[if gte mso 9]>

Re: [edk2] MdePkg: Enhance CompareGuid performance by using CompareMem instead of nested unaligned read calls

Subject: Re: [edk2] MdePkg: Enhance CompareGuid performance by using CompareMem instead of nested unaligned read calls

From: "Kinney, Michael D" <michael.d.kinney@intel.com>

To: "edk2-devel@lists.sourceforge.net" <edk2-devel@lists.sourceforge.net>

Date: 2012-05-30 00:17:57

Eugene,

 

Did you compare the performance of the current and proposed solutions?  An optimizing compiler inlines all of these so I think the current code should be faster on CPUs that support unaligned accesses.

 

Thanks,

 

Mike

 

From: Cohen, Eugene [mailto:eugene@hp.com]
Sent: Tuesday, May 29, 2012 6:25 AM
To: edk2-devel@lists.sourceforge.net
Subject: Re: [edk2] MdePkg: Enhance CompareGuid performance by using CompareMem instead of nested unaligned read calls

 

Corrected patch, sorry for the inconvenience.

 

From: Cohen, Eugene
Sent: Tuesday, May 29, 2012 7:16 AM
To: edk2-devel@lists.sourceforge.net
Subject: MdePkg: Enhance CompareGuid performance by using CompareMem instead of nested unaligned read calls

 

Dear  MdePkg Maintainer,

 

Currently CompareGuid uses ReadUnaligned64 calls which in turn call unaligned 32 and 16 bit calls which is time consuming.  This patch replaces those calls with a single call to CompareMem.

 

MdePkg: Enhance CompareGuid performance by using CompareMem instead of nested unaligned read calls

 

Contributed-under: TianoCore Contribution Agreement 1.0

Signed-off-by: Eugene Cohen eugene@hp.com