]> git.ipfire.org Git - thirdparty/gcc.git/commit
builtins: Fix fold_builtin_query clzg/ctzg side-effects handling [PR112639]
authorJakub Jelinek <jakub@redhat.com>
Tue, 21 Nov 2023 08:32:37 +0000 (09:32 +0100)
committerJakub Jelinek <jakub@redhat.com>
Tue, 21 Nov 2023 08:32:37 +0000 (09:32 +0100)
commit8a8a6d60c670c5153c5109df5fe68deb6fcf466d
treea29901a239bdaa4cb0972956ee2f76f5446a096b
parent7ad308bd4cca871e7509a5eeaef83ee68678820f
builtins: Fix fold_builtin_query clzg/ctzg side-effects handling [PR112639]

As the testcase shows, I've missed one spot where initially the code thinks
it could use 2 argument IFN_CLZ/IFN_CTZ form, but then verifies it can't
because it doesn't have the right target value and turns it into the
arg0 ? arg1 : .C[LT]Z (arg0)
form.  That form evaluates the argument twice though and so needs save_expr,
which I've missed to call in that case.  In other cases where it is known
from the beginning that it will be needed (e.g. the __builtin_clzg case
on types smaller than unsigned int where we'll need to add an addend
to the clz value) or the unsigned __int128 expansion called save_expr
before.

2023-11-21  Jakub Jelinek  <jakub@redhat.com>

PR middle-end/112639
* builtins.cc (fold_builtin_bit_query): If arg0 has side-effects, arg1
is specified but cleared, call save_expr on arg0.

* gcc.dg/torture/pr112639.c: New test.
gcc/builtins.cc
gcc/testsuite/gcc.dg/torture/pr112639.c [new file with mode: 0644]