similar to: Opus's x86 Optimization

Hello guys I have profiled opus code(cpu sampling) and identified functions that are expensive for our use case (mostly x86 silk mode). Those functions could be optimized for x86 using intrinsics. Looking at source tree and there is already optimization work going on ,especially Cisco folks and Jonathan have already contributed for x86. We would like to have some optimization included in future

The direct cause of this issue is that, previously lld converted common symbols to defined symbols before passing input files to LTO, and after r360841 they are passed as common symbols to LTO. Making lld to work as before is easy, as we can convert common symbols to defined symbols as before. Here is a patch to do that, and I confirmed that that restores the original behavior for the reported

Hi Suresh, thanks for reporting this! I thought I fixed this with 8d94d3c3b44c3a27a69b153cef9be4b8e481150e. Did you run before or after that commit? Cheers, Johannes On 5/15/20 7:17 AM, Mani, Suresh via llvm-dev wrote: > [AMD Public Use] > > Hi , > > Please ignore the earlier header of Internal and Official use only. > > Thanks > M Suresh > > From: llvm-dev

Thanks for the info Teresa, Regards M Suresh From: Teresa Johnson <tejohnson at google.com> Sent: Thursday, June 20, 2019 7:15 PM To: Mani, Suresh <Suresh.Mani at amd.com> Cc: Rui Ueyama <ruiu at google.com>; llvm-dev <llvm-dev at lists.llvm.org> Subject: Re: [llvm-dev] Expected behavior of lld during LTO for global symbols (Attr Internal/Common) [CAUTION: External

Please provide the input file that produces this with opus_demo. On 16/04/15 03:24 AM, Suresh Thiriveedi wrote: > Hi Jean-Marc, > > Could you please update if you got a chance to look into. As I > mentioned, I don't see the same issue in 1.1.1, but I don't see any > difference in 1.1.1 other than optimization based on the architecture. > This optimization could have

To be decodable by opus_demo, you'll have to add the 8-byte "header". Just put in the length of the packet followed by "0" for the encoder range (0 means "not present"). That being said, from previous experience, the most likely cause of the crash is a bug in your software causing a corruption in Opus. So it's safe to assume that if you can't reproduce

Hi Teresa, Can you please let me know if there is any update on this issue. Thanks M Suresh From: Teresa Johnson <tejohnson at google.com> Sent: Tuesday, June 11, 2019 7:23 PM To: Rui Ueyama <ruiu at google.com> Cc: Mani, Suresh <Suresh.Mani at amd.com>; llvm-dev <llvm-dev at lists.llvm.org> Subject: Re: [llvm-dev] Expected behavior of lld during LTO for global symbols

Hi, When we pass around 9K samples of only ambient noise (no voice), the encoder which is enabled FEC+DTX is detecting only some 140 frames as non-voice (returning only TOC, no frame content). We were expecting all or more to be identified as non-voice. Our idea was to check how the decoder re-generates the original ambient noise during the silence duration (when we feed NULL to decoder) when

Dear samba team, What I noticed from the below example is , any user who has write access to share are able to change sub folder acls in it. we don't want that. how to restrict this to only admin users in NAS and to AD administrator in windows. ?. Please help . ---------------- 1) Import user from W2K3 R2 Server and set up a secure share. User has Read/Write access. 2)

Red Hat Enterprise Linux Server release 6.4 (Santiago) gcc version 4.4.7 20120313 (Red Hat 4.4.7-3) (GCC) We see the issue in all our Intel based Linux servers. Thanks Suresh On 21 April 2015 at 12:41, Jean-Marc Valin <jmvalin at jmvalin.ca> wrote: > Still can't reproduce. What OS and compiler version? > > Jean-Marc > > On 21/04/15 02:48 AM, Suresh Thiriveedi

Hi, There is no change in the compiler flags. I'm using as it is from the original code. No change in the Makefile and I believe it is using the floating point only by default. We are using 8k samples and mono so the commands is as follows. [root at MEDIA opus-1.1]# ./opus_demo -d 8000 1 opus_encoded_crash.opus opus_encoded_crash.pcm *And segmentation is as below..*. ............ Calling

Hi Jean-Marc, Thanks for your response. Please find the details as below. *Backtrace we got for this crash:* #0 0x0000000000800c54 in opus_decode_frame (st=0x38906b8f99d09c5, data=0xf0aa10b4ef1008ae <Address 0xf0aa10b4ef1008ae out of bounds>, len=-188613428, pcm=0x6e80016085efd57, frame_size=44037315, decode_fec=58716895) at src/opus_decoder.c:384 #1 0x00000000008009c0 in

I have a data file in which one of the columns is country code and NA is the code for Namibia. When I read the data file using read.csv, NA for Namibia is being treated as null or "NA" How can I prevent this from happening? I tried the following but it didn't work input <- read.csv("padded.csv",header = TRUE,as.is = c("code2")) thanks, Suresh [[alternative

Does anyone have any tips on getting hooked up to a local network from win95 client through a linux ppp server? I'm close to getting it working. I've logged on remotely and can ping or telnet to the linux ppp server but I cannot ping to any other machine on the local network. Do I need to add some route commands? This is what my routing table looks like after my ppp connection is started.

Hi, We are able to reproduce the issue with the 1.1 opus_demo (sample file). We captured the frames in our server just before the opus_decode and fed the file to opus_demo (1.1) and it is crashing. Same file is tested with 1.1.1 and it is fine. So this is in line with our server testing observation and I think here we can conclude that the 1.1 library is crashing while handling a specific mode

Dear all, What is the default loop-unroll factor in llvm? How can we specify our own unroll-factor? -Suresh

[AMD Official Use Only - Internal Distribution Only] Hi , There seems to be some issue with attributor, exactly in File Attributor.cpp, Function Attributor::rewriteFunctionSignatures(), Line No: 1600 (approrimate). The llvm source code at above address is as follows: // Eliminate the instructions *after* we visited all of them. for (auto &CallSitePair : CallSitePairs) {

Hi , This Is 'Suresh M' from AMD compiler team. 1) I obtained the source code of llvm and pool-alloc (relase_32 versions ) using the following commands, svn co http://llvm.org/svn/llvm-project/llvm/branches/release_32 llvm cd llvm/projects svn co http://llvm.org/svn/llvm-project/poolalloc/branches/release_32 poolalloc 2) Was able to build the binaries and libraries

Hi all, I used netperf to measure throughput between dom0 & domU. The throughput between dom0 -> domU was 256.00 Mb/sec domU -> dom0 was 401.15 Mb/sec. The throughput between dom0 & domU seems to be very asymmetry. To my surprise the throughput between domU -> dom0 is more. The value which I specified are consistent values. Is

Hi, I'm curious to know when would be the 1.1.1 stable version available. In 1.1, we are facing crash when opus library is trying to decode the CELT-only, full band and 20 ms. So we tried with 1.1.1 beta and it looks to be fine. Is there any open issue regarding this in 1.1 version? Thanks Suresh -------------- next part -------------- An HTML attachment was scrubbed... URL: