Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass StringNumberConversionsTest with msan.

Committed: https://crrev.com/9772d1365b72e15c5a393bad8948d62681145119
Cr-Commit-Position: refs/heads/master@{#420825}

[kcwu, 2016-09-23 18:10:07]

The CQ bit was checked by kcwu@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-09-23 18:10:37]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[kcwu, 2016-09-23 18:11:28]

Description was changed from

==========
Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass
base_unittests StringNumberConversionsTest with msan.
==========

to

==========
Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass
StringNumberConversionsTest with msan.
==========

[scottmg, 2016-09-23 18:12:39]

scottmg@chromium.org changed reviewers:
+ scottmg@chromium.org

[scottmg, 2016-09-23 18:12:40]

Thanks for updating this!

https://codereview.chromium.org/2364123002/diff/1/base/strings/string_number_...
File base/strings/string_number_conversions_unittest.cc (right):

https://codereview.chromium.org/2364123002/diff/1/base/strings/string_number_...
base/strings/string_number_conversions_unittest.cc:904: for (size_t i = 0; i <
arraysize(cases); ++i) {
Since there's only one case, it would be clearer to just have this call the
functions and EXPECT directly without a loop. And perhaps rename the test to
something like StringToDoubleHangBug588726.

[commit-bot: I haz the power, 2016-09-23 18:23:31]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-09-23 18:23:32]

Dry run: Try jobs failed on following builders:
  win_chromium_rel_ng on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[kcwu, 2016-09-23 18:45:46]

The CQ bit was checked by kcwu@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-09-23 18:46:32]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-23 19:12:54]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-09-23 19:12:54]

Dry run: This issue passed the CQ dry run.

[kcwu, 2016-09-23 19:23:05]

kcwu@chromium.org changed reviewers:
+ thestig@chromium.org

[kcwu, 2016-09-23 19:23:06]

PTAL. 

I have no knowledge about the algorithm of dtoa.c. I just blindly merge and
don't know if I break anything.

I have tested base_unittests. string_to_int_fuzzer is running and looks good (so
far, only minutes).

https://codereview.chromium.org/2364123002/diff/1/base/strings/string_number_...
File base/strings/string_number_conversions_unittest.cc (right):

https://codereview.chromium.org/2364123002/diff/1/base/strings/string_number_...
base/strings/string_number_conversions_unittest.cc:904: for (size_t i = 0; i <
arraysize(cases); ++i) {
On 2016/09/23 18:12:40, scottmg wrote:
> Since there's only one case, it would be clearer to just have this call the
> functions and EXPECT directly without a loop. And perhaps rename the test to
> something like StringToDoubleHangBug588726.

Done.

[Lei Zhang, 2016-09-23 19:32:17]

https://codereview.chromium.org/2364123002/diff/20001/base/strings/string_num...
File base/strings/string_number_conversions_unittest.cc (right):

https://codereview.chromium.org/2364123002/diff/20001/base/strings/string_num...
base/strings/string_number_conversions_unittest.cc:893:
TEST(StringNumberConversionsTest, StringToDoubleHangBug588726) {
Just add it to StringNumberConversionsTest.StringToDouble above?

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
File base/third_party/dmg_fp/dtoa.cc (right):

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
base/third_party/dmg_fp/dtoa.cc:1536: static unsigned char hexdig[256] = {
Should this be const?

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
base/third_party/dmg_fp/dtoa.cc:1601: /**** if (!hexdig['0']) hexdig_init();
****/
Hmm

[kcwu, 2016-09-23 19:48:03]

https://codereview.chromium.org/2364123002/diff/20001/base/strings/string_num...
File base/strings/string_number_conversions_unittest.cc (right):

https://codereview.chromium.org/2364123002/diff/20001/base/strings/string_num...
base/strings/string_number_conversions_unittest.cc:893:
TEST(StringNumberConversionsTest, StringToDoubleHangBug588726) {
On 2016/09/23 19:32:16, Lei Zhang wrote:
> Just add it to StringNumberConversionsTest.StringToDouble above?

Done.

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
File base/third_party/dmg_fp/dtoa.cc (right):

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
base/third_party/dmg_fp/dtoa.cc:1536: static unsigned char hexdig[256] = {
On 2016/09/23 19:32:16, Lei Zhang wrote:
> Should this be const?

Done.

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
base/third_party/dmg_fp/dtoa.cc:1601: /**** if (!hexdig['0']) hexdig_init();
****/
On 2016/09/23 19:32:16, Lei Zhang wrote:
> Hmm

This is copied from upstream. I'd like to keep it in order to make future merge
easier.

[Lei Zhang, 2016-09-23 20:33:12]

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
File base/third_party/dmg_fp/dtoa.cc (right):

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
base/third_party/dmg_fp/dtoa.cc:1536: static unsigned char hexdig[256] = {
On 2016/09/23 19:48:03, kcwu wrote:
> On 2016/09/23 19:32:16, Lei Zhang wrote:
> > Should this be const?
> 
> Done.

Thanks, but now we need another .patch file.

[kcwu, 2016-09-24 03:17:07]

Patchset #4 (id:60001) has been deleted

[kcwu, 2016-09-24 03:18:01]

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
File base/third_party/dmg_fp/dtoa.cc (right):

https://codereview.chromium.org/2364123002/diff/20001/base/third_party/dmg_fp...
base/third_party/dmg_fp/dtoa.cc:1536: static unsigned char hexdig[256] = {
On 2016/09/23 20:33:12, Lei Zhang wrote:
> On 2016/09/23 19:48:03, kcwu wrote:
> > On 2016/09/23 19:32:16, Lei Zhang wrote:
> > > Should this be const?
> > 
> > Done.
> 
> Thanks, but now we need another .patch file.

Done.

[Lei Zhang, 2016-09-24 03:42:20]

lgtm

[kcwu, 2016-09-24 04:05:59]

The CQ bit was checked by kcwu@chromium.org

[commit-bot: I haz the power, 2016-09-24 04:06:13]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-24 04:24:05]

Description was changed from

==========
Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass
StringNumberConversionsTest with msan.
==========

to

==========
Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass
StringNumberConversionsTest with msan.
==========

[commit-bot: I haz the power, 2016-09-24 04:24:06]

Committed patchset #4 (id:80001)

[commit-bot: I haz the power, 2016-09-24 04:26:54]

Description was changed from

==========
Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass
StringNumberConversionsTest with msan.
==========

to

==========
Merge latest changes from http://www.netlib.org/fp/dtoa.c into dtoa.cc

This fixed crbug.com/588726 infinite loop issue and also eliminated
previous local patch for crbug.com/123157 crash issue.

BUG=588726
TEST=base_unittests: pass all cases with asan and ubsan_security; pass
StringNumberConversionsTest with msan.

Committed: https://crrev.com/9772d1365b72e15c5a393bad8948d62681145119
Cr-Commit-Position: refs/heads/master@{#420825}
==========

[commit-bot: I haz the power, 2016-09-24 04:26:55]

Patchset 4 (id:??) landed as
https://crrev.com/9772d1365b72e15c5a393bad8948d62681145119
Cr-Commit-Position: refs/heads/master@{#420825}