Re: [oe] [meta-oe][PATCH] mpv: 0.26.0 -> 0.32.0
On Wed, Feb 5, 2020 at 6:32 PM Khem Raj wrote: > On Wed, Feb 5, 2020 at 3:02 PM Trevor Woerner wrote: > > > > On Wed 2020-02-05 @ 10:07:07 PM, Martin Jansa wrote: > > > Thanks, > > > > > > would you mind undoing: > > > > http://git.openembedded.org/meta-openembedded/commit/?id=0c182e58775966699c303c1e2561c273197bf445 > > > in the same commit? > > > > Sure, no problem. > > v2 on the way > > > > > Also was this sent with -M? > > > > haha, I know exactly what you mean! I did the format-patch twice just to > > convince myself I had used the -M (and I had, twice). I don't understand > why > > there's no similarity :-( > > also please take a look at > https://errors.yoctoproject.org/Errors/Details/390294/ > okay -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-python2][RFC] lib/oeqa/runtime/cases/python2_module.py: add test
On Wed, Feb 5, 2020 at 3:18 PM Randy MacLeod wrote: > On 1/25/20 3:25 AM, Tim Orling wrote: > > This test case is intentionally limited to only tesing one > > python package/recipe at a time, because the use case is to > > determine whether dependencies have been met for a minimal > > image with 'only' the package under test installed. This > > dependency checking must include the packages-split python > > standard library modules that OE/YP creates from the python > > manifest. > > > > TODO: Be smarter about PYTHON_IMPORTS_UNDER_TEST, as this > >should be discoverable from /usr/lib/python2.7/site-packages > >or from the rpm/deb/ipk of the python package itself. > >What we do NOT want to do is use pkgdata or some other > >magical Python script that will add dependencies to the > >test image. We need as pristine and minimalist a test > >environment as we can get. > > > > TODO: Use a minimal master test image and install the package under > >test from a package feed using a package manager in the master > >test image. This would allow for much shorter developement > >cycles. > > > > TODO: Parse the ImportError and give hints about what dependencies > >are missing. > > This looks useful. Do you have the python3 version done but not sent > or should Trevor see if he can get that to work? > > > > > Example configuration in local.conf: > > > > IMAGE_CLASSES += " testimage" > > DISTRO_FEAURES_append = " ptest" > > EXTRA_IMAGE_FEATURES = "debug-tweaks" > > TESTIMAGE_AUTO = "1" > > TEST_SUITES = " ping ssh python2 ptest python2_module" > > TEST_QEMUPARAMS += "-smp 4 -m 8192" > > TEST_RUNQEMUPARAMS += "kvm gl" > > IMAGE_ROOTFS_SIZE ?= "8192" > > IMAGE_ROOTFS_EXTRA_SPACE_append = > "${@bb.utils.contains("DISTRO_FEATURES", "systemd", " + 4096", "", d)}" > > IMAGE_INSTALL_append = " ptest-runner dropbear procps coreutils iproute2 > sysstat" > Is this really a minimal set of dependencies and if so, why do you need: > procps coreutils iproute2 sysstat > ? > > Those are required to make the QEMU runner dump any missing things, like TUN/TAP. So no, not required. The host_dump stuff isn't really all that helpful anyway. If you can run "testimage" with KVM, TUN/TAP and virgl/gl then you have all the speed ups that I was trying to make sure were working on the platform running these tests (with gitlab ci runner) The fact that the docker-executor in gitlab does not allow the --add-groups option to the docker command line gets in the way anyway... so you would have to run the container privileged and I'm not going to do that ;) The user in the docker container does not have permissions on /dev/kvm, etc. But that has nothing to do with this code :) > I should probably try it before replying but > it's already been a long day! > > ../Randy > > > PYTHON_PACKAGE_UNDER_TEST = "python-pymongo" > > IMAGE_INSTALL_append = " ${PYTHON_PACKAGE_UNDER_TEST}" > > PYTHON_IMPORTS_UNDER_TEST = "bson gridfs pymongo" > > > > Signed-off-by: Tim Orling > > --- > > lib/oeqa/runtime/cases/python2_module.py | 97 > > 1 file changed, 97 insertions(+) > > create mode 100644 lib/oeqa/runtime/cases/python2_module.py > > > > diff --git a/lib/oeqa/runtime/cases/python2_module.py > b/lib/oeqa/runtime/cases/python2_module.py > > new file mode 100644 > > index ..59ae00fb > > --- /dev/null > > +++ b/lib/oeqa/runtime/cases/python2_module.py > > @@ -0,0 +1,97 @@ > > +# > > +# SPDX-License-Identifier: MIT > > +# > > + > > +import os > > + > > +from oeqa.runtime.case import OERuntimeTestCase > > +from oeqa.core.decorator.depends import OETestDepends > > +from oeqa.core.decorator.data import OETestDataDepends > > +from oeqa.runtime.decorator.package import OEHasPackage > > + > > + > > +class PythonModuleTest(OERuntimeTestCase): > > +"""This test case is intentionally limited to only tesing one > > + python package/recipe at a time, because the use case is to > > + determine whether dependencies have been met for a minimal > > + image with 'only' the package under test installed. This > > + dependency checking must include the packages-split python > > + standard library modules that OE/YP creates from the python > > + manifest. > > + > > + TODO: Be smarter about PYTHON_IMPORTS_UNDER_TEST, as this > > + should be discoverable from > /usr/lib/python2.7/site-packages > > + or from the rpm/deb/ipk of the python package itself. > > + What we do NOT want to do is use pkgdata or some other > > + magical Python script that will add dependencies to the > > + test image. We need as pristine and minimalist a test > > + environment as we can get. > > + > > + TODO: Use a minimal master test image and install the package > under > > + test from a package feed using a package manager in the > master > > +
Re: [oe] [meta-python2][RFC] lib/oeqa/runtime/cases/python2_module.py: add test
On Wed, Feb 5, 2020 at 3:18 PM Randy MacLeod wrote: > On 1/25/20 3:25 AM, Tim Orling wrote: > > This test case is intentionally limited to only tesing one > > python package/recipe at a time, because the use case is to > > determine whether dependencies have been met for a minimal > > image with 'only' the package under test installed. This > > dependency checking must include the packages-split python > > standard library modules that OE/YP creates from the python > > manifest. > > > > TODO: Be smarter about PYTHON_IMPORTS_UNDER_TEST, as this > >should be discoverable from /usr/lib/python2.7/site-packages > >or from the rpm/deb/ipk of the python package itself. > >What we do NOT want to do is use pkgdata or some other > >magical Python script that will add dependencies to the > >test image. We need as pristine and minimalist a test > >environment as we can get. > > > > TODO: Use a minimal master test image and install the package under > >test from a package feed using a package manager in the master > >test image. This would allow for much shorter developement > >cycles. > > > > TODO: Parse the ImportError and give hints about what dependencies > >are missing. > > This looks useful. Do you have the python3 version done but not sent > or should Trevor see if he can get that to work? > > I have not started the python3 version yet. To do what is functional here will be trivial changes. I'd love to see traction on the TODO list and it should be interesting work for Trevor to do. This tool is now in my "daily" workflow. There is one possible issue, which is a change in the variables does not necessarily trigger the tests you expect. The symptom is running testimage tests the last thing you had the variable set to, not the current value in local.conf. There are other places in the code base where variables are watched and trigger changes appropriately. I did not go that far yet. Also, the variables names between py2 and py3 would need to be different, but that is a thing of the past for master, so not a real issue. > > > > Example configuration in local.conf: > > > > IMAGE_CLASSES += " testimage" > > DISTRO_FEAURES_append = " ptest" > > EXTRA_IMAGE_FEATURES = "debug-tweaks" > > TESTIMAGE_AUTO = "1" > > TEST_SUITES = " ping ssh python2 ptest python2_module" > > TEST_QEMUPARAMS += "-smp 4 -m 8192" > > TEST_RUNQEMUPARAMS += "kvm gl" > > IMAGE_ROOTFS_SIZE ?= "8192" > > IMAGE_ROOTFS_EXTRA_SPACE_append = > "${@bb.utils.contains("DISTRO_FEATURES", "systemd", " + 4096", "", d)}" > > IMAGE_INSTALL_append = " ptest-runner dropbear procps coreutils iproute2 > sysstat" > Is this really a minimal set of dependencies and if so, why do you need: > procps coreutils iproute2 sysstat > ? > > I should probably try it before replying but > it's already been a long day! > > ../Randy > > > PYTHON_PACKAGE_UNDER_TEST = "python-pymongo" > > IMAGE_INSTALL_append = " ${PYTHON_PACKAGE_UNDER_TEST}" > > PYTHON_IMPORTS_UNDER_TEST = "bson gridfs pymongo" > > > > Signed-off-by: Tim Orling > > --- > > lib/oeqa/runtime/cases/python2_module.py | 97 > > 1 file changed, 97 insertions(+) > > create mode 100644 lib/oeqa/runtime/cases/python2_module.py > > > > diff --git a/lib/oeqa/runtime/cases/python2_module.py > b/lib/oeqa/runtime/cases/python2_module.py > > new file mode 100644 > > index ..59ae00fb > > --- /dev/null > > +++ b/lib/oeqa/runtime/cases/python2_module.py > > @@ -0,0 +1,97 @@ > > +# > > +# SPDX-License-Identifier: MIT > > +# > > + > > +import os > > + > > +from oeqa.runtime.case import OERuntimeTestCase > > +from oeqa.core.decorator.depends import OETestDepends > > +from oeqa.core.decorator.data import OETestDataDepends > > +from oeqa.runtime.decorator.package import OEHasPackage > > + > > + > > +class PythonModuleTest(OERuntimeTestCase): > > +"""This test case is intentionally limited to only tesing one > > + python package/recipe at a time, because the use case is to > > + determine whether dependencies have been met for a minimal > > + image with 'only' the package under test installed. This > > + dependency checking must include the packages-split python > > + standard library modules that OE/YP creates from the python > > + manifest. > > + > > + TODO: Be smarter about PYTHON_IMPORTS_UNDER_TEST, as this > > + should be discoverable from > /usr/lib/python2.7/site-packages > > + or from the rpm/deb/ipk of the python package itself. > > + What we do NOT want to do is use pkgdata or some other > > + magical Python script that will add dependencies to the > > + test image. We need as pristine and minimalist a test > > + environment as we can get. > > + > > + TODO: Use a minimal master test image and install the package > under > > + test from a package feed using a package
Re: [oe] [meta-oe][PATCH] mpv: 0.26.0 -> 0.32.0
On Wed, Feb 5, 2020 at 3:02 PM Trevor Woerner wrote: > > On Wed 2020-02-05 @ 10:07:07 PM, Martin Jansa wrote: > > Thanks, > > > > would you mind undoing: > > http://git.openembedded.org/meta-openembedded/commit/?id=0c182e58775966699c303c1e2561c273197bf445 > > in the same commit? > > Sure, no problem. > v2 on the way > > > Also was this sent with -M? > > haha, I know exactly what you mean! I did the format-patch twice just to > convince myself I had used the -M (and I had, twice). I don't understand why > there's no similarity :-( also please take a look at https://errors.yoctoproject.org/Errors/Details/390294/ > -- > ___ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.openembedded.org/mailman/listinfo/openembedded-devel -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-python2][RFC] lib/oeqa/runtime/cases/python2_module.py: add test
On 1/25/20 3:25 AM, Tim Orling wrote: This test case is intentionally limited to only tesing one python package/recipe at a time, because the use case is to determine whether dependencies have been met for a minimal image with 'only' the package under test installed. This dependency checking must include the packages-split python standard library modules that OE/YP creates from the python manifest. TODO: Be smarter about PYTHON_IMPORTS_UNDER_TEST, as this should be discoverable from /usr/lib/python2.7/site-packages or from the rpm/deb/ipk of the python package itself. What we do NOT want to do is use pkgdata or some other magical Python script that will add dependencies to the test image. We need as pristine and minimalist a test environment as we can get. TODO: Use a minimal master test image and install the package under test from a package feed using a package manager in the master test image. This would allow for much shorter developement cycles. TODO: Parse the ImportError and give hints about what dependencies are missing. This looks useful. Do you have the python3 version done but not sent or should Trevor see if he can get that to work? Example configuration in local.conf: IMAGE_CLASSES += " testimage" DISTRO_FEAURES_append = " ptest" EXTRA_IMAGE_FEATURES = "debug-tweaks" TESTIMAGE_AUTO = "1" TEST_SUITES = " ping ssh python2 ptest python2_module" TEST_QEMUPARAMS += "-smp 4 -m 8192" TEST_RUNQEMUPARAMS += "kvm gl" IMAGE_ROOTFS_SIZE ?= "8192" IMAGE_ROOTFS_EXTRA_SPACE_append = "${@bb.utils.contains("DISTRO_FEATURES", "systemd", " + 4096", "", d)}" IMAGE_INSTALL_append = " ptest-runner dropbear procps coreutils iproute2 sysstat" Is this really a minimal set of dependencies and if so, why do you need: procps coreutils iproute2 sysstat ? I should probably try it before replying but it's already been a long day! ../Randy PYTHON_PACKAGE_UNDER_TEST = "python-pymongo" IMAGE_INSTALL_append = " ${PYTHON_PACKAGE_UNDER_TEST}" PYTHON_IMPORTS_UNDER_TEST = "bson gridfs pymongo" Signed-off-by: Tim Orling --- lib/oeqa/runtime/cases/python2_module.py | 97 1 file changed, 97 insertions(+) create mode 100644 lib/oeqa/runtime/cases/python2_module.py diff --git a/lib/oeqa/runtime/cases/python2_module.py b/lib/oeqa/runtime/cases/python2_module.py new file mode 100644 index ..59ae00fb --- /dev/null +++ b/lib/oeqa/runtime/cases/python2_module.py @@ -0,0 +1,97 @@ +# +# SPDX-License-Identifier: MIT +# + +import os + +from oeqa.runtime.case import OERuntimeTestCase +from oeqa.core.decorator.depends import OETestDepends +from oeqa.core.decorator.data import OETestDataDepends +from oeqa.runtime.decorator.package import OEHasPackage + + +class PythonModuleTest(OERuntimeTestCase): +"""This test case is intentionally limited to only tesing one + python package/recipe at a time, because the use case is to + determine whether dependencies have been met for a minimal + image with 'only' the package under test installed. This + dependency checking must include the packages-split python + standard library modules that OE/YP creates from the python + manifest. + + TODO: Be smarter about PYTHON_IMPORTS_UNDER_TEST, as this + should be discoverable from /usr/lib/python2.7/site-packages + or from the rpm/deb/ipk of the python package itself. + What we do NOT want to do is use pkgdata or some other + magical Python script that will add dependencies to the + test image. We need as pristine and minimalist a test + environment as we can get. + + TODO: Use a minimal master test image and install the package under + test from a package feed using a package manager in the master + test image. This would allow for much shorter developement + cycles. + + TODO: Parse the ImportError and give hints about what dependencies + are missing. + + + Example configuration in local.conf: + + IMAGE_CLASSES += " testimage" + DISTRO_FEAURES_append = " ptest" + EXTRA_IMAGE_FEATURES = "debug-tweaks" + TESTIMAGE_AUTO = "1" + TEST_SUITES = " ping ssh python2 ptest python2_module" + TEST_QEMUPARAMS += "-smp 4 -m 8192" + TEST_RUNQEMUPARAMS += "kvm gl" + IMAGE_ROOTFS_SIZE ?= "8192" + IMAGE_ROOTFS_EXTRA_SPACE_append = "${@bb.utils.contains("DISTRO_FEATURES", "systemd", " + 4096", "", d)}" + IMAGE_INSTALL_append = " ptest-runner dropbear procps coreutils iproute2 sysstat" + PYTHON_PACKAGE_UNDER_TEST = "python-engineio" + IMAGE_INSTALL_append = " ${PYTHON_PACKAGE_UNDER_TEST}" + PYTHON_IMPORTS_UNDER_TEST = "engineio" +""" + +package = "" +imports = [] + +@classmethod +def setUp(cls): +if 'PYTHON_PACKAGE_UNDER_TEST' in cls.tc.td: +cls.package =
Re: [oe] [meta-oe][PATCH] mpv: 0.26.0 -> 0.32.0
On Wed 2020-02-05 @ 10:07:07 PM, Martin Jansa wrote: > Thanks, > > would you mind undoing: > http://git.openembedded.org/meta-openembedded/commit/?id=0c182e58775966699c303c1e2561c273197bf445 > in the same commit? Sure, no problem. v2 on the way > Also was this sent with -M? haha, I know exactly what you mean! I did the format-patch twice just to convince myself I had used the -M (and I had, twice). I don't understand why there's no similarity :-( -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-networking][PATCH] mdns: improve numerous aspects of Posix backend
Apple's default implementation of the Posix backend for mDNSResponder has a number of weaknesses. Address several of them, most notably: * Improve interface tracking, preventing confusion to mdns's state machine. Prevents spurious removal/republication cycles whenever network interfaces are added or removed. * Support network interfaces whose indeces are great than 31. Indices grow past that range surprisingly quickly, especially with multi- homed, mobile, wifi, Bluetooth, VPN, VLANs, or other interfaces present. * Correctly handle edge cases during removal of a network interface. The fixes are kept as a patch series for clarity. Signed-off-by: Matt Hoosier --- ...utine-for-cleaning-recent-interfaces.patch | 60 + ...outine-for-tearing-down-an-interface.patch | 58 + .../0003-Track-interface-socket-family.patch | 50 + ...0004-Use-list-for-changed-interfaces.patch | 177 +++ .../0005-Handle-noisy-netlink-sockets.patch | 212 ++ .../files/0006-Remove-unneeded-function.patch | 51 + ...cate-loopback-interface-to-mDNS-core.patch | 129 +++ ...-deleted-interfaces-as-being-changed.patch | 39 .../0009-Fix-possible-NULL-dereference.patch | 45 ...0010-Handle-errors-from-socket-calls.patch | 62 + ...ic-allocation-to-file-scope-variable.patch | 51 + .../recipes-protocols/mdns/mdns_878.260.1.bb | 11 + 12 files changed, 945 insertions(+) create mode 100644 meta-networking/recipes-protocols/mdns/files/0001-Create-subroutine-for-cleaning-recent-interfaces.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0002-Create-subroutine-for-tearing-down-an-interface.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0003-Track-interface-socket-family.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0004-Use-list-for-changed-interfaces.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0005-Handle-noisy-netlink-sockets.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0006-Remove-unneeded-function.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0007-Indicate-loopback-interface-to-mDNS-core.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0008-Mark-deleted-interfaces-as-being-changed.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0009-Fix-possible-NULL-dereference.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0010-Handle-errors-from-socket-calls.patch create mode 100644 meta-networking/recipes-protocols/mdns/files/0011-Change-a-dynamic-allocation-to-file-scope-variable.patch diff --git a/meta-networking/recipes-protocols/mdns/files/0001-Create-subroutine-for-cleaning-recent-interfaces.patch b/meta-networking/recipes-protocols/mdns/files/0001-Create-subroutine-for-cleaning-recent-interfaces.patch new file mode 100644 index 0..692c344db --- /dev/null +++ b/meta-networking/recipes-protocols/mdns/files/0001-Create-subroutine-for-cleaning-recent-interfaces.patch @@ -0,0 +1,60 @@ +From 89ea6ac4a8840e8c2be0140a9805c6522c6c5280 Mon Sep 17 00:00:00 2001 +From: Nate Karstens +Date: Wed, 28 Jun 2017 17:30:00 -0500 +Subject: [PATCH 01/11] Create subroutine for cleaning recent interfaces + +Moves functionality for cleaning the list of recent +interfaces into its own subroutine. + +Upstream-Status: Submitted [d...@apple.com] + +Signed-off-by: Nate Karstens +--- + mDNSPosix/mDNSPosix.c | 24 ++-- + 1 file changed, 14 insertions(+), 10 deletions(-) + +diff --git a/mDNSPosix/mDNSPosix.c b/mDNSPosix/mDNSPosix.c +index 0e10bd5..ffc9696 100644 +--- a/mDNSPosix/mDNSPosix.c b/mDNSPosix/mDNSPosix.c +@@ -856,6 +856,19 @@ mDNSlocal int SetupSocket(struct sockaddr *intfAddr, mDNSIPPort port, int interf + return err; + } + ++// Clean up any interfaces that have been hanging around on the RecentInterfaces list for more than a minute ++mDNSlocal void CleanRecentInterfaces(void) ++{ ++PosixNetworkInterface **ri = ++const mDNSs32 utc = mDNSPlatformUTC(); ++while (*ri) ++{ ++PosixNetworkInterface *pi = *ri; ++if (utc - pi->LastSeen < 60) ri = (PosixNetworkInterface **)>coreIntf.next; ++else { *ri = (PosixNetworkInterface *)pi->coreIntf.next; free(pi); } ++} ++} ++ + // Creates a PosixNetworkInterface for the interface whose IP address is + // intfAddr and whose name is intfName and registers it with mDNS core. + mDNSlocal int SetupOneInterface(mDNS *const m, struct sockaddr *intfAddr, struct sockaddr *intfMask, const char *intfName, int intfIndex) +@@ -1010,16 +1023,7 @@ mDNSlocal int SetupInterfaceList(mDNS *const m) + + // Clean up. + if (intfList != NULL) free_ifi_info(intfList); +- +-// Clean up any interfaces that have been hanging around on the RecentInterfaces list for more than a minute +-PosixNetworkInterface **ri = +-const
Re: [oe] [meta-oe][PATCH] mpv: 0.26.0 -> 0.32.0
On Wed, Feb 05, 2020 at 03:30:53PM -0500, Trevor Woerner wrote: > Remove mpv from the PNBLACKLIST by updating to the latest release and using a > newer, python3-aware waf for building (copied from glmark2). > > The name of the license file changed, but the licence of the software itself > did not. > > Added mime-xdg to inherit to handle mime artifacts. > > Added to the FILES command to scoop up additional files being generated by > the build. > > Extensively fixed and cleaned up the PACKAGECONFIG options to reflect the > current options available when configuring mpv. Also cleaned up any configure > options which are no longer available. > > Tested with core-image-x11 on a rock-pi-4 with > PACKAGECONFIG = "libass x11 opengl egl drm gbm vaapi vdpau" > # mpv --vo=help > Available video outputs: > libmpv render API for libmpv > gpu Shader-based GPU Renderer > vdpauVDPAU with X11 > vaapiVA API with X11 > x11 X11 (slow, old crap) > null Null video output > imageWrite video frames to image files > tct true-color terminals > drm Direct Rendering Manager > > Signed-off-by: Trevor Woerner > --- > .../mplayer/mpv/python3.patch | 12 ++ > .../recipes-multimedia/mplayer/mpv_0.26.0.bb | 85 - > .../recipes-multimedia/mplayer/mpv_0.32.0.bb | 115 ++ > 3 files changed, 127 insertions(+), 85 deletions(-) > create mode 100644 meta-oe/recipes-multimedia/mplayer/mpv/python3.patch > delete mode 100644 meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb > create mode 100644 meta-oe/recipes-multimedia/mplayer/mpv_0.32.0.bb Thanks, would you mind undoing: http://git.openembedded.org/meta-openembedded/commit/?id=0c182e58775966699c303c1e2561c273197bf445 in the same commit? Also was this sent with -M? > diff --git a/meta-oe/recipes-multimedia/mplayer/mpv/python3.patch > b/meta-oe/recipes-multimedia/mplayer/mpv/python3.patch > new file mode 100644 > index 0..df208cf87 > --- /dev/null > +++ b/meta-oe/recipes-multimedia/mplayer/mpv/python3.patch > @@ -0,0 +1,12 @@ > +Use Python 3 instead of Python 2. > + > +Upstream-Status: Pending > +Signed-off-by: Ross Burton > + > +diff --git a/waf b/waf > +index 6ce2a24..de3c898 100755 > +--- a/waf > b/waf > +@@ -1 +1 @@ > +-#!/usr/bin/env python > ++#!/usr/bin/env python3 > diff --git a/meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb > b/meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb > deleted file mode 100644 > index c98aa7ab0..0 > --- a/meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb > +++ /dev/null > @@ -1,85 +0,0 @@ > -SUMMARY = "Open Source multimedia player" > -DESCRIPTION = "mpv is a fork of mplayer2 and MPlayer. It shares some > features with the former projects while introducing many more." > -SECTION = "multimedia" > -HOMEPAGE = "http://www.mpv.io/; > -DEPENDS = "zlib ffmpeg jpeg virtual/libx11 xsp libxv \ > - libxscrnsaver libv4l libxinerama \ > -" > - > -REQUIRED_DISTRO_FEATURES = "x11" > - > -LICENSE = "GPLv2+" > -LIC_FILES_CHKSUM = "file://LICENSE;md5=91f1cb870c1cc2d31351a4d2595441cb" > - > -# While this item does not require it, it depends on ffmpeg which does > -LICENSE_FLAGS = "commercial" > - > -SRCREV_mpv = "4db82f04502cbcd20d9a975df9dfde91d8c389e3" > -SRC_URI = " \ > -git://github.com/mpv-player/mpv;name=mpv \ > - > http://www.freehackers.org/~tnagy/release/waf-1.8.12;name=waf;downloadfilename=waf;subdir=git > \ > -" > -SRC_URI[waf.md5sum] = "cef4ee82206b1843db082d0b0506bf71" > -SRC_URI[waf.sha256sum] = > "01bf2beab2106d1558800c8709bc2c8e496d3da4a2ca343fe091f22fca60c98b" > - > -S = "${WORKDIR}/git" > - > -inherit waf pkgconfig python3native features_check > - > -LUA ?= "lua" > -LUA_mips64 = "" > -LUA_aarch64 = "" > -LUA_powerpc64 = "" > -LUA_powerpc64le = "" > -LUA_powerpc = "" > - > -# Note: both lua and libass are required to get on-screen-display (controls) > -PACKAGECONFIG ??= " \ > -${LUA} \ > -libass \ > -${@bb.utils.filter('DISTRO_FEATURES', 'wayland', d)} \ > -" > - > -PACKAGECONFIG_remove_aarch64 = "lua" > -PACKAGECONFIG[drm] = "--enable-drm,--disable-drm,libdrm" > -PACKAGECONFIG[gbm] = "--enable-gbm,--disable-gbm,virtual/mesa" > -PACKAGECONFIG[lua] = "--enable-lua,--disable-lua,lua luajit" > -PACKAGECONFIG[libass] = "--enable-libass,--disable-libass,libass" > -PACKAGECONFIG[libarchive] = > "--enable-libarchive,--disable-libarchive,libarchive" > -PACKAGECONFIG[jack] = "--enable-jack, --disable-jack, jack" > -PACKAGECONFIG[vaapi] = "--enable-vaapi, --disable-vaapi,libva" > -PACKAGECONFIG[vdpau] = "--enable-vdpau, --disable-vdpau,libvdpau" > -PACKAGECONFIG[wayland] = "--enable-wayland, --disable-wayland,wayland > libxkbcommon" > - > -SIMPLE_TARGET_SYS = "${@'${TARGET_SYS}'.replace('${TARGET_VENDOR}', '')}" > - > -EXTRA_OECONF = " \ > -
Re: [oe] [meta-oe][PATCH] mpv: 0.26.0 -> 0.32.0
On Wed, Feb 5, 2020 at 9:31 PM Trevor Woerner wrote: > > Remove mpv from the PNBLACKLIST by updating to the latest release and using a > newer, python3-aware waf for building (copied from glmark2). > > The name of the license file changed, but the licence of the software itself > did not. > > Added mime-xdg to inherit to handle mime artifacts. > > Added to the FILES command to scoop up additional files being generated by > the build. > > Extensively fixed and cleaned up the PACKAGECONFIG options to reflect the > current options available when configuring mpv. Also cleaned up any configure > options which are no longer available. > > Tested with core-image-x11 on a rock-pi-4 with > PACKAGECONFIG = "libass x11 opengl egl drm gbm vaapi vdpau" > # mpv --vo=help > Available video outputs: > libmpv render API for libmpv > gpu Shader-based GPU Renderer > vdpauVDPAU with X11 > vaapiVA API with X11 > x11 X11 (slow, old crap) > null Null video output > imageWrite video frames to image files > tct true-color terminals > drm Direct Rendering Manager > > Signed-off-by: Trevor Woerner Thanks Andreas -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] mpv: 0.26.0 -> 0.32.0
Remove mpv from the PNBLACKLIST by updating to the latest release and using a newer, python3-aware waf for building (copied from glmark2). The name of the license file changed, but the licence of the software itself did not. Added mime-xdg to inherit to handle mime artifacts. Added to the FILES command to scoop up additional files being generated by the build. Extensively fixed and cleaned up the PACKAGECONFIG options to reflect the current options available when configuring mpv. Also cleaned up any configure options which are no longer available. Tested with core-image-x11 on a rock-pi-4 with PACKAGECONFIG = "libass x11 opengl egl drm gbm vaapi vdpau" # mpv --vo=help Available video outputs: libmpv render API for libmpv gpu Shader-based GPU Renderer vdpauVDPAU with X11 vaapiVA API with X11 x11 X11 (slow, old crap) null Null video output imageWrite video frames to image files tct true-color terminals drm Direct Rendering Manager Signed-off-by: Trevor Woerner --- .../mplayer/mpv/python3.patch | 12 ++ .../recipes-multimedia/mplayer/mpv_0.26.0.bb | 85 - .../recipes-multimedia/mplayer/mpv_0.32.0.bb | 115 ++ 3 files changed, 127 insertions(+), 85 deletions(-) create mode 100644 meta-oe/recipes-multimedia/mplayer/mpv/python3.patch delete mode 100644 meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb create mode 100644 meta-oe/recipes-multimedia/mplayer/mpv_0.32.0.bb diff --git a/meta-oe/recipes-multimedia/mplayer/mpv/python3.patch b/meta-oe/recipes-multimedia/mplayer/mpv/python3.patch new file mode 100644 index 0..df208cf87 --- /dev/null +++ b/meta-oe/recipes-multimedia/mplayer/mpv/python3.patch @@ -0,0 +1,12 @@ +Use Python 3 instead of Python 2. + +Upstream-Status: Pending +Signed-off-by: Ross Burton + +diff --git a/waf b/waf +index 6ce2a24..de3c898 100755 +--- a/waf b/waf +@@ -1 +1 @@ +-#!/usr/bin/env python ++#!/usr/bin/env python3 diff --git a/meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb b/meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb deleted file mode 100644 index c98aa7ab0..0 --- a/meta-oe/recipes-multimedia/mplayer/mpv_0.26.0.bb +++ /dev/null @@ -1,85 +0,0 @@ -SUMMARY = "Open Source multimedia player" -DESCRIPTION = "mpv is a fork of mplayer2 and MPlayer. It shares some features with the former projects while introducing many more." -SECTION = "multimedia" -HOMEPAGE = "http://www.mpv.io/; -DEPENDS = "zlib ffmpeg jpeg virtual/libx11 xsp libxv \ - libxscrnsaver libv4l libxinerama \ -" - -REQUIRED_DISTRO_FEATURES = "x11" - -LICENSE = "GPLv2+" -LIC_FILES_CHKSUM = "file://LICENSE;md5=91f1cb870c1cc2d31351a4d2595441cb" - -# While this item does not require it, it depends on ffmpeg which does -LICENSE_FLAGS = "commercial" - -SRCREV_mpv = "4db82f04502cbcd20d9a975df9dfde91d8c389e3" -SRC_URI = " \ -git://github.com/mpv-player/mpv;name=mpv \ - http://www.freehackers.org/~tnagy/release/waf-1.8.12;name=waf;downloadfilename=waf;subdir=git \ -" -SRC_URI[waf.md5sum] = "cef4ee82206b1843db082d0b0506bf71" -SRC_URI[waf.sha256sum] = "01bf2beab2106d1558800c8709bc2c8e496d3da4a2ca343fe091f22fca60c98b" - -S = "${WORKDIR}/git" - -inherit waf pkgconfig python3native features_check - -LUA ?= "lua" -LUA_mips64 = "" -LUA_aarch64 = "" -LUA_powerpc64 = "" -LUA_powerpc64le = "" -LUA_powerpc = "" - -# Note: both lua and libass are required to get on-screen-display (controls) -PACKAGECONFIG ??= " \ -${LUA} \ -libass \ -${@bb.utils.filter('DISTRO_FEATURES', 'wayland', d)} \ -" - -PACKAGECONFIG_remove_aarch64 = "lua" -PACKAGECONFIG[drm] = "--enable-drm,--disable-drm,libdrm" -PACKAGECONFIG[gbm] = "--enable-gbm,--disable-gbm,virtual/mesa" -PACKAGECONFIG[lua] = "--enable-lua,--disable-lua,lua luajit" -PACKAGECONFIG[libass] = "--enable-libass,--disable-libass,libass" -PACKAGECONFIG[libarchive] = "--enable-libarchive,--disable-libarchive,libarchive" -PACKAGECONFIG[jack] = "--enable-jack, --disable-jack, jack" -PACKAGECONFIG[vaapi] = "--enable-vaapi, --disable-vaapi,libva" -PACKAGECONFIG[vdpau] = "--enable-vdpau, --disable-vdpau,libvdpau" -PACKAGECONFIG[wayland] = "--enable-wayland, --disable-wayland,wayland libxkbcommon" - -SIMPLE_TARGET_SYS = "${@'${TARGET_SYS}'.replace('${TARGET_VENDOR}', '')}" - -EXTRA_OECONF = " \ ---prefix=${prefix} \ ---target=${SIMPLE_TARGET_SYS} \ ---confdir=${sysconfdir} \ ---datadir=${datadir} \ ---disable-manpage-build \ ---disable-gl \ ---disable-libsmbclient \ ---disable-encoding \ ---disable-libbluray \ ---disable-dvdread \ ---disable-dvdnav \ ---disable-cdda \ ---disable-uchardet \ ---disable-rubberband \ ---disable-lcms2 \ ---disable-vapoursynth \ ---disable-vapoursynth-lazy \ -
Re: [oe] [PATCH][zeus] libldb: upgrade 1.5.5 -> 1.5.6
On Wed, Feb 5, 2020 at 4:57 PM akuster808 wrote: > > > > On 2/5/20 1:25 AM, Andreas Müller wrote: > > From: Yi Zhao > > > > Signed-off-by: Yi Zhao > > Signed-off-by: Khem Raj > > > > (cherry picked from commit 2b3fd534874aee125a6c79c70c93b1aa498fda3f) > > > > This is necessary for zeus to fix samba build as reported in [1] > > > > [1] > > http://lists.openembedded.org/pipermail/openembedded-devel/2020-January/204168.html > > I already have this > https://git.openembedded.org/meta-openembedded-contrib/commit/?h=stable/zeus-nut=f64c2d691ea597aa6221e3d7706a4d022e07bfea > > thanks for sending the patch. > > - armin Perfect - thanks - did not find that one. Andreas -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-python][zeus][PATCH v2 1/2] python-twisted: add required RDEPENDS for twistd
On 2/5/20 12:01 PM, Martin Jansa wrote: On Wed, Feb 05, 2020 at 10:16:01AM -0500, Trevor Gamblin wrote: On 2/5/20 9:59 AM, Martin Jansa wrote: On Mon, Feb 03, 2020 at 08:45:26PM -0800, Tim Orling wrote: On Mon, Feb 3, 2020 at 8:30 PM Martin Jansa wrote: Please send this fix to normal meta-python in meta-oe/zeus: http://git.openembedded.org/meta-openembedded/tree/meta-python/recipes-devtools/python/python-twisted.inc?h=zeus BTW: Why are you using meta-python2 with zeus branch? It imho doesn't have anything which would be missing in oe-core+meta-oe zeus branch and fixes like this should be resolved also for people using zeus branch without overlaying them by meta-python2. Because, as I believe I have said before, I need a prior release to test the infrastructure on. zeus allows me to work on more than just master until we have the next stable release branching. The question was for Trevor, I understood that you wanted zeus branch to test the infrastructure, but I still don't understand why any "project" should use meta-python2/zeus for the python2 bits which are still in oe-core/zeus + meta-python/zeus. I'm not sure if I'm misunderstanding, or if I have been the cause for confusion in my summary. When the reporter filed https://bugzilla.yoctoproject.org/show_bug.cgi?id=13741, it was clear he was using the python 2.x version of Twisted, on the zeus branch of meta-openembedded (based on the version of Twisted that he reported). The patch was written and tested using meta-oe/meta-python zeus and submitted for that layer. I suppose it could be applied to meta-python2 as well. Thanks for reply, not it's more clear that my brain somehow failed to parse "[oe] [meta-python][zeus][PATCH v2 1/2]", all I was seeing was "meta-python" and "2", since my first reply I was seeing this as meta-python2 and assumed that you're submitting it only there, not to meta-python in meta-oe/zeus which is completely reasonable. Sorry for noise I just need more sleep or coffee or both actually. Now when I see that you're working on meta-python in meta-oe, why not cherry-pick the commits from master as they are? Similar change in master: http://git.openembedded.org/meta-openembedded/commit/meta-python/recipes-devtools/python/python-twisted.inc?id=7077abdf29a56896ef4dbad94b28201689b70ee3 http://git.openembedded.org/meta-openembedded/commit/meta-python/recipes-devtools/python/python-twisted.inc?id=53e62e4e57c701dd546a2e27d8b72b7a4698551a it has a bit fewer modules in RDEPENDS. Is it because of 19.10.0 version in master while zeus has only 19.7.0 or are they actually missing in master as well? When testing for both zeus and master, there was a discrepancy between the number of RDEPENDS that needed to be added for /usr/bin/twistd to get to its usage/help without issues - master's Twisted 19.10.0 only needed unixadmin, debugger, and pyhamcrest, but the zeus build (with Twisted version 19.7.0) was still complaining until I added misc, pkg-resources, pkgutil, plistlib, pyserial, and setuptools as well, hence the separate patch. It's all making me realize that I haven't been getting some other patches for Twisted's RDEPENDS backported, though... Similarly what about the RDEPENDS of conch fixed in master with: http://git.openembedded.org/meta-openembedded/commit/meta-python/recipes-devtools/python/python-twisted.inc?id=c3a3470b1ba3768e04e0a938870ad487bd848120 isn't this needed in zeus as well? I don't use python-twisted anywhere, I was just curious why this isn't straight backport from master. I suppose it could be applied to meta-python2 as well. Yes, it looks like these changes are missing in meta-python2/master as well: https://git.openembedded.org/meta-python2/tree/recipes-devtools/python/python-twisted_19.10.0.bb#n66 Cheers, I see that there are some differences in the recipes in zeus, e.g.: martin@jama:/OE/meta-python2$ diff recipes-devtools/python/python-configparser_3.8.1.bb ../meta-openembedded/meta-python/recipes-devtools/python/python-configparser_3.8.1.bb 1c1 < inherit setuptools python-backports-init --- inherit setuptools martin@jama:/OE/meta-python2$ diff recipes-devtools/python/python-dbus_1.2.12.bb ../meta-openembedded/meta-python/recipes-devtools/python/python-dbus_1.2.12.bb 18,19c18,19 < # documentation needs python-sphinx, which is not in oe-core or meta-python2 for now < # change to use PACKAGECONFIG when python-sphinx is added to oe-core or meta-python2 --- # documentation needs python-sphinx, which is not in oe-core or meta-python for now # change to use PACKAGECONFIG when python-sphinx is added to oe-core or meta-python But README nor your reply explained to me, what's the benefit of using meta-python2/zeus instead of the existing python2 recipes which are in oe-core and meta-oe zeus. That's why I've never tried to use it with zeus, nor submitted the python-numpy and python-nose recipes which got removed from oe-core/zeus, but weren't re-introduced in meta-python2
Re: [oe] [meta-python][zeus][PATCH v2 1/2] python-twisted: add required RDEPENDS for twistd
On Wed, Feb 05, 2020 at 10:16:01AM -0500, Trevor Gamblin wrote: > > On 2/5/20 9:59 AM, Martin Jansa wrote: > > On Mon, Feb 03, 2020 at 08:45:26PM -0800, Tim Orling wrote: > > > On Mon, Feb 3, 2020 at 8:30 PM Martin Jansa > > > wrote: > > > > > > > Please send this fix to normal meta-python in meta-oe/zeus: > > > > > > > > http://git.openembedded.org/meta-openembedded/tree/meta-python/recipes-devtools/python/python-twisted.inc?h=zeus > > > > > > > > BTW: Why are you using meta-python2 with zeus branch? It imho doesn't > > > > have > > > > anything which would be missing in oe-core+meta-oe zeus branch and fixes > > > > like this should be resolved also for people using zeus branch without > > > > overlaying them by meta-python2. > > > > > > > > > > > Because, as I believe I have said before, I need a prior release to test > > > the infrastructure on. zeus allows me to work on more than just master > > > until we have the next stable release branching. > > The question was for Trevor, I understood that you wanted zeus branch to > > test the infrastructure, but I still don't understand why any "project" > > should use meta-python2/zeus for the python2 bits which are still in > > oe-core/zeus + meta-python/zeus. > I'm not sure if I'm misunderstanding, or if I have been the cause for > confusion in my summary. When the reporter filed > https://bugzilla.yoctoproject.org/show_bug.cgi?id=13741, it was clear he was > using the python 2.x version of Twisted, on the zeus branch of > meta-openembedded (based on the version of Twisted that he reported). The > patch was written and tested using meta-oe/meta-python zeus and submitted > for that layer. I suppose it could be applied to meta-python2 as well. Thanks for reply, not it's more clear that my brain somehow failed to parse "[oe] [meta-python][zeus][PATCH v2 1/2]", all I was seeing was "meta-python" and "2", since my first reply I was seeing this as meta-python2 and assumed that you're submitting it only there, not to meta-python in meta-oe/zeus which is completely reasonable. Sorry for noise I just need more sleep or coffee or both actually. Now when I see that you're working on meta-python in meta-oe, why not cherry-pick the commits from master as they are? Similar change in master: http://git.openembedded.org/meta-openembedded/commit/meta-python/recipes-devtools/python/python-twisted.inc?id=7077abdf29a56896ef4dbad94b28201689b70ee3 http://git.openembedded.org/meta-openembedded/commit/meta-python/recipes-devtools/python/python-twisted.inc?id=53e62e4e57c701dd546a2e27d8b72b7a4698551a it has a bit fewer modules in RDEPENDS. Is it because of 19.10.0 version in master while zeus has only 19.7.0 or are they actually missing in master as well? Similarly what about the RDEPENDS of conch fixed in master with: http://git.openembedded.org/meta-openembedded/commit/meta-python/recipes-devtools/python/python-twisted.inc?id=c3a3470b1ba3768e04e0a938870ad487bd848120 isn't this needed in zeus as well? I don't use python-twisted anywhere, I was just curious why this isn't straight backport from master. > I suppose it could be applied to meta-python2 as well. Yes, it looks like these changes are missing in meta-python2/master as well: https://git.openembedded.org/meta-python2/tree/recipes-devtools/python/python-twisted_19.10.0.bb#n66 Cheers, > > I see that there are some differences in the recipes in zeus, e.g.: > > > > martin@jama:/OE/meta-python2$ diff > > recipes-devtools/python/python-configparser_3.8.1.bb > > ../meta-openembedded/meta-python/recipes-devtools/python/python-configparser_3.8.1.bb > > 1c1 > > < inherit setuptools python-backports-init > > --- > > > inherit setuptools > > martin@jama:/OE/meta-python2$ diff > > recipes-devtools/python/python-dbus_1.2.12.bb > > ../meta-openembedded/meta-python/recipes-devtools/python/python-dbus_1.2.12.bb > > 18,19c18,19 > > < # documentation needs python-sphinx, which is not in oe-core or > > meta-python2 for now > > < # change to use PACKAGECONFIG when python-sphinx is added to oe-core or > > meta-python2 > > --- > > > # documentation needs python-sphinx, which is not in oe-core or > > > meta-python for now > > > # change to use PACKAGECONFIG when python-sphinx is added to oe-core or > > > meta-python > > But README nor your reply explained to me, what's the benefit of using > > meta-python2/zeus > > instead of the existing python2 recipes which are in oe-core and meta-oe > > zeus. That's why > > I've never tried to use it with zeus, nor submitted the python-numpy and > > python-nose > > recipes which got removed from oe-core/zeus, but weren't re-introduced in > > meta-python2 > > python-nose was removed in: > > http://git.openembedded.org/openembedded-core/commit/?h=zeus=9f07b2836405d86e869780f8f1ae00843eadc409 > > python-numpy in: > > http://git.openembedded.org/openembedded-core/commit/?h=zeus=4e9659f24e2f699effadcbe378b6a746d77ccdbd > > from oe-core/zeus. > > > > > It has nothing to do with what is
Re: [oe] [meta-oe][PATCH] smem: use python 3
Hi André Thanks for your patch. This seems more complete and should go in in favour of this one. No I didn't test with --source, I only tested using smem directly on the embedded target which worked well for all but the charting options. I'm aware that the charting option gracefully declines to work when the plugins you now added with RRECOMMEND are missing. Max Am Mi., 5. Feb. 2020 um 17:39 Uhr schrieb André Draszik : > Hi, > > Interestingly I had been looking at this today as well... > > On Wed, 2020-02-05 at 00:40 +0100, Max Krummenacher wrote: > > There seems to be no user of python-textutils python-shell python-codecs > > in the script, so these rdepends are dropped. > > > > On a i.MX6 based machine smem runs with just python3-compression and > > its dependencies installed. > > Upstream has released v1.5 with almost complete python3 support. > I'd say any patches should go on top of that. > > Also, I don't think smem --source= will work with your version? Did you > test that? > > > > > Signed-off-by: Max Krummenacher > > --- > > .../smem/0001-smem-convert-to-python-3.patch | 161 ++ > > meta-oe/recipes-support/smem/smem_1.4.bb | 7 +- > > 2 files changed, 166 insertions(+), 2 deletions(-) > > create mode 100644 > meta-oe/recipes-support/smem/smem/0001-smem-convert-to-python-3.patch > > > > diff --git > a/meta-oe/recipes-support/smem/smem/0001-smem-convert-to-python-3.patch > b/meta-oe/recipes- > > support/smem/smem/0001-smem-convert-to-python-3.patch > > new file mode 100644 > > index 0..85f35be44 > > --- /dev/null > > +++ > b/meta-oe/recipes-support/smem/smem/0001-smem-convert-to-python-3.patch > > @@ -0,0 +1,161 @@ > > +From 5535a524f6db57c0b2e4a539db733fd9e840c97e Mon Sep 17 00:00:00 2001 > > +From: Max Krummenacher > > +Date: Mon, 3 Feb 2020 21:04:42 + > > +Subject: [PATCH] smem: convert to python 3 > > + > > +Upstream-Status: Pending > > + > > +Signed-off-by: Max Krummenacher > > +--- > > + smem | 41 ++--- > > + 1 file changed, 22 insertions(+), 19 deletions(-) > > + > > +diff --git a/smem b/smem > > +index 0cbd925..577d27b 100755 > > +--- a/smem > > b/smem > > +@@ -1,4 +1,4 @@ > > +-#!/usr/bin/env python > > ++#!/usr/bin/env python3 > > + # > > + # smem - a tool for meaningful memory reporting > > + # > > +@@ -21,7 +21,10 @@ class procdata(object): > > + def _list(self): > > + return os.listdir(self.source + "/proc") > > + def _read(self, f): > > +-return file(self.source + '/proc/' + f).read() > > ++afile = open(self.source + '/proc/' + f) > > ++ret = afile.read() > > ++afile.close() > > ++return ret > > + def _readlines(self, f): > > + return self._read(f).splitlines(True) > > + def _stat(self, f): > > +@@ -209,7 +212,7 @@ def fromunits(x): > > + s = dict(k=2**10, K=2**10, kB=2**10, KB=2**10, > > + M=2**20, MB=2**20, G=2**30, GB=2**30, > > + T=2**40, TB=2**40) > > +-for k,v in s.items(): > > ++for k,v in list(s.items()): > > + if x.endswith(k): > > + return int(float(x[:-len(k)])*v) > > + sys.stderr.write("Memory size should be written with units, for > example 1024M\n") > > +@@ -240,7 +243,7 @@ def pidtotals(pid): > > + maps = pidmaps(pid) > > + t = dict(size=0, rss=0, pss=0, shared_clean=0, shared_dirty=0, > > + private_clean=0, private_dirty=0, referenced=0, swap=0) > > +-for m in maps.iterkeys(): > > ++for m in maps.keys(): > > + for k in t: > > + t[k] += maps[m].get(k, 0) > > + > > +@@ -296,7 +299,7 @@ def showpids(): > > + ) > > + columns = options.columns or 'pid user command swap uss pss rss' > > + > > +-showtable(pt.keys(), fields, columns.split(), options.sort or > 'pss') > > ++showtable(list(pt.keys()), fields, columns.split(), options.sort > or 'pss') > > + > > + def maptotals(pids): > > + totals = {} > > +@@ -307,7 +310,7 @@ def maptotals(pids): > > + try: > > + maps = pidmaps(pid) > > + seen = {} > > +-for m in maps.iterkeys(): > > ++for m in maps.keys(): > > + name = maps[m]['name'] > > + if name not in totals: > > + t = dict(size=0, rss=0, pss=0, shared_clean=0, > > +@@ -361,7 +364,7 @@ def showmaps(): > > + ) > > + columns = options.columns or 'map pids avgpss pss' > > + > > +-showtable(pt.keys(), fields, columns.split(), options.sort or > 'pss') > > ++showtable(list(pt.keys()), fields, columns.split(), options.sort > or 'pss') > > + > > + def usertotals(pids): > > + totals = {} > > +@@ -383,7 +386,7 @@ def usertotals(pids): > > + else: > > + t = totals[user] > > + > > +-for m in maps.iterkeys(): > > ++for m in maps.keys(): > > + for k in t: > > + t[k] += maps[m].get(k, 0) > > + > > +@@ -419,7
Re: [oe] [meta-oe][PATCH] smem: use python 3
Hi, Interestingly I had been looking at this today as well... On Wed, 2020-02-05 at 00:40 +0100, Max Krummenacher wrote: > There seems to be no user of python-textutils python-shell python-codecs > in the script, so these rdepends are dropped. > > On a i.MX6 based machine smem runs with just python3-compression and > its dependencies installed. Upstream has released v1.5 with almost complete python3 support. I'd say any patches should go on top of that. Also, I don't think smem --source= will work with your version? Did you test that? > > Signed-off-by: Max Krummenacher > --- > .../smem/0001-smem-convert-to-python-3.patch | 161 ++ > meta-oe/recipes-support/smem/smem_1.4.bb | 7 +- > 2 files changed, 166 insertions(+), 2 deletions(-) > create mode 100644 > meta-oe/recipes-support/smem/smem/0001-smem-convert-to-python-3.patch > > diff --git > a/meta-oe/recipes-support/smem/smem/0001-smem-convert-to-python-3.patch > b/meta-oe/recipes- > support/smem/smem/0001-smem-convert-to-python-3.patch > new file mode 100644 > index 0..85f35be44 > --- /dev/null > +++ b/meta-oe/recipes-support/smem/smem/0001-smem-convert-to-python-3.patch > @@ -0,0 +1,161 @@ > +From 5535a524f6db57c0b2e4a539db733fd9e840c97e Mon Sep 17 00:00:00 2001 > +From: Max Krummenacher > +Date: Mon, 3 Feb 2020 21:04:42 + > +Subject: [PATCH] smem: convert to python 3 > + > +Upstream-Status: Pending > + > +Signed-off-by: Max Krummenacher > +--- > + smem | 41 ++--- > + 1 file changed, 22 insertions(+), 19 deletions(-) > + > +diff --git a/smem b/smem > +index 0cbd925..577d27b 100755 > +--- a/smem > b/smem > +@@ -1,4 +1,4 @@ > +-#!/usr/bin/env python > ++#!/usr/bin/env python3 > + # > + # smem - a tool for meaningful memory reporting > + # > +@@ -21,7 +21,10 @@ class procdata(object): > + def _list(self): > + return os.listdir(self.source + "/proc") > + def _read(self, f): > +-return file(self.source + '/proc/' + f).read() > ++afile = open(self.source + '/proc/' + f) > ++ret = afile.read() > ++afile.close() > ++return ret > + def _readlines(self, f): > + return self._read(f).splitlines(True) > + def _stat(self, f): > +@@ -209,7 +212,7 @@ def fromunits(x): > + s = dict(k=2**10, K=2**10, kB=2**10, KB=2**10, > + M=2**20, MB=2**20, G=2**30, GB=2**30, > + T=2**40, TB=2**40) > +-for k,v in s.items(): > ++for k,v in list(s.items()): > + if x.endswith(k): > + return int(float(x[:-len(k)])*v) > + sys.stderr.write("Memory size should be written with units, for example > 1024M\n") > +@@ -240,7 +243,7 @@ def pidtotals(pid): > + maps = pidmaps(pid) > + t = dict(size=0, rss=0, pss=0, shared_clean=0, shared_dirty=0, > + private_clean=0, private_dirty=0, referenced=0, swap=0) > +-for m in maps.iterkeys(): > ++for m in maps.keys(): > + for k in t: > + t[k] += maps[m].get(k, 0) > + > +@@ -296,7 +299,7 @@ def showpids(): > + ) > + columns = options.columns or 'pid user command swap uss pss rss' > + > +-showtable(pt.keys(), fields, columns.split(), options.sort or 'pss') > ++showtable(list(pt.keys()), fields, columns.split(), options.sort or > 'pss') > + > + def maptotals(pids): > + totals = {} > +@@ -307,7 +310,7 @@ def maptotals(pids): > + try: > + maps = pidmaps(pid) > + seen = {} > +-for m in maps.iterkeys(): > ++for m in maps.keys(): > + name = maps[m]['name'] > + if name not in totals: > + t = dict(size=0, rss=0, pss=0, shared_clean=0, > +@@ -361,7 +364,7 @@ def showmaps(): > + ) > + columns = options.columns or 'map pids avgpss pss' > + > +-showtable(pt.keys(), fields, columns.split(), options.sort or 'pss') > ++showtable(list(pt.keys()), fields, columns.split(), options.sort or > 'pss') > + > + def usertotals(pids): > + totals = {} > +@@ -383,7 +386,7 @@ def usertotals(pids): > + else: > + t = totals[user] > + > +-for m in maps.iterkeys(): > ++for m in maps.keys(): > + for k in t: > + t[k] += maps[m].get(k, 0) > + > +@@ -419,7 +422,7 @@ def showusers(): > + ) > + columns = options.columns or 'user count swap uss pss rss' > + > +-showtable(pt.keys(), fields, columns.split(), options.sort or 'pss') > ++showtable(list(pt.keys()), fields, columns.split(), options.sort or > 'pss') > + > + def showsystem(): > + t = totalmem() > +@@ -460,14 +463,14 @@ def showsystem(): > + 'area not reclaimable')) > + > + columns = options.columns or 'area used cache noncache' > +-showtable(range(len(l)), fields, columns.split(), options.sort or > 'order') > ++showtable(list(range(len(l))), fields, columns.split(), options.sort or >
[oe] [meta-oe][master-next][PATCH 2/2] smem: update to v1.5 (python3 compatibility)
smem generally works using the upstream code, but the --source argument still throws exceptions - add a patch to resolve those as well. There is no release on the downloads URL, but a tag in mercurial - so this now downloads the snapshot that tag is pointing to. Signed-off-by: André Draszik --- ...ix-support-for-source-option-python3.patch | 53 +++ .../smem/{smem_1.4.bb => smem_1.5.bb} | 20 +-- 2 files changed, 68 insertions(+), 5 deletions(-) create mode 100644 meta-oe/recipes-support/smem/smem/0001-smem-fix-support-for-source-option-python3.patch rename meta-oe/recipes-support/smem/{smem_1.4.bb => smem_1.5.bb} (56%) diff --git a/meta-oe/recipes-support/smem/smem/0001-smem-fix-support-for-source-option-python3.patch b/meta-oe/recipes-support/smem/smem/0001-smem-fix-support-for-source-option-python3.patch new file mode 100644 index 0..5c1be5a0a --- /dev/null +++ b/meta-oe/recipes-support/smem/smem/0001-smem-fix-support-for-source-option-python3.patch @@ -0,0 +1,53 @@ +From 3ff78f1f00973393d1a7ee4e467a2bacf1c807f3 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Andr=C3=A9=20Draszik?= +Date: Wed, 5 Feb 2020 16:14:21 + +Subject: [PATCH] smem: fix support for --source option (python3) +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +Using --source doesn't work without this patch: +Traceback (most recent call last): + File "./smem", line 727, in +showpids() + File "./smem", line 299, in showpids +showtable(pt.keys(), fields, columns.split(), options.sort or 'pss') + File "./smem", line 519, in showtable +mt = totalmem() + File "./smem", line 118, in totalmem +_totalmem = memory()['memtotal'] + File "./smem", line 193, in memory +m = f.match(l) +TypeError: cannot use a string pattern on a bytes-like object + +python3's tarfile returns bytes, whereas all of the rest of +the code assumes str. + +Fix the tarfile usage to convert to str before returning the +results. + +Signed-off-by: André Draszik +Upstream-Status: Inappropriate [upstream wants to support python2 & python3] +--- + smem | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/smem b/smem +index 46a3189..54d40dd 100755 +--- a/smem b/smem +@@ -90,9 +90,9 @@ class tardata(procdata): + d,f = ti.name.split('/') + yield d + def _read(self, f): +-return self.tar.extractfile(f).read() ++return self.tar.extractfile(f).read().decode() + def _readlines(self, f): +-return self.tar.extractfile(f).readlines() ++return [l.decode() for l in self.tar.extractfile(f).readlines()] + def piduser(self, p): + t = self.tar.getmember("%d" % p) + if t.uname: +-- +2.23.0.rc1 + diff --git a/meta-oe/recipes-support/smem/smem_1.4.bb b/meta-oe/recipes-support/smem/smem_1.5.bb similarity index 56% rename from meta-oe/recipes-support/smem/smem_1.4.bb rename to meta-oe/recipes-support/smem/smem_1.5.bb index 947c47b0f..90db9c3f3 100644 --- a/meta-oe/recipes-support/smem/smem_1.4.bb +++ b/meta-oe/recipes-support/smem/smem_1.5.bb @@ -9,9 +9,16 @@ SECTION = "Applications/System" LICENSE = "GPLv2+" LIC_FILES_CHKSUM = "file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263" -SRC_URI = "http://www.selenic.com/${BPN}/download/${BP}.tar.gz; -SRC_URI[md5sum] = "fe79435c3930389bfdb560255c802162" -SRC_URI[sha256sum] = "2ea9f878f4cf3c276774c3f7e2a41977a1f2d64f98d2dcb6a15f1f3d84df61ec" +HG_CHANGESET = "98273ce331bb" +SRC_URI = "https://selenic.com/repo/${BPN}/archive/${HG_CHANGESET}.tar.bz2;downloadfilename=${BP}.tar.bz2 \ + file://0001-smem-fix-support-for-source-option-python3.patch" +SRC_URI[md5sum] = "51c3989779360f42b42ef46b2831be3a" +SRC_URI[sha256sum] = "161131c686a6d9962a0e96912526dd46308e022d62e3f8acaed5a56fda8e08ce" + +UPSTREAM_CHECK_URI = "https://selenic.com/repo/smem/tags; +UPSTREAM_CHECK_REGEX = "(?P\d+(\.\d+)+)" + +S = "${WORKDIR}/${BPN}-${HG_CHANGESET}" do_compile() { ${CC} ${CFLAGS} ${LDFLAGS} smemcap.c -o smemcap @@ -21,12 +28,15 @@ do_install() { install -d ${D}/${bindir}/ install -d ${D}/${mandir}/man8 install -m 0755 ${S}/smem ${D}${bindir}/ +sed -i -e '1s,#!.*python.*,#!${USRBINPATH}/env python3,' ${D}${bindir}/smem install -m 0755 ${S}/smemcap ${D}${bindir}/ install -m 0644 ${S}/smem.8 ${D}/${mandir}/man8/ } -RDEPENDS_${PN} += "python-textutils python-compression python-shell python-codecs" -PACKAGES =+ "smemcap" +RDEPENDS_${PN} = "python3-core python3-compression" +RRECOMMENDS_${PN} = "python3-matplotlib python3-numpy" + +PACKAGE_BEFORE_PN = "smemcap" FILES_smemcap = "${bindir}/smemcap" -- 2.23.0.rc1 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][master-next][PATCH 1/2] python3-matplotlib: provide a -native variant
From: André Draszik This is useful e.g. for smem-native. Signed-off-by: André Draszik --- meta-python/recipes-devtools/python/python3-matplotlib_3.1.1.bb | 2 ++ 1 file changed, 2 insertions(+) diff --git a/meta-python/recipes-devtools/python/python3-matplotlib_3.1.1.bb b/meta-python/recipes-devtools/python/python3-matplotlib_3.1.1.bb index 9b86ccc83..824680c24 100644 --- a/meta-python/recipes-devtools/python/python3-matplotlib_3.1.1.bb +++ b/meta-python/recipes-devtools/python/python3-matplotlib_3.1.1.bb @@ -15,3 +15,5 @@ SRC_URI[sha256sum] = "1febd22afe1489b13c6749ea059d392c03261b2950d1d45c17e3aed812 PYPI_PACKAGE = "matplotlib" inherit pypi setuptools3 + +BBCLASSEXTEND = "native" -- 2.23.0.rc1 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [PATCH][zeus] libldb: upgrade 1.5.5 -> 1.5.6
On 2/5/20 1:25 AM, Andreas Müller wrote: > From: Yi Zhao > > Signed-off-by: Yi Zhao > Signed-off-by: Khem Raj > > (cherry picked from commit 2b3fd534874aee125a6c79c70c93b1aa498fda3f) > > This is necessary for zeus to fix samba build as reported in [1] > > [1] > http://lists.openembedded.org/pipermail/openembedded-devel/2020-January/204168.html I already have this https://git.openembedded.org/meta-openembedded-contrib/commit/?h=stable/zeus-nut=f64c2d691ea597aa6221e3d7706a4d022e07bfea thanks for sending the patch. - armin > > Signed-off-by: Andreas Müller > --- > .../libldb/{libldb_1.5.5.bb => libldb_1.5.6.bb} | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > rename meta-networking/recipes-support/libldb/{libldb_1.5.5.bb => > libldb_1.5.6.bb} (95%) > > diff --git a/meta-networking/recipes-support/libldb/libldb_1.5.5.bb > b/meta-networking/recipes-support/libldb/libldb_1.5.6.bb > similarity index 95% > rename from meta-networking/recipes-support/libldb/libldb_1.5.5.bb > rename to meta-networking/recipes-support/libldb/libldb_1.5.6.bb > index b19c964d8..99eb6f9ce 100644 > --- a/meta-networking/recipes-support/libldb/libldb_1.5.5.bb > +++ b/meta-networking/recipes-support/libldb/libldb_1.5.6.bb > @@ -33,8 +33,8 @@ LIC_FILES_CHKSUM = > "file://pyldb.h;endline=24;md5=dfbd238cecad76957f7f860fbe9ada > > file://man/ldb.3.xml;beginline=261;endline=262;md5=137f9fd61040c1505d1aa1019663fd08 > \ > > file://tools/ldbdump.c;endline=19;md5=a7d4fc5d1f75676b49df491575a86a42" > > -SRC_URI[md5sum] = "339a41ac9efaec680e6d1469dd9aa35b" > -SRC_URI[sha256sum] = > "199f5861aa863f538ec66d5fa95ecc13254a2030c53daf0e47363fa9ba235c68" > +SRC_URI[md5sum] = "fc58ef432c1fcb03fc3bb6e08977" > +SRC_URI[sha256sum] = > "ff82474d0bf109e415a2d50334bde5715f486a53ff4bb8c7f74459dd229e975b" > > inherit waf-samba distro_features_check > REQUIRED_DISTRO_FEATURES = "pam" -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-python][zeus][PATCH v2 1/2] python-twisted: add required RDEPENDS for twistd
On 2/5/20 9:59 AM, Martin Jansa wrote: On Mon, Feb 03, 2020 at 08:45:26PM -0800, Tim Orling wrote: On Mon, Feb 3, 2020 at 8:30 PM Martin Jansa wrote: Please send this fix to normal meta-python in meta-oe/zeus: http://git.openembedded.org/meta-openembedded/tree/meta-python/recipes-devtools/python/python-twisted.inc?h=zeus BTW: Why are you using meta-python2 with zeus branch? It imho doesn't have anything which would be missing in oe-core+meta-oe zeus branch and fixes like this should be resolved also for people using zeus branch without overlaying them by meta-python2. Because, as I believe I have said before, I need a prior release to test the infrastructure on. zeus allows me to work on more than just master until we have the next stable release branching. The question was for Trevor, I understood that you wanted zeus branch to test the infrastructure, but I still don't understand why any "project" should use meta-python2/zeus for the python2 bits which are still in oe-core/zeus + meta-python/zeus. I'm not sure if I'm misunderstanding, or if I have been the cause for confusion in my summary. When the reporter filed https://bugzilla.yoctoproject.org/show_bug.cgi?id=13741, it was clear he was using the python 2.x version of Twisted, on the zeus branch of meta-openembedded (based on the version of Twisted that he reported). The patch was written and tested using meta-oe/meta-python zeus and submitted for that layer. I suppose it could be applied to meta-python2 as well. I see that there are some differences in the recipes in zeus, e.g.: martin@jama:/OE/meta-python2$ diff recipes-devtools/python/python-configparser_3.8.1.bb ../meta-openembedded/meta-python/recipes-devtools/python/python-configparser_3.8.1.bb 1c1 < inherit setuptools python-backports-init --- inherit setuptools martin@jama:/OE/meta-python2$ diff recipes-devtools/python/python-dbus_1.2.12.bb ../meta-openembedded/meta-python/recipes-devtools/python/python-dbus_1.2.12.bb 18,19c18,19 < # documentation needs python-sphinx, which is not in oe-core or meta-python2 for now < # change to use PACKAGECONFIG when python-sphinx is added to oe-core or meta-python2 --- # documentation needs python-sphinx, which is not in oe-core or meta-python for now # change to use PACKAGECONFIG when python-sphinx is added to oe-core or meta-python But README nor your reply explained to me, what's the benefit of using meta-python2/zeus instead of the existing python2 recipes which are in oe-core and meta-oe zeus. That's why I've never tried to use it with zeus, nor submitted the python-numpy and python-nose recipes which got removed from oe-core/zeus, but weren't re-introduced in meta-python2 python-nose was removed in: http://git.openembedded.org/openembedded-core/commit/?h=zeus=9f07b2836405d86e869780f8f1ae00843eadc409 python-numpy in: http://git.openembedded.org/openembedded-core/commit/?h=zeus=4e9659f24e2f699effadcbe378b6a746d77ccdbd from oe-core/zeus. It has nothing to do with what is missing from meta-oe, it is a viable alternative that also helps prove out my workflow as I have EXTREMELY LIMITED TIME to support this. Understood, thanks for doing this work, I'm happy to use and contribute to master branch, it's just that I still don't understand the purpose of zeus branch (other than testing the infrastructure by you) and I hope you agree that bug fixes for python2 issues like the one in this thread belong to meta-oe/meta-python zeus in first place (because that's the layer which what most people are probably still using with zeus). Cheers, On Fri, Jan 31, 2020 at 10:29 AM Trevor Gamblin < trevor.gamb...@windriver.com> wrote: From: Trevor Gamblin python-twisted is currently missing some packages for its /usr/bin/twistd executable to function. Add these to the RDEPENDS list, and also clean up formatting since the list is getting long. Note that a similar patch has already been submitted for master branch, but the python2 version of twisted needs more explicit RDEPENDS than the python3 version submitted to master in order for /usr/bin/twistd to work properly. Signed-off-by: Trevor Gamblin --- .../recipes-devtools/python/python-twisted.inc | 18 +- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/meta-python/recipes-devtools/python/python-twisted.inc b/meta-python/recipes-devtools/python/python-twisted.inc index d70104dc2..024a9fa7c 100644 --- a/meta-python/recipes-devtools/python/python-twisted.inc +++ b/meta-python/recipes-devtools/python/python-twisted.inc @@ -56,7 +56,23 @@ RDEPENDS_${PN} = "\ ${PN}-zsh \ " -RDEPENDS_${PN}-core = "${PYTHON_PN}-core ${PYTHON_PN}-zopeinterface ${PYTHON_PN}-incremental ${PYTHON_PN}-constantly ${PYTHON_PN}-hyperlink ${PYTHON_PN}-automat" +RDEPENDS_${PN}-core = "${PYTHON_PN}-automat \ + ${PYTHON_PN}-constantly \ + ${PYTHON_PN}-core \ + ${PYTHON_PN}-debugger \ +
Re: [oe] [meta-python][zeus][PATCH v2 1/2] python-twisted: add required RDEPENDS for twistd
On Mon, Feb 03, 2020 at 08:45:26PM -0800, Tim Orling wrote: > On Mon, Feb 3, 2020 at 8:30 PM Martin Jansa wrote: > > > Please send this fix to normal meta-python in meta-oe/zeus: > > > > http://git.openembedded.org/meta-openembedded/tree/meta-python/recipes-devtools/python/python-twisted.inc?h=zeus > > > > BTW: Why are you using meta-python2 with zeus branch? It imho doesn't have > > anything which would be missing in oe-core+meta-oe zeus branch and fixes > > like this should be resolved also for people using zeus branch without > > overlaying them by meta-python2. > > > > > Because, as I believe I have said before, I need a prior release to test > the infrastructure on. zeus allows me to work on more than just master > until we have the next stable release branching. The question was for Trevor, I understood that you wanted zeus branch to test the infrastructure, but I still don't understand why any "project" should use meta-python2/zeus for the python2 bits which are still in oe-core/zeus + meta-python/zeus. I see that there are some differences in the recipes in zeus, e.g.: martin@jama:/OE/meta-python2$ diff recipes-devtools/python/python-configparser_3.8.1.bb ../meta-openembedded/meta-python/recipes-devtools/python/python-configparser_3.8.1.bb 1c1 < inherit setuptools python-backports-init --- > inherit setuptools martin@jama:/OE/meta-python2$ diff recipes-devtools/python/python-dbus_1.2.12.bb ../meta-openembedded/meta-python/recipes-devtools/python/python-dbus_1.2.12.bb 18,19c18,19 < # documentation needs python-sphinx, which is not in oe-core or meta-python2 for now < # change to use PACKAGECONFIG when python-sphinx is added to oe-core or meta-python2 --- > # documentation needs python-sphinx, which is not in oe-core or meta-python > for now > # change to use PACKAGECONFIG when python-sphinx is added to oe-core or > meta-python But README nor your reply explained to me, what's the benefit of using meta-python2/zeus instead of the existing python2 recipes which are in oe-core and meta-oe zeus. That's why I've never tried to use it with zeus, nor submitted the python-numpy and python-nose recipes which got removed from oe-core/zeus, but weren't re-introduced in meta-python2 python-nose was removed in: http://git.openembedded.org/openembedded-core/commit/?h=zeus=9f07b2836405d86e869780f8f1ae00843eadc409 python-numpy in: http://git.openembedded.org/openembedded-core/commit/?h=zeus=4e9659f24e2f699effadcbe378b6a746d77ccdbd from oe-core/zeus. > It has nothing to do with what is missing from meta-oe, it is a viable > alternative that also helps prove out my workflow as I have EXTREMELY > LIMITED TIME to support this. Understood, thanks for doing this work, I'm happy to use and contribute to master branch, it's just that I still don't understand the purpose of zeus branch (other than testing the infrastructure by you) and I hope you agree that bug fixes for python2 issues like the one in this thread belong to meta-oe/meta-python zeus in first place (because that's the layer which what most people are probably still using with zeus). Cheers, > On Fri, Jan 31, 2020 at 10:29 AM Trevor Gamblin < > > trevor.gamb...@windriver.com> wrote: > > > > > From: Trevor Gamblin > > > > > > python-twisted is currently missing some packages for its /usr/bin/twistd > > > executable to function. Add these to the RDEPENDS list, and also clean up > > > formatting since the list is getting long. > > > > > > Note that a similar patch has already been submitted for master branch, > > > but the python2 version of twisted needs more explicit RDEPENDS than the > > > python3 version submitted to master in order for /usr/bin/twistd to work > > > properly. > > > > > > Signed-off-by: Trevor Gamblin > > > --- > > > .../recipes-devtools/python/python-twisted.inc | 18 +- > > > 1 file changed, 17 insertions(+), 1 deletion(-) > > > > > > diff --git a/meta-python/recipes-devtools/python/python-twisted.inc > > > b/meta-python/recipes-devtools/python/python-twisted.inc > > > index d70104dc2..024a9fa7c 100644 > > > --- a/meta-python/recipes-devtools/python/python-twisted.inc > > > +++ b/meta-python/recipes-devtools/python/python-twisted.inc > > > @@ -56,7 +56,23 @@ RDEPENDS_${PN} = "\ > > > ${PN}-zsh \ > > > " > > > > > > -RDEPENDS_${PN}-core = "${PYTHON_PN}-core ${PYTHON_PN}-zopeinterface > > > ${PYTHON_PN}-incremental ${PYTHON_PN}-constantly ${PYTHON_PN}-hyperlink > > > ${PYTHON_PN}-automat" > > > +RDEPENDS_${PN}-core = "${PYTHON_PN}-automat \ > > > + ${PYTHON_PN}-constantly \ > > > + ${PYTHON_PN}-core \ > > > + ${PYTHON_PN}-debugger \ > > > + ${PYTHON_PN}-hyperlink \ > > > + ${PYTHON_PN}-incremental \ > > > + ${PYTHON_PN}-misc \ > > > + ${PYTHON_PN}-pkg-resources \ > > > + ${PYTHON_PN}-pkgutil \ > > > +
[oe] [meta-python2][PATCH] python2: add ntpath
From: Armin Kuster python3 has this but python is missing this. [Yocto #13740] Signed-off-by: Armin Kuster Signed-off-by: Tim Orling --- recipes-devtools/python/python/python2-manifest.json | 1 + 1 file changed, 1 insertion(+) diff --git a/recipes-devtools/python/python/python2-manifest.json b/recipes-devtools/python/python/python2-manifest.json index eb52e862..fd98774d 100644 --- a/recipes-devtools/python/python/python2-manifest.json +++ b/recipes-devtools/python/python/python2-manifest.json @@ -267,6 +267,7 @@ "${libdir}/python2.7/lib-dynload/xreadlines.so", "${libdir}/python2.7/linecache.py", "${libdir}/python2.7/new.py", +"${libdir}/python2.7/ntpath.py", "${libdir}/python2.7/os.py", "${libdir}/python2.7/platform.py", "${libdir}/python2.7/posixpath.py", -- 2.25.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
Re: [oe] [meta-java][g0hl1n/wip-jdk8u242ga][PATCH] Fix building openjre-8 242
Hi Richard and Jacob, I just stumbled across the very same error, but on aarch64. I tried to switch to the wip-branch, but I get ParseError at /[...]/meta-java/recipes-core/openjdk/openjdk-7_99b00-2.6.5.bb:7: Could not inherit file classes/features_check.bbclass I am on zeus... Regards, Matthias On 1/27/20 6:22 AM, Richard Leitner wrote: On Sun, Jan 26, 2020 at 09:03:45PM +0100, Jacob Kroon wrote: On 1/26/20 12:26 PM, Jacob Kroon wrote: ... Btw, as I mentioned my build host is an up2date Fedora 31, x86-64. Any idea why these errors aren't showing up on your test builds ? In my tests I only built the openjdk package, therefore the "find" issue didn't trigger. The linker errors were fixed in aarch32 [1] and aarch64 [2] upstream. I just double-checked my tests and it seems I only built for those two architectures... Time for me to rework my tests ;-) So I will adapt and apply your PATCH and push it to the branch. Thanks for your work! If you have any further feedback/tests for the 242 update please feel free to share ;-) regards;rl [1] http://hg.openjdk.java.net/aarch32-port/jdk8u/jdk/rev/4a51a4ac7fc2 [2] https://hg.openjdk.java.net/shenandoah/jdk8/jdk/rev/391be061dfc7 Cheers, Jacob -- Dr.-Ing. Matthias Schöpfer matthias.schoep...@googlemail.com -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [meta-oe][PATCH] xmlsec1: fix for multilib
With previous change of nss/nspr in this recipe, multilib build failure at do_configure. Fix it by not using hardcoded /lib/. Signed-off-by: Chen Qi --- .../0001-nss-nspr-fix-for-multilib.patch | 52 +++ .../recipes-support/xmlsec1/xmlsec1_1.2.29.bb | 3 +- 2 files changed, 54 insertions(+), 1 deletion(-) create mode 100644 meta-oe/recipes-support/xmlsec1/xmlsec1/0001-nss-nspr-fix-for-multilib.patch diff --git a/meta-oe/recipes-support/xmlsec1/xmlsec1/0001-nss-nspr-fix-for-multilib.patch b/meta-oe/recipes-support/xmlsec1/xmlsec1/0001-nss-nspr-fix-for-multilib.patch new file mode 100644 index 0..53c119765 --- /dev/null +++ b/meta-oe/recipes-support/xmlsec1/xmlsec1/0001-nss-nspr-fix-for-multilib.patch @@ -0,0 +1,52 @@ +From 9ed3b4e628f9254d5c86006fe63c33a1eb02aee5 Mon Sep 17 00:00:00 2001 +From: Chen Qi +Date: Tue, 4 Feb 2020 23:39:49 -0800 +Subject: [PATCH] nss/nspr: fix for multilib + +Upstream-Status: Pending + +Signed-off-by: Chen Qi +--- + configure.ac | 16 + 1 file changed, 8 insertions(+), 8 deletions(-) + +diff --git a/configure.ac b/configure.ac +index b46d97d..39f4318 100644 +--- a/configure.ac b/configure.ac +@@ -920,24 +920,24 @@ fi + dnl Priority 1: User specifies the path to installation + if test "z$NSPR_FOUND" = "zno" -a "z$with_nspr" != "z" -a "z$with_nspr" != "zyes" ; then + AC_MSG_CHECKING(for nspr library installation in "$with_nspr" folder) +-if test -f "$with_nspr/include/$NSPR_INCLUDE_MARKER" -a -f "$with_nspr/lib/$NSPR_LIB_MARKER" ; then +-NSPR_INCLUDE_PATH="$with_nspr/include" +-NSPR_LIB_PATH="$with_nspr/lib" ++if test -f "$with_nspr/${includedir}/$NSPR_INCLUDE_MARKER" -a -f "$with_nspr/${libdir}/$NSPR_LIB_MARKER" ; then ++NSPR_INCLUDE_PATH="$with_nspr/${includedir}" ++NSPR_LIB_PATH="$with_nspr/${libdir}" + NSPR_FOUND="yes" + AC_MSG_RESULT([yes]) + else +-AC_MSG_ERROR([not found: "$with_nspr/include/$NSPR_INCLUDE_MARKER" and/or "$with_nspr/lib/$NSPR_LIB_MARKER" files don't exist), typo?]) ++AC_MSG_ERROR([not found: "$with_nspr/${includedir}/$NSPR_INCLUDE_MARKER" and/or "$with_nspr/${libdir}/$NSPR_LIB_MARKER" files don't exist), typo?]) + fi + fi + if test "z$NSS_FOUND" = "zno" -a "z$with_nss" != "z" -a "z$with_nss" != "zyes" ; then + AC_MSG_CHECKING(for nss library installation in "$with_nss" folder) +-if test -f "$with_nss/include/$NSS_INCLUDE_MARKER" -a -f "$with_nss/lib/$NSS_LIB_MARKER" ; then +-NSS_INCLUDE_PATH="$with_nss/include" +-NSS_LIB_PATH="$with_nss/lib" ++if test -f "$with_nss/${includedir}/$NSS_INCLUDE_MARKER" -a -f "$with_nss/${libdir}/$NSS_LIB_MARKER" ; then ++NSS_INCLUDE_PATH="$with_nss/${includedir}" ++NSS_LIB_PATH="$with_nss/${libdir}" + NSS_FOUND="yes" + AC_MSG_RESULT([yes]) + else +-AC_MSG_ERROR([not found: "$with_nss/include/$NSS_INCLUDE_MARKER" and/or "$with_nss/lib/$NSS_LIB_MARKER" files don't exist), typo?]) ++AC_MSG_ERROR([not found: "$with_nss/${includedir}/$NSS_INCLUDE_MARKER" and/or "$with_nss/${libdir}/$NSS_LIB_MARKER" files don't exist), typo?]) + fi + fi + +-- +2.21.0 + diff --git a/meta-oe/recipes-support/xmlsec1/xmlsec1_1.2.29.bb b/meta-oe/recipes-support/xmlsec1/xmlsec1_1.2.29.bb index 3e25e8d35..8fc611a27 100644 --- a/meta-oe/recipes-support/xmlsec1/xmlsec1_1.2.29.bb +++ b/meta-oe/recipes-support/xmlsec1/xmlsec1_1.2.29.bb @@ -17,6 +17,7 @@ SRC_URI = "http://www.aleksey.com/xmlsec/download/${BP}.tar.gz \ file://change-finding-path-of-nss.patch \ file://makefile-ptest.patch \ file://xmlsec1-examples-allow-build-in-separate-dir.patch \ +file://0001-nss-nspr-fix-for-multilib.patch \ file://run-ptest \ " @@ -31,7 +32,7 @@ inherit autotools-brokensep ptest pkgconfig PACKAGECONFIG ??= "gnutls libgcrypt nss openssl des" PACKAGECONFIG[gnutls] = ",,gnutls" PACKAGECONFIG[libgcrypt] = ",,libgcrypt" -PACKAGECONFIG[nss] = "--with-nss=${STAGING_LIBDIR}/.. --with-nspr=${STAGING_LIBDIR}/..,,nss nspr" +PACKAGECONFIG[nss] = "--with-nss=${STAGING_DIR_HOST} --with-nspr=${STAGING_DIR_HOST},,nss nspr" PACKAGECONFIG[openssl] = ",,openssl" PACKAGECONFIG[des] = ",--disable-des,," -- 2.21.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel
[oe] [PATCH][zeus] libldb: upgrade 1.5.5 -> 1.5.6
From: Yi Zhao Signed-off-by: Yi Zhao Signed-off-by: Khem Raj (cherry picked from commit 2b3fd534874aee125a6c79c70c93b1aa498fda3f) This is necessary for zeus to fix samba build as reported in [1] [1] http://lists.openembedded.org/pipermail/openembedded-devel/2020-January/204168.html Signed-off-by: Andreas Müller --- .../libldb/{libldb_1.5.5.bb => libldb_1.5.6.bb} | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) rename meta-networking/recipes-support/libldb/{libldb_1.5.5.bb => libldb_1.5.6.bb} (95%) diff --git a/meta-networking/recipes-support/libldb/libldb_1.5.5.bb b/meta-networking/recipes-support/libldb/libldb_1.5.6.bb similarity index 95% rename from meta-networking/recipes-support/libldb/libldb_1.5.5.bb rename to meta-networking/recipes-support/libldb/libldb_1.5.6.bb index b19c964d8..99eb6f9ce 100644 --- a/meta-networking/recipes-support/libldb/libldb_1.5.5.bb +++ b/meta-networking/recipes-support/libldb/libldb_1.5.6.bb @@ -33,8 +33,8 @@ LIC_FILES_CHKSUM = "file://pyldb.h;endline=24;md5=dfbd238cecad76957f7f860fbe9ada file://man/ldb.3.xml;beginline=261;endline=262;md5=137f9fd61040c1505d1aa1019663fd08 \ file://tools/ldbdump.c;endline=19;md5=a7d4fc5d1f75676b49df491575a86a42" -SRC_URI[md5sum] = "339a41ac9efaec680e6d1469dd9aa35b" -SRC_URI[sha256sum] = "199f5861aa863f538ec66d5fa95ecc13254a2030c53daf0e47363fa9ba235c68" +SRC_URI[md5sum] = "fc58ef432c1fcb03fc3bb6e08977" +SRC_URI[sha256sum] = "ff82474d0bf109e415a2d50334bde5715f486a53ff4bb8c7f74459dd229e975b" inherit waf-samba distro_features_check REQUIRED_DISTRO_FEATURES = "pam" -- 2.21.0 -- ___ Openembedded-devel mailing list Openembedded-devel@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-devel