similar to: floating point exceptions, and questions.

Hi all, The following is a trivial patch to psy.c, but combining the two loops saves a percent or two (according to gprof). ================================================================= --- psy.c Mon Jun 19 12:05:57 2000 +++ - Wed Jun 28 15:14:17 2000 @@ -550,12 +550,11 @@ frameno++; memset(flr,0,n*sizeof(double)); - for(i=0;i<n;i++)work[i]=fabs(f[i]); - - /* find the highest

Hi, I've read the Vorbis stereo documentation on square polar mapping and currently reading the source code to understand it. But there are some things which I don't quite understand and hope I can get some guidance on. I understand the decoding/decoupling part as it is the same as the one described in the stereo docs: From mapping0.c: /* channel coupling */

I posted a similar issue, but this relates to how to keep a function call from being lowered to an ISD:FABS. In this case, I don't want FABS (floating point absolute) to be converted into an operation). In this case, what happens is that clang generates the code for floating point absolute (fabs) as a subroutine call: %5 = load douuble %x, align 8 %call = call double @fabs(double %5)

you probably don't want the ~2% perf improvement below... but i figured i'd post it anyhow :) well... i do suggest the use of fabsf() rather than fabs() -- gcc doesn't seem smart enough to figure out fabs() when you ask it to compile "-mfpmath=sse". but the #else part uses properties of ieee-754 float layout which you may or may not be interested in (ab)using. basically

On Thu, 28 Apr 2005, Morten Ofstad wrote: > I have been gone for a while, finishing work on my Master's thesis... Hi Morten, congrats! :) > Now that I'm back I updated LLVM to the most recent version and found > that my FP_ABS SelectionDAGNode type and code generation was now > conflicting with the new FABS node type. I brought the rest of my local > modifications in

Chris Lattner wrote: > On Wed, 16 Mar 2005, Morten Ofstad wrote: >> The case which emits code for the special case of comparing against >> constant 0.0 does not return after generating it's code, so the normal >> compare is also generated! As far as I can tell it should return right >> after this: >> >> BuildMI(BB, X86::SAHF, 1); >> >>

Hello, I have been gone for a while, finishing work on my Master's thesis... Now that I'm back I updated LLVM to the most recent version and found that my FP_ABS SelectionDAGNode type and code generation was now conflicting with the new FABS node type. I brought the rest of my local modifications in line with the FABS implementation, so here is my patch that includes sqrt, sin and cos

Chris Lattner wrote: > The patches I didn't apply are these: > > 1. Match (Y < 0) ? -Y : Y -> FABS in the SelectionDAGISel.cpp file. We > already catch this at the DAG level. If we aren't, please let me know. OK, no problem - I was just told last time I tried to get my patch in that this was needed because the C++ frontend generated this code, I'm generating

Hi again! I have some other suggestions/modifications. 1. Bitrate scaling by frequency in vorbis_encode_init: bpch=(long)((float)nominal_bitrate/(float)channels*44100.0f/(float)rate); With this modification, if we use -b128 at 22khz files, the OGG file will be 128kbit/s, not 64kbit/s ... 2. Probable quality bug: I tested some floating point constants (recalculated them (see below) and encoded

Hello, Is there a way of generating unary floating point operations (like ISD::FABS, ISD::FSIN, ...) from C code using clang? I am building a backend for a machine that has hw support for these ops and I need a way to test them. Thanks, Alex -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hello, I'm not quite sure if this is the right list for my questions, please excuse if it isn't. 1. While using vorbislib (Win2K/Borland C++ Builder 5), I'm getting reproducible floating point exceptions at lots of sample rate/bit rate combinations. For example, the original encoder_example.c, with just the initialization line changed to

Solaris 5.8 has a quirky qsort that requires the ability to recognize elements as equal. here is a patch I have created to deal w/ this problem. I apologize if the patch is in the wrong format and would love to be corrected if wrong. I used the following to create the patch libvorbis-1.0> diff -u lib/psy.c lib/psy_new.c > libv.patch <p><p><p><p> --------------

Looking at the llvm code I have the impression that ordered fp comparisons should use exception throwing instructions. Is it true? Thanks, Rafael

On Fri, 13 Oct 2006, [UTF-8] Rafael Esp?ndola wrote: > Looking at the llvm code I have the impression that ordered fp > comparisons should use exception throwing instructions. Is it true? Nope, you want non-trapping instructions. If you use trapping instructions, C99 functions like isgreater will not work correctly with NAN's. -Chris -- http://nondot.org/sabre/ http://llvm.org/

> Nope, you want non-trapping instructions. If you use trapping > instructions, C99 functions like isgreater will not work correctly with > NAN's. How do I know when to use a trapping instruction? For example consider this C function: ------------------------ int f(float a, float b) { return a <= b; } ------------------------ gcc uses fcmpes instead of fcmps... >

#include <float.h> should define DBL_MIN and FLT_MIN, as well as DBL_MIN_EXP and FLT_MIN_EXP (such that 10eFLT_MIN_EXP is a valid float) This is a standard header file. But gcc now hides it away (mine was in /usr/lib/gcc-lib/i386-linux/<compiler-version>/include/ ) BTW - any corrections for my surf.chimie.uqam.ca/~kruus/vorbis/ stuff? Erik Kruus <kruus@on2.com> --- >8

On Tue, Apr 15, 2014 at 12:33 PM, Peter Meerwald <pmeerw at pmeerw.net> wrote: > Signed-off-by: Peter Meerwald <pmeerw at pmeerw.net> > --- > libspeexdsp/resample.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) > > diff --git a/libspeexdsp/resample.c b/libspeexdsp/resample.c > index e32ca45..a19b997 100644 > ---

Signed-off-by: Peter Meerwald <pmeerw at pmeerw.net> --- libspeexdsp/resample.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/libspeexdsp/resample.c b/libspeexdsp/resample.c index e32ca45..a19b997 100644 --- a/libspeexdsp/resample.c +++ b/libspeexdsp/resample.c @@ -85,7 +85,7 @@ static void speex_free (void *ptr) {free(ptr);} #ifdef FIXED_POINT #define

Well, here you are. 24k; sorry if I'm not supposed to put this size things in your mailbox, didn't know where else to put it. And you all are subscribed to vorbis-dev, after all. I'm not that good at breaking patches apart, so it's one big patch. Sorry. Overview: configure.in make profiling easier & more useful decoder-example.c (#if 0'ed) dither output;

On Wed, 16 Mar 2005, Morten Ofstad wrote: > Hello, > > I didn't get any reply to my previous mail about adding floating point > intrinsics to the X86 pattern instruction selector... And I could really need > some help. Sorry about that, it slipped through the cracks. :( > Anyway, I think my confusion was caused partly by an already > existing bug in the instruction