similar to: libvorbis: NO_FLOAT_MATH_LIB patch

Apps building against libvorbis 1.0b4 will fail to link on Solaris 2.7 because libvorbis uses atanf, sqrtf, etc., which aren't available on that platform. In lib/os.h, there are a couple of workarounds (ifdef DARWIN, ifndef GNUC), but I'm building on Solaris with gcc, so this doesn't help me any. :) Attached is a patch to configure.in which checks for sqrtf in libm, and if it finds

This is also pretty simple. Just a small patch to configure.in to set the right compiler flags and changes to os.h to (a) not have deeply nested ifdefs, making the file needlessly hard to read, (b) break up the sections by functionality and (c) don't assume that every non-Win32 has the floating versions of math library functions. -tim Index: configure.in

We've got both an sqrtf and an expf opcode in our architecture. If I call sqrt() on the C side, I see the sqrtf opcode show up in the generated assembly. However, if I call exp() on the C side, I don't see the expf opcode show up on the generated assembly, I see a call to an exp function from libm. Here's what we've got in our TargetInstrinfo.td file for both of these

Hi I'm having a problem compiling opus-tools-0.1.7. Version opus-tools-0.1.6 seems to compile OK. I've tried with opus-1.0.3 and opus-1.1-beta. The errors are like this:- "undefined reference to `sqrtf'" etc. This OS is Peppermint Three, similar to Ubuntu 12.04. It uses:- gcc --version gcc (Ubuntu/Linaro 4.6.3-1ubuntu5) 4.6.3 Google says it's maybe something to do

sqrtf is detected by code in SelectionDAGBuilder.cpp. This gets turns into a FSQRT ISD node type that the target can handle just like any other ISD node. If the target doesn't mark ISD::FSQRT as Legal or Custom then ExpandNode in LegalizeDAG.cpp turns it back into a sqrtf libcall. On Sun, Jan 20, 2013 at 11:34 AM, David Given <dg at cowlark.com> wrote: > On 20/01/13 19:20,

On 20/01/13 19:20, Caldarale, Charles R wrote: [...] > That's because there is no llvm.ceil.* intrinsic defined in include/llvm/Intrinsics.td for 3.2 Ah. Yes, that would explain it... does this mean that I can rely on all the intrinsics listed existing for the common types (int, float, double)? Or should I be trying to follow the libc call route? I've noticed that llc is successfully

Hi, Am trying to run beta4 encoder on a Sun machine. Want to compare floor0 and floor1. Do not have autoconf and automake. Got the encoder to compile with a simple makefile. The only define i am using is NO_FLOAT_MATH_LIB. The encoded file when played(winamp) gives out plain noise. Am i missing any other flags/defs? -Cathy <p>--------------------------------- Do you Yahoo!? Yahoo!

Hi! in the llvm code there are several places with hardcoded function names for e.g. sin, sinf, sqrt, sqrtf etc., namely ConstantFolding.cpp InlineCost.cpp SelectionDAGBuilder.cpp IntrinsicLowering.cpp TargetLowering.cpp my question is: wouldn't it be beneficial to use intrinsics for this? for example a c/c++ frontend (clang) could translate the function calls to intrinsics and then in a

Eliminate undefined references to powf, sqrtf and friends. Index: MultiSource/Benchmarks/Bullet/Makefile =================================================================== --- MultiSource/Benchmarks/Bullet/Makefile (revision 92512) +++ MultiSource/Benchmarks/Bullet/Makefile (working copy) @@ -1,7 +1,7 @@ LEVEL = ../../../ PROG = bullet CPPFLAGS += -I$(PROJ_SRC_DIR)/include -DNO_TIME -LDFLAGS

Hi, >> in the llvm code there are several places with hardcoded function >> names for e.g. sin, sinf, sqrt, sqrtf etc., namely >> ConstantFolding.cpp >> InlineCost.cpp >> SelectionDAGBuilder.cpp >> IntrinsicLowering.cpp >> TargetLowering.cpp >> >> my question is: wouldn't it be beneficial to use intrinsics for this? >> for example a

Hi, Seems like it's not linking with libm. I suspect it has to do with linking statically with libopus.a (is that intended?). Maybe opus-tools relies on the fact that libopus is linked with libm and doesn't explicitly add it? Greg? Jean-Marc On 09/24/2013 06:09 AM, bat guano wrote: > Hi > I'm having a problem compiling opus-tools-0.1.7. > Version opus-tools-0.1.6 seems to

On Fri, 02 Mar 2012 13:55:18 +0100 Jochen Wilhelmy <jochen.wilhelmy at googlemail.com> wrote: > Hi! > > in the llvm code there are several places with hardcoded function > names for e.g. sin, sinf, sqrt, sqrtf etc., namely > ConstantFolding.cpp > InlineCost.cpp > SelectionDAGBuilder.cpp > IntrinsicLowering.cpp > TargetLowering.cpp > > my question is:

There are at least three cases: * Linux: check credentials and pid from client; restrict permissions from server * BSD: check credentials only from client; restrict permissions from server * Solaris: wide open --- configure.in | 4 ++-- src/control.c | 11 ++++++++++- src/control_common.h | 1 + src/tincctl.c | 38 ++++++++++++++++++++++++++++++++++---- 4

Dan, and anyone else interested. I am not sure if this has been discussed before, but I do have a case when the following logic fails to work: lib/Analysis/ConstantFolding.cpp static Constant *ConstantFoldBinaryFP(double (*NativeFP)(double, double), double V, double W, Type *Ty) { sys::llvm_fenv_clearexcept(); V = NativeFP(V, W); if

On Fri, 02 Mar 2012 16:05:17 +0100 Duncan Sands <baldrick at free.fr> wrote: > Hi, > > >> in the llvm code there are several places with hardcoded function > >> names for e.g. sin, sinf, sqrt, sqrtf etc., namely > >> ConstantFolding.cpp > >> InlineCost.cpp > >> SelectionDAGBuilder.cpp > >> IntrinsicLowering.cpp > >>

On 27 June 2011 18:58, Chris Lattner <clattner at apple.com> wrote: > > On Jun 25, 2011, at 11:59 AM, Jay Foad wrote: > >> I've played with the Clang bits of this. The biggest problem I've >> found is that Clang uses LLVM's type resolution not just for >> forward-declared structs/classes/unions, which convert >> straightforwardly to the new

Hi Micheal, Thanks for the blog post. Just like to point out few things that I thought is related to FP Numerics. LLVM could do some additional transformation with "sqrt" and "division" under fast math on X86 like 1/sqrt(x)* 1/sqrt(x) to 1/x. These are long latency instructions and could get benefit if enabled under unsafe math. Also are we considering doing such FP

Hi Sergei, The degree to which LLVM actually makes any guarantees about IEEE arithmetic precision is ambiguous. LangRef, for one, doesn't even mention it (it mentions formats, but nothing else). The de-facto way of interpreting holes in LangRef is to consider how the IR is used by clang and follow the path up into the C and/or C++ standards and then work from there. C describes a binding to

Hi, I have compiled the Vorbis libraries and tools on Solaris 8 using Forte 6 update 2. I am now trying to run ogg123 but nothing happens: $ ogg123 -v Pateras.ogg Audio Device: Sun audio driver output Author: Christian Weisgerber <naddy@openbsd.org> Comments: Outputs to the sun audio system. I have also worked around bug #191 by defining HAVE_SQRTF:

Hello! Testing for a library is not sufficient if you also need the headers. Many distributions install libraries without headers by default. This is the case for Fedora Core 3 and libsysfs. The library is in sysfsutils, but the headers are in sysfsutils-devel. The patch for configure.in tests for both the headers and the library and makes the configure script print a warning if libsysfs