[patch] Fix for excessive stack use (gdb/10457)

[patch] Fix for excessive stack use (gdb/10457)

[Paul Pluzhnikov]

[-- Attachment #1: Type: text/plain, Size: 261 bytes --]

Greetings,

Calling alloca in a loop is a bad idea.
Here is a fix for gdb/10457; rather on the obvious side.

Thanks,
-- 
Paul Pluzhnikov

2009-10-07  Paul Pluzhnikov  <ppluzhnikov@google.com>

        * elfread.c (elf_symtab_read): Don't use alloca in a loop.

[-- Attachment #2: gdb-pr10457-20091007.txt --]
[-- Type: text/plain, Size: 857 bytes --]

Index: elfread.c
===================================================================
RCS file: /cvs/src/src/gdb/elfread.c,v
retrieving revision 1.77
diff -p -u -r1.77 elfread.c
--- elfread.c	30 Apr 2009 21:59:03 -0000	1.77
+++ elfread.c	8 Oct 2009 05:22:49 -0000
@@ -535,7 +535,7 @@ elf_symtab_read (struct objfile *objfile
 
 	      if (len > 4 && strcmp (sym->name + len - 4, "@plt") == 0)
 		{
-		  char *base_name = alloca (len - 4 + 1);
+		  char *base_name = xmalloc (len - 4 + 1);
 		  struct minimal_symbol *mtramp;
 
 		  memcpy (base_name, sym->name, len - 4);
@@ -543,6 +543,7 @@ elf_symtab_read (struct objfile *objfile
 		  mtramp = record_minimal_symbol (base_name, symaddr,
 						  mst_solib_trampoline,
 						  sym->section, objfile);
+		  xfree (base_name);
 		  if (mtramp)
 		    {
 		      MSYMBOL_SIZE (mtramp) = MSYMBOL_SIZE (msym);

Re: [patch] Fix for excessive stack use (gdb/10457)

[Joel Brobecker]

> 2009-10-07  Paul Pluzhnikov  <ppluzhnikov@google.com>
> 
>         * elfread.c (elf_symtab_read): Don't use alloca in a loop.

alloca was very convenient because it allows us to not worry about
memory leaks that could happen if record_minimal_symbol throws an error,
for instance. Thanks to Jan's excellent eyes, we know that
record_minimal_symbol eventually results in a call to XCALLOC, which
can possibly cause an internal_error(). It's very unlikely to happen.
So I wouldn't have complained if you had used a cleanup to make sure
that the memory is always released. But in this case, I don't see
record_minimal_symbol ever needing to thrown an exception on its own,
and I don't see that such a small leak can make a difference if we run
out of memory.

The patch is approved.

-- 
Joel

Re: [patch] Fix for excessive stack use (gdb/10457)

[Paul Pluzhnikov]

On Thu, Oct 8, 2009 at 9:20 AM, Joel Brobecker <brobecker@adacore.com> wrote:

> The patch is approved.

So committed.

Should I also check this in on the 7.0 branch?

Thanks,
-- 
Paul Pluzhnikov

Re: [patch] Fix for excessive stack use (gdb/10457)

[Joel Brobecker]

> Should I also check this in on the 7.0 branch?

Sure, the patch looks very safe to me.

-- 
Joel

Re: [patch] Fix for excessive stack use (gdb/10457)

[Holger Freyther]

On Thursday 08 October 2009 19:35:14 Joel Brobecker wrote:
> > Should I also check this in on the 7.0 branch?
> 
> Sure, the patch looks very safe to me.
> 

Thanks a lot.