PING for LTO ld -r patches

PING for LTO ld -r patches

[Andi Kleen]

Review ping for:

http://permalink.gmane.org/gmane.comp.gcc.patches/212187
http://permalink.gmane.org/gmane.comp.gcc.patches/212185
http://permalink.gmane.org/gmane.comp.gcc.patches/212186
http://permalink.gmane.org/gmane.comp.gcc.patches/212226

The first three patches should be quite straight forward? Ok
to commit those after rebootstrap/retest?

The fourth is a bit more complicated, but should get reviewed
too please.

Thanks,

-Andi

Re: PING for LTO ld -r patches

[Diego Novillo]

On 10-07-22 04:22 , Andi Kleen wrote:
> Review ping for:
>
> http://permalink.gmane.org/gmane.comp.gcc.patches/212187

OK.

> http://permalink.gmane.org/gmane.comp.gcc.patches/212185

This is OK, but I think it should be possible to use the 
lto_section_name[] array when returning the concat'd values in 
lto_get_secion_name.  Would you be willing to work a patch for that?

> http://permalink.gmane.org/gmane.comp.gcc.patches/212186

OK.


> http://permalink.gmane.org/gmane.comp.gcc.patches/212226

OK.  What is the difficulty with the testcase in this one?


Diego.

Re: PING for LTO ld -r patches

[Andi Kleen]

Diego Novillo <dnovillo@google.com> writes:

Thanks for the review.

> On 10-07-22 04:22 , Andi Kleen wrote:
>> Review ping for:
>>
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212187
>
> OK.
>
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212185
>
> This is OK, but I think it should be possible to use the
> lto_section_name[] array when returning the concat'd values in
> lto_get_secion_name.  Would you be willing to work a patch for that?

The fourth patch already does this. I didn't try to untangle
it from the ld -r changes.

>
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212186
>
> OK.
>
>
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212226
>
> OK.  What is the difficulty with the testcase in this one?

The LTO test case is a make file with some custom rules and I haven't 
figured out how to express that in dejagnu. If someone helps me with the 
dejagnu conversion the test case could be comitted.

The test case with Makefile is http://halobates.de/lto-r.tar.gz

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: PING for LTO ld -r patches

[H.J. Lu]

On Thu, Jul 22, 2010 at 1:22 AM, Andi Kleen <andi@firstfloor.org> wrote:
> Review ping for:
>
> http://permalink.gmane.org/gmane.comp.gcc.patches/212187
> http://permalink.gmane.org/gmane.comp.gcc.patches/212185
> http://permalink.gmane.org/gmane.comp.gcc.patches/212186
> http://permalink.gmane.org/gmane.comp.gcc.patches/212226
>
> The first three patches should be quite straight forward? Ok
> to commit those after rebootstrap/retest?
>
> The fourth is a bit more complicated, but should get reviewed
> too please.

This caused:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45042


-- 
H.J.

Re: PING for LTO ld -r patches

[H.J. Lu]

On Fri, Jul 23, 2010 at 6:25 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Thu, Jul 22, 2010 at 1:22 AM, Andi Kleen <andi@firstfloor.org> wrote:
>> Review ping for:
>>
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212187
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212185
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212186
>> http://permalink.gmane.org/gmane.comp.gcc.patches/212226
>>
>> The first three patches should be quite straight forward? Ok
>> to commit those after rebootstrap/retest?
>>
>> The fourth is a bit more complicated, but should get reviewed
>> too please.
>
> This caused:
>
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45042
>

I checked in this as an obvious fix.


-- 
H.J.
---
Index: ChangeLog
===================================================================
--- ChangeLog	(revision 162464)
+++ ChangeLog	(working copy)
@@ -1,5 +1,10 @@
-2010-07-10   Andi Kleen <ak@linux.intel.com>
+2010-07-23  H.J. Lu  <hongjiu.lu@intel.com>

+	PR bootstrap/45042
+	* lto-plugin.c (translate): Cast to unsigned long.
+
+2010-07-23   Andi Kleen <ak@linux.intel.com>
+
 	PR lto/44992
 	* lto-plugin.c (sym_aux): Add.
 	(plugin_symtab): Remove slots. Add aux and id.
Index: lto-plugin.c
===================================================================
--- lto-plugin.c	(revision 162464)
+++ lto-plugin.c	(working copy)
@@ -212,7 +212,8 @@
       data = parse_table_entry (data, &syms[n], &aux[n]);
     }

-  fprintf (stderr, "n = %d len = %d end-data=%lu\n", n, len, end-data);
+  fprintf (stderr, "n = %d len = %d end-data=%lu\n", n, len,
+	   (unsigned long) (end - data));
   assert(n < len);

   out->nsyms = n;

Re: PING for LTO ld -r patches

[Andi Kleen]

On Fri, Jul 23, 2010 at 06:25:11AM -0700, H.J. Lu wrote:
> On Thu, Jul 22, 2010 at 1:22 AM, Andi Kleen <andi@firstfloor.org> wrote:
> > Review ping for:
> >
> > http://permalink.gmane.org/gmane.comp.gcc.patches/212187
> > http://permalink.gmane.org/gmane.comp.gcc.patches/212185
> > http://permalink.gmane.org/gmane.comp.gcc.patches/212186
> > http://permalink.gmane.org/gmane.comp.gcc.patches/212226
> >
> > The first three patches should be quite straight forward? Ok
> > to commit those after rebootstrap/retest?
> >
> > The fourth is a bit more complicated, but should get reviewed
> > too please.
> 
> This caused:
> 
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45042

Thanks for the fix.

-Andi