From steve.white at aei.mpg.de Fri Mar 3 02:57:20 2006 From: steve.white at aei.mpg.de (Steve White) Date: Fri, 3 Mar 2006 09:57:20 +0100 Subject: [Patches] Re: snprintf.c : reduce compiler warnings In-Reply-To: <20060213134453.GA22546@xeon01.aei.mpg.de> References: <20060213134453.GA22546@xeon01.aei.mpg.de> Message-ID: <20060303085720.GC23795@xeon01.aei.mpg.de> Applied 2.02.06, following discussion in Cactus call. -- Steve White : Programmer Max-Planck-Institut f?r Gravitationsphysik Albert-Einstein-Institut Am M?hlenberg 1, D-14476 Golm, Germany +49-331-567-7625 From steve.white at aei.mpg.de Fri Mar 3 02:57:48 2006 From: steve.white at aei.mpg.de (Steve White) Date: Fri, 3 Mar 2006 09:57:48 +0100 Subject: [Patches] Re: ParseFile.c: reduce compiler warnings In-Reply-To: <20060213134840.GB22546@xeon01.aei.mpg.de> References: <20060213134840.GB22546@xeon01.aei.mpg.de> Message-ID: <20060303085748.GD23795@xeon01.aei.mpg.de> Applied 2.02.06, following discussion in Cactus call. -- Steve White : Programmer Max-Planck-Institut f?r Gravitationsphysik Albert-Einstein-Institut Am M?hlenberg 1, D-14476 Golm, Germany +49-331-567-7625 From steve.white at aei.mpg.de Fri Mar 3 02:58:10 2006 From: steve.white at aei.mpg.de (Steve White) Date: Fri, 3 Mar 2006 09:58:10 +0100 Subject: [Patches] Re: DefaultTimers.c: reduce compiler warnings In-Reply-To: <20060213135216.GC22546@xeon01.aei.mpg.de> References: <20060213135216.GC22546@xeon01.aei.mpg.de> Message-ID: <20060303085810.GE23795@xeon01.aei.mpg.de> Applied 2.02.06, following discussion in Cactus call. -- Steve White : Programmer Max-Planck-Institut f?r Gravitationsphysik Albert-Einstein-Institut Am M?hlenberg 1, D-14476 Golm, Germany +49-331-567-7625 From steve.white at aei.mpg.de Fri Mar 3 02:58:34 2006 From: steve.white at aei.mpg.de (Steve White) Date: Fri, 3 Mar 2006 09:58:34 +0100 Subject: [Patches] Re: ScheduleSorter.c: reduce compiler warnings In-Reply-To: <20060213140024.GD22546@xeon01.aei.mpg.de> References: <20060213140024.GD22546@xeon01.aei.mpg.de> Message-ID: <20060303085834.GF23795@xeon01.aei.mpg.de> Applied 2.02.06, following discussion in Cactus call. -- Steve White : Programmer Max-Planck-Institut f?r Gravitationsphysik Albert-Einstein-Institut Am M?hlenberg 1, D-14476 Golm, Germany +49-331-567-7625 From steve.white at aei.mpg.de Fri Mar 3 02:59:37 2006 From: steve.white at aei.mpg.de (Steve White) Date: Fri, 3 Mar 2006 09:59:37 +0100 Subject: [Patches] Re: Complex.c: reduce compiler warnings In-Reply-To: <20060214152452.GB3733@xeon01.aei.mpg.de> References: <20060214152452.GB3733@xeon01.aei.mpg.de> Message-ID: <20060303085937.GG23795@xeon01.aei.mpg.de> Applied 2.02.06, following discussion in Cactus call. T.G. suggests that the origins of code to the complex sqrt function be tracked down and reported. -- Steve White : Programmer Max-Planck-Institut f?r Gravitationsphysik Albert-Einstein-Institut Am M?hlenberg 1, D-14476 Golm, Germany +49-331-567-7625 From steve.white at aei.mpg.de Fri Mar 3 04:52:33 2006 From: steve.white at aei.mpg.de (Steve White) Date: Fri, 3 Mar 2006 11:52:33 +0100 Subject: [Patches] CartGrid3D: reduce compiler warnings Message-ID: <20060303105233.GA25120@xeon01.aei.mpg.de> Hello, This is part of a project to reduce the compiler warnings generated by Cactus Flesh code. Intel icc complains about lots of questionabale type conversions in CartGrid3D/src/Symmetry.c. Find attached a patch, to be applied from within the CartGrid3D directory. Notes ----- * Code is macros within macros within macros that depend on outside entities. Sometimes a type name is passed. * The innermost macro performs the type conversion * A typename is passed, but unfortunately, it isn't always the type being converted. Sometimes it's a CCTK_COMPLEX type, but the type being converted is only a component of the complex. Solution -------- Invented a new preprocessor symbol, NUMBER_TYPE, which indicates the of a scalar multiplication. Tests ----- Ran full set of distributions testsuites both with and without the changes. Results were identical. -- Steve White : Programmer Max-Planck-Institut f?r Gravitationsphysik Albert-Einstein-Institut Am M?hlenberg 1, D-14476 Golm, Germany +49-331-567-7625 -------------- next part -------------- ? CartGrid3D.diff Index: src/Symmetry.c =================================================================== RCS file: /cactusdevcvs/CactusBase/CartGrid3D/src/Symmetry.c,v retrieving revision 1.41 diff -u -r1.41 Symmetry.c --- src/Symmetry.c 29 Jan 2005 17:11:13 -0000 1.41 +++ src/Symmetry.c 3 Mar 2006 10:34:02 -0000 @@ -311,8 +311,9 @@ { \ for (ii = 0; ii < gdata.nghostzones[dir/2]; ii++) \ { \ - _var[INDEX_3D (lsh, i, j, k)] NUMBER_PART = GFSym[vindex][dir] * \ - _var[INDEX_3D (lsh, iii, jjj, kkk)] NUMBER_PART; \ + _var[INDEX_3D (lsh, i, j, k)] NUMBER_PART \ + = (NUMBER_TYPE)(GFSym[vindex][dir] \ + * _var[INDEX_3D (lsh, iii, jjj, kkk)] NUMBER_PART); \ } \ } \ } \ @@ -326,8 +327,9 @@ { \ for (ii = lsh[dir/2]-gdata.nghostzones[dir/2]; ii < lsh[dir/2]; ii++) \ { \ - _var[INDEX_3D (lsh, i, j, k)] NUMBER_PART = GFSym[vindex][dir] * \ - _var[INDEX_3D (lsh, iii, jjj, kkk)] NUMBER_PART; \ + _var[INDEX_3D (lsh, i, j, k)] NUMBER_PART \ + = (NUMBER_TYPE)(GFSym[vindex][dir] \ + * _var[INDEX_3D (lsh, iii, jjj, kkk)] NUMBER_PART); \ } \ } \ } \ @@ -446,55 +448,91 @@ */ #define NUMBER_PART .Re +#define NUMBER_TYPE CCTK_REAL SYMMETRY_FUNCTION(CCTK_COMPLEX,R) +#undef NUMBER_TYPE #ifdef HAVE_CCTK_COMPLEX8 +#define NUMBER_TYPE CCTK_REAL4 SYMMETRY_FUNCTION(CCTK_COMPLEX8,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_COMPLEX16 +#define NUMBER_TYPE CCTK_REAL8 SYMMETRY_FUNCTION(CCTK_COMPLEX16,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_COMPLEX32 +#define NUMBER_TYPE CCTK_REAL16 SYMMETRY_FUNCTION(CCTK_COMPLEX32,R) +#undef NUMBER_TYPE #endif #undef NUMBER_PART #define NUMBER_PART .Im +#define NUMBER_TYPE CCTK_REAL SYMMETRY_FUNCTION(CCTK_COMPLEX,I) +#undef NUMBER_TYPE +#define NUMBER_TYPE CCTK_REAL4 #ifdef HAVE_CCTK_COMPLEX8 SYMMETRY_FUNCTION(CCTK_COMPLEX8,I) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_COMPLEX16 +#define NUMBER_TYPE CCTK_REAL8 SYMMETRY_FUNCTION(CCTK_COMPLEX16,I) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_COMPLEX32 +#define NUMBER_TYPE CCTK_REAL16 SYMMETRY_FUNCTION(CCTK_COMPLEX32,I) +#undef NUMBER_TYPE #endif #undef NUMBER_PART #define NUMBER_PART +#define NUMBER_TYPE CCTK_BYTE SYMMETRY_FUNCTION(CCTK_BYTE,R) +#undef NUMBER_TYPE +#define NUMBER_TYPE CCTK_INT SYMMETRY_FUNCTION(CCTK_INT,R) +#undef NUMBER_TYPE #ifdef HAVE_CCTK_INT1 +#define NUMBER_TYPE CCTK_INT1 SYMMETRY_FUNCTION(CCTK_INT1,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_INT2 +#define NUMBER_TYPE CCTK_INT2 SYMMETRY_FUNCTION(CCTK_INT2,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_INT4 +#define NUMBER_TYPE CCTK_INT4 SYMMETRY_FUNCTION(CCTK_INT4,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_INT8 +#define NUMBER_TYPE CCTK_INT8 SYMMETRY_FUNCTION(CCTK_INT8,R) +#undef NUMBER_TYPE #endif +#define NUMBER_TYPE CCTK_REAL SYMMETRY_FUNCTION(CCTK_REAL,R) +#undef NUMBER_TYPE #ifdef HAVE_CCTK_REAL4 +#define NUMBER_TYPE CCTK_REAL4 SYMMETRY_FUNCTION(CCTK_REAL4,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_REAL8 +#define NUMBER_TYPE CCTK_REAL8 SYMMETRY_FUNCTION(CCTK_REAL8,R) +#undef NUMBER_TYPE #endif #ifdef HAVE_CCTK_REAL16 +#define NUMBER_TYPE CCTK_REAL16 SYMMETRY_FUNCTION(CCTK_REAL16,R) +#undef NUMBER_TYPE #endif #undef NUMBER_PART From schnetter at cct.lsu.edu Mon Mar 6 01:09:04 2006 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Mon, 6 Mar 2006 01:09:04 -0600 Subject: [Patches] Define HAVE_CCTK_BYTE etc. Message-ID: The flesh defines HAVE_CCTK_* macros to indicate what types are available. This patch completes this by defining such macros for all Cactus types. This makes it easier to treat all types in the same way and simplifies type-generic "templates" in C, e.g. for LocalReduce. -erik -- Erik Schnetter My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: have_cctk_byte.diff Type: application/octet-stream Size: 1694 bytes Desc: not available Url : http://www.cactuscode.org/pipermail/patches/attachments/20060306/319ae1ee/attachment.obj From tradke at aei.mpg.de Thu Mar 9 07:14:31 2006 From: tradke at aei.mpg.de (Thomas Radke) Date: Thu, 09 Mar 2006 14:14:31 +0100 Subject: [Patches] fix order of external/system libs on the linker command line Message-ID: <44102A37.7040504@aei.mpg.de> With the new mechanism to configure Cactus with external packages, external libraries (as listed in bindings/Configuration/make.link) are appended to the linker command line _after_ all other libraries (including any system/compiler-specific libs). This is a problem if the external libraries themself need compiler-specific libs. The old mechanism worked fine though: scripts in lib/make/extra/ listed their external libs in config-data/make.extra.defn which is included in config-data/make.config.defn which _then_ appends system/compiler-specific libs. The attached patch makes sure that definitions in the bindings/Configuration/make.link file are treated equivalently to definitions in the config-data/make.extra.defn file. -- Cheers, Thomas. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: patch Url: http://www.cactuscode.org/pipermail/patches/attachments/20060309/05ec7fda/attachment.pl From tradke at aei.mpg.de Thu Mar 9 09:34:12 2006 From: tradke at aei.mpg.de (Thomas Radke) Date: Thu, 09 Mar 2006 16:34:12 +0100 Subject: [Patches] fix order of external/system libs on the linker command line Message-ID: <44104AF4.1030403@aei.mpg.de> Applied on 9 March 2006. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: patch Url: http://www.cactuscode.org/pipermail/patches/attachments/20060309/334e7f28/attachment.pl