diff mbox

[01/89] Const-correctness fixes for some gimple accessors

Message ID 1398118294.26834.80.camel@surprise
State New
Headers show

Commit Message

David Malcolm April 21, 2014, 10:11 p.m. UTC 
On Mon, 2014-04-21 at 12:43 -0600, Jeff Law wrote:
> On 04/21/14 10:56, David Malcolm wrote:
> > gcc/
> > 	* gimple.h (gimple_assign_single_p): Accept a const_gimple rather
> > 	than a gimple.
> > 	(gimple_store_p): Likewise.
> > 	(gimple_assign_load_p): Likewise.
> > 	(gimple_assign_cast_p): Likewise.
> > 	(gimple_clobber_p): Likewise.
> I know you bootstrapped the entire 89 patch series.  Generally when we 
> have a nice little independent patch like this, it should be separately 
> bootstrapped and tested so that it can go forward independent of 
> everything else.
> 
> Anyway, this is clearly OK and non-controversial.  OK for the trunk.
> 
> In fact, I would consider any const-correctness patch like this to be 
> pre-approved.

Thanks.

It was pointed out to me off-list that this patch series lacks
documentation changes.  I'm working on fixing that, though am not sure I
want to fill everyone inboxes with an updated set of patches yet.
Should I send a combined patch for the documentation changes?  (I can
break it up and merge it into the individual changes on commit, or if
these need editing).

In any case I fixed up the corresponding entries in gcc/doc/gimple.texi,
double-checked the bootstrap/regtest/HTML generation of this one (on top
of r209545), and committed it to trunk as r209548.  I'm attaching what I
actually committed.

Dave

Comments

Gerald Pfeifer April 27, 2014, 12:21 a.m. UTC | #1 
On Mon, 21 Apr 2014, David Malcolm wrote:
> It was pointed out to me off-list that this patch series lacks
> documentation changes.  I'm working on fixing that, though am not sure I
> want to fill everyone inboxes with an updated set of patches yet.
> Should I send a combined patch for the documentation changes?  (I can
> break it up and merge it into the individual changes on commit, or if
> these need editing).
> 
> In any case I fixed up the corresponding entries in gcc/doc/gimple.texi,
> double-checked the bootstrap/regtest/HTML generation of this one (on top
> of r209545), and committed it to trunk as r209548.  I'm attaching what I
> actually committed.

You certainly can consider any doc changes that mirror code changes
as pre-approved.

Based on what Jeff said, about changes like this one (const-correctness)
going in right away, perhaps get these off your table together with their
associated documentation aspects?  And then do one doc change for all the
rest -- or individual ones, I don't have a strong preference.

Thanks for working on this!

Gerald
diff mbox

Patch

Index: gcc/doc/gimple.texi
===================================================================
--- gcc/doc/gimple.texi	(revision 209547)
+++ gcc/doc/gimple.texi	(revision 209548)
@@ -701,7 +701,7 @@ 
 Return true if the code of g is @code{GIMPLE_DEBUG}.
 @end deftypefn
 
-@deftypefn {GIMPLE function} bool gimple_assign_cast_p (gimple g)
+@deftypefn {GIMPLE function} bool gimple_assign_cast_p (const_gimple g)
 Return true if g is a @code{GIMPLE_ASSIGN} that performs a type cast
 operation.
 @end deftypefn
@@ -1102,7 +1102,7 @@ 
 statement @code{G}.
 @end deftypefn
 
-@deftypefn {GIMPLE function} bool gimple_assign_cast_p (gimple s)
+@deftypefn {GIMPLE function} bool gimple_assign_cast_p (const_gimple s)
 Return true if @code{S} is a type-cast assignment.
 @end deftypefn
 
Index: gcc/ChangeLog
===================================================================
--- gcc/ChangeLog	(revision 209547)
+++ gcc/ChangeLog	(revision 209548)
@@ -1,3 +1,16 @@ 
+2014-04-21  David Malcolm  <dmalcolm@redhat.com>
+
+	* gimple.h (gimple_assign_single_p): Accept a const_gimple rather
+	than a gimple.
+	(gimple_store_p): Likewise.
+	(gimple_assign_load_p): Likewise.
+	(gimple_assign_cast_p): Likewise.
+	(gimple_clobber_p): Likewise.
+
+	* doc/gimple.texi (gimple_assign_cast_p): Accept a const_gimple
+	rather than a gimple.
+	(gimple_assign_cast_p): Likewise.
+
 2014-04-21  Michael Meissner  <meissner@linux.vnet.ibm.com>
 
 	PR target/60735
Index: gcc/gimple.h
===================================================================
--- gcc/gimple.h	(revision 209547)
+++ gcc/gimple.h	(revision 209548)
@@ -2330,7 +2330,7 @@ 
    and do not have the semantics of a copy, such as COND_EXPR.  */
 
 static inline bool
-gimple_assign_single_p (gimple gs)
+gimple_assign_single_p (const_gimple gs)
 {
   return (is_gimple_assign (gs)
           && gimple_assign_rhs_class (gs) == GIMPLE_SINGLE_RHS);
@@ -2339,7 +2339,7 @@ 
 /* Return true if GS performs a store to its lhs.  */
 
 static inline bool
-gimple_store_p (gimple gs)
+gimple_store_p (const_gimple gs)
 {
   tree lhs = gimple_get_lhs (gs);
   return lhs && !is_gimple_reg (lhs);
@@ -2348,7 +2348,7 @@ 
 /* Return true if GS is an assignment that loads from its rhs1.  */
 
 static inline bool
-gimple_assign_load_p (gimple gs)
+gimple_assign_load_p (const_gimple gs)
 {
   tree rhs;
   if (!gimple_assign_single_p (gs))
@@ -2365,7 +2365,7 @@ 
 /* Return true if S is a type-cast assignment.  */
 
 static inline bool
-gimple_assign_cast_p (gimple s)
+gimple_assign_cast_p (const_gimple s)
 {
   if (is_gimple_assign (s))
     {
@@ -2381,7 +2381,7 @@ 
 /* Return true if S is a clobber statement.  */
 
 static inline bool
-gimple_clobber_p (gimple s)
+gimple_clobber_p (const_gimple s)
 {
   return gimple_assign_single_p (s)
          && TREE_CLOBBER_P (gimple_assign_rhs1 (s));