[PATCH] PR middle-end/103059: reload: Also accept ASHIFT with indexed addressing

Maciej W. Rozycki macro@embecosm.com
Sun Nov 7 21:24:23 GMT 2021


On Wed, 3 Nov 2021, Maciej W. Rozycki wrote:

>  I have not yet tracked down which change after commit c605a8bf9270 made 
> regressions appear in the test suites, however clearly the commit wasn't 
> as complete a change as it should have been.  I'll see if I can find it 
> and will mention it in the final commit description if there is anything 
> useful in that.

 So regrettably it's my original commit c605a8bf9270 that caused these 
regressions.  I don't know how it happened that it slipped through in 
testing; I may have had the change left in the tree that enabled LRA and 
did not notice it.  Sorry about that, umm.

>  As noted in the commit description this has been regression-tested with 
> commit 4a960d548b7d^.  I'm running regression-testing with GCC 11 right 
> now as well and expect results by the end of week.

 For the record the change does fix numerous regressions in GCC 11 too, 
e.g.:

		=== gcc Summary ===

# of expected passes		122244
# of unexpected failures	1852
# of expected passes		122428
# of unexpected failures	1484
# of unexpected successes	5
# of expected failures		721
# of unresolved testcases	10

So either this change, if deemed safe enough, will have to go into GCC 11, 
or, if we're going to keep old reload yet for GCC 12, then the best course 
of action for GCC 11 might be reverting commit c605a8bf9270, although at 
the cost of regressing code quality a bit where it does build.

 Suggestions?

  Maciej


More information about the Gcc-patches mailing list