Skip to content

Commit eb48619

Browse files
committed
Fix optimization hazard in gram.y's makeOrderedSetArgs(), redux.
It appears that commit cf63c64, which intended to prevent misoptimization of the result-building step in makeOrderedSetArgs, didn't go far enough: buildfarm member hornet's version of xlc is now optimizing back to the old, broken behavior in which list_length(directargs) is fetched only after list_concat() has changed that value. I'm not entirely convinced whether that's an undeniable compiler bug or whether it can be justified by a sufficiently aggressive interpretation of C sequence points. So let's just change the code to make it harder to misinterpret. Back-patch to all supported versions, just in case. Discussion: https://postgr.es/m/1830491.1601944935@sss.pgh.pa.us
1 parent 710c0a6 commit eb48619

File tree

1 file changed

+3
-3
lines changed

1 file changed

+3
-3
lines changed

src/backend/parser/gram.y

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -14533,7 +14533,7 @@ makeOrderedSetArgs(List *directargs, List *orderedargs,
1453314533
core_yyscan_t yyscanner)
1453414534
{
1453514535
FunctionParameter *lastd = (FunctionParameter *) llast(directargs);
14536-
int ndirectargs;
14536+
Value *ndirectargs;
1453714537

1453814538
/* No restriction unless last direct arg is VARIADIC */
1453914539
if (lastd->mode == FUNC_PARAM_VARIADIC)
@@ -14557,10 +14557,10 @@ makeOrderedSetArgs(List *directargs, List *orderedargs,
1455714557
}
1455814558

1455914559
/* don't merge into the next line, as list_concat changes directargs */
14560-
ndirectargs = list_length(directargs);
14560+
ndirectargs = makeInteger(list_length(directargs));
1456114561

1456214562
return list_make2(list_concat(directargs, orderedargs),
14563-
makeInteger(ndirectargs));
14563+
ndirectargs);
1456414564
}
1456514565

1456614566
/* insertSelectOptions()

0 commit comments

Comments
 (0)