[2.6.19 patch] i386/x86_64: remove the unused EXPORT_SYMBOL(dump_trace)

[2.6.19 patch] i386/x86_64: remove the unused EXPORT_SYMBOL(dump_trace)

[Adrian Bunk]

This patch removes the unused EXPORT_SYMBOL(dump_trace) added on i386 
and x86_64 in 2.6.19-rc.

By removing them before the final 2.6.19 we avoid the possibility of 
people later whining that we removed exports they started using.

Signed-off-by: Adrian Bunk <bunk@stusta.de>

---

 arch/i386/kernel/traps.c   |    1 -
 arch/x86_64/kernel/traps.c |    1 -
 2 files changed, 2 deletions(-)

--- linux-2.6.19-rc5-mm2/arch/i386/kernel/traps.c.old	2006-11-21 19:45:10.000000000 +0100
+++ linux-2.6.19-rc5-mm2/arch/i386/kernel/traps.c	2006-11-21 19:45:26.000000000 +0100
@@ -247,7 +247,6 @@
 			break;
 	}
 }
-EXPORT_SYMBOL(dump_trace);
 
 static void
 print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
--- linux-2.6.19-rc5-mm2/arch/x86_64/kernel/traps.c.old	2006-11-21 19:45:33.000000000 +0100
+++ linux-2.6.19-rc5-mm2/arch/x86_64/kernel/traps.c	2006-11-21 19:45:59.000000000 +0100
@@ -377,7 +377,6 @@
 out:
 	put_cpu();
 }
-EXPORT_SYMBOL(dump_trace);
 
 static void
 print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)

Re: [2.6.19 patch] i386/x86_64: remove the unused EXPORT_SYMBOL(dump_trace)

[Andi Kleen]

On Tuesday 21 November 2006 20:41, Adrian Bunk wrote:
> This patch removes the unused EXPORT_SYMBOL(dump_trace) added on i386 
> and x86_64 in 2.6.19-rc.
> 
> By removing them before the final 2.6.19 we avoid the possibility of 
> people later whining that we removed exports they started using.

I exported it for systemtap so that they can stop using the broken
hack they currently use as unwinder.

-Andi

Re: [2.6.19 patch] i386/x86_64: remove the unused EXPORT_SYMBOL(dump_trace)

[Christoph Hellwig]

On Tue, Nov 21, 2006 at 08:47:30PM +0100, Andi Kleen wrote:
> On Tuesday 21 November 2006 20:41, Adrian Bunk wrote:
> > This patch removes the unused EXPORT_SYMBOL(dump_trace) added on i386 
> > and x86_64 in 2.6.19-rc.
> > 
> > By removing them before the final 2.6.19 we avoid the possibility of 
> > people later whining that we removed exports they started using.
> 
> I exported it for systemtap so that they can stop using the broken
> hack they currently use as unwinder.

Nack, dump_trace is nothing that should be export for broken out of tree
junk.

Re: [2.6.19 patch] i386/x86_64: remove the unused EXPORT_SYMBOL(dump_trace)

[Alan]

On Tue, 21 Nov 2006 20:18:44 +0000
Christoph Hellwig <hch@infradead.org> wrote:

> On Tue, Nov 21, 2006 at 08:47:30PM +0100, Andi Kleen wrote:
> > On Tuesday 21 November 2006 20:41, Adrian Bunk wrote:
> > > This patch removes the unused EXPORT_SYMBOL(dump_trace) added on i386 
> > > and x86_64 in 2.6.19-rc.
> > > 
> > > By removing them before the final 2.6.19 we avoid the possibility of 
> > > people later whining that we removed exports they started using.
> > 
> > I exported it for systemtap so that they can stop using the broken
> > hack they currently use as unwinder.
> 
> Nack, dump_trace is nothing that should be export for broken out of tree
> junk.

It is exported for systemtap not random broken out of tree junk, and the
result is a good deal prettier. Systemtap guys really ought to get their
stuff merged too, although how we merge a dynamic module writing tool I'm
not so sure ?

Alan

Re: [2.6.19 patch] i386/x86_64: remove the unused EXPORT_SYMBOL(dump_trace)

[Vara Prasad]

Alan wrote:

>On Tue, 21 Nov 2006 20:18:44 +0000
>Christoph Hellwig <hch@infradead.org> wrote:
>
>  
>
>>On Tue, Nov 21, 2006 at 08:47:30PM +0100, Andi Kleen wrote:
>>    
>>
>>>On Tuesday 21 November 2006 20:41, Adrian Bunk wrote:
>>>      
>>>
>>>>This patch removes the unused EXPORT_SYMBOL(dump_trace) added on i386
>>>>and x86_64 in 2.6.19-rc.
>>>>
>>>>By removing them before the final 2.6.19 we avoid the possibility of
>>>>people later whining that we removed exports they started using.
>>>>        
>>>>
>>>I exported it for systemtap so that they can stop using the broken
>>>hack they currently use as unwinder.
>>>      
>>>
>>Nack, dump_trace is nothing that should be export for broken out of tree
>>junk.
>>    
>>
>
>It is exported for systemtap not random broken out of tree junk, and the
>result is a good deal prettier. Systemtap guys really ought to get their
>stuff merged too, although how we merge a dynamic module writing tool I'm
>not so sure ?
>
>  
>
As you all know SystemTap uses kprobes and relayfs as the basis which 
are already merged into the mainline.
We are looking at all the other pieces of SystemTap that can be merged 
to mainline but as Alan mentioned it is not easy and obvious. We think 
we can merge transport part of the runtime, here is the initial patch 
under review in this thread
http://sources.redhat.com/ml/systemtap/2006-q4/msg00031.html
http://sources.redhat.com/ml/systemtap/2006-q4/msg00030.html

We are open for other suggestions as well.

bye,
Vara Prasad

>Alan
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/
>
>  
>