diff options
| author | jdduke <jdduke@chromium.org> | 2015-05-05 00:34:05 -0700 |
|---|---|---|
| committer | Commit bot <commit-bot@chromium.org> | 2015-05-05 07:34:51 +0000 |
| commit | 9f355f21c4049c622c71fc66a50e0288c5211335 (patch) | |
| tree | b6d24d34e4c2a71d5502c110d17899964cfcd3a3 /base/strings/string_number_conversions_unittest.cc | |
| parent | 3e8bda613486ae81f53428ab9f41a7dbb7f9f0c3 (diff) | |
| download | chromium_src-9f355f21c4049c622c71fc66a50e0288c5211335.zip chromium_src-9f355f21c4049c622c71fc66a50e0288c5211335.tar.gz chromium_src-9f355f21c4049c622c71fc66a50e0288c5211335.tar.bz2 | |
Reland "Enable libc++ on Android"
Switch all Android builds to use libc++ instead of stlport.
This change originally landed in crrev.com/315085, but was reverted in
crrev.com/315174 due to CQ execution time regressions.
The root cause of the test time regression was determined to be an issue
with MD5 hash computation, where hashing failed to produce a consistent
result in turn preventing deployment optimizations. This was resolved in
crbug.com/456396, leaving libc++ test runs comparable with stlport runs.
BUG=456396,427718
Review URL: https://codereview.chromium.org/951983002
Cr-Commit-Position: refs/heads/master@{#328296}
Diffstat (limited to 'base/strings/string_number_conversions_unittest.cc')
0 files changed, 0 insertions, 0 deletions