<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <meta name="Generator" content="Microsoft Word 14 (filtered medium)"> <style></style> </head> <body lang="ZH-CN" link="blue" vlink="purple"> <div class="WordSection1"> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">I am not an expert on the Cache related thing, basically constant cache is part of the Read-only cache lies in L3.<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">From the code in src/intel/intel_gpgpu.c, below logic is for IvyBridge:<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">If (slmMode) <o:p></o:p></span></p> <p class="MsoNormal" style="text-indent:21.0pt"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">allocate 64KB constant cache<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">Else<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D"> Allocate 32KB constant cache<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">I am not sure is there any big performance difference between less than or greater than the real constant cache size in L3.<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">I simply wrote a random-selected number 512KB as the up limit in driver API.<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">But it did deserve to investigate the performance change according to used constant size.<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">If we use too much constant larger than the constant cache allocated from L3,<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">I think it will definitely cause constant cache data swap in-out frequently. Right?<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">If you would like to contribute any performance test to beignet, or any other open source test suite, it would be really appreciated!<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">Thanks!<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1F497D">Ruiling<o:p></o:p></span></p> <div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt"> <div> <div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm"> <p class="MsoNormal"><b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> Beignet [mailto:beignet-bounces@lists.freedesktop.org] <b>On Behalf Of </b>Tony Moore<br> <b>Sent:</b> Wednesday, November 26, 2014 6:45 AM<br> <b>To:</b> beignet@lists.freedesktop.org<br> <b>Subject:</b> Re: [Beignet] Combine Loads from __constant space<o:p></o:p></span></p> </div> </div> <p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p> <p class="MsoNormal"><span lang="EN-US">Another question I had about __constant, was there seems to be no limit. I'm using __constant for every read-only parameter now totalling 1500Kb and this test now runs in 32ms. So, is there a limit? Is this method reliable? Can driver do this implicitly on all read-only buffers?<o:p></o:p></span></p> <div> <p class="MsoNormal"><span lang="EN-US">thanks<o:p></o:p></span></p> </div> <p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p> <div> <p class="MsoNormal"><span lang="EN-US">On Tue Nov 25 2014 at 2:11:26 PM Tony Moore <<a href="mailto:tonywmnix@gmail.com">tonywmnix@gmail.com</a>> wrote:<o:p></o:p></span></p> <p class="MsoNormal"><span lang="EN-US">Hello,<o:p></o:p></span></p> <div> <p class="MsoNormal"><span lang="EN-US">I notice that reads are not being combined when I use __constant on a read-only kernel buffer. Is this something that can be improved?<o:p></o:p></span></p> </div> <div> <p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p> </div> <div> <div> <p class="MsoNormal"><span lang="EN-US">In my kernel there are many loads from a read-only data structure. When I use the __global specifier for the memory space I see a total of 33 send instructions and a runtime of 81ms. When I use the __constant specifier, I see 43 send instructions and a runtime of 40ms. I'm hoping that combining the loads could improve performance further.<o:p></o:p></span></p> </div> </div> <div> <p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p> </div> <div> <p class="MsoNormal"><span lang="EN-US">thanks!<o:p></o:p></span></p> </div> <div> <p class="MsoNormal"><span lang="EN-US">tony<o:p></o:p></span></p> </div> </div> </div> </div> </body> </html>