<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=us-ascii"> <meta name="Generator" content="Microsoft Word 15 (filtered medium)"> <style></style> </head> <body lang="EN-US" link="#0563C1" vlink="#954F72"> <div class="WordSection1"> Thanks for replying!<o:p></o:p> <o:p> </o:p> I’m not writing any OpenCL code directly. I’m using the ArrayFire C++ library, which JIT generates OpenCL and compiles it when the process runs. That is, I have no OpenCL source code, and I don’t write any C or C++ code that interacts with kernels directly or even knows what they are.<o:p></o:p> <o:p> </o:p> My unit tests pass, though, and so it seems that the calculations are being done correctly.<o:p></o:p> <o:p> </o:p> I wonder how ArrayFire decides on sizes of chunks of work, and I wonder how much of a performance hit I’m taking on Bay Trail because of this.<o:p></o:p> <o:p> </o:p> Can you explain what “curbe” stands for and what’s really going on in __gen_gpgpu_get_curbe_size()?<o:p></o:p> <o:p> </o:p> <div> --<o:p></o:p> Thomas E. Vaughan<o:p></o:p> 720 201 7058 (cell)<o:p></o:p> </div> <o:p> </o:p> <div style="border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in 4.0pt"> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in"> From: Yang, Rong R [mailto:rong.r.yang@intel.com] Sent: January 31, 2018 02:08 To: Vaughan, Thomas E <tevaugha@ball.com>; beignet@lists.freedesktop.org Subject: RE: warning, curbe size exceed limitation.<o:p></o:p> </div> </div> <o:p> </o:p> Do you use this array as kernel’s argument?<o:p></o:p> If the argument’s size exceeds the argument buffer’s size, beignet will print this warning.<o:p></o:p> And IVB’s argument buffer size is larger than BayTrail.<o:p></o:p> <o:p> </o:p> Thanks,<o:p></o:p> <a name="_MailEndCompose">Yang Rong</a><o:p></o:p> <div style="border:none;border-left:solid blue 1.5pt;padding:0in 0in 0in 4.0pt"> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in"> <a name="_____replyseparator"></a>From: Beignet [<a href="mailto:beignet-bounces@lists.freedesktop.org">mailto:beignet-bounces@lists.freedesktop.org</a>] On Behalf Of Vaughan, Thomas E Sent: Wednesday, January 31, 2018 2:04 AM To: <a href="mailto:beignet@lists.freedesktop.org">beignet@lists.freedesktop.org</a> Subject: [Beignet] warning, curbe size exceed limitation.<o:p></o:p> </div> </div> <o:p> </o:p> Thanks to Rebecca Palmer and to Mark Thompson for replying to my last plea for help over the weekend.<o:p></o:p> <o:p> </o:p> Now that I have started to get things working with the Atom GPU on Debian, I have noticed an odd message when I run some ArrayFire code that was tested on Ivy Bridge.<o:p></o:p> <o:p> </o:p> My application prints out “warning, curbe size exceed limitation” in various places.<o:p></o:p> <o:p> </o:p> For example, at one point I try to tile a 7x176 array into a 7x176x7 array, and that message is issued twice.<o:p></o:p> <o:p> </o:p> Does anyone have a clue about what’s going on?<o:p></o:p> <o:p> </o:p> --<o:p></o:p> Thomas E. Vaughan<o:p></o:p> 720 201 7058 (cell)<o:p></o:p> <o:p> </o:p> <div> This message and any enclosures are intended only for the addressee. Please notify the sender by email if you are not the intended recipient. If you are not the intended recipient, you may not use, copy, disclose, or distribute this message or its contents or enclosures to any other person and any such actions may be unlawful. Ball reserves the right to monitor and review all messages and enclosures sent to or from this email address.<o:p></o:p> </div> </div> </div> </div> <DIV> This message and any enclosures are intended only for the addressee. Please notify the sender by email if you are not the intended recipient. If you are not the intended recipient, you may not use, copy, disclose, or distribute this message or its contents or enclosures to any other person and any such actions may be unlawful. Ball reserves the right to monitor and review all messages and enclosures sent to or from this email address. </DIV></body> </html>