[Bug 1709727] Re: asan causes hangs on arm64

[Matthias Klose]

** Description changed:

  [Impact]
  Binaries built w/ asan support hang on arm64. This causes many timeouts during the test phase of a gcc build, causing builds to need ~24 hours to complete.
  
  [Test Case]
  $ cat test.c
  #include <stdio.h>
  
  void main() {
-   printf("hi.\n");
+   printf("hi.\n");
  }
  $ gcc test.c -g -fsanitize=address -fno-omit-frame-pointer test.c -o test
  $ ./test
  ==46644==AddressSanitizer CHECK failed: ../../../../src/libsanitizer/sanitizer_common/sanitizer_allocator.h:844 "((res)) < ((kNumPossibleRegions))" (0xffffb4b, 0x8000000)
  [... HANG ...]
  
+ Plus see the tests in the libasan testsuite succeed.
+ 
  [Regression Risk]
+ None, unless the 48bit-vma support is removed again from the kernel. Can't find the corresponding kernel SRU which explained the regression potential in this kernel change.

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to gcc-5 in Ubuntu.
https://bugs.launchpad.net/bugs/1709727

Title:
  asan causes hangs on arm64

Status in gcc-5 package in Ubuntu:
  Fix Released
Status in gcc-6 package in Ubuntu:
  Fix Released
Status in gcc-5 source package in Xenial:
  Triaged
Status in gcc-6 source package in Xenial:
  Invalid
Status in gcc-5 source package in Zesty:
  Triaged
Status in gcc-6 source package in Zesty:
  Triaged

Bug description:
  [Impact]
  Binaries built w/ asan support hang on arm64. This causes many timeouts during the test phase of a gcc build, causing builds to need ~24 hours to complete.

  [Test Case]
  $ cat test.c
  #include <stdio.h>

  void main() {
    printf("hi.\n");
  }
  $ gcc test.c -g -fsanitize=address -fno-omit-frame-pointer test.c -o test
  $ ./test
  ==46644==AddressSanitizer CHECK failed: ../../../../src/libsanitizer/sanitizer_common/sanitizer_allocator.h:844 "((res)) < ((kNumPossibleRegions))" (0xffffb4b, 0x8000000)
  [... HANG ...]

  Plus see the tests in the libasan testsuite succeed.

  [Regression Risk]
  None, unless the 48bit-vma support is removed again from the kernel. Can't find the corresponding kernel SRU which explained the regression potential in this kernel change.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/gcc-5/+bug/1709727/+subscriptions