[rfa/testsuite/threads] fix gcc signed/unsigned warning

[rfa/testsuite/threads] fix gcc signed/unsigned warning

[Michael Chastain]

This patch takes care of one set of gcc HEAD warnings in
schedlock.c and thread-specific.c.

The code uses "unsigned int" and "signed int" for the same data, but I'm
not into changing that right now.  I'm just killing the compiler warning
so the code can keep doing what it's doing.

Tested on native i686-pc-linux-gnu with gcc 2.95.3, gcc 3.3.4,
gcc 3.4.1, and gcc HEAD 20040824, dwarf-2 and stabs+.

Okay to commit?

2004-08-26  Michael Chastain  <mec.gnu@mindspring.com>

	Fix PR testsuite/1735.
	* gdb.threads/schedlock.c (thread_function): Add a cast
	to suppress a gcc warning.
	* gdb.threads/thread-specific.c (thread_function): Likewise.

Index: gdb.threads/schedlock.c
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.threads/schedlock.c,v
retrieving revision 1.4
diff -c -3 -p -r1.4 schedlock.c
*** gdb.threads/schedlock.c	26 Aug 2004 13:10:19 -0000	1.4
--- gdb.threads/schedlock.c	26 Aug 2004 16:40:48 -0000
***************
*** 1,6 ****
  /* This testcase is part of GDB, the GNU debugger.
  
!    Copyright 2002, 2003 Free Software Foundation, Inc.
  
     This program is free software; you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
--- 1,6 ----
  /* This testcase is part of GDB, the GNU debugger.
  
!    Copyright 2002, 2003, 2004 Free Software Foundation, Inc.
  
     This program is free software; you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
*************** int main() {
*** 52,58 ****
  
  void *thread_function(void *arg) {
      int my_number =  (long) arg;
!     int *myp = &args[my_number];
  
      /* Don't run forever.  Run just short of it :)  */
      while (*myp > 0)
--- 52,58 ----
  
  void *thread_function(void *arg) {
      int my_number =  (long) arg;
!     int *myp = (int *) &args[my_number];
  
      /* Don't run forever.  Run just short of it :)  */
      while (*myp > 0)
Index: gdb.threads/thread-specific.c
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.threads/thread-specific.c,v
retrieving revision 1.1
diff -c -3 -p -r1.1 thread-specific.c
*** gdb.threads/thread-specific.c	1 Feb 2004 18:04:37 -0000	1.1
--- gdb.threads/thread-specific.c	26 Aug 2004 16:40:48 -0000
*************** int main() {
*** 53,59 ****
  
  void *thread_function(void *arg) {
      int my_number =  (long) arg;
!     int *myp = &args[my_number];
  
      /* Don't run forever.  Run just short of it :)  */
      while (*myp > 0)
--- 53,59 ----
  
  void *thread_function(void *arg) {
      int my_number =  (long) arg;
!     int *myp = (int *) &args[my_number];
  
      /* Don't run forever.  Run just short of it :)  */
      while (*myp > 0)

Re: [rfa/testsuite/threads] fix gcc signed/unsigned warning

[Michael Snyder]

Michael Chastain wrote:
> This patch takes care of one set of gcc HEAD warnings in
> schedlock.c and thread-specific.c.
> 
> The code uses "unsigned int" and "signed int" for the same data, but I'm
> not into changing that right now.  I'm just killing the compiler warning
> so the code can keep doing what it's doing.
> 
> Tested on native i686-pc-linux-gnu with gcc 2.95.3, gcc 3.3.4,
> gcc 3.4.1, and gcc HEAD 20040824, dwarf-2 and stabs+.
> 
> Okay to commit?

Well, it really looks to me as if either args should be
signed, or myp should be unsigned.  The test will work
either way, although signed would "feel" safer.

But if you want to just do this, it's OK with me.

> 
> 2004-08-26  Michael Chastain  <mec.gnu@mindspring.com>
> 
> 	Fix PR testsuite/1735.
> 	* gdb.threads/schedlock.c (thread_function): Add a cast
> 	to suppress a gcc warning.
> 	* gdb.threads/thread-specific.c (thread_function): Likewise.
> 
> Index: gdb.threads/schedlock.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/testsuite/gdb.threads/schedlock.c,v
> retrieving revision 1.4
> diff -c -3 -p -r1.4 schedlock.c
> *** gdb.threads/schedlock.c	26 Aug 2004 13:10:19 -0000	1.4
> --- gdb.threads/schedlock.c	26 Aug 2004 16:40:48 -0000
> ***************
> *** 1,6 ****
>   /* This testcase is part of GDB, the GNU debugger.
>   
> !    Copyright 2002, 2003 Free Software Foundation, Inc.
>   
>      This program is free software; you can redistribute it and/or modify
>      it under the terms of the GNU General Public License as published by
> --- 1,6 ----
>   /* This testcase is part of GDB, the GNU debugger.
>   
> !    Copyright 2002, 2003, 2004 Free Software Foundation, Inc.
>   
>      This program is free software; you can redistribute it and/or modify
>      it under the terms of the GNU General Public License as published by
> *************** int main() {
> *** 52,58 ****
>   
>   void *thread_function(void *arg) {
>       int my_number =  (long) arg;
> !     int *myp = &args[my_number];
>   
>       /* Don't run forever.  Run just short of it :)  */
>       while (*myp > 0)
> --- 52,58 ----
>   
>   void *thread_function(void *arg) {
>       int my_number =  (long) arg;
> !     int *myp = (int *) &args[my_number];
>   
>       /* Don't run forever.  Run just short of it :)  */
>       while (*myp > 0)
> Index: gdb.threads/thread-specific.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/testsuite/gdb.threads/thread-specific.c,v
> retrieving revision 1.1
> diff -c -3 -p -r1.1 thread-specific.c
> *** gdb.threads/thread-specific.c	1 Feb 2004 18:04:37 -0000	1.1
> --- gdb.threads/thread-specific.c	26 Aug 2004 16:40:48 -0000
> *************** int main() {
> *** 53,59 ****
>   
>   void *thread_function(void *arg) {
>       int my_number =  (long) arg;
> !     int *myp = &args[my_number];
>   
>       /* Don't run forever.  Run just short of it :)  */
>       while (*myp > 0)
> --- 53,59 ----
>   
>   void *thread_function(void *arg) {
>       int my_number =  (long) arg;
> !     int *myp = (int *) &args[my_number];
>   
>       /* Don't run forever.  Run just short of it :)  */
>       while (*myp > 0)
> 

Re: [rfa/testsuite/threads] fix gcc signed/unsigned warning

[Michael Chastain]

Michael Snyder <msnyder@redhat.com> wrote:
> But if you want to just do this, it's OK with me.

Yay, committed.

> Well, it really looks to me as if either args should be
> signed, or myp should be unsigned.  The test will work
> either way, although signed would "feel" safer.

I fooled around with this.  I couldn't find something that was type-safe
and felt exactly like the old test.  So I decided "one thing at a time"
-- first make it compile like it did with old compilers, then later, fix
the signed/unsigned muddle.

I've filed a PR for the signed/unsigned part.

===

2004-08-26  Michael Chastain  <mec.gnu@mindspring.com>

	Fix PR testsuite/1735.
	* gdb.threads/schedlock.c (thread_function): Add a cast
	to suppress a gcc warning.
	* gdb.threads/thread-specific.c (thread_function): Likewise.