Re: [google][4.6]Compiler Directed Multiversioning with new -mvarch option (issue 5490054)

[tmsriram at google dot com]

I have uploaded a new patch set with all the mentioned changes made. If
a function has the target attribute it will not be touched by the
autoclone pass.

Also fixed some test cases which were broken because the clone names
used '_' instead of '.' for suffixing.

On 2011/12/16 19:39:47, davidxl wrote:
> http://codereview.appspot.com/5490054/diff/1011/config/i386/i386.c
> File config/i386/i386.c (right):


http://codereview.appspot.com/5490054/diff/1011/config/i386/i386.c#newcode26569
> config/i386/i386.c:26569: +mversion_for_core2 (tree
*optimization_node,
> -> mversionable_for_core2_p ?

> http://codereview.appspot.com/5490054/diff/1011/mversn-dispatch.c
> File mversn-dispatch.c (right):


http://codereview.appspot.com/5490054/diff/1011/mversn-dispatch.c#newcode931
> mversn-dispatch.c:931: DECL_STATIC_DESTRUCTOR (new_decl) = 0;
> Should you assert it instead? Should not clone ctor/dtors.


http://codereview.appspot.com/5490054/diff/1011/mversn-dispatch.c#newcode2221
> mversn-dispatch.c:2221: VEC_truncate (edge, EXIT_BLOCK_PTR->preds, 0);
> {} --> remove


http://codereview.appspot.com/5490054/diff/1011/mversn-dispatch.c#newcode2389
> mversn-dispatch.c:2389:
> How does it interact with manual multi-versioning from user? You
probably don't
> want to clone functions that are marked with target attributes
(explicitly --
> not implied from command line).



http://codereview.appspot.com/5490054/