Commit e69e9d4a authored by HATAYAMA Daisuke's avatar HATAYAMA Daisuke Committed by Linus Torvalds

vmalloc: introduce remap_vmalloc_range_partial

We want to allocate ELF note segment buffer on the 2nd kernel in vmalloc
space and remap it to user-space in order to reduce the risk that memory
allocation fails on system with huge number of CPUs and so with huge ELF
note segment that exceeds 11-order block size.

Although there's already remap_vmalloc_range for the purpose of
remapping vmalloc memory to user-space, we need to specify user-space
range via vma.
 Mmap on /proc/vmcore needs to remap range across multiple objects, so
the interface that requires vma to cover full range is problematic.

This patch introduces remap_vmalloc_range_partial that receives user-space
range as a pair of base address and size and can be used for mmap on
/proc/vmcore case.

remap_vmalloc_range is rewritten using remap_vmalloc_range_partial.

[akpm@linux-foundation.org: use PAGE_ALIGNED()]
Signed-off-by: default avatarHATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Atsushi Kumagai <kumagai-atsushi@mxc.nes.nec.co.jp>
Cc: Lisa Mitchell <lisa.mitchell@hp.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent cef2ac3f
...@@ -82,6 +82,10 @@ extern void *vmap(struct page **pages, unsigned int count, ...@@ -82,6 +82,10 @@ extern void *vmap(struct page **pages, unsigned int count,
unsigned long flags, pgprot_t prot); unsigned long flags, pgprot_t prot);
extern void vunmap(const void *addr); extern void vunmap(const void *addr);
extern int remap_vmalloc_range_partial(struct vm_area_struct *vma,
unsigned long uaddr, void *kaddr,
unsigned long size);
extern int remap_vmalloc_range(struct vm_area_struct *vma, void *addr, extern int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
unsigned long pgoff); unsigned long pgoff);
void vmalloc_sync_all(void); void vmalloc_sync_all(void);
......
...@@ -1476,10 +1476,9 @@ static void __vunmap(const void *addr, int deallocate_pages) ...@@ -1476,10 +1476,9 @@ static void __vunmap(const void *addr, int deallocate_pages)
if (!addr) if (!addr)
return; return;
if ((PAGE_SIZE-1) & (unsigned long)addr) { if (WARN(!PAGE_ALIGNED(addr), "Trying to vfree() bad address (%p)\n",
WARN(1, KERN_ERR "Trying to vfree() bad address (%p)\n", addr); addr));
return; return;
}
area = remove_vm_area(addr); area = remove_vm_area(addr);
if (unlikely(!area)) { if (unlikely(!area)) {
...@@ -2148,42 +2147,43 @@ long vwrite(char *buf, char *addr, unsigned long count) ...@@ -2148,42 +2147,43 @@ long vwrite(char *buf, char *addr, unsigned long count)
} }
/** /**
* remap_vmalloc_range - map vmalloc pages to userspace * remap_vmalloc_range_partial - map vmalloc pages to userspace
* @vma: vma to cover (map full range of vma) * @vma: vma to cover
* @addr: vmalloc memory * @uaddr: target user address to start at
* @pgoff: number of pages into addr before first page to map * @kaddr: virtual address of vmalloc kernel memory
* @size: size of map area
* *
* Returns: 0 for success, -Exxx on failure * Returns: 0 for success, -Exxx on failure
* *
* This function checks that addr is a valid vmalloc'ed area, and * This function checks that @kaddr is a valid vmalloc'ed area,
* that it is big enough to cover the vma. Will return failure if * and that it is big enough to cover the range starting at
* that criteria isn't met. * @uaddr in @vma. Will return failure if that criteria isn't
* met.
* *
* Similar to remap_pfn_range() (see mm/memory.c) * Similar to remap_pfn_range() (see mm/memory.c)
*/ */
int remap_vmalloc_range(struct vm_area_struct *vma, void *addr, int remap_vmalloc_range_partial(struct vm_area_struct *vma, unsigned long uaddr,
unsigned long pgoff) void *kaddr, unsigned long size)
{ {
struct vm_struct *area; struct vm_struct *area;
unsigned long uaddr = vma->vm_start;
unsigned long usize = vma->vm_end - vma->vm_start;
if ((PAGE_SIZE-1) & (unsigned long)addr) size = PAGE_ALIGN(size);
if (!PAGE_ALIGNED(uaddr) || !PAGE_ALIGNED(kaddr))
return -EINVAL; return -EINVAL;
area = find_vm_area(addr); area = find_vm_area(kaddr);
if (!area) if (!area)
return -EINVAL; return -EINVAL;
if (!(area->flags & VM_USERMAP)) if (!(area->flags & VM_USERMAP))
return -EINVAL; return -EINVAL;
if (usize + (pgoff << PAGE_SHIFT) > area->size - PAGE_SIZE) if (kaddr + size > area->addr + area->size)
return -EINVAL; return -EINVAL;
addr += pgoff << PAGE_SHIFT;
do { do {
struct page *page = vmalloc_to_page(addr); struct page *page = vmalloc_to_page(kaddr);
int ret; int ret;
ret = vm_insert_page(vma, uaddr, page); ret = vm_insert_page(vma, uaddr, page);
...@@ -2191,14 +2191,37 @@ int remap_vmalloc_range(struct vm_area_struct *vma, void *addr, ...@@ -2191,14 +2191,37 @@ int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
return ret; return ret;
uaddr += PAGE_SIZE; uaddr += PAGE_SIZE;
addr += PAGE_SIZE; kaddr += PAGE_SIZE;
usize -= PAGE_SIZE; size -= PAGE_SIZE;
} while (usize > 0); } while (size > 0);
vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP; vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP;
return 0; return 0;
} }
EXPORT_SYMBOL(remap_vmalloc_range_partial);
/**
* remap_vmalloc_range - map vmalloc pages to userspace
* @vma: vma to cover (map full range of vma)
* @addr: vmalloc memory
* @pgoff: number of pages into addr before first page to map
*
* Returns: 0 for success, -Exxx on failure
*
* This function checks that addr is a valid vmalloc'ed area, and
* that it is big enough to cover the vma. Will return failure if
* that criteria isn't met.
*
* Similar to remap_pfn_range() (see mm/memory.c)
*/
int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
unsigned long pgoff)
{
return remap_vmalloc_range_partial(vma, vma->vm_start,
addr + (pgoff << PAGE_SHIFT),
vma->vm_end - vma->vm_start);
}
EXPORT_SYMBOL(remap_vmalloc_range); EXPORT_SYMBOL(remap_vmalloc_range);
/* /*
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment