March 2023 - openSUSE Kernel Bugs - openSUSE Mailing Lists

[Bug 1208822] [Build 20230301] openQA test fails in cve-2022-0185 (fsconfig03)
by bugzilla_noreply＠suse.com 28 Mar '23

28 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1208822 https://bugzilla.suse.com/show_bug.cgi?id=1208822#c12 --- Comment #12 from WEI GAO <wegao(a)suse.com> --- (In reply to Petr Vorel from comment #11) > (In reply to WEI GAO from comment #10) > > In my local env, the case will pass with mkfs.exfat installed, BUT in jeos > > env mkfs.exfat can not be installed(caused by can not install exfatprogs). > > I'm sorry, it must be something else. The original report shows mkfs.exfat > installed via exfat-utils: > > The following package is recommended, but will not be installed due to > conflicts or dependency issues: > exfatprogs > > The following 167 NEW packages are going to be installed: > NetworkManager-dns-bind NetworkManager-dns-dnsmasq apparmor-utils attr > babeltrace bc bind bind-utils binutils cpp cpp12 ctags cyrus-sasl > cyrus-sasl-gssapi debuginfod-profile device-mapper dhcp dhcp-client > dhcp-server dnsmasq e2fsprogs ethtool evmctl exfat-utils > (note exfat-utils) > ... > > # fsconfig03; echo "### TEST cve-2022-0185 COMPLETE >>> $?." > ... > tst_supported_fs_types.c:120: TINFO: FUSE does support exfat > tst_supported_fs_types.c:55: TINFO: mkfs.exfat does exist > ... > tst_test.c:1634: TINFO: === Testing on btrfs === > tst_test.c:1093: TINFO: Formatting /dev/loop0 with btrfs opts='' extra > opts='' > fsconfig03.c:72: TPASS: kernel seems to be fine on btrfs > tst_test.c:1634: TINFO: === Testing on exfat === > tst_test.c:1093: TINFO: Formatting /dev/loop0 with exfat opts='' extra > opts='' > fsconfig03.c:38: TBROK: fsopen() failed: ENODEV (19) > > I also verified exfat-utils package locally: > rpm -qlp exfat-utils-1.4.0-1.2.x86_64.rpm > ... > /usr/sbin/mkfs.exfat > ... > => it's not a problem of mkfs.exfat dependency. > > Sometimes some kernel modules are missing, because we test with only > kernel-default-base package. But 1) we run with fuse 2) although exfat is > module (CONFIG_EXFAT_FS=m see > https://openqa.opensuse.org/tests/3172507#step/install_ltp/75) LTP > confirmed the support (thus module was loaded). > > I searched in dmesg. The usual dmesg file in openQA don't contain dmesg > output: > https://openqa.opensuse.org/tests/3172507/file/serial0.txt > > Looking into another file, which contains dmesg > https://openqa.opensuse.org/tests/3172507/file/shutdown_ltp-dmesg.txt > is without "OpenQA::run_ltp.pm: Starting ..." mark in dmesg > https://github.com/os-autoinst/os-autoinst-distri-opensuse/blob/ > 63a66f99ef7b274c94ac546d3807c89e5a9a0a9c/tests/kernel/run_ltp.pm#L306-L318 > which helps finding which dmesg part was caused by particular test (but test > is using virtio console). > > There are errors in the end, but because the latest message is btrfs, it > [ 1517.688092] operation not supported error, dev loop0, sector 98310 op > 0x9:(WRITE_ZEROES) flags 0x8000800 phys_seg 0 prio class 2 > [ 1517.688196] operation not supported error, dev loop0, sector 115212 op > 0x9:(WRITE_ZEROES) flags 0x8000800 phys_seg 0 prio class 2 > [ 1517.688300] operation not supported error, dev loop0, sector 131078 op > 0x9:(WRITE_ZEROES) flags 0x8000800 phys_seg 0 prio class 2 > [ 1517.809762] BTRFS: device fsid ea5c84b6-3547-494d-a4b5-599c5c2e7ec3 devid > 1 transid 6 /dev/loop0 scanned by mkfs.btrfs (13689) > > => It would help if you download JeOS image, start it locally with > virt-manager and watch dmesg. Or even start it with strace to see more. Download openSUSE-Tumbleweed-Minimal-VM.x86_64-1.0.0-kvm-and-xen-Snapshot20230326.qcow2 and do some check: kernel config has CONFIG_EXFAT_FS=m but you can not find exfat related modules installed so running system actually not support exfat, that's why LTP case failed on fsopen(exfat) For LTP side, should we also need some fix? Since current LTP case still think exfat is support on system but actually not. ===output from openSUSE-Tumbleweed-Minimal-VM.x86_64-1.0.0-kvm-and-xen-Snapshot20230326.qcow2=== uefi-http-vm:~ # cat /boot/config-6.2.8-1-default | grep FAT # DOS/FAT/EXFAT/NT Filesystems CONFIG_FAT_FS=m CONFIG_VFAT_FS=m CONFIG_FAT_DEFAULT_CODEPAGE=437 CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" # CONFIG_FAT_DEFAULT_UTF8 is not set CONFIG_EXFAT_FS=m CONFIG_EXFAT_DEFAULT_IOCHARSET="utf8" # end of DOS/FAT/EXFAT/NT Filesystems uefi-http-vm:~ # uefi-http-vm:~ # uname -r 6.2.8-1-default uefi-http-vm:~ # cat /lib/modules/6.2.8-1-default/modules.builtin | grep fat uefi-http-vm:~ # modprobe exfat modprobe: FATAL: Module exfat not found in directory /lib/modules/6.2.8-1-default uefi-http-vm:~ # cat /proc/filesystems | grep fat vfat uefi-http-vm:~ # cat /proc/filesystems nodev sysfs nodev tmpfs nodev bdev nodev proc nodev cgroup nodev cgroup2 nodev cpuset nodev devtmpfs nodev debugfs nodev tracefs nodev securityfs nodev sockfs nodev bpf nodev pipefs nodev ramfs nodev hugetlbfs nodev devpts nodev autofs nodev mqueue nodev pstore nodev efivarfs btrfs nodev configfs fuseblk nodev fuse nodev fusectl vfat -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209805] Kernel regression: Possible integer underflow in input pipeline of ThinkPad trackpoint.
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1209805 http://bugzilla.opensuse.org/show_bug.cgi?id=1209805#c6 Stefan Dirsch <sndirsch(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Component|X.Org |Kernel Assignee|gfx-bugs(a)suse.de |kernel-bugs(a)opensuse.org Summary|Possible integer underflow |Kernel regression: Possible |in input pipeline of |integer underflow in input |ThinkPad trackpoint. |pipeline of ThinkPad | |trackpoint. QA Contact|gfx-bugs(a)suse.de |qa-bugs(a)suse.de --- Comment #6 from Stefan Dirsch <sndirsch(a)suse.com> --- Thanks. That already helps a lot! Reassigning ... -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209724] kernel can not be built for s390x because of outside array bounds
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1209724 https://bugzilla.suse.com/show_bug.cgi?id=1209724#c10 --- Comment #10 from Sarah Kriesch <ada.lovelace(a)gmx.de> --- Thank you! I will test this patch. I did not know about such a gcc version dependency for the kernel and it is looking like that should have really happened for all architectures. I took a look into the submit request (change) by the SUSE Developers. They have removed only "CONFIG_SERIAL_FSL_LPUART_CONSOLE=y" for the x86 config. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1208822] [Build 20230301] openQA test fails in cve-2022-0185 (fsconfig03)
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1208822 Petr Vorel <petr.vorel(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- QA Contact|qa-bugs(a)suse.de |wegao(a)suse.com -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1208822] [Build 20230301] openQA test fails in cve-2022-0185 (fsconfig03)
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1208822 https://bugzilla.suse.com/show_bug.cgi?id=1208822#c11 --- Comment #11 from Petr Vorel <petr.vorel(a)suse.com> --- (In reply to WEI GAO from comment #10) > In my local env, the case will pass with mkfs.exfat installed, BUT in jeos > env mkfs.exfat can not be installed(caused by can not install exfatprogs). I'm sorry, it must be something else. The original report shows mkfs.exfat installed via exfat-utils: The following package is recommended, but will not be installed due to conflicts or dependency issues: exfatprogs The following 167 NEW packages are going to be installed: NetworkManager-dns-bind NetworkManager-dns-dnsmasq apparmor-utils attr babeltrace bc bind bind-utils binutils cpp cpp12 ctags cyrus-sasl cyrus-sasl-gssapi debuginfod-profile device-mapper dhcp dhcp-client dhcp-server dnsmasq e2fsprogs ethtool evmctl exfat-utils (note exfat-utils) ... # fsconfig03; echo "### TEST cve-2022-0185 COMPLETE >>> $?." ... tst_supported_fs_types.c:120: TINFO: FUSE does support exfat tst_supported_fs_types.c:55: TINFO: mkfs.exfat does exist ... tst_test.c:1634: TINFO: === Testing on btrfs === tst_test.c:1093: TINFO: Formatting /dev/loop0 with btrfs opts='' extra opts='' fsconfig03.c:72: TPASS: kernel seems to be fine on btrfs tst_test.c:1634: TINFO: === Testing on exfat === tst_test.c:1093: TINFO: Formatting /dev/loop0 with exfat opts='' extra opts='' fsconfig03.c:38: TBROK: fsopen() failed: ENODEV (19) I also verified exfat-utils package locally: rpm -qlp exfat-utils-1.4.0-1.2.x86_64.rpm ... /usr/sbin/mkfs.exfat ... => it's not a problem of mkfs.exfat dependency. Sometimes some kernel modules are missing, because we test with only kernel-default-base package. But 1) we run with fuse 2) although exfat is module (CONFIG_EXFAT_FS=m see https://openqa.opensuse.org/tests/3172507#step/install_ltp/75) LTP confirmed the support (thus module was loaded). I searched in dmesg. The usual dmesg file in openQA don't contain dmesg output: https://openqa.opensuse.org/tests/3172507/file/serial0.txt Looking into another file, which contains dmesg https://openqa.opensuse.org/tests/3172507/file/shutdown_ltp-dmesg.txt is without "OpenQA::run_ltp.pm: Starting ..." mark in dmesg https://github.com/os-autoinst/os-autoinst-distri-opensuse/blob/63a66f99ef7… which helps finding which dmesg part was caused by particular test (but test is using virtio console). There are errors in the end, but because the latest message is btrfs, it [ 1517.688092] operation not supported error, dev loop0, sector 98310 op 0x9:(WRITE_ZEROES) flags 0x8000800 phys_seg 0 prio class 2 [ 1517.688196] operation not supported error, dev loop0, sector 115212 op 0x9:(WRITE_ZEROES) flags 0x8000800 phys_seg 0 prio class 2 [ 1517.688300] operation not supported error, dev loop0, sector 131078 op 0x9:(WRITE_ZEROES) flags 0x8000800 phys_seg 0 prio class 2 [ 1517.809762] BTRFS: device fsid ea5c84b6-3547-494d-a4b5-599c5c2e7ec3 devid 1 transid 6 /dev/loop0 scanned by mkfs.btrfs (13689) => It would help if you download JeOS image, start it locally with virt-manager and watch dmesg. Or even start it with strace to see more. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209724] kernel can not be built for s390x because of outside array bounds
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1209724 https://bugzilla.suse.com/show_bug.cgi?id=1209724#c9 --- Comment #9 from LTC BugProxy <bugproxy(a)us.ibm.com> --- Created attachment 865870 --> https://bugzilla.suse.com/attachment.cgi?id=865870&action=edit a patch to disable -Warray-bounds for gcc-13 same as for gcc-11 and 12 -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209724] kernel can not be built for s390x because of outside array bounds
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1209724 https://bugzilla.suse.com/show_bug.cgi?id=1209724#c8 --- Comment #8 from LTC BugProxy <bugproxy(a)us.ibm.com> --- ------- Comment From vasily.gorbik(a)de.ibm.com 2023-03-27 12:21 EDT------- Neither the gcc 13 nor the kernel built with it have been sufficiently tested for any use other than pure development needs. The '-Warray-bounds' warnings are explicitly disabled in the kernel for gcc 11 and 12. It is yet unclear how '-Warray-bounds' will be handled for gcc 13 because it has not been released yet. We do not know if other architectures will have to disable it or if it will only apply to s390. In other words, support for gcc 13 in the kernel has not been settled yet, and it will most likely be changed like the patch attached. The '-Warray-bounds' warning is more problematic for s390 due to the way our architecture-specific lowcore is accessed. The particular build error you encountered stands out from array-bounds warnings, but that is only because the MTK_T7XX driver is using 'ccflags-y += -Werror.' Disabling the '-Warray-bounds' warning also addresses that issue. However, I do not know the reasons for your early adoption of gcc 13. I noticed that there are already gcc 13 build-specific patches available. [ 245s] + patch -s -F0 -E -p1 --no-backup-if-mismatch -i ../patches.kernel.org/6.2.3-662-gcc-plugins-drop-std-gnu-11-to-fix-GCC-13-build.patch If urgent and if desired, you could pick up the attached patch until it is superseded by kernel changes later. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209724] kernel can not be built for s390x because of outside array bounds
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1209724 https://bugzilla.suse.com/show_bug.cgi?id=1209724#c7 --- Comment #7 from Sarah Kriesch <ada.lovelace(a)gmx.de> --- If you create an openSUSE account and login into our Open Build System, you can receive this (unreleased) gcc version as a package from here: https://build.opensuse.org/package/binaries/openSUSE:Factory:zSystems/gcc13… -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209724] kernel can not be built for s390x because of outside array bounds
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1209724 https://bugzilla.suse.com/show_bug.cgi?id=1209724#c6 --- Comment #6 from Sarah Kriesch <ada.lovelace(a)gmx.de> --- (In reply to LTC BugProxy from comment #5) > ------- Comment From vasily.gorbik(a)de.ibm.com 2023-03-27 09:58 EDT------- > are you trying to use unreleased gcc 13 for the build? Yes. But we do not receive such error messages for the other architectures with that. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1209526] WOL not working with RTL8111/8168/8411 with 6.2.x kernel
by bugzilla_noreply＠suse.com 27 Mar '23

27 Mar '23

https://bugzilla.suse.com/show_bug.cgi?id=1209526 https://bugzilla.suse.com/show_bug.cgi?id=1209526#c8 Takashi Iwai <tiwai(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Flags| |needinfo?(jiivee(a)iki.fi) --- Comment #8 from Takashi Iwai <tiwai(a)suse.com> --- Another test kernel is being built for the revised fix patch in OBS home:tiwai:bsc1209526-3 repo. Please try it once after the build finishes. -- You are receiving this mail because: You are the assignee for the bug.

1 0