Rafael Espindola b4b0b34181 Write relocation sections contiguously.
Linkers normally read all the relocations upfront to compute the references
between sections. Putting them together is a bit more cache friendly.

I benchmarked linking a Release+Asserts clang with gold on a vm. I tried all
4 combinations of --gc-sections/no --gc-section hot and cold cache.

I cleared the cache with

echo 3 > /proc/sys/vm/drop_caches

and warmed it up by running the link once before timing the subsequent ones.

With cold cache and --gc-sections the time goes from

1.86130781665 +- 0.01713126697463843 seconds
to
1.82370735105 +- 0.014127522318814516 seconds

With cold cache and no --gc-sections the time goes from

1.6087245435500002 +- 0.012999066825178644 seconds
to
1.5687122041500001 +- 0.013145850126026619 seconds

With hot cache and no --gc-sections the time goes from

0.926200939 ( +-  0.33% ) seconds
to
0.907200079 ( +-  0.31% ) seconds

With hot cache and gc sections the time goes from

1.183038049 ( +-  0.34% ) seconds
to
1.147355862 ( +-  0.39% ) seconds

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235165 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-17 08:11:38 +00:00
..
2013-10-29 09:47:35 +00:00
2014-11-05 12:40:21 +00:00
2014-06-19 16:35:19 +00:00
2014-01-30 04:46:24 +00:00
2014-05-05 17:58:46 +00:00
2014-01-24 17:20:08 +00:00
2014-01-28 23:13:30 +00:00
2013-09-18 09:46:49 +00:00
2014-04-03 11:29:15 +00:00
2013-08-28 16:39:20 +00:00
2013-08-28 16:39:20 +00:00
2014-01-17 13:53:08 +00:00
2014-04-16 16:45:18 +00:00
2015-04-15 17:28:23 +00:00
2013-06-24 09:11:53 +00:00