Daniel Dunbar
84e383fc42
Don't allow formatted_ostream to be unbuffered, even if its underlying buffer
...
is.
- The problem is that formatted_ostream forces its underlying buffer to be
unbuffered, so if some client happens to wrap a formatted_ostream around
something, but still use the underlying stream, then we can end up writing on
a fully unbuffered output (which was never intended to be unbuffered).
- This makes clang (and presumably llvm-gcc) -emit-llvm -S a mere 10x faster.
llvm-svn: 82434
2009-09-21 03:56:00 +00:00
..
2009-02-20 22:51:36 +00:00
2009-07-23 18:34:13 +00:00
2009-09-03 02:02:59 +00:00
2009-08-23 07:19:13 +00:00
2009-08-27 23:44:33 +00:00
2009-09-20 06:21:43 +00:00
2009-09-17 23:27:10 +00:00
2009-09-03 23:34:49 +00:00
2009-08-23 06:03:38 +00:00
2009-04-30 17:09:22 +00:00
2009-08-10 19:44:53 +00:00
2009-08-10 19:44:53 +00:00
2009-08-23 08:50:52 +00:00
2009-08-28 23:24:31 +00:00
2009-06-24 17:37:09 +00:00
2009-08-21 21:03:57 +00:00
2009-05-30 21:05:47 +00:00
2009-08-20 01:46:24 +00:00
2009-02-20 22:51:36 +00:00
2009-08-23 20:31:39 +00:00
2009-09-21 03:56:00 +00:00
2009-08-27 23:44:33 +00:00
2009-09-20 04:03:25 +00:00
2009-07-15 22:43:51 +00:00
2009-09-16 18:16:11 +00:00
2009-09-08 22:20:35 +00:00
2009-08-19 17:07:46 +00:00
2009-07-09 17:26:16 +00:00
2009-09-18 16:57:42 +00:00
2009-08-28 05:48:04 +00:00
2009-08-24 09:29:24 +00:00
2009-09-09 22:49:13 +00:00
2009-09-03 23:34:49 +00:00
2009-06-07 21:22:38 +00:00
2009-08-23 06:03:38 +00:00
2009-09-06 12:56:52 +00:00
2009-05-01 23:48:33 +00:00
2009-04-22 08:46:33 +00:00
2009-07-16 06:17:45 +00:00
2009-08-24 04:02:06 +00:00
2009-09-15 20:31:46 +00:00
2009-09-06 08:55:57 +00:00
2009-08-30 08:24:09 +00:00
2009-09-06 07:23:28 +00:00
2009-08-11 17:49:14 +00:00
2009-09-18 22:03:29 +00:00
2009-07-23 18:17:34 +00:00
2009-08-23 21:36:09 +00:00
2009-09-03 23:34:49 +00:00
2009-09-15 15:08:33 +00:00
2009-08-12 22:10:57 +00:00
2009-09-09 05:04:01 +00:00
2009-09-19 20:40:05 +00:00