ben@epmooch.UUCP (Rev. Ben A. Mesander) (04/06/91)
>In article <1991Apr7.142405.4581@ux1.cso.uiuc.edu> gerber@sirius.astro.uiuc.edu (Richard Gerber) writes: >Inspired by the recent discussion on optimization, I decided to run a >program I'm working on through the SAS/C 5.10a profiler. I found that >40% of the program time is being spent in the routine _CXM55. How do I >find out what function this is? Odds are that's a bit of the internal library division routine, or a bit of the internal library transcendental function code. The _CXMxx functions seem to be internal math subroutines used by the SAS/C compiler. What floating point format are you using? -- | ben@epmooch.UUCP (Ben Mesander) | "Cash is more important than | | ben%servalan.UUCP@uokmax.ecn.uoknor.edu | your mother." - Al Shugart, | | !chinet!uokmax!servalan!epmooch!ben | CEO, Seagate Technologies |
gerber@sirius.astro.uiuc.edu (Richard Gerber) (04/07/91)
Inspired by the recent discussion on optimization, I decided to run a program I'm working on through the SAS/C 5.10a profiler. I found that 40% of the program time is being spent in the routine _CXM55. How do I find out what function this is? I assume it's related to one of the many math function calls I'm using, but I can't figure out which one. If it's in the manuals, I can't find it. Thanks, Richard gerber@rigel.astro.uiuc.edu
bombadil@diku.dk (Kristian Nielsen) (04/08/91)
gerber@sirius.astro.uiuc.edu (Richard Gerber) writes: >Inspired by the recent discussion on optimization, I decided to run a >program I'm working on through the SAS/C 5.10a profiler. I found that >40% of the program time is being spent in the routine _CXM55. How do I >find out what function this is? From long hours of disassembly, I am pretty sure this is a function to multiply two double's (M for multiply, 5 for double). >I assume it's related to one of the many math function calls I'm using, >but I can't figure out which one. If it's in the manuals, I can't find >it. Yes, it wouldn't hurt anyone to have these documented in the manuals. >Thanks, >Richard >gerber@rigel.astro.uiuc.edu Kristian