This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [gomp4] rewrite simd clone argument adjustment to avoid regimplification
- From: Aldy Hernandez <aldyh at redhat dot com>
- To: Jakub Jelinek <jakub at redhat dot com>, gcc-patches <gcc-patches at gcc dot gnu dot org>, Martin Jambor <mjambor at suse dot cz>
- Date: Tue, 12 Nov 2013 09:27:26 -0700
- Subject: Re: [gomp4] rewrite simd clone argument adjustment to avoid regimplification
- Authentication-results: sourceware.org; auth=none
- References: <5277BAF0 dot 6050506 at redhat dot com> <20131104154435 dot GS27813 at tucnak dot zalov dot cz> <527AC1A8 dot 70904 at redhat dot com> <20131106224819 dot GJ27813 at tucnak dot zalov dot cz> <527AE0AF dot 2010106 at redhat dot com> <20131107175836 dot GF3572 at virgil dot suse> <528110AC dot 2000203 at redhat dot com> <20131112135607 dot GC10643 at virgil dot suse>
On 11/12/13 06:56, Martin Jambor wrote:
+
+/* Given an expression, return an adjustment entry specifying the
+ transformation to be done on EXPR. If no suitable adjustment entry
+ was found, returns NULL.
+
+ If IGNORE_DEFAULT_DEF is set, consider SSA_NAMEs which are not a
+ default def, otherwise bail on them.
+
+ If CONVERT is non-NULL,
This is not true, convert can be set even when the function might
return NULL afterwards.
Fixed comment.
+ this function will set *CONVERT if the
+ expression provided is a component reference that must be converted
+ upon return. ADJUSTMENTS is the adjustments vector. */
+
+ipa_parm_adjustment *
+ipa_get_adjustment_candidate (tree *&expr, bool *convert,
+ ipa_parm_adjustment_vec adjustments,
+ bool ignore_default_def)
I find changing parameters passed by reference confusing and
error-prone, I would very much prefer if this was "tree **expr".
Either way, you should document that expr can be changed in the
function comment.
I absolutely agree. That was sloppy on my part. Fixed.
The patch is OK with me but please at least fix the comments.
Committed to branch.
Thanks.