | - · line 100

Quoted 7 lines

Documentation/x86/pat.txt +7 -0

+continue to use the above sequence, with either pgprot_noncached() or
+pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or WC in memtype
+list in order to ensure no conflicting mapping.
+
+Note that this set of APIs only work with IO (non RAM) regions. If a driver

only works with IO (non-RAM)

Quoted 3 lines

Documentation/x86/pat.txt +3 -0

+wants to export a RAM region, it has to do set_memory_uc() or set_memory_wc()
+as step 0 above and also track the usage of those pages and use set_memory_wb()
+before the page is freed to free pool.

Signature · 1 line

~Randy

Pallipadi, Venkatesh

author

2008-12-18 22:02 UTC #22 lore raw

RE: [patch 7/7] x86 PAT: update documentation to cover pgprot and remap_pfn related changes - v3

Quoted 64 lines

-----Original Message-----
From: Randy Dunlap [mailto:[email protected]]
Sent: Thursday, December 18, 2008 1:53 PM
To: Pallipadi, Venkatesh
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Siddha, Suresh B
Subject: Re: [patch 7/7] x86 PAT: update documentation to cover pgprot and remap_pfn related changes - v3
Quoted 6 lines · Venkatesh
Documentation/x86/pat.txt pci proc | -- | -
Quoted 5 lines · Randy Dunlap
[email protected] wrote:
Documentation/x86/pat.txt +3 -0
+Advanced APIs for drivers
+-------------------------
+A. Exporting pages to user with remap_pfn_range, 
io_remap_pfn_range,
Quoted 7 lines
Documentation/x86/pat.txt pci proc | -- | -
Quoted 7 lines
to users
or to userspace
Documentation/x86/pat.txt +3 -0
+vm_insert_pfn
+
+Drivers wanting to export some pages to userspace, do it 
by using mmap
Quoted 8 lines
Documentation/x86/pat.txt pci proc | -- | -
Quoted 8 lines
Drop comma.
Documentation/x86/pat.txt +5 -0
+interface and a combination of
+1) pgprot_noncached()
+2) io_remap_pfn_range() or remap_pfn_range() or vm_insert_pfn()
+
+With pat support, a new API pgprot_writecombine is being 
added. So, driver can
Quoted 5 lines
Quoted 5 lines
s/pat/PAT/
s/driver/drivers/ or s/driver/a driver/
+continue to use the above sequence, with either
pgprot_noncached() or
Quoted 3 lines
Documentation/x86/pat.txt pci proc | -- | -
Quoted 3 lines
Documentation/x86/pat.txt +3 -0
+pgprot_writecombine() in step 1, followed by step 2.
+
+In addition, step 2 internally tracks the region as UC or 
WC in memtype
Quoted 3 lines
Documentation/x86/pat.txt pci proc | -- | -
Quoted 3 lines
Documentation/x86/pat.txt +3 -0
+list in order to ensure no conflicting mapping.
+
+Note that this set of APIs only work with IO (non RAM) 
regions. If driver
Quoted 2 lines
Quoted 2 lines
works with IO (non-RAM)
regions. If a driver
Quoted 2 lines
Quoted 2 lines
+wants to export RAM region, it has to do set_memory_uc()
or set_memory_wc()
Quoted 4 lines
Quoted 4 lines
export a RAM region,
+as step 0 above and also track the usage of those pages
and use set_memory_wb()

Quoted 8 lines

Quoted 5 lines
Quoted 3 lines
+before the page is freed to free pool.
refreshed patch below
Some of them were changed, others not changed, with no explanation... More changes below:

Sorry. Not sure how I missed it :( (note to self - do not send patches soon after lunch). Will promise to do all the changes as an incremental patch along with other comments.

Thanks,
Venki

Pallipadi, Venkatesh

author

2008-12-18 22:11 UTC #3 lore raw

Re: [patch 1/7] x86 PAT: store vm_pgoff for all linear_over_vma_region mappings - v3

Quoted 76 lines · Nick Piggin

Quoted 75 lines · On Thu, Dec 18, 2008 at 11:41:27AM -0800, [email protected] wrote:
Drivers use mmap followed by pgprot_* and remap_pfn_range or vm_insert_pfn, in order to export reserved memory to userspace. Currently, such mappings are not tracked and hence not kept consistent with other mappings (/dev/mem, pci resource, ioremap) for the sme memory, that may exist in the system.
The following patchset adds x86 PAT attribute tracking and untracking for pfnmap related APIs.
First three patches in the patchset are changing the generic mm code to fit in this tracking. Last four patches are x86 specific to make things work with x86 PAT code. The patchset aso introduces pgprot_writecombine interface, which gives writecombine mapping when enabled, falling back to pgprot_noncached otherwise.
This patch:
While working on x86 PAT, we faced some hurdles with trackking remap_pfn_range() regions, as we do not have any information to say whether that PFNMAP mapping is linear for the entire vma range or it is smaller granularity regions within the vma.
A simple solution to this is to use vm_pgoff as an indicator for linear mapping over the vma region. Currently, remap_pfn_range only sets vm_pgoff for COW mappings. Below patch changes the logic and sets the vm_pgoff irrespective of COW. This will still not be enough for the case where pfn is zero (vma region mapped to physical address zero). But, for all the other cases, we can look at pfnmap VMAs and say whether the mappng is for the entire vma region or not.
Signed-off-by Venkatesh Pallipadi <[email protected]>
Signed-off-by Suresh Siddha <[email protected]>
---
 include/linux/mm.h |    9 +++++++++
 mm/memory.c        |    7 +++----
 2 files changed, 12 insertions(+), 4 deletions(-)
Index: linux-2.6/mm/memory.c
===================================================================
mm/memory.c +3 -4
@@ -1575,11 +1575,10 @@ int remap_pfn_range(struct vm_area_struc
 	 * behaviour that some programs depend on. We mark the "original"
 	 * un-COW'ed pages by matching them up with "vma->vm_pgoff".
 	 */
-	if (is_cow_mapping(vma->vm_flags)) {
-		if (addr != vma->vm_start || end != vma->vm_end)
-			return -EINVAL;
+	if (addr == vma->vm_start && end == vma->vm_end)
 		vma->vm_pgoff = pfn;
-	}
+	else if (is_cow_mapping(vma->vm_flags))
+		return -EINVAL;
 
 	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
 
Index: linux-2.6/include/linux/mm.h
===================================================================
include/linux/mm.h +9 -0
@@ -145,6 +145,15 @@ extern pgprot_t protection_map[16];
 #define FAULT_FLAG_WRITE	0x01	/* Fault was a write access */
 #define FAULT_FLAG_NONLINEAR	0x02	/* Fault was via a nonlinear mapping */
 
+static inline int is_linear_pfn_mapping(struct vm_area_struct *vma)
+{
+	return ((vma->vm_flags & VM_PFNMAP) && vma->vm_pgoff);
+}
+
+static inline int is_pfn_mapping(struct vm_area_struct *vma)
+{
+	return (vma->vm_flags & VM_PFNMAP);
+}
 

Quoted 6 lines

include/linux/mm.h extern pgprot_t protection_map[16]; · line 159
Quoted 2 lines
include/linux/mm.h +0 -0
 /*
  * vm_fault is filled by the the pagefault handler and passed to the vma's
This is fine by me, however:
1. Can you add some comments to say "this is not for core vm but for pat, oh and a pgoff of zero is not going to work".

OK. Will add comments about both the points.

Quoted 4 lines

2. Can you please justify to me (or the changelog) roughly why PAT wants to know if the mapping is linear or not? Presumably it has to handle both types? If performance wasn't an issue, then you could manually scan the ptes to verify (which would solve your zero-offset bug). etc.

The main reason is performance. If we know it is linear, we can track the entire region as one block and do the reserve free for entire region. But, if it is not linear, then we have to reserve memtype of physical addresses page by page. This will not be optimal as it will result in reserve and free becoming slower. Almost all users that we find in kernel today (atleast in x86) are all linear.

Thanks,
Venki

Pallipadi, Venkatesh

author

2008-12-18 22:15 UTC #7 lore raw

Re: [patch 2/7] x86 PAT: Add follow_pfnmp_pte routine to help tracking pfnmap pages - v3

Quoted 5 lines · Nick Piggin

Quoted 4 lines · On Thu, Dec 18, 2008 at 11:41:28AM -0800, [email protected] wrote:
Add a generic interface to follow pfn in a pfnmap vma range. This is used by one of the subsequent x86 PAT related patch to keep track of memory types for vma regions across vma copy and free.

Quoted 7 lines

Quoted 2 lines
Signed-off-by Venkatesh Pallipadi <[email protected]>
Signed-off-by Suresh Siddha <[email protected]>
Can you please reuse follow_phys for this? (preferably use the same API even if it requires some modification, otherwise if not possible, then at least can you implement a common core for both APIs).

Yes. Hadn't noticed the presence of follow_phys before. Will send a patch to handle this.

Thanks,
Venki

Pallipadi, Venkatesh

author

2008-12-18 22:23 UTC #10 lore raw

Re: [patch 3/7] x86 PAT: hooks in generic vm code to help archs to track pfnmap regions - v3