public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [PATCH] Fix PR 108582: ICE due to PHI-OPT removing a still in use ssa_name.
@ 2023-01-28 22:24 Andrew Pinski
  2023-01-30  7:44 ` Richard Biener
  0 siblings, 1 reply; 2+ messages in thread
From: Andrew Pinski @ 2023-01-28 22:24 UTC (permalink / raw)
  To: gcc-patches; +Cc: Andrew Pinski

This patch adds a check in match_simplify_replacement to make sure the middlebb
does not have any phi-nodes as we don't currently move those.
This was just a thinko from before.

Ok? Bootstrapped and tested on x86_64-linux-gnu with no regressions?

	PR tree-optimization/108582

gcc/ChangeLog:

	* tree-ssa-phiopt.cc (match_simplify_replacement): Add check
	for middlebb to have no phi nodes.

gcc/testsuite/ChangeLog:

	* gcc.dg/pr108582-1.c: New test.
---
 gcc/testsuite/gcc.dg/pr108582-1.c | 58 +++++++++++++++++++++++++++++++
 gcc/tree-ssa-phiopt.cc            |  5 +++
 2 files changed, 63 insertions(+)
 create mode 100644 gcc/testsuite/gcc.dg/pr108582-1.c

diff --git a/gcc/testsuite/gcc.dg/pr108582-1.c b/gcc/testsuite/gcc.dg/pr108582-1.c
new file mode 100644
index 00000000000..88c2de369ad
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/pr108582-1.c
@@ -0,0 +1,58 @@
+/* { dg-do compile } */
+/* { dg-options "-O3 -fno-tree-ccp -fno-tree-dce" } */
+
+/*
+  PHI-OPT via match_simplify_replacement used to transform:
+  if (_25 != 0)
+    goto <bb 8>; [25.00%]
+  else
+    goto <bb 9>; [75.00%]
+
+  <bb 8> [local count: 11649864]:
+  # iftmp.5_13 = PHI <2(7)>
+  k_22 = k_11 | iftmp.5_13;
+
+  <bb 9> [local count: 105655256]:
+  # g_9 = PHI <1(2), 0(8), g_8(7)>
+  # k_12 = PHI <k_20(D)(2), k_22(8), k_11(7)>
+
+into:
+
+  _15 = (int) _25;
+  _28 = -_15;
+  _4 = _13 & _28;
+  _6 = _4 | k_11;
+
+  <bb 8> [local count: 105655256]:
+  # g_9 = PHI <1(2), g_8(7)>
+  # k_12 = PHI <k_20(D)(2), _6(7)>
+
+Removing the phi-node/assignment of _13.
+
+ */
+
+int a, c, d, e, f;
+char b;
+int main() {
+  int g = 1;
+  char h[1] = {0};
+  while (a) {
+    if (f) {
+      b = 0;
+      if (d)
+        continue;
+    }
+    if (a < 1) {
+      g = 0;
+      goto L;
+    }
+  }
+  while (c) {
+    char *j = h;
+    int k;
+  L:
+    if (e && !g)
+      k |= 2 | (*j < 0);
+  }
+  return 0;
+}
diff --git a/gcc/tree-ssa-phiopt.cc b/gcc/tree-ssa-phiopt.cc
index c3a889dc593..a7ab6ce4ad9 100644
--- a/gcc/tree-ssa-phiopt.cc
+++ b/gcc/tree-ssa-phiopt.cc
@@ -1002,6 +1002,11 @@ match_simplify_replacement (basic_block cond_bb, basic_block middle_bb,
       if (!single_pred_p (middle_bb))
 	return false;
 
+      /* The middle bb cannot have phi nodes as we don't
+	 move those assignments yet. */
+      if (!gimple_seq_empty_p (phi_nodes (middle_bb)))
+	return false;
+
       stmt_to_move = last_and_only_stmt (middle_bb);
       if (!stmt_to_move)
 	return false;
-- 
2.31.1


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] Fix PR 108582: ICE due to PHI-OPT removing a still in use ssa_name.
  2023-01-28 22:24 [PATCH] Fix PR 108582: ICE due to PHI-OPT removing a still in use ssa_name Andrew Pinski
@ 2023-01-30  7:44 ` Richard Biener
  0 siblings, 0 replies; 2+ messages in thread
From: Richard Biener @ 2023-01-30  7:44 UTC (permalink / raw)
  To: Andrew Pinski; +Cc: gcc-patches

On Sat, Jan 28, 2023 at 11:25 PM Andrew Pinski via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> This patch adds a check in match_simplify_replacement to make sure the middlebb
> does not have any phi-nodes as we don't currently move those.
> This was just a thinko from before.
>
> Ok? Bootstrapped and tested on x86_64-linux-gnu with no regressions?

OK.

>         PR tree-optimization/108582
>
> gcc/ChangeLog:
>
>         * tree-ssa-phiopt.cc (match_simplify_replacement): Add check
>         for middlebb to have no phi nodes.
>
> gcc/testsuite/ChangeLog:
>
>         * gcc.dg/pr108582-1.c: New test.
> ---
>  gcc/testsuite/gcc.dg/pr108582-1.c | 58 +++++++++++++++++++++++++++++++
>  gcc/tree-ssa-phiopt.cc            |  5 +++
>  2 files changed, 63 insertions(+)
>  create mode 100644 gcc/testsuite/gcc.dg/pr108582-1.c
>
> diff --git a/gcc/testsuite/gcc.dg/pr108582-1.c b/gcc/testsuite/gcc.dg/pr108582-1.c
> new file mode 100644
> index 00000000000..88c2de369ad
> --- /dev/null
> +++ b/gcc/testsuite/gcc.dg/pr108582-1.c
> @@ -0,0 +1,58 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O3 -fno-tree-ccp -fno-tree-dce" } */
> +
> +/*
> +  PHI-OPT via match_simplify_replacement used to transform:
> +  if (_25 != 0)
> +    goto <bb 8>; [25.00%]
> +  else
> +    goto <bb 9>; [75.00%]
> +
> +  <bb 8> [local count: 11649864]:
> +  # iftmp.5_13 = PHI <2(7)>
> +  k_22 = k_11 | iftmp.5_13;
> +
> +  <bb 9> [local count: 105655256]:
> +  # g_9 = PHI <1(2), 0(8), g_8(7)>
> +  # k_12 = PHI <k_20(D)(2), k_22(8), k_11(7)>
> +
> +into:
> +
> +  _15 = (int) _25;
> +  _28 = -_15;
> +  _4 = _13 & _28;
> +  _6 = _4 | k_11;
> +
> +  <bb 8> [local count: 105655256]:
> +  # g_9 = PHI <1(2), g_8(7)>
> +  # k_12 = PHI <k_20(D)(2), _6(7)>
> +
> +Removing the phi-node/assignment of _13.
> +
> + */
> +
> +int a, c, d, e, f;
> +char b;
> +int main() {
> +  int g = 1;
> +  char h[1] = {0};
> +  while (a) {
> +    if (f) {
> +      b = 0;
> +      if (d)
> +        continue;
> +    }
> +    if (a < 1) {
> +      g = 0;
> +      goto L;
> +    }
> +  }
> +  while (c) {
> +    char *j = h;
> +    int k;
> +  L:
> +    if (e && !g)
> +      k |= 2 | (*j < 0);
> +  }
> +  return 0;
> +}
> diff --git a/gcc/tree-ssa-phiopt.cc b/gcc/tree-ssa-phiopt.cc
> index c3a889dc593..a7ab6ce4ad9 100644
> --- a/gcc/tree-ssa-phiopt.cc
> +++ b/gcc/tree-ssa-phiopt.cc
> @@ -1002,6 +1002,11 @@ match_simplify_replacement (basic_block cond_bb, basic_block middle_bb,
>        if (!single_pred_p (middle_bb))
>         return false;
>
> +      /* The middle bb cannot have phi nodes as we don't
> +        move those assignments yet. */
> +      if (!gimple_seq_empty_p (phi_nodes (middle_bb)))
> +       return false;
> +
>        stmt_to_move = last_and_only_stmt (middle_bb);
>        if (!stmt_to_move)
>         return false;
> --
> 2.31.1
>

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2023-01-30  7:44 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-01-28 22:24 [PATCH] Fix PR 108582: ICE due to PHI-OPT removing a still in use ssa_name Andrew Pinski
2023-01-30  7:44 ` Richard Biener

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).