| Submitter | Paul Menzel |
|---|---|
| Date | April 17, 2011, 5:47 p.m. |
| Message ID | <1303062425.23572.31.camel@mattotaupa> |
| Download | mbox | patch |
| Permalink | /patch/2439/ |
| State | New, archived |
| Headers | show |
Comments
Try to compile schroedinger_1.0.9 with orc_0.4.10 instead of orc_0.4.12 seems that the int[8,16,32]_t definitions are gone int orc_0.4.12. h
On Mon, Apr 18, 2011 at 11:47 PM, eric weiss <eric@eric-weiss.de> wrote: > Try to compile schroedinger_1.0.9 with orc_0.4.10 instead of orc_0.4.12 > seems that the int[8,16,32]_t definitions are gone int orc_0.4.12. > > h > > _______________________________________________ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel > I compiled it just hacking the orc-stdint.h header like this http://code.google.com/p/ossbuild/source/browse/trunk/Shared/Build/Windows/Win32/include/orc-0.4/orc/orc-stdint.h?r=857 Andrea
eric weiss <eric <at> eric-weiss.de> writes: > > Try to compile schroedinger_1.0.9 with orc_0.4.10 instead of orc_0.4.12 > seems that the int[8,16,32]_t definitions are gone int orc_0.4.12. > > h > Hello Eric, As being newbie to the OpenEmbedded and facing the same problem would you please instruct me how I can change the orc version and rebuild the package ? Thanks,
Dear DmitryZ, Am Samstag, den 23.04.2011, 09:33 +0000 schrieb DmitryZ: > eric weiss <eric <at> eric-weiss.de> writes: > > Try to compile schroedinger_1.0.9 with orc_0.4.10 instead of orc_0.4.12 > > seems that the int[8,16,32]_t definitions are gone int orc_0.4.12. > As being newbie to the OpenEmbedded and facing the same problem would you please > instruct me how I can change the orc version and rebuild the package ? please update to latest master and make a new build from scratch, i. e., remove your build directory beforehand. You could also use `bitbake -c clean yourprogramname` before the upgrade to try to get rid of the wrong versions. Thanks, Paul
Paul Menzel <paulepanter <at> users.sourceforge.net> writes: > please update to latest master and make a new build from scratch, i. e., > remove your build directory beforehand. > > You could also use `bitbake -c clean yourprogramname` before the upgrade > to try to get rid of the wrong versions. > > Thanks, > > Paul > Dear Paul, Thanks for advice. I'm just trying your suggestions. But just to let you know that I was updated to the master about three days ago. I just double checked which of the orc versions the OE downloaded in order to build the problematic package and it was 0.4.13. At the beginning of this thread it was said to try to build it instead of 0.4.12 with 0.4.10 where the defines for different types still exist. However, it looks that the OE trying to build it with 0.4.13 where the types still omitted. Do you know about some patches that were applied in a last three days or something. Just to let you know, I have been building the Angstrom console-image with eglibc for mini6410 machine. The preffered kernel version that came with OE git is 2.6.28 with stable patch 6, however it is impossible to build it since the make oldconfig fails with ERROR that says that it mixes the implicit and normal rules, so I just needed to upgrade to some newer kernel. So I just upgraded it to the 2.6.37 and specifically kernel build succeeded. As being newbie to the whole stuff I was wondering if the all things somehow have been in connection with one to other.
Dear DmitryZ, Am Samstag, den 23.04.2011, 12:58 +0000 schrieb DmitryZ: > Paul Menzel <paulepanter <at> users.sourceforge.net> writes: > > > please update to latest master and make a new build from scratch, i. e., > > remove your build directory beforehand. > > > > You could also use `bitbake -c clean yourprogramname` before the upgrade > > to try to get rid of the wrong versions. > Thanks for advice. I'm just trying your suggestions. did it work? > But just to let you know > that I was updated to the master about three days ago. I just double checked > which of the orc versions the OE downloaded in order to build the problematic > package and it was 0.4.13. It is still there, but it will not be used by default. > At the beginning of this thread it was said to try to > build it instead of 0.4.12 with 0.4.10 where the defines for different types > still exist. However, it looks that the OE trying to build it with 0.4.13 where > the types still omitted. Do you know about some patches that were applied in a > last three days or something. No patches were applied as far as I know. Only Orc version 0.4.10 is still used by default now. > Just to let you know, I have been building the Angstrom console-image with > eglibc for mini6410 machine. The preffered kernel version that came with OE git > is 2.6.28 with stable patch 6, however it is impossible to build it since the > make oldconfig fails with ERROR that says that it mixes the implicit and normal > rules, so I just needed to upgrade to some newer kernel. So I just upgraded it > to the 2.6.37 and specifically kernel build succeeded. > As being newbie to the whole stuff I was wondering if the all things somehow > have been in connection with one to other. Thank you for the detailed information. But I do not think that it is related. Thanks, Paul
Dear Paul, Sorry for the long delay, holidays you know and I have a very slow machine for building (32 bit etc.) available for me without Psyco JIT, since I have Python 2.7 installed by default that current sources of Psyco JIT Compiler doesn't support. So, answering your questions, yes after updating to the last master I successfully build the schroedinger package and the whole Angstrom console-image for the mini6410 with the parameters described earlier in this thread. Thanks a lot for the assistance and help. Just to let you know I haven't tested the image on the target board, so I can confirm only the build. Again, as being newbie to the whole stuff I simply don't know where I should post the information and problems regarding the components compatibility issues and build problems (for example those described above for the kernel version 2.6.28 for mini6410 and I have a few more). Would you please kindly instruct me on which of the OpenEmbedded mailing lists I should post this information. Simply according to the information given at openembedded.org for the Release 2011.03 Tom Rini has the responsibily for the build for mini6410, so I want that this information somehow will reach the right address. Sorry for junking the thread with irrelevant info. Thanks, Dmitry.
Dear DmitryZ, Am Montag, den 25.04.2011, 10:07 +0000 schrieb DmitryZ: […] > So, answering your questions, yes after updating to the last master I > successfully build the schroedinger package and the whole Angstrom > console-image for the mini6410 with the parameters described earlier in this > thread. Thanks a lot for the assistance and help. > Just to let you know I haven't tested the image on the target board, so I can > confirm only the build. it is great to hear that. > Again, as being newbie to the whole stuff I simply don't know where I should > post the information and problems regarding the components compatibility issues and > build problems (for example those described above for the kernel version 2.6.28 > for mini6410 and I have a few more). Would you please kindly instruct me on > which of the OpenEmbedded mailing lists I should post this information. Simply > according to the information given at openembedded.org for the Release 2011.03 > Tom Rini has the responsibily for the build for mini6410, so I want that this > information somehow will reach the right address. > Sorry for junking the thread with irrelevant info. Well I do not know, but I suggest to open a new thread describing the problem and your fix. Hopefully Tom or someone else will respond and suggest a fix. You can even put him into CC and if he complains about that, then you can blame me and not put him into CC next time. Thanks, Paul
Patch
diff --git a/recipes/schroedinger/gst-plugin-schroedinger_1.0.10.bb b/recipes/schroedinger/gst-plugin-schroedinger_1.0.10.bb new file mode 100644 index 0000000..c2768fe --- /dev/null +++ b/recipes/schroedinger/gst-plugin-schroedinger_1.0.10.bb @@ -0,0 +1,15 @@ +require schroedinger.inc +PR = "${INC_PR}.0" + +DEPENDS += "gstreamer gst-plugins-base" + +SRC_URI[schroedingertargz.md5sum] = "9de088ccc314bb9e766cb3aa6510a0ef" +SRC_URI[schroedingertargz.sha256sum] = "9a45c4f8d6197a641a9b06ab9b59ec02ad9986723fd855528a00ec3477a71964" + +S = "${WORKDIR}/schroedinger-${PV}" + +PACKAGES = "gst-plugin-schroedinger gst-plugin-schroedinger-dev gst-plugin-schroedinger-dbg" +FILES_gst-plugin-schroedinger += "${libdir}/gstreamer-0.10/libgstschro.so" +FILES_gst-plugin-schroedinger-dev += "${libdir}/gstreamer-0.10/libgstschro.*" +FILES_gst-plugin-schroedinger-dbg += "${libdir}/gstreamer-0.10/.debug/libgstschro.so" + diff --git a/recipes/schroedinger/schroedinger_1.0.10.bb b/recipes/schroedinger/schroedinger_1.0.10.bb new file mode 100644 index 0000000..962256f --- /dev/null +++ b/recipes/schroedinger/schroedinger_1.0.10.bb @@ -0,0 +1,13 @@ +require schroedinger.inc + +PR = "${INC_PR}.0" + +inherit autotools pkgconfig + +SRC_URI[schroedingertargz.md5sum] = "9de088ccc314bb9e766cb3aa6510a0ef" +SRC_URI[schroedingertargz.sha256sum] = "9a45c4f8d6197a641a9b06ab9b59ec02ad9986723fd855528a00ec3477a71964" + +PACKAGES =+ "gst-plugin-schroedinger-dbg gst-plugin-schroedinger-dev gst-plugin-schroedinger" +FILES_gst-plugin-schroedinger = "${libdir}/gstreamer-0.10/*.so" +FILES_gst-plugin-schroedinger-dbg = "${libdir}/gstreamer-0.10/.debug" +FILES_gst-plugin-schroedinger-dev = "${libdir}/gstreamer-0.10/*.*a"
Date: Sun, 17 Apr 2011 17:09:50 +0200 Task compile fails with the following error. /bin/sh ../arm-angstrom-linux-gnueabi-libtool --tag=CC --mode=compile arm-angstrom-linux-gnueabi-gcc -march=armv7-a -mtune=cortex-a8 -mfpu=neon -mfloat-abi=softfp -mthumb-interwork -mno-thumb --sysroot=/oe/build-angstrom-next/angstrom-dev/sysroots/armv7a-angstrom-linux-gnueabi -DHAVE_CONFIG_H -I. -I.. -Wall -I.. -I/oe/build-angstrom-next/angstrom-dev/sysroots/armv7a-angstrom-linux-gnueabi/usr/include/orc-0.4 -DSCHRO_ENABLE_UNSTABLE_API -I./cuda -I./opengl -fexpensive-optimizations -fomit-frame-pointer -frename-registers -O3 -ggdb2 -MT libschroedinger_1.0_la-tmp-orc.lo -MD -MP -MF .deps/libschroedinger_1.0_la-tmp-orc.Tpo -c -o libschroedinger_1.0_la-tmp-orc.lo `test -f 'tmp-orc.c' || echo './'`tmp-orc.c tmp-orc.c:57:45: error: expected ')' before '*' token tmp-orc.c:58:42: error: expected ')' before '*' token tmp-orc.c:59:45: error: expected ')' before '*' token tmp-orc.c:60:42: error: expected ')' before '*' token tmp-orc.c:61:45: error: expected ')' before '*' token tmp-orc.c:62:42: error: expected ')' before '*' token tmp-orc.c:63:45: error: expected ')' before '*' token tmp-orc.c:64:42: error: expected ')' before '*' token tmp-orc.c:65:43: error: expected ')' before '*' token tmp-orc.c:66:40: error: expected ')' before '*' token tmp-orc.c:67:27: error: expected ')' before '*' token tmp-orc.c:68:30: error: expected ')' before '*' token tmp-orc.c:69:35: error: expected ')' before '*' token tmp-orc.c:70:31: error: expected ')' before '*' token tmp-orc.c:71:31: error: expected ')' before '*' token tmp-orc.c:72:33: error: expected ')' before '*' token tmp-orc.c:73:35: error: expected ')' before '*' token tmp-orc.c:74:35: error: expected ')' before '*' token tmp-orc.c:75:35: error: expected ')' before '*' token tmp-orc.c:76:35: error: expected ')' before '*' token tmp-orc.c:77:47: error: expected ')' before '*' token tmp-orc.c:78:47: error: expected ')' before '*' token tmp-orc.c:79:47: error: expected ')' before '*' token tmp-orc.c:80:47: error: expected ')' before '*' token tmp-orc.c:81:32: error: expected ')' before '*' token tmp-orc.c:82:30: error: expected ')' before '*' token tmp-orc.c:83:33: error: expected ')' before '*' token tmp-orc.c:85:48: warning: type defaults to 'int' in declaration of 'int16_t' tmp-orc.c:85:56: error: expected ';', ',' or ')' before '*' token […] Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net> --- I thought maybe to get rid of the build errors of 1.0.9 [1] by upgrading [2]. – Build fixes on various platforms But I was not lucky. Please find the log attached. I just sent this patch to the list if someone else has time to take a shot. [2] http://diracvideo.org/2010/10/schroedinger-1-0-10/ --- .../schroedinger/gst-plugin-schroedinger_1.0.10.bb | 15 +++++++++++++++ recipes/schroedinger/schroedinger_1.0.10.bb | 13 +++++++++++++ 2 files changed, 28 insertions(+), 0 deletions(-) create mode 100644 recipes/schroedinger/gst-plugin-schroedinger_1.0.10.bb create mode 100644 recipes/schroedinger/schroedinger_1.0.10.bb