Jean-Marc, I can try both of the things you suggested. Maybe this coming weekend. I installed the Raspbian libopus-dev package instead of building it myself and that's been working great. Do I use --disable-intrinsics on ./configure, or on make? On Thu, May 25, 2017 at 9:28 PM, Jean-Marc Valin <jmvalin at jmvalin.ca> wrote:> Hi Samuel, > > Can you see if this commit fixes the problem: > https://git.xiph.org/?p=opus.git;a=commitdiff;h=4507637cc > > Jean-Marc > > On 20/05/17 02:55 AM, Samuel Bearg wrote: > > Hello, > > > > I'm following the directions that 'make check' gave me: to e-mail you > > log files. > > > > I'm attempting to build opus on a raspberry pi zero w, which I think has > > more or less the original hardware as an original raspberry pi + WiFi. > > > > I'm running a clean install of the latest version of raspbian lite. > > Note that lite has no gui, but also has less packages in general, so it > > might be missing something important for building opus. > > > > I tried building opus twice. Well, it seemed like I succeeded in > > building opus twice, but both times the 'make check' tests failed. > > > > Time one I downloaded the tar.gz for 1.1.4. > > > > ./configure gave the following output: > > > > ------------------------------------------------------------------------ > > opus 1.1.4: Automatic configuration OK. > > > > Compiler support: > > > > C99 var arrays: ................ yes > > C99 lrintf: .................... yes > > Use alloca: .................... no (using var arrays) > > > > General configuration: > > > > Floating point support: ........ yes > > Fast float approximations: ..... no > > Fixed point debugging: ......... no > > Inline Assembly Optimizations: . No inline ASM for your platform, > please send patches > > External Assembly Optimizations: > > Intrinsics Optimizations.......: ARM (NEON) > > Run-time CPU detection: ........ ARM (NEON Intrinsics) > > Custom modes: .................. no > > Assertion checking: ............ no > > Fuzzing: ....................... no > > Ambisonics support: .............no > > > > API documentation: ............. yes > > Extra programs: ................ yes > > ------------------------------------------------------------------------ > > > > It appeared to build ok, but 'make check' failed. The log is attached > as gittest-suite.log > > > > I then cloned the git repository and built from that. I had to install > git, autoconf, and libtools with apt-get. > > > > I ran both autogen.sh and ./configure with the --enable-fixed-point > flag. I wasn't sure which one took that argument. > > > > ./configure gave the following output. Once again it seemed to build > ok, but 'make check' failed. I've attached that as gittest-suite.log > > > > ------------------------------------------------------------------------ > > opus 1.2-alpha2-38-g43db562: Automatic configuration OK. > > > > > > > > Compiler support: > > > > C99 var arrays: ................ yes > > C99 lrintf: .................... yes > > Use alloca: .................... no (using var arrays) > > > > General configuration: > > > > Floating point support: ........ no > > Fast float approximations: ..... no > > Fixed point debugging: ......... no > > Inline Assembly Optimizations: . ARM (EDSP) (Media) > > External Assembly Optimizations: ARM (EDSP) (Media) > > Intrinsics Optimizations.......: ARM (NEON) > > Run-time CPU detection: ........ ARM (NEON) (NEON Intrinsics) > > Custom modes: .................. no > > Assertion checking: ............ no > > Fuzzing: ....................... no > > Check ASM: ..................... no > > Ambisonics support: ............ no > > > > API documentation: ............. yes > > Extra programs: ................ yes > > ------------------------------------------------------------------------ > > > > > > -- > > Samuel Bearg > > (617) 816-2835 > > DNA Medicine Institute > > 727 Massachusetts Ave. > > Cambridge MA 02139 > > > > > > _______________________________________________ > > opus mailing list > > opus at xiph.org > > http://lists.xiph.org/mailman/listinfo/opus > > >-- Samuel Bearg (617) 816-2835 DNA Medicine Institute 727 Massachusetts Ave. Cambridge MA 02139 -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.xiph.org/pipermail/opus/attachments/20170525/3a4046ce/attachment-0001.html>
On 25/05/17 10:03 PM, Samuel Bearg wrote:> Do I use --disable-intrinsics on ./configure, or on make?--disable-intrinsics is a configure option. It will not compile any of the Neon code. By default, the library builds with Neon support, but only enables it when it detects that it can at run-time. The run-time detection didn't work with the unit tests and that's what my patch fixes. Cheers, Jean-Marc> On Thu, May 25, 2017 at 9:28 PM, Jean-Marc Valin <jmvalin at jmvalin.ca > <mailto:jmvalin at jmvalin.ca>> wrote: > > Hi Samuel, > > Can you see if this commit fixes the problem: > https://git.xiph.org/?p=opus.git;a=commitdiff;h=4507637cc > <https://git.xiph.org/?p=opus.git;a=commitdiff;h=4507637cc> > > Jean-Marc > > On 20/05/17 02:55 AM, Samuel Bearg wrote: > > Hello, > > > > I'm following the directions that 'make check' gave me: to e-mail you > > log files. > > > > I'm attempting to build opus on a raspberry pi zero w, which I > think has > > more or less the original hardware as an original raspberry pi + WiFi. > > > > I'm running a clean install of the latest version of raspbian lite. > > Note that lite has no gui, but also has less packages in general, > so it > > might be missing something important for building opus. > > > > I tried building opus twice. Well, it seemed like I succeeded in > > building opus twice, but both times the 'make check' tests failed. > > > > Time one I downloaded the tar.gz for 1.1.4. > > > > ./configure gave the following output: > > > > > ------------------------------------------------------------------------ > > opus 1.1.4: Automatic configuration OK. > > > > Compiler support: > > > > C99 var arrays: ................ yes > > C99 lrintf: .................... yes > > Use alloca: .................... no (using var arrays) > > > > General configuration: > > > > Floating point support: ........ yes > > Fast float approximations: ..... no > > Fixed point debugging: ......... no > > Inline Assembly Optimizations: . No inline ASM for your > platform, please send patches > > External Assembly Optimizations: > > Intrinsics Optimizations.......: ARM (NEON) > > Run-time CPU detection: ........ ARM (NEON Intrinsics) > > Custom modes: .................. no > > Assertion checking: ............ no > > Fuzzing: ....................... no > > Ambisonics support: .............no > > > > API documentation: ............. yes > > Extra programs: ................ yes > > > ------------------------------------------------------------------------ > > > > It appeared to build ok, but 'make check' failed. The log is > attached as gittest-suite.log > > > > I then cloned the git repository and built from that. I had to > install git, autoconf, and libtools with apt-get. > > > > I ran both autogen.sh and ./configure with the > --enable-fixed-point flag. I wasn't sure which one took that argument. > > > > ./configure gave the following output. Once again it seemed to > build ok, but 'make check' failed. I've attached that as > gittest-suite.log > > > > > ------------------------------------------------------------------------ > > opus 1.2-alpha2-38-g43db562: Automatic configuration OK. > > > > > > > > Compiler support: > > > > C99 var arrays: ................ yes > > C99 lrintf: .................... yes > > Use alloca: .................... no (using var arrays) > > > > General configuration: > > > > Floating point support: ........ no > > Fast float approximations: ..... no > > Fixed point debugging: ......... no > > Inline Assembly Optimizations: . ARM (EDSP) (Media) > > External Assembly Optimizations: ARM (EDSP) (Media) > > Intrinsics Optimizations.......: ARM (NEON) > > Run-time CPU detection: ........ ARM (NEON) (NEON Intrinsics) > > Custom modes: .................. no > > Assertion checking: ............ no > > Fuzzing: ....................... no > > Check ASM: ..................... no > > Ambisonics support: ............ no > > > > API documentation: ............. yes > > Extra programs: ................ yes > > > ------------------------------------------------------------------------ > > > > > > -- > > Samuel Bearg > > (617) 816-2835 > > DNA Medicine Institute > > 727 Massachusetts Ave. > > Cambridge MA 02139 > > > > > > _______________________________________________ > > opus mailing list > > opus at xiph.org <mailto:opus at xiph.org> > > http://lists.xiph.org/mailman/listinfo/opus > <http://lists.xiph.org/mailman/listinfo/opus> > > > > > > > -- > Samuel Bearg > (617) 816-2835 > DNA Medicine Institute > 727 Massachusetts Ave. > Cambridge MA 02139
Jean-Marc, I built the git commit you linked using just ./configure and it passed all the 'make check' tests. Looks like your update fixed things. Nice work, Samuel On Thu, May 25, 2017 at 10:39 PM, Jean-Marc Valin <jmvalin at jmvalin.ca> wrote:> On 25/05/17 10:03 PM, Samuel Bearg wrote: > > Do I use --disable-intrinsics on ./configure, or on make? > > --disable-intrinsics is a configure option. It will not compile any of > the Neon code. By default, the library builds with Neon support, but > only enables it when it detects that it can at run-time. The run-time > detection didn't work with the unit tests and that's what my patch fixes. > > Cheers, > > Jean-Marc > > > > > On Thu, May 25, 2017 at 9:28 PM, Jean-Marc Valin <jmvalin at jmvalin.ca > > <mailto:jmvalin at jmvalin.ca>> wrote: > > > > Hi Samuel, > > > > Can you see if this commit fixes the problem: > > https://git.xiph.org/?p=opus.git;a=commitdiff;h=4507637cc > > <https://git.xiph.org/?p=opus.git;a=commitdiff;h=4507637cc> > > > > Jean-Marc > > > > On 20/05/17 02:55 AM, Samuel Bearg wrote: > > > Hello, > > > > > > I'm following the directions that 'make check' gave me: to e-mail > you > > > log files. > > > > > > I'm attempting to build opus on a raspberry pi zero w, which I > > think has > > > more or less the original hardware as an original raspberry pi + > WiFi. > > > > > > I'm running a clean install of the latest version of raspbian lite. > > > Note that lite has no gui, but also has less packages in general, > > so it > > > might be missing something important for building opus. > > > > > > I tried building opus twice. Well, it seemed like I succeeded in > > > building opus twice, but both times the 'make check' tests failed. > > > > > > Time one I downloaded the tar.gz for 1.1.4. > > > > > > ./configure gave the following output: > > > > > > > > ------------------------------------------------------------ > ------------ > > > opus 1.1.4: Automatic configuration OK. > > > > > > Compiler support: > > > > > > C99 var arrays: ................ yes > > > C99 lrintf: .................... yes > > > Use alloca: .................... no (using var arrays) > > > > > > General configuration: > > > > > > Floating point support: ........ yes > > > Fast float approximations: ..... no > > > Fixed point debugging: ......... no > > > Inline Assembly Optimizations: . No inline ASM for your > > platform, please send patches > > > External Assembly Optimizations: > > > Intrinsics Optimizations.......: ARM (NEON) > > > Run-time CPU detection: ........ ARM (NEON Intrinsics) > > > Custom modes: .................. no > > > Assertion checking: ............ no > > > Fuzzing: ....................... no > > > Ambisonics support: .............no > > > > > > API documentation: ............. yes > > > Extra programs: ................ yes > > > > > ------------------------------------------------------------ > ------------ > > > > > > It appeared to build ok, but 'make check' failed. The log is > > attached as gittest-suite.log > > > > > > I then cloned the git repository and built from that. I had to > > install git, autoconf, and libtools with apt-get. > > > > > > I ran both autogen.sh and ./configure with the > > --enable-fixed-point flag. I wasn't sure which one took that > argument. > > > > > > ./configure gave the following output. Once again it seemed to > > build ok, but 'make check' failed. I've attached that as > > gittest-suite.log > > > > > > > > ------------------------------------------------------------ > ------------ > > > opus 1.2-alpha2-38-g43db562: Automatic configuration OK. > > > > > > > > > > > > Compiler support: > > > > > > C99 var arrays: ................ yes > > > C99 lrintf: .................... yes > > > Use alloca: .................... no (using var arrays) > > > > > > General configuration: > > > > > > Floating point support: ........ no > > > Fast float approximations: ..... no > > > Fixed point debugging: ......... no > > > Inline Assembly Optimizations: . ARM (EDSP) (Media) > > > External Assembly Optimizations: ARM (EDSP) (Media) > > > Intrinsics Optimizations.......: ARM (NEON) > > > Run-time CPU detection: ........ ARM (NEON) (NEON Intrinsics) > > > Custom modes: .................. no > > > Assertion checking: ............ no > > > Fuzzing: ....................... no > > > Check ASM: ..................... no > > > Ambisonics support: ............ no > > > > > > API documentation: ............. yes > > > Extra programs: ................ yes > > > > > ------------------------------------------------------------ > ------------ > > > > > > > > > -- > > > Samuel Bearg > > > (617) 816-2835 > > > DNA Medicine Institute > > > 727 Massachusetts Ave. > > > Cambridge MA 02139 > > > > > > > > > _______________________________________________ > > > opus mailing list > > > opus at xiph.org <mailto:opus at xiph.org> > > > http://lists.xiph.org/mailman/listinfo/opus > > <http://lists.xiph.org/mailman/listinfo/opus> > > > > > > > > > > > > > -- > > Samuel Bearg > > (617) 816-2835 > > DNA Medicine Institute > > 727 Massachusetts Ave. > > Cambridge MA 02139 >-- Samuel Bearg (617) 816-2835 DNA Medicine Institute 727 Massachusetts Ave. Cambridge MA 02139 -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.xiph.org/pipermail/opus/attachments/20170526/38cc6f96/attachment.html>