|
Inclusive Language Proposal for YP/OE folllow-up
Jon Mason
This is a follow up to the Inclusive Language email I sent out on
January 24 (see https://lore.kernel.org/yocto/CAPoiz9wL16OTzxNHdw_5-L72GOHkhhM0--WZF7An071cx6sr_A@mail.gmail.com/). I'm adding a couple of additional mailing lists to this email that were not on the original distribution, in case there are developers for those areas which didn't see the original email. I really appreciate the positive feedback received on this. I've updated the wiki with the suggested changes by Ross (acked by Randy) and Mikko, and extrapolated that to be the following list: CVE_CHECK_PN_WHITELIST -> CVE_CHECK_SKIPRECIPE -> CVE_CHECK_SKIP_RECIPE CVE_CHECK_WHITELIST -> CVE_CHECK_IGNORECVE -> CVE_CHECK_IGNORE INHERIT_BLACKLIST -> INHERIT_RECIPESKIP -> INHERIT_RECIPE_SKIP SDK_LOCAL_CONF_BLACKLIST -> ESDK_LOCALCONF_REMOVE -> ESDK_LOCAL_CONF_REMOVE If anyone wants to nack the changes above, please let me know (or modify https://wiki.yoctoproject.org/wiki/Inclusive_language). Similarly, I've added all of the proposed names to the "approved name" column. On the same wiki, I've changed the tables to now have a volunteer developer column (from the "Assigned Developer" name) to make it more clear that we need help in doing these tasks. If there is anything you are interested in, please feel free to put your name down there and have at it. Thanks, Jon
|
||
|
|
||
|
Re: wget - The certificate has not yet been activated (does also happen in qemuarm "virt" machine)
On Thu, Feb 3, 2022 at 9:14 AM Matthias Klein
<matthias.klein@optimeas.de> wrote: this means you are choosing openSSL to provide TLS protocol implementation instead of gnuTLS
|
||
|
|
||
|
Re: wget - The certificate has not yet been activated (does also happen in qemuarm "virt" machine)
Matthias Klein
Hello together,
I can "fix" the bug by switching from gnutls to openssl: PACKAGECONFIG:remove = "gnutls" PACKAGECONFIG:append = " openssl" Can anyone explain this? What exactly does the change to openssl mean? Many greetings, Matthias
|
||
|
|
||
|
Minutes: Yocto Project Weekly Triage Meeting 2/3/2022
Trevor Gamblin
Wiki: https://wiki.yoctoproject.org/wiki/Bug_Triage Attendees: Alejandro, Alexandre, Andrei, Armin, Bruce,
Daiane, Michael, Pavel, Randy, Richard, Saul, Stephen, Steve,
Tim, Trevor ARs: - Armin/Bruce/Trevor/Saul to move Old Milestone bugs to M3 or
later Notes:
- ~50% of AB workers have been switched to SSDs. Failure rate
appears lower, but still TBD Medium+ 3.5 Unassigned Enhancements/Bugs: 76 (Last week
76) AB Bugs: 73
(Last week 76)
|
||
|
|
||
|
Re: IGC build issue with devtoolset-8 (GNU 8.3.1)
Monsees, Steven C (US)
Is there a CPP conflict due to :
./tmp/work/corei7-64-poky-linux/intel-graphics-compiler/1.0.11-r0/git/inc/common/UFO/portable_compiler.h
Is this a name space issue ?
From: yocto@... <yocto@...>
On Behalf Of Monsees, Steven C (US) via lists.yoctoproject.org
Sent: Thursday, February 3, 2022 8:33 AM To: yocto@... Subject: Re: [yocto] IGC build issue with devtoolset-8 (GNU 8.3.1)
Anybody ?
It appears to be a CPP issue in the IGC code… but not sure why. The “__syscall_slong_t” definition appears to be getting resolved out correctly when I replace the array with individual variables of the same type (and the IGC is working)…
I’d feel more comfortable patching the IGC, but am still trying to fully understand the root cause…
/usr/include/bits>diff stat.h stat.h_HOLD 106c106,108 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 164c166,168 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 08:29 smonsees@yix465383 /usr/include/bits>
Note:
tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/3d/common/iStdLib/File.h:
#elif defined(__GNUC__) # include <libgen.h> # include <sys/stat.h> <<<---Line #47 #endif
sys/stat.h:
#include <bits/types.h> /* For __mode_t and __dev_t. */
#include <bits/stat.h>
bits/stat.h: __syscall_slong_t __unused[3];
bits/types.h:__STD_TYPE __SYSCALL_SLONG_TYPE __syscall_slong_t;
#if defined __x86_64__ && defined __ILP32__ bits/typesizes.h:# define __SYSCALL_SLONG_TYPE __SQUAD_TYPE #else bits/typesizes.h:# define __SYSCALL_SLONG_TYPE __SLONGWORD_TYPE
#if __WORDSIZE == 32 bits/types.h:# define __SQUAD_TYPE __quad_t #elif __WORDSIZE == 64 bits/types.h:# define __SQUAD_TYPE long int
bits/types.h:
#define __SLONGWORD_TYPE long int
From: Monsees, Steven C (US)
I am building zeus with basic OpenCL support for centos 7.x, and using GNU 8.3.1 compiler and see the following Error when IGC is built, I see the same error when building with GNU 5.3.1…
Is this a known issue, is a patch available ? Any ideas why I might be seeing this ?
cpp.o -MF IGC/Compiler/CMakeFiles/Compiler.dir/CISACodeGen/DebugInfo.cpp.o.d -o IGC/Compiler/CMakeFiles/Compiler.dir/CISACodeGen/DebugInfo.cpp.o -c /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.cpp | In file included from /usr/include/sys/stat.h:106, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/../3d/common/iStdLib/File.h:47, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.hpp:42, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.cpp:26: | /usr/include/bits/stat.h:106:31: error: expected unqualified-id before ‘[’ token | __syscall_slong_t __unused[3]; | ^ | /usr/include/bits/stat.h:164:31: error: expected unqualified-id before ‘[’ token | __syscall_slong_t __unused[3]; |
I am using bitbake –k, and everything else builds correctly, and no other components references this…
If I modify the header like so:
/usr/include/bits>diff stat.h stat.h_HOLD 106c106,108 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 164c166,168 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 08:29 smonsees@yix465383 /usr/include/bits>
IGC builds clean, and test show it appears to working correctly…
I really shouldn’t be modifying the header, and would like to know what the real issue issue is…
Thanks, Steve
|
||
|
|
||
|
Re: IGC build issue with devtoolset-8 (GNU 8.3.1)
Monsees, Steven C (US)
Anybody ?
It appears to be a CPP issue in the IGC code… but not sure why. The “__syscall_slong_t” definition appears to be getting resolved out correctly when I replace the array with individual variables of the same type (and the IGC is working)…
I’d feel more comfortable patching the IGC, but am still trying to fully understand the root cause…
/usr/include/bits>diff stat.h stat.h_HOLD 106c106,108 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 164c166,168 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 08:29 smonsees@yix465383 /usr/include/bits>
Note:
tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/3d/common/iStdLib/File.h:
#elif defined(__GNUC__) # include <libgen.h> # include <sys/stat.h> <<<---Line #47 #endif
sys/stat.h:
#include <bits/types.h> /* For __mode_t and __dev_t. */
#include <bits/stat.h>
bits/stat.h: __syscall_slong_t __unused[3];
bits/types.h:__STD_TYPE __SYSCALL_SLONG_TYPE __syscall_slong_t;
#if defined __x86_64__ && defined __ILP32__ bits/typesizes.h:# define __SYSCALL_SLONG_TYPE __SQUAD_TYPE #else bits/typesizes.h:# define __SYSCALL_SLONG_TYPE __SLONGWORD_TYPE
#if __WORDSIZE == 32 bits/types.h:# define __SQUAD_TYPE __quad_t #elif __WORDSIZE == 64 bits/types.h:# define __SQUAD_TYPE long int
bits/types.h:
#define __SLONGWORD_TYPE long int
From: Monsees, Steven C (US)
Sent: Wednesday, February 2, 2022 8:55 AM To: yocto@... Subject: IGC build issue with devtoolset-8 (GNU 8.3.1)
I am building zeus with basic OpenCL support for centos 7.x, and using GNU 8.3.1 compiler and see the following Error when IGC is built, I see the same error when building with GNU 5.3.1…
Is this a known issue, is a patch available ? Any ideas why I might be seeing this ?
cpp.o -MF IGC/Compiler/CMakeFiles/Compiler.dir/CISACodeGen/DebugInfo.cpp.o.d -o IGC/Compiler/CMakeFiles/Compiler.dir/CISACodeGen/DebugInfo.cpp.o -c /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.cpp | In file included from /usr/include/sys/stat.h:106, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/../3d/common/iStdLib/File.h:47, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.hpp:42, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.cpp:26: | /usr/include/bits/stat.h:106:31: error: expected unqualified-id before ‘[’ token | __syscall_slong_t __unused[3]; | ^ | /usr/include/bits/stat.h:164:31: error: expected unqualified-id before ‘[’ token | __syscall_slong_t __unused[3]; |
I am using bitbake –k, and everything else builds correctly, and no other components references this…
If I modify the header like so:
/usr/include/bits>diff stat.h stat.h_HOLD 106c106,108 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 164c166,168 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 08:29 smonsees@yix465383 /usr/include/bits>
IGC builds clean, and test show it appears to working correctly…
I really shouldn’t be modifying the header, and would like to know what the real issue issue is…
Thanks, Steve
|
||
|
|
||
|
Re: [meta-lts-mixins][dunfell/go PATCH 1/4] Initial commit: add license, readme and layer config.
Alexander Kanavin
Thanks Michael! Adrian, the branches are now available: so I'm ready to take any patches you would like to have there, pls send them via this mailing list. Otherwise I'll get the branches updated to latest component versions. Alex
On Thu, 3 Feb 2022 at 00:50, Michael Halstead <mhalstead@...> wrote:
|
||
|
|
||
|
Sourabh Hegde
Hello Bernd,
I am facing same issue. Did the patch worked for you? I tried to apply the patch and still have the same issue. Thanks in advance
|
||
|
|
||
|
Re: Build error:undefined reference to `stime' on Yocto Warrior
Sourabh Hegde
Hello,
I applied the patch https://patchwork.openembedded.org/patch/172282/ like below: Downloaded patch file from above link and deleted below part from patch file: diff --git a/meta/recipes-devtools/qemu/qemu.inc b/meta/recipes-devtools/qemu/qemu.inc index f451017f6d..ad4ff52892 100644 --- a/meta/recipes-devtools/qemu/qemu.inc +++ b/meta/recipes-devtools/qemu/qemu.inc @@ -27,6 +27,7 @@ SRC_URI = "https://download.qemu.org/${BPN}-${PV}.tar.xz \ file://0008-linux-user-Fix-webkitgtk-hangs-on-32-bit-x86-target.patch \ file://0009-Fix-webkitgtk-builds.patch \ file://0010-configure-Add-pkg-config-handling-for-libgcrypt.patch \ + file://0011-linux-user-remove-host-stime-syscall.patch \ file://CVE-2019-15890.patch \ file://CVE-2019-12068.patch \ file://CVE-2020-1711.patch \ Because,my /meta/recipes-devtools/qemu/qemu.inc file already has other patches under the anme "0011". Then I copied the patch file to "poky/meta/recipes-devtools/qemu/qemu/" And appended SRC_URI of "qemu-native.inc" with "file://0015-linux-user-remove-host-stime-syscall.patch \". After doing bitbake -c cleansstate <image> and bitbake <image> I still get below errors which is related to "x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime'" and some warnings related to utime and stime. i have attached complete log file for reference. Can anyone please let me know if patch has been correctly applied and how to resolve this issue? Thanks in advance. Your help will be much appreciated.
|
||
|
|
||
|
Re: Private: Re: [yocto] Fetch private gitlab repo using ssh with Yocto recipe
#bitbake
Nicolas Jeker
Re-adding the list.
On Wed, 2022-02-02 at 11:51 -0800, Sourabh Hegde wrote: Hi @Nicolas,Hi Sourabh Thanks for the detailed answer.Glad to hear it works now. Regarding my build environment: I am using Docker Desktop for WindowsI see, I never used Docker Desktop for Windows, so I can't give you advice about that specifically. If the ssh connection works correctly with the configuration you created, you shouldn't need to forward the ssh agent. Thanks in advance.
|
||
|
|
||
|
wget - The certificate has not yet been activated (does also happen in qemuarm "virt" machine)
Matthias Klein
Hello,
another addition to the problem: The problem does not seem to be directly related to the custom kernel or the one yocto, as I can recreate the problem in a second yocto. The yocto is for the qemuarm "virt" machine. I have the complete sources for it here: https://github.com/matthiasklein/meta-yocto-coffee-qemuarm Is it possible that it is somehow related to 32 bit ARM? # uname -a Linux qemuarm 5.15.16-yocto-standard #1 SMP PREEMPT Tue Jan 25 14:32:58 UTC 2022 armv7l GNU/Linux # wget -4 https://speed.hetzner.de/100MB.bin --2022-02-03 09:18:37-- https://speed.hetzner.de/100MB.bin SSL_INIT Resolving speed.hetzner.de... 88.198.248.254 Connecting to speed.hetzner.de|88.198.248.254|:443... connected. The certificate has not yet been activated # wget --version GNU Wget 1.21.2 built on linux-gnueabi. -cares +digest -gpgme +https +ipv6 -iri +large-file -metalink +nls +ntlm +opie -psl +ssl/gnutls Wgetrc: /etc/wgetrc (system) Locale: /usr/share/locale Compile: arm-poky-linux-gnueabi-gcc -mthumb -mfpu=neon -mfloat-abi=hard -mcpu=cortex-a15 -fstack-protector-strong -O2 -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security -Werror=format-security -DHAVE_CONFIG_H -DSYSTEM_WGETRC="/etc/wgetrc" -DLOCALEDIR="/usr/share/locale" -I. -I../../wget-1.21.2/src -I../lib -I../../wget-1.21.2/lib -DNDEBUG -O2 -pipe -g -feliminate-unused-debug-types Link: arm-poky-linux-gnueabi-gcc -mthumb -mfpu=neon -mfloat-abi=hard -mcpu=cortex-a15 -fstack-protector-strong -O2 -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security -Werror=format-security -DNDEBUG -O2 -pipe -g -feliminate-unused-debug-types -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now -lpcre -lnettle -lgnutls -lz ftp-opie.o gnutls.o http-ntlm.o ../lib/libgnu.a -lunistring Copyright (C) 2015 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://www.gnu.org/licenses/gpl.html>. This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Originally written by Hrvoje Niksic <hniksic@xemacs.org>. Please send bug reports and questions to <bug-wget@gnu.org>. Best regards, Matthias
|
||
|
|
||
|
Re: wget - The certificate has not yet been activated
Matthias Klein
Hi,
Thank you fort he tip, but I am using the real wget (not from busybox):
wget --version GNU Wget 1.21.2 built on linux-gnueabi.
-cares +digest -gpgme +https +ipv6 -iri +large-file -metalink +nls +ntlm +opie -psl +ssl/gnutls
Best regards, Matthias Klein
|
||
|
|
||
|
Re: wget - The certificate has not yet been activated
Markus Volk
Hi Matthias, check if busybox is built with this option CONFIG_FEATURE_WGET_OPENSSL=y
On Thu, Feb 3 2022 at 06:37:01 AM +0000, Matthias Klein <matthias.klein@...> wrote:
|
||
|
|
||
|
wget - The certificate has not yet been activated
Matthias Klein
Hello,
I have with the current master branch in a IMX6 Yocto the problem that with wget no HTTPS downloads work: # wget -4 https://speed.hetzner.de/100MB.bin --2022-02-03 06:23:25-- https://speed.hetzner.de/100MB.bin SSL_INIT Resolving speed.hetzner.de... 88.198.248.254 Connecting to speed.hetzner.de|88.198.248.254|:443... connected. The certificate has not yet been activated I do not understand why the validation of the certificate does not work. With curl it works. It is also not due to the time of the system. I use my own kernel based on the mainline kernel 5.10.47 with PREEMPT_RT. TARGET_SYS = "arm-poky-linux-gnueabi" TUNE_FEATURES = "arm vfp cortexa9 neon thumb callconvention-hard" TARGET_FPU = "hard" Can it be that something is missing in the kernel configuration? Does anyone have any idea what the problem could be? # wget --version GNU Wget 1.21.2 built on linux-gnueabi. -cares +digest -gpgme +https +ipv6 -iri +large-file -metalink +nls +ntlm +opie -psl +ssl/gnutls Wgetrc: /etc/wgetrc (system) Locale: /usr/share/locale Compile: arm-poky-linux-gnueabi-gcc -mthumb -mfpu=neon -mfloat-abi=hard -mcpu=cortex-a9 -fstack-protector-strong -O2 -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security -Werror=format-security -DHAVE_CONFIG_H -DSYSTEM_WGETRC="/etc/wgetrc" -DLOCALEDIR="/usr/share/locale" -I. -I../../wget-1.21.2/src -I../lib -I../../wget-1.21.2/lib -DNDEBUG -O2 -pipe -g -feliminate-unused-debug-types Link: arm-poky-linux-gnueabi-gcc -mthumb -mfpu=neon -mfloat-abi=hard -mcpu=cortex-a9 -fstack-protector-strong -O2 -D_FORTIFY_SOURCE=2 -Wformat -Wformat-security -Werror=format-security -DNDEBUG -O2 -pipe -g -feliminate-unused-debug-types -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -Wl,-z,relro,-z,now -lpcre -lnettle -lgnutls -lz ftp-opie.o gnutls.o http-ntlm.o ../lib/libgnu.a -lunistring Copyright (C) 2015 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later http://www.gnu.org/licenses/gpl.html. This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Originally written by Hrvoje Niksic mailto:hniksic@xemacs.org. Please send bug reports and questions to mailto:bug-wget@gnu.org. A test with gnutls-cli seems to work though: root@smartrail-8037:~# gnutls-cli -d 1 imap.gmail.com -p 993 Processed 127 CA certificate(s). Resolving 'imap.gmail.com:993'... Connecting to '2a00:1450:4013:c02::6d:993'... - Certificate type: X.509 - Got a certificate list of 3 certificates. - Certificate[0] info: - subject `CN=imap.gmail.com', issuer `CN=GTS CA 1C3,O=Google Trust Services LLC,C=US', serial 0x65fa03b5a71a05070a000000012e04f6, EC/ECDSA key 256 bits, signed using RSA-SHA256, activated `2022-01-10 03:07:10 UTC', expires `2022-04-04 03:07:09 UTC', pin-sha256="ZrSVXSwpcGu6oCbquwHwx6H2FM7DjzANRxMjQUC/Ng8=" Public Key ID: sha1:ae10ac489504779956e7acfc17631471be3e20d6 sha256:66b4955d2c29706bbaa026eabb01f0c7a1f614cec38f300d4713234140bf360f Public Key PIN: pin-sha256:ZrSVXSwpcGu6oCbquwHwx6H2FM7DjzANRxMjQUC/Ng8= - Certificate[1] info: - subject `CN=GTS CA 1C3,O=Google Trust Services LLC,C=US', issuer `CN=GTS Root R1,O=Google Trust Services LLC,C=US', serial 0x0203bc53596b34c718f5015066, RSA key 2048 bits, signed using RSA-SHA256, activated `2020-08-13 00:00:42 UTC', expires `2027-09-30 00:00:42 UTC', pin-sha256="zCTnfLwLKbS9S2sbp+uFz4KZOocFvXxkV06Ce9O5M2w=" - Certificate[2] info: - subject `CN=GTS Root R1,O=Google Trust Services LLC,C=US', issuer `CN=GlobalSign Root CA,OU=Root CA,O=GlobalSign nv-sa,C=BE', serial 0x77bd0d6cdb36f91aea210fc4f058d30d, RSA key 4096 bits, signed using RSA-SHA256, activated `2020-06-19 00:00:42 UTC', expires `2028-01-28 00:00:42 UTC', pin-sha256="hxqRlPTu1bMS/0DITB1SSu0vd4u/8l8TjPgfaAp63Gc=" - Status: The certificate is trusted. - Description: (TLS1.3-X.509)-(ECDHE-X25519)-(ECDSA-SECP256R1-SHA256)-(AES-256-GCM) - Session ID: FB:E7:27:9D:B0:8F:4C:2D:0C:5C:E9:17:0F:5C:9B:28:EE:3F:C0:38:0C:43:15:8D:9B:73:A7:AA:BD:AA:F9:87 - Options: - Handshake was completed - Simple Client Mode: * OK Gimap ready for requests from 2a02:908:4c16:7960:20c:c6ff:fe81:e7fa k10mb249481499edf Best regargds, Matthias
|
||
|
|
||
|
Re: [meta-lts-mixins][dunfell/go PATCH 1/4] Initial commit: add license, readme and layer config.
Michael Halstead
Thank you for the ping. These branches are ready for you.
On Tue, Feb 1, 2022 at 6:23 AM Alexander Kanavin <alex.kanavin@...> wrote:
--
Michael Halstead Linux Foundation / Yocto Project Systems Operations Engineer
|
||
|
|
||
|
Issues setting SELinux file contexts during build
#selinux
Ansh
Hi all,
I am currently working on enabling SELinux on an x86 system (Yocto version 2.4). I was able to successfully enable SELinux on the target but need to set SELinux file contexts during the build. I have inherited the selinux-image.bbclass in my image recipe and can confirm that it's executed during the build. But, when I boot up the image on target all the files have the wrong (or default) file contexts. For e.g. all files in /etc have system_u:object_r:root_t:s0 labels. When I run restorecon, the files get the correct labels. I have managed to narrow it down to the following two issues: 1. setfiles fails during the build but doesn't return any error or error msg causing the build to pass I looked into the pseudo database table xattr to check if the xattrs are set during the build and found the table empty. So I modified the selinux-image.bbclass and added a setfattr to set xattr for a single file in /etc before it runs setfiles. After the build, I found one entry for the same file for which I was using the setfattr command in the xattr table of pseudo db and was also able to get the xattr using getfattr in the modified selinux-image.bbclass. So that means setfattr was working as expected but not setfiles. Next, I enabled some debug logs for pseudo and found the following that points to setfiles failing after it tries to read security.restorecon_last. Any pointer as to why it is trying to read security.restorecon_last xattr (even though we are not using -d option for setfiles) and fails if it doesn't find it? 27286: wrapper called: getxattr 27286: getxattr - signals blocked, obtaining lock 27286: <nil> + build/tmp/work/x86_ctm-poky-linux-musl/custom-image/1.0-r0/rootfs => <build/tmp/work/x86_ctm-poky-linux-musl/custom-image/1.0-r0/rootfs> 2. xattrs set during the build are not present on the target As mentioned above, I modified to selinux-image.bbclass to set xattr for a single file in /etc. This works during the build since I can see the corresponding entry in pseudo db and also can get the set xattr using getfattr in the same selinux-image.bbclass. But when I boot up the image on target the xattr is replaced by (I am guessing) the default SELinux label of system_u:object_r:root_t:s0. I am using initramfs for packaging the root filesystem and I found that cpio ignores xattrs and also initramfs doesn't support xattrs. Does that mean I can't set the file contexts during the build because I am using initramfs and cpio? Thanks, Ansh
|
||
|
|
||
|
Re: Build error:undefined reference to `stime' on Yocto Warrior
Sourabh Hegde
Hello @Khem, Thanks for the quick response. Can you please let me know where and how to include this patch in poky/meta? Thanks in advance
On Wed, Feb 2, 2022, 21:19 Khem Raj <raj.khem@...> wrote: you need something like this
|
||
|
|
||
|
Re: Build error:undefined reference to `stime' on Yocto Warrior
toggle quoted messageShow quoted text
On Wed, Feb 2, 2022 at 11:43 AM Sourabh Hegde <hrsourabh011@gmail.com> wrote:
|
||
|
|
||
|
Build error:undefined reference to `stime' on Yocto Warrior
Sourabh Hegde
Hello All,
When I execute the bitbake for my custom image recipe I hit the below issue related to "qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime'". DEBUG: Executing shell function do_compile NOTE: make -j 8 LD=ld AR=ar OBJCOPY=objcopy LDFLAGS=-L/root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/recipe-sysroot-native/usr/lib -L/root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/recipe-sysroot-native/lib -Wl,--enable-new-dtags -Wl,-rpath-link,/root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/recipe-sysroot-native/usr/lib -Wl,-rpath-link,/root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/recipe-sysroot-native/lib -Wl,-rpath,/root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/recipe-sysroot-native/usr/lib -Wl,-rpath,/root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/recipe-sysroot-native/lib -Wl,-O1 -fuse-ld=bfd LINK i386-linux-user/qemu-i386 LINK mipsel-linux-user/qemu-mipsel LINK arm-linux-user/qemu-arm LINK mips-linux-user/qemu-mips LINK ppc-linux-user/qemu-ppc GEN x86_64-linux-user/config-target.h LINK sh4-linux-user/qemu-sh4 CC x86_64-linux-user/exec.o CC x86_64-linux-user/tcg/tcg.o CC x86_64-linux-user/tcg/tcg-op.o /root/build-rauc/tmp-glibc/hosttools/ld.bfd: linux-user/syscall.o: in function `do_syscall1': /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:199: qemu-arm] Error 1 make: *** [Makefile:483: subdir-arm-linux-user] Error 2 make: *** Waiting for unfinished jobs.... CC x86_64-linux-user/tcg/tcg-op-vec.o /root/build-rauc/tmp-glibc/hosttools/ld.bfd: linux-user/syscall.o: in function `do_syscall1': /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:199: qemu-i386] Error 1 make: *** [Makefile:483: subdir-i386-linux-user] Error 2 CC x86_64-linux-user/tcg/tcg-op-gvec.o /root/build-rauc/tmp-glibc/hosttools/ld.bfd: linux-user/syscall.o: in function `do_syscall1': /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:199: qemu-sh4] Error 1 make: *** [Makefile:483: subdir-sh4-linux-user] Error 2 CC x86_64-linux-user/tcg/tcg-common.o /root/build-rauc/tmp-glibc/hosttools/ld.bfd: linux-user/syscall.o: in function `do_syscall1': /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:199: qemu-ppc] Error 1 make: *** [Makefile:483: subdir-ppc-linux-user] Error 2 CC x86_64-linux-user/tcg/optimize.o /root/build-rauc/tmp-glibc/hosttools/ld.bfd: linux-user/syscall.o: in function `do_syscall1': /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:199: qemu-mipsel] Error 1 make: *** [Makefile:483: subdir-mipsel-linux-user] Error 2 CC x86_64-linux-user/fpu/softfloat.o /root/build-rauc/tmp-glibc/hosttools/ld.bfd: linux-user/syscall.o: in function `do_syscall1': /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/syscall.c:7404: undefined reference to `stime' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:199: qemu-mips] Error 1 make: *** [Makefile:483: subdir-mips-linux-user] Error 2 CC x86_64-linux-user/disas.o GEN x86_64-linux-user/gdbstub-xml.c CC x86_64-linux-user/gdbstub.o CC x86_64-linux-user/thunk.o CC x86_64-linux-user/accel/stubs/hax-stub.o CC x86_64-linux-user/accel/stubs/hvf-stub.o CC x86_64-linux-user/accel/stubs/whpx-stub.o CC x86_64-linux-user/accel/stubs/kvm-stub.o CC x86_64-linux-user/accel/tcg/tcg-runtime.o CC x86_64-linux-user/accel/tcg/tcg-runtime-gvec.o CC x86_64-linux-user/accel/tcg/cpu-exec.o CC x86_64-linux-user/accel/tcg/cpu-exec-common.o CC x86_64-linux-user/accel/tcg/translate-all.o CC x86_64-linux-user/accel/tcg/translator.o CC x86_64-linux-user/accel/tcg/user-exec.o CC x86_64-linux-user/accel/tcg/user-exec-stub.o CC x86_64-linux-user/linux-user/main.o CC x86_64-linux-user/linux-user/syscall.o CC x86_64-linux-user/linux-user/strace.o CC x86_64-linux-user/linux-user/mmap.o CC x86_64-linux-user/linux-user/signal.o CC x86_64-linux-user/linux-user/elfload.o In file included from /usr/include/string.h:495, from /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/include/qemu/osdep.h:84, from /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/elfload.c:2: In function ‘strncpy’, inlined from ‘fill_psinfo’ at /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/elfload.c:3158:12, inlined from ‘fill_note_info’ at /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/elfload.c:3340:5, inlined from ‘elf_core_dump’ at /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/qemu-3.1.1.1/linux-user/elfload.c:3489:9: /usr/include/x86_64-linux-gnu/bits/string_fortified.h:106:10: warning: ‘__builtin_strncpy’ specified bound 16 equals destination size [-Wstringop-truncation] 106 | return __builtin___strncpy_chk (__dest, __src, __len, __bos (__dest)); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CC x86_64-linux-user/linux-user/linuxload.o CC x86_64-linux-user/linux-user/uaccess.o CC x86_64-linux-user/linux-user/uname.o CCAS x86_64-linux-user/linux-user/safe-syscall.o CC x86_64-linux-user/linux-user/x86_64/signal.o CC x86_64-linux-user/linux-user/x86_64/cpu_loop.o CC x86_64-linux-user/linux-user/exit.o CC x86_64-linux-user/linux-user/fd-trans.o CC x86_64-linux-user/target/i386/helper.o CC x86_64-linux-user/target/i386/cpu.o CC x86_64-linux-user/target/i386/gdbstub.o CC x86_64-linux-user/target/i386/xsave_helper.o CC x86_64-linux-user/target/i386/translate.o CC x86_64-linux-user/target/i386/bpt_helper.o CC x86_64-linux-user/target/i386/cc_helper.o CC x86_64-linux-user/target/i386/excp_helper.o CC x86_64-linux-user/target/i386/fpu_helper.o CC x86_64-linux-user/target/i386/int_helper.o CC x86_64-linux-user/target/i386/mem_helper.o CC x86_64-linux-user/target/i386/misc_helper.o CC x86_64-linux-user/target/i386/mpx_helper.o CC x86_64-linux-user/target/i386/seg_helper.o CC x86_64-linux-user/target/i386/smm_helper.o CC x86_64-linux-user/target/i386/svm_helper.o CC x86_64-linux-user/target/i386/sev-stub.o GEN trace/generated-helpers.c CC x86_64-linux-user/trace/control-target.o CC x86_64-linux-user/gdbstub-xml.o CC x86_64-linux-user/trace/generated-helpers.o LINK x86_64-linux-user/qemu-x86_64 ERROR: oe_runmake failed WARNING: exit code 1 from a shell command. ERROR: Function failed: do_compile (log file is located at /root/build-rauc/tmp-glibc/work/x86_64-linux/qemu-native/3.1.1.1-r0/temp/log.do_compile.5584) I am using "Warrior" version due to limitation from other meta-layers. I have read that above issue is fixed in newer versions of Yocto. But how can i fix this in Warrior? Is there any workaround? Can someone please help me resolve it? Your help will be much appreciated. Thanks in advance.
|
||
|
|
||
|
IGC build issue with devtoolset-8 (GNU 8.3.1)
Monsees, Steven C (US)
I am building zeus with basic OpenCL support for centos 7.x, and using GNU 8.3.1 compiler and see the following Error when IGC is built, I see the same error when building with GNU 5.3.1…
Is this a known issue, is a patch available ? Any ideas why I might be seeing this ?
cpp.o -MF IGC/Compiler/CMakeFiles/Compiler.dir/CISACodeGen/DebugInfo.cpp.o.d -o IGC/Compiler/CMakeFiles/Compiler.dir/CISACodeGen/DebugInfo.cpp.o -c /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.cpp | In file included from /usr/include/sys/stat.h:106, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/../3d/common/iStdLib/File.h:47, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.hpp:42, | from /disk0/scratch/yocto_user/yocto/workspace_zeus/builds/sbcb-default/tmp/work/x86_64-linux/intel-graphics-compiler-native/1.0.11-r0/git/IGC/Compiler/CISACodeGen/DebugInfo.cpp:26: | /usr/include/bits/stat.h:106:31: error: expected unqualified-id before ‘[’ token | __syscall_slong_t __unused[3]; | ^ | /usr/include/bits/stat.h:164:31: error: expected unqualified-id before ‘[’ token | __syscall_slong_t __unused[3]; |
I am using bitbake –k, and everything else builds correctly, and no other components references this…
If I modify the header like so:
/usr/include/bits>diff stat.h stat.h_HOLD 106c106,108 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 164c166,168 < __syscall_slong_t __unused[3]; --- > __syscall_slong_t __unused1; > __syscall_slong_t __unused2; > __syscall_slong_t __unused3; 08:29 smonsees@yix465383 /usr/include/bits>
IGC builds clean, and test show it appears to working correctly…
I really shouldn’t be modifying the header, and would like to know what the real issue issue is…
Thanks, Steve
|
||
|
|