Files
archived-llvm/test/CodeGen/ARM
James Molloy bd7c3fb3bf [LSR] Don't try and create post-inc expressions on non-rotated loops
If a loop is not rotated (for example when optimizing for size), the latch is not the backedge. If we promote an expression to post-inc form, we not only increase register pressure and add a COPY for that IV expression but for all IVs!

Motivating testcase:

    void f(float *a, float *b, float *c, int n) {
      while (n-- > 0)
        *c++ = *a++ + *b++;
    }

It's imperative that the pointer increments be located in the latch block and not the header block; if not, we cannot use post-increment loads and stores and we have to keep both the post-inc and pre-inc values around until the end of the latch which bloats register usage.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@278658 91177308-0d34-0410-b5e6-96231b3b80d8
2016-08-15 07:53:03 +00:00
..
2016-01-26 00:03:25 +00:00
2014-04-03 17:35:22 +00:00
2016-06-16 16:09:53 +00:00
2016-06-16 16:09:53 +00:00
2016-01-28 18:59:04 +00:00
2016-05-19 12:59:17 +00:00
2016-05-19 12:59:17 +00:00
2014-02-16 07:31:05 +00:00
2014-08-26 12:47:26 +00:00
2014-01-29 11:50:56 +00:00
2014-01-29 11:50:56 +00:00
2016-06-16 16:09:53 +00:00
2015-10-26 20:49:49 +00:00
2016-06-20 17:45:33 +00:00
2016-01-26 00:03:25 +00:00
2016-04-08 18:15:37 +00:00
2014-04-25 17:51:25 +00:00
2014-11-17 14:08:57 +00:00
2016-07-25 10:11:00 +00:00
2014-04-03 17:35:22 +00:00
2014-12-04 19:34:50 +00:00
2016-06-16 16:09:53 +00:00
2016-03-31 19:42:04 +00:00
2015-12-17 01:29:08 +00:00
2014-04-03 17:35:22 +00:00
2014-04-03 17:35:22 +00:00
2015-01-19 15:16:06 +00:00
2014-04-23 01:09:29 +00:00
2016-01-26 00:03:25 +00:00
2015-08-13 17:28:16 +00:00