implement adaptive redzones

[GoogleCodeExporter]

When instrumenting global arrays, we add fixed-size redzones (32 bytes) to them.
If the array has two or more dimensions, we're likely to miss the redzone even 
if the array index is off by one:

char arr[40][40];
char arr2[40];

arr[40][0] = 'a';  // this will hit arr2

We may want to vary the size for two-dimensional arrays depending on their 
sizes.

Original issue reported on code.google.com by ramosian.glider@gmail.com on 14 Oct 2011 at 6:51

[GoogleCodeExporter]

Original comment by ramosian.glider@gmail.com on 14 Oct 2011 at 6:51

[GoogleCodeExporter]

We may want to implement adaptive redzones for globals and for heap. 
Maybe even for stack. 
The hypothesis is that large objects tend to be overflown by larger amounts. 
I don't think this should have anything to do with two-dimensional arrays. 

Original comment by konstant...@gmail.com on 31 Oct 2011 at 5:44

• Changed title: _implement adaptive redzones _

[GoogleCodeExporter]

Original comment by ramosian.glider@gmail.com on 2 Feb 2012 at 12:14

• Added labels: Type-Enhancement
• Removed labels: Type-Defect

[GoogleCodeExporter]

- adaptive redzones for globals are committed as LLVM r173335/r173336
- adaptive redzones for heap are a part of the new asan allocator

Need to think what to do with stack. 

Original comment by konstant...@gmail.com on 24 Jan 2013 at 10:38

[GoogleCodeExporter]

Original comment by ramosian.glider@gmail.com on 7 Feb 2013 at 4:02

[GoogleCodeExporter]

A somewhat related problem: http://llvm.org/bugs/show_bug.cgi?id=16277
(can't handle stack objects aligned by more than 32 bytes).
This may potentially be solved together with adaptive redzones for stack. 

Original comment by konstant...@gmail.com on 26 Jun 2013 at 9:56

[GoogleCodeExporter]

Fixed, see https://code.google.com/p/address-sanitizer/issues/detail?id=245#c2

Original comment by konstant...@gmail.com on 6 Dec 2013 at 9:12

• Changed state: Fixed