Patchwork x86-64: remove duplicate line on PREFETCH_ONE_SET macro

login
register
mail settings
Submitter leonardo.sandoval.gonzalez@linux.intel.com
Date May 15, 2018, 5:13 p.m.
Message ID <20180515171303.11312-1-leonardo.sandoval.gonzalez@linux.intel.com>
Download mbox | patch
Permalink /patch/27273/
State New
Headers show

Comments

From: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>

Tested on 64-bit AVX machine

       	  * sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S (MEMMOVE):
 	  remove duplicate line on PREFETCH_ONE_SET macro

Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
---
 sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S | 1 -
 1 file changed, 1 deletion(-)
Florian Weimer - May 15, 2018, 6:27 p.m.
* leonardo sandoval gonzalez:

> From: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
>
> Tested on 64-bit AVX machine
>
>        	  * sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S (MEMMOVE):
>  	  remove duplicate line on PREFETCH_ONE_SET macro

This should be:

	* sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
	(PREFETCH_ONE_SET): Remove duplicate line.

> Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>

We don't use DCO, so this line is very misleading and should not be
present.

Looks good otherwise.  Do you have commit access?
On Tue, 2018-05-15 at 20:27 +0200, Florian Weimer wrote:
> * leonardo sandoval gonzalez:
> 
> > From: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com
> > >
> > 
> > Tested on 64-bit AVX machine
> > 
> >        	  * sysdeps/x86_64/multiarch/memmove-vec-unaligned-
> > erms.S (MEMMOVE):
> >  	  remove duplicate line on PREFETCH_ONE_SET macro
> 
> This should be:
> 
> 	* sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
> 	(PREFETCH_ONE_SET): Remove duplicate line.
> 
> > Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.
> > intel.com>
> 
> We don't use DCO, so this line is very misleading and should not be
> present.
> 

I will send a v2.

> Looks good otherwise.  Do you have commit access?

No
Florian Weimer - May 16, 2018, 5:01 p.m.
* Leonardo Sandoval:

> Florian, can someone check (perhaps you) if I now have commit access?

You will receive email from the overseers once this is set up, I
think.
On Tue, 2018-05-15 at 20:27 +0200, Florian Weimer wrote:
> * leonardo sandoval gonzalez:
> 
> > From: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com
> > >
> > 
> > Tested on 64-bit AVX machine
> > 
> >        	  * sysdeps/x86_64/multiarch/memmove-vec-unaligned-
> > erms.S (MEMMOVE):
> >  	  remove duplicate line on PREFETCH_ONE_SET macro
> 
> This should be:
> 
> 	* sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
> 	(PREFETCH_ONE_SET): Remove duplicate line.
> 
> > Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.
> > intel.com>
> 
> We don't use DCO, so this line is very misleading and should not be
> present.
> 
> Looks good otherwise.  Do you have commit access?

Florian, can someone check (perhaps you) if I now have commit access?

Leo

Patch

diff --git a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
index cbd0d077cf6..1404131bb71 100644
--- a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
+++ b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
@@ -90,7 +90,6 @@ 
 #  define PREFETCH_ONE_SET(dir, base, offset) \
 	PREFETCH ((offset)base); \
 	PREFETCH ((offset + dir * PREFETCH_SIZE)base); \
-	PREFETCH ((offset + dir * PREFETCH_SIZE)base); \
 	PREFETCH ((offset + dir * PREFETCH_SIZE * 2)base); \
 	PREFETCH ((offset + dir * PREFETCH_SIZE * 3)base)
 # else