Comments on: ARM compiler shoot-out http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/ Everything is broken Mon, 30 Aug 2010 09:33:38 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: Mans http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-178 Mans Wed, 19 Aug 2009 12:37:28 +0000 http://hardwarebug.org/?p=150#comment-178 I measured the CPU time needed to decode the samples. I measured the CPU time needed to decode the samples.

]]> By: Fredrik http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-175 Fredrik Wed, 19 Aug 2009 11:06:58 +0000 http://hardwarebug.org/?p=150#comment-175 Hi, How did you measure the performance? Hi,
How did you measure the performance?

]]> By: Mans http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-88 Mans Wed, 12 Aug 2009 14:02:06 +0000 http://hardwarebug.org/?p=150#comment-88 Yes, I'm running Linux on the board. Yes, I’m running Linux on the board.

]]> By: asn http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-87 asn Wed, 12 Aug 2009 13:38:28 +0000 http://hardwarebug.org/?p=150#comment-87 Hi, How did you measure the performance ? Have you put any Linux distro on the board ? Hi,
How did you measure the performance ? Have you put any Linux distro on the board ?

]]> By: Mans http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-85 Mans Tue, 11 Aug 2009 02:26:42 +0000 http://hardwarebug.org/?p=150#comment-85 I enabled gcc compatibility in RVCT so those options are exactly what I used. I'm not sure what RVCT-native options they map to. I did not use --vectorize for the simple reason that it fails to compile FFmpeg with vectorisation enabled. I enabled gcc compatibility in RVCT so those options are exactly what I used. I’m not sure what RVCT-native options they map to.

I did not use –vectorize for the simple reason that it fails to compile FFmpeg with vectorisation enabled.

]]> By: mh http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-84 mh Tue, 11 Aug 2009 02:12:55 +0000 http://hardwarebug.org/?p=150#comment-84 I doubt that the RVCT options you quoted were the same you actually used. Looks more like a copy paste issue to me. Did you use --vectorize with RVCT? Otherwise you might not be able to take full advantage of NEON. The capability of the compiler to vectorize code depends quite a bit on the C code structure, though. I doubt that the RVCT options you quoted were the same you actually used. Looks more like a copy paste issue to me.

Did you use –vectorize with RVCT? Otherwise you might not be able to take full advantage of NEON. The capability of the compiler to vectorize code depends quite a bit on the C code structure, though.

]]> By: kert http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-82 kert Wed, 05 Aug 2009 15:22:44 +0000 http://hardwarebug.org/?p=150#comment-82 Does anyone know if any other optflags beyond -Os now work reliably in ARM code ? A year or so back, -O2 produced very weird assembler-level mess here and there in the code ( a large c++ lib ) and from newsgroups i basically got this: dont use anything beyond -Os Does anyone know if any other optflags beyond -Os now work reliably in ARM code ?
A year or so back, -O2 produced very weird assembler-level mess here and there in the code ( a large c++ lib ) and from newsgroups i basically got this: dont use anything beyond -Os

]]> By: Mans http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-81 Mans Wed, 05 Aug 2009 08:42:17 +0000 http://hardwarebug.org/?p=150#comment-81 I've updated the post with the flags used. I’ve updated the post with the flags used.

]]> By: Mans http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-80 Mans Wed, 05 Aug 2009 08:32:53 +0000 http://hardwarebug.org/?p=150#comment-80 With inline asm enabled, the difference for MP3 is more in line with the other tests. I don't remember the exact numbers. With inline asm enabled, the difference for MP3 is more in line with the other tests. I don’t remember the exact numbers.

]]> By: Mans http://hardwarebug.org/2009/08/05/arm-compiler-shoot-out/comment-page-1/#comment-79 Mans Wed, 05 Aug 2009 08:25:48 +0000 http://hardwarebug.org/?p=150#comment-79 I have already reported a number of errors to TI, and the compiler team is working on fixing them. I have already reported a number of errors to TI, and the compiler team is working on fixing them.

]]>