similar to: [PATCH 4/5]

On 01/18/16 07:42 AM, lvqcl wrote: > Dave Yeo wrote: > >> >Seems that the default binutils on OS/2 is too old to support AVX2, >> >attached patch works around this. Not the best solution as best would be >> >configure tests, but simple. > Are you sure that these binutils support AVX and FMA? (Currently libFLAC > doesn't contain AVX and FMA instructions).

Dave Yeo wrote: > Seems that the default binutils on OS/2 is too old to support AVX2, > attached patch works around this. Not the best solution as best would be > configure tests, but simple. Are you sure that these binutils support AVX and FMA? (Currently libFLAC doesn't contain AVX and FMA instructions). If they aren't supported then it's better to include them too into

Dave Yeo wrote: > The nature of the error implies AVX2 support that is missing but I'm not > much up on assembly, > > Best to be safe so updated patch attached. > I've also opened a ticket, http://trac.netlabs.org/rpm/ticket/165#ticket > Dave I cannot find information what version of binutils supports AVX/AVX2/FMA instructions, but IIRC OS/2 doesn't support AVX

Dave Yeo wrote: >> I cannot find information what version of binutils supports AVX/AVX2/FMA >> instructions, but IIRC OS/2 doesn't support AVX instructions anyway, >> so it doesn't matter much. > > Surprisingly, I've yet to have a report of an AVX related crash or trap > (used in FFmpeg and projects based on it, Mozilla, probably others). > As I

This patch adds two AVX2 files and adds AVX2 support code into init_stream_internal_() in stream_encoder.c. -------------- next part -------------- A non-text attachment was scrubbed... Name: 05_avx2.zip Type: application/zip Size: 7279 bytes Desc: not available Url : http://lists.xiph.org/pipermail/flac-dev/attachments/20141003/1471b3d6/attachment-0001.zip

Kashyap: I have not tried integrating your guest xml but I will look over it today when I get a chance. Thank you. Martin: Below is the output from /proc/cpuinfo. Let me know if there is anything else that would be helpful in debugging this. Thank you, Michael Giardino processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 60 model name : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz stepping

Attached is a patch to fix the incorrect CPU feature detection: On 2.1.2017 16.39, Janne Hyvärinen wrote: > Something seems to be wrong with cpu.c CPU detection code. When I > compile things with MSVC all instructions except FMA is detected as > missing, even though they are present in my CPU. That of course > results in awful performance. > > >

Hi all, I am trying to configure a kvm guest using pci passthrough to pass it a wireless pci adapter (host is my personal laptop). But when I try to start it: error: Failed to start domain obsdfw error: unsupported configuration: host doesn't support passthrough of host PCI devices PCI passthrough is enabled: [ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-3.10.0-327.28.3.el7.x86_64

It is relatively easy to convert some SSE2/3/4 code into AVX2: just use AVX2 intrinsics instead of SSE and the logic of the functions. Unfortunately my CPU doesn't have AVX2. But today I managed to briefly test AVX2 code on i5 Haswell CPU. Unfortunately I wasn't able to run full test suite on Haswell, but it seems that the new code works correctly. The results of a quick performance test

I am using nested virtualization on a Google Cloud Compute instance. Things are generally working fine with libvirt using qemu+kvm, however I observed that <cpu mode='host-model'> is not exposing avx and avx2 instruction set to the guest Linux instance. Google Cloud platform claims the CPU model of the host compute instance is Broadwell, however libvirt capabilities maps it to

Op 02-02-16 om 16:54 schreef lvqcl: > BTW, 64-bit flac can benefit from 64-bit words in > bitreader/bitwriter routines. Currently it requires > --enable-64-bit-words switch in ./configure parameters. It > would be interesting to test its effect, especially on ARM. Okay, I might test that as well. I've checked performance with the brand new laptop my employer gave me, which is

hello , thanks for comments . I tried now with force options for CPU flag which were not supported . Now the command with non fully supported CPU model gets executed , But i am surprised to see that still Guest cpu model is not changed and still same as host cpu model(SAndy Bridge) Why don't i see the model as HAswell now , could you please comment. Command used : virt-install

On 5.10.2014 10:22, Erik de Castro Lopo wrote: > lvqcl wrote: > >> This patch adds two AVX2 files and adds AVX2 support code into >> init_stream_internal_() in stream_encoder.c. > Ste of 5 patches all applied. Thanks > > I did run this on my machine with 4 i5-4440 cores wich accoring to /proc/cpu > has svx2 support. However I'm not sure if flac was actually using

On 05/11/2017 12:45 PM, Leon Fauster wrote: >> Am 11.05.2017 um 16:29 schrieb Leon Fauster <leonfauster at googlemail.com>: >> >>> Am 11.05.2017 um 14:48 schrieb Leon Fauster <leonfauster at googlemail.com>: >>> >>> https://access.redhat.com/support/policy/intel >>> >>> shows mainly Xeon CPUs. What about >>> >>>

On 01/24/16 12:29 PM, Erik de Castro Lopo wrote:> Dave Yeo wrote: > >> After this the build dies with, >> util.c: In function 'benchmark_function': >> util.c:124:17: error: 'CLOCK_PROCESS_CPUTIME_ID' undeclared (first use >> in this function) >> clock_gettime (CLOCK_PROCESS_CPUTIME_ID, &start) ; >> >> Would using

lvqcl wrote: > It doesn't know about uint32_t type, so the definition of cpu_xgetbv_x86() fails. > It can be fixed by adding "#include share/compat.h" to cpu.c (or by using > FLAC__uint32 from FLAC/ordinals.h). Ok, added share/compat.h. > When I fix this, the following problem occurs: > > error LNK2019: unresolved external symbol ___cpuidex referenced in

On Fri, Jul 10, 2020 at 11:45 PM H.J. Lu via Gcc <gcc at gcc.gnu.org> wrote: > > On Fri, Jul 10, 2020 at 10:30 AM Florian Weimer <fweimer at redhat.com> wrote: > > > > Most Linux distributions still compile against the original x86-64 > > baseline that was based on the AMD K8 (minus the 3DNow! parts, for Intel > > EM64T compatibility). > > > >

Hi, In the following instruction sequence, llvm.objectsize.i64(p) returns 6 (the entire *.ll is attached to the mail). Is this correct? Shouldn't the "object" refer to the entire block of memory being allocated? (char*) p = malloc(56) llvm.objectisize.i32(p+50); Thanks Shuxin This question is related to PR14988 (failure in bootstrap build with LTO). Part of the

> Here's mine. Interesting differences: If you disable Intel Speedstep in the BIOS it should lock the CPU to its fastest speed, but you lose power saving during idle. On Thu, May 11, 2017 at 3:48 PM, ken <gebser at mousecar.com> wrote: > On 05/11/2017 12:45 PM, Leon Fauster wrote: > >> Am 11.05.2017 um 16:29 schrieb Leon Fauster <leonfauster at googlemail.com>:

HI all, I think its time for a new release. The current code base is stable and I've been building it for x86_64/linux, powerpc/linux, armhf/linux, x86_64/darwin in a Jenkins build bot. I'm pretty sure others have been building regularly on their platforms of interest. I made a first pass on an update to the changelog.html. I' also putting toether a release checklist that I will add