similar to: [LLVMdev] ARM NEON intrinsics in clang

On 26 September 2013 17:52, Stanislav Manilov <stanislav.manilov at gmail.com>wrote: > To answer your question I am testing on a pandaboard currently, which has > an arm cortex-a9 processor, which I think is 64-bit. > Cortex-A9 is still 32-bits, so you'll have all support you need. ;) however it doesn't if I remove the -ffreestanding flag. I need to figure > this out

> To answer your question I am testing on a pandaboard currently, which has >> an arm cortex-a9 processor, which I think is 64-bit. >> > > Cortex-A9 is still 32-bits, so you'll have all support you need. ;) > Ah, Okay, embarrassing... however it doesn't if I remove the -ffreestanding flag. I need to figure >> this out next. >> > > Can you at

Hello LLVM Devs, I am starting my PhD on Automatic Parallelization for DSP and want to play with some ARM NEON intrinsics for a start. I spent the last three days trying to compile a version of LLVM that would allow me to compile sources that contain these intrinsics, but with no success. In the process I found out that clang doesn't support NEON (as per

Hello Tim, > I spent the last three days trying to compile a version of LLVM that would > > allow me to compile sources that contain these intrinsics, but with no > success. > > Ok. This we can probably help with. Did you manage to build a version > of Clang (preferably from git/subversion)? > Yes, I managed to build the latest (r191291) svn revision of LLVM + clang. If

Hello LLVM Devs, Thanks for helping me previously to cross-compile for ARM, I managed to get a working toolchain and am currently having fun compiling different toy problems and running them on a pandaboard. As part of my research I am trying to implement the ARM NEON Intrinsics in the PowerPC LLVM backend. I am still at the beginning of my efforts and am not yet familiar with either the ARM or

On Wed, May 4, 2016 at 3:10 PM, Stanislav Manilov < stanislav.manilov at gmail.com> wrote: > As in "look at the source of clang" or as in "look at the -S -emit-llvm" > output? If you mean the former, then would that be easy for someone who > hasn't seen the clang source before? > Generally the latter - then potentially set some breakpoints & look at

Hi Philip, I forgot to mention that I was ignoring loop-independent dependences. If I don't I get an inconsistent, ordered, anti, loop-independent dependence and an inconsistent, ordered, flow, loop-carried dependence for example A. At the same time I get just a consistent, ordered, anti, loop-independent dependence for example B. Here's the .ll code for example A: *; Function Attrs:

Hello Hal, I am not very familiar with the DSP capabilities of PowerPC, but I imagine there will be instructions for simple vector operations like vector addition, multiplication, etc. so for these I imagine the implementation would consist of just outputting the correct instruction. However, for NEON instructions like the reciprocal step (see

Optimize kf_bfly4 function using ARM NEON intrinsics for SoCs that have NEON VFP unit As initial step, only targetting ARMv7-VFP based SoCs. To enable this optimization, use --enable-armv7-neon-float when running configure command. This is disabled by default. --- Makefile.am | 16 ++++ celt/_kiss_fft_guts.h | 13 +++ celt/arm/kiss_fft_neon.c | 211

Viswanath Puttagunta wrote: > + SUMM = vdupq_n_f32(0); It kills me that there's no intrinsic for VMOV.F32 d0, #0 (or at least I couldn't find one), so this takes two instructions instead of one. > + /* Consume 4 elements in x vector and 8 elements in y > + * vector. However, the 8'th element in y never really gets > + * touched in this loop. So, if len == 4,

Oh, I see, that makes a lot of sense. How do I build the pass pipeline? On Thu, Feb 11, 2016 at 5:54 PM, Mehdi Amini <mehdi.amini at apple.com> wrote: > > On Feb 11, 2016, at 9:49 AM, Stanislav Manilov via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > > > Hello, > > > > I am used to specifying dependence on other LLVM passes in the >

On 26 September 2013 18:13, Stanislav Manilov <stanislav.manilov at gmail.com>wrote: > which I suspect has something to do with the fact that in /usr/include I > have a folder called x86_64-linux-gnu but not one > called arm-linux-gnueabihf. Am I even remotely right? > Yes, you are, and the docs should (hopefully) have all the information you need to get past that, and other

Optimize celt_pitch_xcorr function (for floating point) using ARM NEON intrinsics for SoCs that have NEON VFP unit. To enable this optimization, use --enable-intrinsics configure option. Compile time and runtime checks are also supported to make sure this optimization is only enabled when the compiler supports neon intrinsics. --- Makefile.am | 12 ++

Optimize celt_pitch_xcorr function (for floating point) using ARM NEON intrinsics for SoCs that have NEON VFP unit. To enable this optimization, use --enable-intrinsics configure option. Compile time and runtime checks are also supported to make sure this optimization is only enabled when the compiler supports neon intrinsics. --- Makefile.am | 12 ++

The usual advice I provide people is "see what Clang does with an equivalent C construct" On Wed, May 4, 2016 at 12:18 PM, Stanislav Manilov < stanislav.manilov at gmail.com> wrote: > Hi, > > There is another benefit to keeping the CppBackend: it's great for > learning how to use the IR and the C++ API in particular, as can be seen > from this SO Q&A: >

Optimize celt_pitch_xcorr function (for floating point) using ARM NEON intrinsics for SoCs that have NEON VFP unit. To enable this optimization, use --enable-intrinsics configure option. Compile time and runtime checks are also supported to make sure this optimization is only enabled when the compiler supports neon intrinsics. --- Makefile.am | 11 ++

Hi, I'm trying to compile half precision program for ARM, while it seems LLVM fails to automatically generate fused-multiply-add instructions for c += a * b. I'm wondering whether I did something wrong, if not, is it a missing feature that will be supported later? (I know there're fp16 FMLA intrinsics though) Test programs and outputs, $ clang -O3 -march=armv8.2-a+fp16fml

On 2 October 2013 12:17, Renato Golin <renato.golin at linaro.org> wrote: > On 2 October 2013 10:12, Steven Newbury <steve at snewbury.org.uk> wrote: > >> How does this make any sense? >> > > I have to agree with you that this doesn't make much sense, but there is a > case where you would want something like that: when the original source > uses NEON

On 19 December 2014 at 17:25, Viswanath Puttagunta <viswanath.puttagunta at linaro.org> wrote: > Optimize celt_pitch_xcorr function (for floating point) > using ARM NEON intrinsics for SoCs that have NEON VFP unit. > > To enable this optimization, use --enable-intrinsics > configure option. > > Compile time and runtime checks are also supported to make sure > this

Hi Philip, Thanks for checking! I'm running my own Foo pass that registers DependenceAnalysisWrapperPass as a prerequisite and then I run it like so: opt -load libfoo.so -foo example.bc This is LLVM 3.9. Cheers, - Stan On Wed, Jun 21, 2017 at 5:40 PM, Philip Pfaffe <philip.pfaffe at gmail.com> wrote: > Hi Stan, > > in both cases I get a consistent anti result. Can you