pass_manager: Fix ICE with -fdump-passes -fdisable-tree-* [PR93874]

Message ID 20200225083640.GI2155@tucnak
State New
Headers show
Series
  • pass_manager: Fix ICE with -fdump-passes -fdisable-tree-* [PR93874]
Related show

Commit Message

Jakub Jelinek Feb. 25, 2020, 8:36 a.m.
Hi!

dump_passes pushes a dummy function for which it evaluates the gates
and checks whether the pass is enabled or disabled.
Unfortunately, if any -fdisable-*-*/-fenable-*-* options were seen,
we ICE during is_pass_explicitly_enabled_or_disabled because slot
is non-NULL then and the code will do:
  cgraph_uid = func ? cgraph_node::get (func)->get_uid () : 0;
but the dummy function doesn't have a cgraph node.

So, either we need to create and then remove a cgraph node for the dummy
function like the following patch, or function.c would need to export the
in_dummy_function flag (or have some way to query that flag from other TUs)
and we'd need to check it in is_pass_explicitly_enabled_or_disabled.

Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

2020-02-25  Jakub Jelinek  <jakub@redhat.com>

	PR middle-end/93874
	* passes.c (pass_manager::dump_passes): Create a cgraph node for the
	dummy function and remove it at the end.

	* gcc.dg/pr93874.c: New test.


	Jakub

Comments

Richard Biener Feb. 25, 2020, 9:11 a.m. | #1
On Tue, 25 Feb 2020, Jakub Jelinek wrote:

> Hi!

> 

> dump_passes pushes a dummy function for which it evaluates the gates

> and checks whether the pass is enabled or disabled.

> Unfortunately, if any -fdisable-*-*/-fenable-*-* options were seen,

> we ICE during is_pass_explicitly_enabled_or_disabled because slot

> is non-NULL then and the code will do:

>   cgraph_uid = func ? cgraph_node::get (func)->get_uid () : 0;

> but the dummy function doesn't have a cgraph node.

> 

> So, either we need to create and then remove a cgraph node for the dummy

> function like the following patch, or function.c would need to export the

> in_dummy_function flag (or have some way to query that flag from other TUs)

> and we'd need to check it in is_pass_explicitly_enabled_or_disabled.

> 

> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?


OK.

Richard.

> 2020-02-25  Jakub Jelinek  <jakub@redhat.com>

> 

> 	PR middle-end/93874

> 	* passes.c (pass_manager::dump_passes): Create a cgraph node for the

> 	dummy function and remove it at the end.

> 

> 	* gcc.dg/pr93874.c: New test.

> 

> --- gcc/passes.c.jj	2020-01-12 11:54:36.693409184 +0100

> +++ gcc/passes.c	2020-02-24 10:34:24.081281789 +0100

> @@ -950,6 +950,7 @@ void

>  pass_manager::dump_passes () const

>  {

>    push_dummy_function (true);

> +  cgraph_node *node = cgraph_node::get_create (current_function_decl);

>  

>    create_pass_tab ();

>  

> @@ -959,6 +960,7 @@ pass_manager::dump_passes () const

>    dump_pass_list (all_late_ipa_passes, 1);

>    dump_pass_list (all_passes, 1);

>  

> +  node->remove ();

>    pop_dummy_function ();

>  }

>  

> --- gcc/testsuite/gcc.dg/pr93874.c.jj	2020-02-24 10:39:43.417514220 +0100

> +++ gcc/testsuite/gcc.dg/pr93874.c	2020-02-24 10:41:26.506976191 +0100

> @@ -0,0 +1,6 @@

> +/* PR middle-end/93874 */

> +/* { dg-do compile } */

> +/* { dg-options "-fdisable-tree-dse3 -fdump-passes" } */

> +/* { dg-prune-output ".*" } */

> +

> +int i;

> 

> 	Jakub

> 

> 


-- 
Richard Biener <rguenther@suse.de>
SUSE Software Solutions Germany GmbH, Maxfeldstrasse 5, 90409 Nuernberg,
Germany; GF: Felix Imendörffer; HRB 36809 (AG Nuernberg)
Jan Hubicka Feb. 25, 2020, 10:17 a.m. | #2
> Hi!

> 

> dump_passes pushes a dummy function for which it evaluates the gates

> and checks whether the pass is enabled or disabled.

> Unfortunately, if any -fdisable-*-*/-fenable-*-* options were seen,

> we ICE during is_pass_explicitly_enabled_or_disabled because slot

> is non-NULL then and the code will do:

>   cgraph_uid = func ? cgraph_node::get (func)->get_uid () : 0;

> but the dummy function doesn't have a cgraph node.

> 

> So, either we need to create and then remove a cgraph node for the dummy

> function like the following patch, or function.c would need to export the

> in_dummy_function flag (or have some way to query that flag from other TUs)

> and we'd need to check it in is_pass_explicitly_enabled_or_disabled.

> 

> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

> 

> 2020-02-25  Jakub Jelinek  <jakub@redhat.com>

> 

> 	PR middle-end/93874

> 	* passes.c (pass_manager::dump_passes): Create a cgraph node for the

> 	dummy function and remove it at the end.

> 

> 	* gcc.dg/pr93874.c: New test.


OK,
Honza

Patch

--- gcc/passes.c.jj	2020-01-12 11:54:36.693409184 +0100
+++ gcc/passes.c	2020-02-24 10:34:24.081281789 +0100
@@ -950,6 +950,7 @@  void
 pass_manager::dump_passes () const
 {
   push_dummy_function (true);
+  cgraph_node *node = cgraph_node::get_create (current_function_decl);
 
   create_pass_tab ();
 
@@ -959,6 +960,7 @@  pass_manager::dump_passes () const
   dump_pass_list (all_late_ipa_passes, 1);
   dump_pass_list (all_passes, 1);
 
+  node->remove ();
   pop_dummy_function ();
 }
 
--- gcc/testsuite/gcc.dg/pr93874.c.jj	2020-02-24 10:39:43.417514220 +0100
+++ gcc/testsuite/gcc.dg/pr93874.c	2020-02-24 10:41:26.506976191 +0100
@@ -0,0 +1,6 @@ 
+/* PR middle-end/93874 */
+/* { dg-do compile } */
+/* { dg-options "-fdisable-tree-dse3 -fdump-passes" } */
+/* { dg-prune-output ".*" } */
+
+int i;