Re: [PATCH] mm: larger stack guard gap, between vmas

From: Ben Hutchings
Date: Thu Jun 22 2017 - 08:31:33 EST

Here's my attempt at a backport to 3.2. This is only tested on
x86_64 and I think I should introduce local variables for
vma_start_gap() in a few places. I had to cherry-pick commit
09884964335e "mm: do not grow the stack vma just because of an overrun
on preceding vma" before this one (which was a clean cherry-pick).

Ben.

---
From: Hugh Dickins <hughd@xxxxxxxxxx>
Date: Mon, 19 Jun 2017 20:32:47 +0200
Subject: mm: larger stack guard gap, between vmas

commit 1be7107fbe18eed3e319a6c3e83c78254b693acb upstream.

Stack guard page is a useful feature to reduce a risk of stack smashing
into a different mapping. We have been using a single page gap which
is sufficient to prevent having stack adjacent to a different mapping.
But this seems to be insufficient in the light of the stack usage in
userspace. E.g. glibc uses as large as 64kB alloca() in many commonly
used functions. Others use constructs liks gid_t buffer[NGROUPS_MAX]
which is 256kB or stack strings with MAX_ARG_STRLEN.

This will become especially dangerous for suid binaries and the default
no limit for the stack size limit because those applications can be
tricked to consume a large portion of the stack and a single glibc call
could jump over the guard page. These attacks are not theoretical,
unfortunatelly.

Make those attacks less probable by increasing the stack guard gap
to 1MB (on systems with 4k pages; but make it depend on the page size
because systems with larger base pages might cap stack allocations in
the PAGE_SIZE units) which should cover larger alloca() and VLA stack
allocations. It is obviously not a full fix because the problem is
somehow inherent, but it should reduce attack space a lot.

One could argue that the gap size should be configurable from userspace,
but that can be done later when somebody finds that the new 1MB is wrong
for some special case applications. For now, add a kernel command line
option (stack_guard_gap) to specify the stack gap size (in page units).

Implementation wise, first delete all the old code for stack guard page:
because although we could get away with accounting one extra page in a
stack vma, accounting a larger gap can break userspace - case in point,
a program run with "ulimit -S -v 20000" failed when the 1MB gap was
counted for RLIMIT_AS; similar problems could come with RLIMIT_MLOCK
and strict non-overcommit mode.

Instead of keeping gap inside the stack vma, maintain the stack guard
gap as a gap between vmas: using vm_start_gap() in place of vm_start
(or vm_end_gap() in place of vm_end if VM_GROWSUP) in just those few
places which need to respect the gap - mainly arch_get_unmapped_area(),
and and the vma tree's subtree_gap support for that.

Original-patch-by: Oleg Nesterov <oleg@xxxxxxxxxx>
Original-patch-by: Michal Hocko <mhocko@xxxxxxxx>
Signed-off-by: Hugh Dickins <hughd@xxxxxxxxxx>
[wt: backport to 4.11: adjust context]
[wt: backport to 4.9: adjust context ; kernel doc was not in admin-guide]
[wt: backport to 4.4: adjust context ; drop ppc hugetlb_radix changes]
[wt: backport to 3.18: adjust context ; no FOLL_POPULATE ;
s390 uses generic arch_get_unmapped_area()]
[wt: backport to 3.16: adjust context]
[wt: backport to 3.10: adjust context ; code logic in PARISC's
arch_get_unmapped_area() wasn't found ; code inserted into
expand_upwards() and expand_downwards() runs under anon_vma lock;
changes for gup.c:faultin_page go to memory.c:__get_user_pages()]
Signed-off-by: Willy Tarreau <w@xxxxxx>
---
Some of these suggested adjustments below are just what comparing mine
and yours showed up, and I'm being anal in passing them on e.g. I do
like your blank line in mm.h, but Michal chose to leave it out, and
I think that the closer we keep these sources to each other,
the less trouble we shall have patching on top in future.

Hugh.
---
[bwh: Backported to 3.2:
- Drop changes for arc (doesn't exist here), xtensa (doesn't implement
arch_get_unmapped_area() here)
- There's no rb_subtree_gap, so patch the loop in each
arch_get_unmapped_area{,_topdown}()
- Adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -2463,6 +2463,13 @@ bytes respectively. Such letter suffixes
spia_pedr=
spia_peddr=

+ stack_guard_gap= [MM]
+ override the default stack gap protection. The value
+ is in page units and it defines how many pages prior
+ to (for stacks growing down) resp. after (for stacks
+ growing up) the main stack are reserved for no other
+ mapping. Default value is 256 pages.
+
stacktrace [FTRACE]
Enabled the stack tracer on boot up.

--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -101,7 +101,7 @@ arch_get_unmapped_area(struct file *filp

vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}
if (len > mm->cached_hole_size) {
@@ -131,15 +131,15 @@ full_search:
}
return -ENOMEM;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if (!vma || addr + len <= vm_start_gap(vma)) {
/*
* Remember the place where we stopped the search:
*/
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;
addr = vma->vm_end;
if (do_align)
addr = COLOUR_ALIGN(addr, pgoff);
@@ -183,7 +183,7 @@ arch_get_unmapped_area_topdown(struct fi
addr = PAGE_ALIGN(addr);
vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -222,19 +222,19 @@ arch_get_unmapped_area_topdown(struct fi
* return with success:
*/
vma = find_vma(mm, addr);
- if (!vma || addr+len <= vma->vm_start)
+ if (!vma || addr + len <= vm_start_gap(vma))
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr);

/* remember the largest hole we saw so far */
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = vma->vm_start - len;
+ addr = vm_start_gap(vma) - len;
if (do_align)
addr = COLOUR_ALIGN_DOWN(addr, pgoff);
- } while (len < vma->vm_start);
+ } while (len < vm_start_gap(vma));

bottomup:
/*
--- a/arch/frv/mm/elf-fdpic.c
+++ b/arch/frv/mm/elf-fdpic.c
@@ -74,7 +74,7 @@ unsigned long arch_get_unmapped_area(str
addr = PAGE_ALIGN(addr);
vma = find_vma(current->mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
goto success;
}

@@ -89,7 +89,7 @@ unsigned long arch_get_unmapped_area(str
for (; vma; vma = vma->vm_next) {
if (addr > limit)
break;
- if (addr + len <= vma->vm_start)
+ if (addr + len <= vm_start_gap(vma))
goto success;
addr = vma->vm_end;
}
@@ -104,7 +104,7 @@ unsigned long arch_get_unmapped_area(str
for (; vma; vma = vma->vm_next) {
if (addr > limit)
break;
- if (addr + len <= vma->vm_start)
+ if (addr + len <= vm_start_gap(vma))
goto success;
addr = vma->vm_end;
}
--- a/arch/ia64/kernel/sys_ia64.c
+++ b/arch/ia64/kernel/sys_ia64.c
@@ -27,7 +27,7 @@ arch_get_unmapped_area (struct file *fil
long map_shared = (flags & MAP_SHARED);
unsigned long start_addr, align_mask = PAGE_SIZE - 1;
struct mm_struct *mm = current->mm;
- struct vm_area_struct *vma;
+ struct vm_area_struct *vma, *prev;

if (len > RGN_MAP_LIMIT)
return -ENOMEM;
@@ -58,7 +58,7 @@ arch_get_unmapped_area (struct file *fil
full_search:
start_addr = addr = (addr + align_mask) & ~align_mask;

- for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
+ for (vma = find_vma_prev(mm, addr, &prev); ; vma = vma->vm_next) {
/* At this point: (!vma || addr < vma->vm_end). */
if (TASK_SIZE - len < addr || RGN_MAP_LIMIT - len < REGION_OFFSET(addr)) {
if (start_addr != TASK_UNMAPPED_BASE) {
@@ -68,12 +68,13 @@ arch_get_unmapped_area (struct file *fil
}
return -ENOMEM;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if ((!vma || addr + len <= vm_start_gap(vma)) &&
+ (!prev || addr >= vm_end_gap(prev))) {
/* Remember the address where we stopped this search: */
mm->free_area_cache = addr + len;
return addr;
}
- addr = (vma->vm_end + align_mask) & ~align_mask;
+ addr = (vm_end_gap(vma) + align_mask) & ~align_mask;
}
}

--- a/arch/mips/mm/mmap.c
+++ b/arch/mips/mm/mmap.c
@@ -103,7 +103,7 @@ static unsigned long arch_get_unmapped_a

vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -118,7 +118,7 @@ static unsigned long arch_get_unmapped_a
/* At this point: (!vma || addr < vma->vm_end). */
if (TASK_SIZE - len < addr)
return -ENOMEM;
- if (!vma || addr + len <= vma->vm_start)
+ if (!vma || addr + len <= vm_start_gap(vma))
return addr;
addr = vma->vm_end;
if (do_color_align)
@@ -145,7 +145,7 @@ static unsigned long arch_get_unmapped_a
/* make sure it can fit in the remaining address space */
if (likely(addr > len)) {
vma = find_vma(mm, addr - len);
- if (!vma || addr <= vma->vm_start) {
+ if (!vma || addr <= vm_start_gap(vma)) {
/* cache the address as a hint for next time */
return mm->free_area_cache = addr - len;
}
@@ -165,20 +165,20 @@ static unsigned long arch_get_unmapped_a
* return with success:
*/
vma = find_vma(mm, addr);
- if (likely(!vma || addr + len <= vma->vm_start)) {
+ if (likely(!vma || addr + len <= vm_start_gap(vma))) {
/* cache the address as a hint for next time */
return mm->free_area_cache = addr;
}

/* remember the largest hole we saw so far */
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
addr = vma->vm_start - len;
if (do_color_align)
addr = COLOUR_ALIGN_DOWN(addr, pgoff);
- } while (likely(len < vma->vm_start));
+ } while (likely(len < vm_start_gap(vma)));

bottomup:
/*
--- a/arch/parisc/kernel/sys_parisc.c
+++ b/arch/parisc/kernel/sys_parisc.c
@@ -35,17 +35,18 @@

static unsigned long get_unshared_area(unsigned long addr, unsigned long len)
{
- struct vm_area_struct *vma;
+ struct vm_area_struct *vma, *prev;

addr = PAGE_ALIGN(addr);

- for (vma = find_vma(current->mm, addr); ; vma = vma->vm_next) {
+ for (vma = find_vma_prev(current->mm, addr, &prev); ; vma = vma->vm_next) {
/* At this point: (!vma || addr < vma->vm_end). */
if (TASK_SIZE - len < addr)
return -ENOMEM;
- if (!vma || addr + len <= vma->vm_start)
+ if ((!vma || addr + len <= vma->vm_start) &&
+ (!prev || addr >= vm_end_gap(prev)))
return addr;
- addr = vma->vm_end;
+ addr = vm_end_gap(vma);
}
}

@@ -70,21 +71,22 @@ static int get_offset(struct address_spa
static unsigned long get_shared_area(struct address_space *mapping,
unsigned long addr, unsigned long len, unsigned long pgoff)
{
- struct vm_area_struct *vma;
+ struct vm_area_struct *vma, *prev;
int offset = mapping ? get_offset(mapping) : 0;

offset = (offset + (pgoff << PAGE_SHIFT)) & 0x3FF000;

addr = DCACHE_ALIGN(addr - offset) + offset;

- for (vma = find_vma(current->mm, addr); ; vma = vma->vm_next) {
+ for (vma = find_vma_prev(current->mm, addr, &prev); ; vma = vma->vm_next) {
/* At this point: (!vma || addr < vma->vm_end). */
if (TASK_SIZE - len < addr)
return -ENOMEM;
- if (!vma || addr + len <= vma->vm_start)
+ if ((!vma || addr + len <= vma->vm_start) &&
+ (!prev || addr >= vm_end_gap(prev)))
return addr;
- addr = DCACHE_ALIGN(vma->vm_end - offset) + offset;
- if (addr < vma->vm_end) /* handle wraparound */
+ addr = DCACHE_ALIGN(vm_end_gap(vma) - offset) + offset;
+ if (addr < vm_end_gap(vma)) /* handle wraparound */
return -ENOMEM;
}
}
--- a/arch/powerpc/mm/slice.c
+++ b/arch/powerpc/mm/slice.c
@@ -98,7 +98,7 @@ static int slice_area_is_free(struct mm_
if ((mm->task_size - len) < addr)
return 0;
vma = find_vma(mm, addr);
- return (!vma || (addr + len) <= vma->vm_start);
+ return (!vma || (addr + len) <= vm_start_gap(vma));
}

static int slice_low_has_vma(struct mm_struct *mm, unsigned long slice)
@@ -256,7 +256,7 @@ full_search:
addr = _ALIGN_UP(addr + 1, 1ul << SLICE_HIGH_SHIFT);
continue;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if (!vma || addr + len <= vm_start_gap(vma)) {
/*
* Remember the place where we stopped the search:
*/
@@ -264,8 +264,8 @@ full_search:
mm->free_area_cache = addr + len;
return addr;
}
- if (use_cache && (addr + mm->cached_hole_size) < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (use_cache && (addr + mm->cached_hole_size) < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;
addr = vma->vm_end;
}

@@ -336,7 +336,7 @@ static unsigned long slice_find_area_top
* return with success:
*/
vma = find_vma(mm, addr);
- if (!vma || (addr + len) <= vma->vm_start) {
+ if (!vma || (addr + len) <= vm_start_gap(vma)) {
/* remember the address as a hint for next time */
if (use_cache)
mm->free_area_cache = addr;
@@ -344,11 +344,11 @@ static unsigned long slice_find_area_top
}

/* remember the largest hole we saw so far */
- if (use_cache && (addr + mm->cached_hole_size) < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (use_cache && (addr + mm->cached_hole_size) < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = vma->vm_start;
+ addr = vm_start_gap(vma);
}

/*
--- a/arch/sh/mm/mmap.c
+++ b/arch/sh/mm/mmap.c
@@ -75,7 +75,7 @@ unsigned long arch_get_unmapped_area(str

vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -106,15 +106,15 @@ full_search:
}
return -ENOMEM;
}
- if (likely(!vma || addr + len <= vma->vm_start)) {
+ if (likely(!vma || addr + len <= vm_start_gap(vma))) {
/*
* Remember the place where we stopped the search:
*/
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

addr = vma->vm_end;
if (do_colour_align)
@@ -158,7 +158,7 @@ arch_get_unmapped_area_topdown(struct fi

vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -179,7 +179,7 @@ arch_get_unmapped_area_topdown(struct fi
/* make sure it can fit in the remaining address space */
if (likely(addr > len)) {
vma = find_vma(mm, addr-len);
- if (!vma || addr <= vma->vm_start) {
+ if (!vma || addr <= vm_start_gap(vma)) {
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr-len);
}
@@ -199,20 +199,20 @@ arch_get_unmapped_area_topdown(struct fi
* return with success:
*/
vma = find_vma(mm, addr);
- if (likely(!vma || addr+len <= vma->vm_start)) {
+ if (likely(!vma || addr + len <= vm_start_gap(vma))) {
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr);
}

/* remember the largest hole we saw so far */
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = vma->vm_start-len;
+ addr = vm_start_gap(vma) - len;
if (do_colour_align)
addr = COLOUR_ALIGN_DOWN(addr, pgoff);
- } while (likely(len < vma->vm_start));
+ } while (likely(len < vm_start_gap(vma)));

bottomup:
/*
--- a/arch/sparc/kernel/sys_sparc_64.c
+++ b/arch/sparc/kernel/sys_sparc_64.c
@@ -147,7 +147,7 @@ unsigned long arch_get_unmapped_area(str

vma = find_vma(mm, addr);
if (task_size - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -181,15 +181,15 @@ full_search:
}
return -ENOMEM;
}
- if (likely(!vma || addr + len <= vma->vm_start)) {
+ if (likely(!vma || addr + len <= vm_start_gap(vma))) {
/*
* Remember the place where we stopped the search:
*/
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

addr = vma->vm_end;
if (do_color_align)
@@ -237,7 +237,7 @@ arch_get_unmapped_area_topdown(struct fi

vma = find_vma(mm, addr);
if (task_size - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -258,7 +258,7 @@ arch_get_unmapped_area_topdown(struct fi
/* make sure it can fit in the remaining address space */
if (likely(addr > len)) {
vma = find_vma(mm, addr-len);
- if (!vma || addr <= vma->vm_start) {
+ if (!vma || addr <= vm_start_gap(vma)) {
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr-len);
}
@@ -278,20 +278,20 @@ arch_get_unmapped_area_topdown(struct fi
* return with success:
*/
vma = find_vma(mm, addr);
- if (likely(!vma || addr+len <= vma->vm_start)) {
+ if (likely(!vma || addr + len <= vm_start_gap(vma))) {
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr);
}

/* remember the largest hole we saw so far */
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = vma->vm_start-len;
+ addr = vm_start_gap(vma) - len;
if (do_color_align)
addr = COLOUR_ALIGN_DOWN(addr, pgoff);
- } while (likely(len < vma->vm_start));
+ } while (likely(len < vm_start_gap(vma)));

bottomup:
/*
--- a/arch/sparc/mm/hugetlbpage.c
+++ b/arch/sparc/mm/hugetlbpage.c
@@ -67,15 +67,15 @@ full_search:
}
return -ENOMEM;
}
- if (likely(!vma || addr + len <= vma->vm_start)) {
+ if (likely(!vma || addr + len <= vm_start_gap(vma))) {
/*
* Remember the place where we stopped the search:
*/
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

addr = ALIGN(vma->vm_end, HPAGE_SIZE);
}
@@ -182,7 +182,7 @@ hugetlb_get_unmapped_area(struct file *f
addr = ALIGN(addr, HPAGE_SIZE);
vma = find_vma(mm, addr);
if (task_size - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}
if (mm->get_unmapped_area == arch_get_unmapped_area)
--- a/arch/tile/mm/hugetlbpage.c
+++ b/arch/tile/mm/hugetlbpage.c
@@ -185,12 +185,12 @@ full_search:
}
return -ENOMEM;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if (!vma || addr + len <= vm_start_gap(vma)) {
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;
addr = ALIGN(vma->vm_end, huge_page_size(h));
}
}
@@ -236,7 +236,7 @@ try_again:
* new region fits between prev_vma->vm_end and
* vma->vm_start, use it:
*/
- if (addr + len <= vma->vm_start &&
+ if (addr + len <= vm_start_gap(vma) &&
(!prev_vma || (addr >= prev_vma->vm_end))) {
/* remember the address as a hint for next time */
mm->cached_hole_size = largest_hole;
@@ -251,13 +251,13 @@ try_again:
}

/* remember the largest hole we saw so far */
- if (addr + largest_hole < vma->vm_start)
- largest_hole = vma->vm_start - addr;
+ if (addr + largest_hole < vm_start_gap(vma))
+ largest_hole = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = (vma->vm_start - len) & huge_page_mask(h);
+ addr = (vm_start_gap(vma) - len) & huge_page_mask(h);

- } while (len <= vma->vm_start);
+ } while (len <= vm_start_gap(vma));

fail:
/*
@@ -312,7 +312,7 @@ unsigned long hugetlb_get_unmapped_area(
addr = ALIGN(addr, huge_page_size(h));
vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}
if (current->mm->get_unmapped_area == arch_get_unmapped_area)
--- a/arch/x86/kernel/sys_x86_64.c
+++ b/arch/x86/kernel/sys_x86_64.c
@@ -141,7 +141,7 @@ arch_get_unmapped_area(struct file *filp
addr = PAGE_ALIGN(addr);
vma = find_vma(mm, addr);
if (end - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}
if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32))
@@ -172,15 +172,15 @@ full_search:
}
return -ENOMEM;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if (!vma || addr + len <= vm_start_gap(vma)) {
/*
* Remember the place where we stopped the search:
*/
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

addr = vma->vm_end;
addr = align_addr(addr, filp, 0);
@@ -213,7 +213,7 @@ arch_get_unmapped_area_topdown(struct fi
addr = PAGE_ALIGN(addr);
vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -232,7 +232,7 @@ arch_get_unmapped_area_topdown(struct fi
ALIGN_TOPDOWN);

vma = find_vma(mm, tmp_addr);
- if (!vma || tmp_addr + len <= vma->vm_start)
+ if (!vma || tmp_addr + len <= vm_start_gap(vma))
/* remember the address as a hint for next time */
return mm->free_area_cache = tmp_addr;
}
@@ -251,17 +251,17 @@ arch_get_unmapped_area_topdown(struct fi
* return with success:
*/
vma = find_vma(mm, addr);
- if (!vma || addr+len <= vma->vm_start)
+ if (!vma || addr + len <= vm_start_gap(vma))
/* remember the address as a hint for next time */
return mm->free_area_cache = addr;

/* remember the largest hole we saw so far */
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = vma->vm_start-len;
- } while (len < vma->vm_start);
+ addr = vm_start_gap(vma) - len;
+ } while (len < vm_start_gap(vma));

bottomup:
/*
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -303,12 +303,12 @@ full_search:
}
return -ENOMEM;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if (!vma || addr + len <= vm_start_gap(vma)) {
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;
addr = ALIGN(vma->vm_end, huge_page_size(h));
}
}
@@ -351,7 +351,7 @@ try_again:
* new region fits between prev_vma->vm_end and
* vma->vm_start, use it:
*/
- if (addr + len <= vma->vm_start &&
+ if (addr + len <= vm_start_gap(vma) &&
(!prev_vma || (addr >= prev_vma->vm_end))) {
/* remember the address as a hint for next time */
mm->cached_hole_size = largest_hole;
@@ -365,12 +365,12 @@ try_again:
}

/* remember the largest hole we saw so far */
- if (addr + largest_hole < vma->vm_start)
- largest_hole = vma->vm_start - addr;
+ if (addr + largest_hole < vm_start_gap(vma))
+ largest_hole = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = (vma->vm_start - len) & huge_page_mask(h);
- } while (len <= vma->vm_start);
+ addr = (vm_start_gap(vma) - len) & huge_page_mask(h);
+ } while (len <= vm_start_gap(vma));

fail:
/*
@@ -426,7 +426,7 @@ hugetlb_get_unmapped_area(struct file *f
addr = ALIGN(addr, huge_page_size(h));
vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}
if (mm->get_unmapped_area == arch_get_unmapped_area)
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -150,7 +150,7 @@ hugetlb_get_unmapped_area(struct file *f
addr = ALIGN(addr, huge_page_size(h));
vma = find_vma(mm, addr);
if (TASK_SIZE - len >= addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)))
return addr;
}

@@ -176,7 +176,7 @@ full_search:
return -ENOMEM;
}

- if (!vma || addr + len <= vma->vm_start)
+ if (!vma || addr + len <= vm_start_gap(vma))
return addr;
addr = ALIGN(vma->vm_end, huge_page_size(h));
}
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -230,11 +230,7 @@ static void show_map_vma(struct seq_file

/* We don't show the stack guard page in /proc/maps */
start = vma->vm_start;
- if (stack_guard_page_start(vma, start))
- start += PAGE_SIZE;
end = vma->vm_end;
- if (stack_guard_page_end(vma, end))
- end -= PAGE_SIZE;

seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
start,
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1015,34 +1015,6 @@ int set_page_dirty(struct page *page);
int set_page_dirty_lock(struct page *page);
int clear_page_dirty_for_io(struct page *page);

-/* Is the vma a continuation of the stack vma above it? */
-static inline int vma_growsdown(struct vm_area_struct *vma, unsigned long addr)
-{
- return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN);
-}
-
-static inline int stack_guard_page_start(struct vm_area_struct *vma,
- unsigned long addr)
-{
- return (vma->vm_flags & VM_GROWSDOWN) &&
- (vma->vm_start == addr) &&
- !vma_growsdown(vma->vm_prev, addr);
-}
-
-/* Is the vma a continuation of the stack vma below it? */
-static inline int vma_growsup(struct vm_area_struct *vma, unsigned long addr)
-{
- return vma && (vma->vm_start == addr) && (vma->vm_flags & VM_GROWSUP);
-}
-
-static inline int stack_guard_page_end(struct vm_area_struct *vma,
- unsigned long addr)
-{
- return (vma->vm_flags & VM_GROWSUP) &&
- (vma->vm_end == addr) &&
- !vma_growsup(vma->vm_next, addr);
-}
-
extern unsigned long move_page_tables(struct vm_area_struct *vma,
unsigned long old_addr, struct vm_area_struct *new_vma,
unsigned long new_addr, unsigned long len);
@@ -1462,6 +1434,7 @@ unsigned long ra_submit(struct file_ra_s
struct address_space *mapping,
struct file *filp);

+extern unsigned long stack_guard_gap;
/* Generic expand stack which grows the stack according to GROWS{UP,DOWN} */
extern int expand_stack(struct vm_area_struct *vma, unsigned long address);

@@ -1490,6 +1463,30 @@ static inline struct vm_area_struct * fi
return vma;
}

+static inline unsigned long vm_start_gap(struct vm_area_struct *vma)
+{
+ unsigned long vm_start = vma->vm_start;
+
+ if (vma->vm_flags & VM_GROWSDOWN) {
+ vm_start -= stack_guard_gap;
+ if (vm_start > vma->vm_start)
+ vm_start = 0;
+ }
+ return vm_start;
+}
+
+static inline unsigned long vm_end_gap(struct vm_area_struct *vma)
+{
+ unsigned long vm_end = vma->vm_end;
+
+ if (vma->vm_flags & VM_GROWSUP) {
+ vm_end += stack_guard_gap;
+ if (vm_end < vma->vm_end)
+ vm_end = -PAGE_SIZE;
+ }
+ return vm_end;
+}
+
static inline unsigned long vma_pages(struct vm_area_struct *vma)
{
return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1605,12 +1605,6 @@ no_page_table:
return page;
}

-static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr)
-{
- return stack_guard_page_start(vma, addr) ||
- stack_guard_page_end(vma, addr+PAGE_SIZE);
-}
-
/**
* __get_user_pages() - pin user pages in memory
* @tsk: task_struct of target task
@@ -1761,11 +1755,6 @@ int __get_user_pages(struct task_struct
int ret;
unsigned int fault_flags = 0;

- /* For mlock, just skip the stack guard page. */
- if (foll_flags & FOLL_MLOCK) {
- if (stack_guard_page(vma, start))
- goto next_page;
- }
if (foll_flags & FOLL_WRITE)
fault_flags |= FAULT_FLAG_WRITE;
if (nonblocking)
@@ -3122,40 +3111,6 @@ out_release:
}

/*
- * This is like a special single-page "expand_{down|up}wards()",
- * except we must first make sure that 'address{-|+}PAGE_SIZE'
- * doesn't hit another vma.
- */
-static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address)
-{
- address &= PAGE_MASK;
- if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) {
- struct vm_area_struct *prev = vma->vm_prev;
-
- /*
- * Is there a mapping abutting this one below?
- *
- * That's only ok if it's the same stack mapping
- * that has gotten split..
- */
- if (prev && prev->vm_end == address)
- return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM;
-
- return expand_downwards(vma, address - PAGE_SIZE);
- }
- if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) {
- struct vm_area_struct *next = vma->vm_next;
-
- /* As VM_GROWSDOWN but s/below/above/ */
- if (next && next->vm_start == address + PAGE_SIZE)
- return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM;
-
- return expand_upwards(vma, address + PAGE_SIZE);
- }
- return 0;
-}
-
-/*
* We enter with non-exclusive mmap_sem (to exclude vma changes,
* but allow concurrent faults), and pte mapped but not yet locked.
* We return with mmap_sem still held, but pte unmapped and unlocked.
@@ -3174,10 +3129,6 @@ static int do_anonymous_page(struct mm_s
if (vma->vm_flags & VM_SHARED)
return VM_FAULT_SIGBUS;

- /* Check if we need to add a guard page to the stack */
- if (check_stack_guard_page(vma, address) < 0)
- return VM_FAULT_SIGSEGV;
-
/* Use the zero-page for reads */
if (!(flags & FAULT_FLAG_WRITE)) {
entry = pte_mkspecial(pfn_pte(my_zero_pfn(address),
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -245,6 +245,7 @@ SYSCALL_DEFINE1(brk, unsigned long, brk)
unsigned long rlim, retval;
unsigned long newbrk, oldbrk;
struct mm_struct *mm = current->mm;
+ struct vm_area_struct *next;
unsigned long min_brk;

down_write(&mm->mmap_sem);
@@ -289,7 +290,8 @@ SYSCALL_DEFINE1(brk, unsigned long, brk)
}

/* Check against existing mmap mappings. */
- if (find_vma_intersection(mm, oldbrk, newbrk+PAGE_SIZE))
+ next = find_vma(mm, oldbrk);
+ if (next && newbrk + PAGE_SIZE > vm_start_gap(next))
goto out;

/* Ok, looks good - let it rip. */
@@ -1368,7 +1370,7 @@ arch_get_unmapped_area(struct file *filp
unsigned long len, unsigned long pgoff, unsigned long flags)
{
struct mm_struct *mm = current->mm;
- struct vm_area_struct *vma;
+ struct vm_area_struct *vma, *prev;
unsigned long start_addr;

if (len > TASK_SIZE - mmap_min_addr)
@@ -1379,9 +1381,10 @@ arch_get_unmapped_area(struct file *filp

if (addr) {
addr = PAGE_ALIGN(addr);
- vma = find_vma(mm, addr);
+ vma = find_vma_prev(mm, addr, &prev);
if (TASK_SIZE - len >= addr && addr >= mmap_min_addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)) &&
+ (!prev || addr >= vm_end_gap(prev)))
return addr;
}
if (len > mm->cached_hole_size) {
@@ -1407,16 +1410,16 @@ full_search:
}
return -ENOMEM;
}
- if (!vma || addr + len <= vma->vm_start) {
+ if (!vma || addr + len <= vm_start_gap(vma)) {
/*
* Remember the place where we stopped the search:
*/
mm->free_area_cache = addr + len;
return addr;
}
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
- addr = vma->vm_end;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;
+ addr = vm_end_gap(vma);
}
}
#endif
@@ -1442,7 +1445,7 @@ arch_get_unmapped_area_topdown(struct fi
const unsigned long len, const unsigned long pgoff,
const unsigned long flags)
{
- struct vm_area_struct *vma;
+ struct vm_area_struct *vma, *prev;
struct mm_struct *mm = current->mm;
unsigned long addr = addr0;
unsigned long low_limit = max(PAGE_SIZE, mmap_min_addr);
@@ -1457,9 +1460,10 @@ arch_get_unmapped_area_topdown(struct fi
/* requesting a specific address */
if (addr) {
addr = PAGE_ALIGN(addr);
- vma = find_vma(mm, addr);
+ vma = find_vma_prev(mm, addr, &prev);
if (TASK_SIZE - len >= addr && addr >= mmap_min_addr &&
- (!vma || addr + len <= vma->vm_start))
+ (!vma || addr + len <= vm_start_gap(vma)) &&
+ (!prev || addr >= vm_end_gap(prev)))
return addr;
}

@@ -1475,7 +1479,7 @@ arch_get_unmapped_area_topdown(struct fi
/* make sure it can fit in the remaining address space */
if (addr >= low_limit + len) {
vma = find_vma(mm, addr-len);
- if (!vma || addr <= vma->vm_start)
+ if (!vma || addr <= vm_start_gap(vma))
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr-len);
}
@@ -1492,17 +1496,17 @@ arch_get_unmapped_area_topdown(struct fi
* return with success:
*/
vma = find_vma(mm, addr);
- if (!vma || addr+len <= vma->vm_start)
+ if (!vma || addr + len <= vm_start_gap(vma))
/* remember the address as a hint for next time */
return (mm->free_area_cache = addr);

/* remember the largest hole we saw so far */
- if (addr + mm->cached_hole_size < vma->vm_start)
- mm->cached_hole_size = vma->vm_start - addr;
+ if (addr + mm->cached_hole_size < vm_start_gap(vma))
+ mm->cached_hole_size = vm_start_gap(vma) - addr;

/* try just below the current vma->vm_start */
- addr = vma->vm_start-len;
- } while (vma->vm_start >= low_limit + len);
+ addr = vm_start_gap(vma) - len;
+ } while (vm_start_gap(vma) >= low_limit + len);

bottomup:
/*
@@ -1647,21 +1651,19 @@ out:
* update accounting. This is shared with both the
* grow-up and grow-down cases.
*/
-static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow)
+static int acct_stack_growth(struct vm_area_struct *vma,
+ unsigned long size, unsigned long grow)
{
struct mm_struct *mm = vma->vm_mm;
struct rlimit *rlim = current->signal->rlim;
- unsigned long new_start, actual_size;
+ unsigned long new_start;

/* address space limit tests */
if (!may_expand_vm(mm, grow))
return -ENOMEM;

/* Stack limit test */
- actual_size = size;
- if (size && (vma->vm_flags & (VM_GROWSUP | VM_GROWSDOWN)))
- actual_size -= PAGE_SIZE;
- if (actual_size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur))
+ if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur))
return -ENOMEM;

/* mlock limit tests */
@@ -1703,32 +1705,40 @@ static int acct_stack_growth(struct vm_a
*/
int expand_upwards(struct vm_area_struct *vma, unsigned long address)
{
- int error;
+ struct vm_area_struct *next;
+ unsigned long gap_addr;
+ int error = 0;

if (!(vma->vm_flags & VM_GROWSUP))
return -EFAULT;

- /*
- * We must make sure the anon_vma is allocated
- * so that the anon_vma locking is not a noop.
- */
+ /* Guard against wrapping around to address 0. */
+ address &= PAGE_MASK;
+ address += PAGE_SIZE;
+ if (!address)
+ return -ENOMEM;
+
+ /* Enforce stack_guard_gap */
+ gap_addr = address + stack_guard_gap;
+ if (gap_addr < address)
+ return -ENOMEM;
+ next = vma->vm_next;
+ if (next && next->vm_start < gap_addr) {
+ if (!(next->vm_flags & VM_GROWSUP))
+ return -ENOMEM;
+ /* Check that both stack segments have the same anon_vma? */
+ }
+
+ /* We must make sure the anon_vma is allocated. */
if (unlikely(anon_vma_prepare(vma)))
return -ENOMEM;
- vma_lock_anon_vma(vma);

/*
* vma->vm_start/vm_end cannot change under us because the caller
* is required to hold the mmap_sem in read mode. We need the
* anon_vma lock to serialize against concurrent expand_stacks.
- * Also guard against wrapping around to address 0.
*/
- if (address < PAGE_ALIGN(address+4))
- address = PAGE_ALIGN(address+4);
- else {
- vma_unlock_anon_vma(vma);
- return -ENOMEM;
- }
- error = 0;
+ vma_lock_anon_vma(vma);

/* Somebody else might have raced and expanded it already */
if (address > vma->vm_end) {
@@ -1758,27 +1768,36 @@ int expand_upwards(struct vm_area_struct
int expand_downwards(struct vm_area_struct *vma,
unsigned long address)
{
+ struct vm_area_struct *prev;
+ unsigned long gap_addr;
int error;

- /*
- * We must make sure the anon_vma is allocated
- * so that the anon_vma locking is not a noop.
- */
- if (unlikely(anon_vma_prepare(vma)))
- return -ENOMEM;
-
address &= PAGE_MASK;
error = security_file_mmap(NULL, 0, 0, 0, address, 1);
if (error)
return error;

- vma_lock_anon_vma(vma);
+ /* Enforce stack_guard_gap */
+ gap_addr = address - stack_guard_gap;
+ if (gap_addr > address)
+ return -ENOMEM;
+ prev = vma->vm_prev;
+ if (prev && prev->vm_end > gap_addr) {
+ if (!(prev->vm_flags & VM_GROWSDOWN))
+ return -ENOMEM;
+ /* Check that both stack segments have the same anon_vma? */
+ }
+
+ /* We must make sure the anon_vma is allocated. */
+ if (unlikely(anon_vma_prepare(vma)))
+ return -ENOMEM;

/*
* vma->vm_start/vm_end cannot change under us because the caller
* is required to hold the mmap_sem in read mode. We need the
* anon_vma lock to serialize against concurrent expand_stacks.
*/
+ vma_lock_anon_vma(vma);

/* Somebody else might have raced and expanded it already */
if (address < vma->vm_start) {
@@ -1802,28 +1821,25 @@ int expand_downwards(struct vm_area_stru
return error;
}

-/*
- * Note how expand_stack() refuses to expand the stack all the way to
- * abut the next virtual mapping, *unless* that mapping itself is also
- * a stack mapping. We want to leave room for a guard page, after all
- * (the guard page itself is not added here, that is done by the
- * actual page faulting logic)
- *
- * This matches the behavior of the guard page logic (see mm/memory.c:
- * check_stack_guard_page()), which only allows the guard page to be
- * removed under these circumstances.
- */
+/* enforced gap between the expanding stack and other mappings. */
+unsigned long stack_guard_gap = 256UL<<PAGE_SHIFT;
+
+static int __init cmdline_parse_stack_guard_gap(char *p)
+{
+ unsigned long val;
+ char *endptr;
+
+ val = simple_strtoul(p, &endptr, 10);
+ if (!*endptr)
+ stack_guard_gap = val << PAGE_SHIFT;
+
+ return 0;
+}
+__setup("stack_guard_gap=", cmdline_parse_stack_guard_gap);
+
#ifdef CONFIG_STACK_GROWSUP
int expand_stack(struct vm_area_struct *vma, unsigned long address)
{
- struct vm_area_struct *next;
-
- address &= PAGE_MASK;
- next = vma->vm_next;
- if (next && next->vm_start == address + PAGE_SIZE) {
- if (!(next->vm_flags & VM_GROWSUP))
- return -ENOMEM;
- }
return expand_upwards(vma, address);
}

@@ -1846,14 +1862,6 @@ find_extend_vma(struct mm_struct *mm, un
#else
int expand_stack(struct vm_area_struct *vma, unsigned long address)
{
- struct vm_area_struct *prev;
-
- address &= PAGE_MASK;
- prev = vma->vm_prev;
- if (prev && prev->vm_end == address) {
- if (!(prev->vm_flags & VM_GROWSDOWN))
- return -ENOMEM;
- }
return expand_downwards(vma, address);
}


--
Ben Hutchings
Sturgeon's Law: Ninety percent of everything is crap.

Attachment: signature.asc
Description: Digital signature