[PATCH] D63793: Treat the range of representable values of floating-point types as [-inf, +inf] not as [-max, +max].

[Oliver Stannard (Linaro) via Phabricator via cfe-commits]

ostannard added a comment.

There are a number of buildbot failures which look related to this, e.g.

- http://lab.llvm.org:8011/builders/clang-cmake-armv7-full/builds/6730/steps/ninja%20check%201/logs/FAIL%3A%20UBSan-AddressSanitizer-armhf%3A%3Adiv-zero.cpp
- http://lab.llvm.org:8011/builders/clang-cmake-armv7-full/builds/6730/steps/ninja%20check%201/logs/FAIL%3A%20UBSan-AddressSanitizer-armhf%3A%3Adiv-zero.cpp

I think that removing the float-divide-by-zero sanitizer from -fsanitize=undefined has also caused it to be marked as not supported, so it can't be enabled even explicitly.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63793/new/

https://reviews.llvm.org/D63793