[opensuse-factory] New Tumbleweed snapshot 20180210 released!
Please note that this mail was generated by a script. The described changes are computed based on the x86_64 DVD. The full online repo contains too many changes to be listed here. Please check the known defects of this snapshot before upgrading: https://openqa.opensuse.org/tests/overview?distri=opensuse&groupid=1&version=Tumbleweed&build=20180210 When you reply to report some issues, make sure to change the subject. It is not helpful to keep the release announcement subject in a thread while discussing a specific problem. Packages changed: aaa_base (84.87+git20180205.2d2832f -> 84.87+git20180208.8eeab90) bash bind cfitsio (3.410 -> 3.420) freetype2 (2.7.1 -> 2.9) ft2demos (2.7.1 -> 2.9) iproute2 (4.14.1 -> 4.15.0) iptables (1.6.1 -> 1.6.2) kernel-source (4.15.1 -> 4.15.2) libcue (2.1.0 -> 2.2.0) libisoburn (1.4.6 -> 1.4.8) libshout lz4 (1.8.0 -> 1.8.1.2) mono-core (5.4.1 -> 5.8.0) mpfr (4.0.0 -> 4.0.1) openblas_pthreads perl-DBI (1.639 -> 1.640) perl-DateTime-TimeZone (2.15 -> 2.17) perl-MailTools (2.19 -> 2.20) php7 (7.2.1 -> 7.2.2) postgresql10 (10.1 -> 10.2) python-lxml (4.0.0 -> 4.1.1) smartmontools (6.5 -> 6.6) unzip vulkan (1.0.65.0 -> 1.0.68.0) === Details === ==== aaa_base ==== Version update (84.87+git20180205.2d2832f -> 84.87+git20180208.8eeab90) Subpackages: aaa_base-extras - Update to version 84.87+git20180208.8eeab90: * Don't call fillup for removed sysconfig.news * Adjust path for script converting sysctl config * For ksh use builtin keyword 'function' to make sure that the keyword 'typeset' really set the variable IFS to be local within the function _ls. ==== bash ==== Subpackages: bash-doc bash-lang libreadline7 readline-devel readline-doc - Add patch 19 to bash-4.4-patches.tar.bz2 * With certain values for PS1, especially those that wrap onto three or more lines, readline will miscalculate the number of invisible characters, leading to crashes and core dumps. ==== bind ==== Subpackages: bind-chrootenv bind-doc bind-utils libbind9-160 libdns169 libirs160 libisc166 libisccc160 libisccfg160 liblwres160 python3-bind - Implement systemd init scripts for bind and lwresd (fate#323155) ==== cfitsio ==== Version update (3.410 -> 3.420) Subpackages: cfitsio-devel libcfitsio5 - Update to version 3.420: * added https support to the collection of drivers handled in cfileio.c and drvrnet.c. This also handles the case where http transfers are rerouted to https. Note that this enhancement introduces a dependency on the libcurl development package. If this package is absent, CFITSIO will still build but will not have https capability. * made fix to imcomp_init_table function in imcompress.c. It now writes ZSIMPLE keyword only to a compressed image that will be placed in the primary header. * fix made to fits_get_col_display_width for case of a vector column of strings. * in ftp_open_network and in ftp_file_exist, added code to repeatedly attempt to make a ftp connection if the ftp server does not respond to the first request. (some ftp servers don't appear to be 100% reliable). * in drvrnet.c added many calls to 'fclose' to close unneeded files, to avoid exceeding the maximum allowed number of files that can be open at once. * made substantial changes to the ftp_checkfile and http_checkfile routines to streamline the process of checking for the existence of a .gz or .Z compressed version of the file before opening the uncompressed file (when using http or ftp to open the file). * modified the code in ftp_open_network to send "\r\n" as end-of-line characters instead of just "\n". Some ftp servers (in particular, at heasarc.gsfc.nasa.gov) now require both characters, otherwise the network connection simply hangs. * modified the http_open_network routine to handle HTTP 301 or 302 redirects to a FTP url. This is needed to support the new configuration on the heasarc HTTP server which sometimes redirects http URLS to a ftp URL. - Added new build requirement libcurl-devel. ==== freetype2 ==== Version update (2.7.1 -> 2.9) Subpackages: freetype2-devel libfreetype6 libfreetype6-32bit - Added patch: * enable-long-family-names-by-default.patch + Define PCF_CONFIG_OPTION_LONG_FAMILY_NAMES to obtain 2.7.1 behaviour - Added patches: * 0001-src-truetype-ttinterp.c-Ins_GETVARIATION-Avoid-NULL-.patch + Upstream fix for bsc#1079603: Avoid NULL reference in src/truetype/ttinterp.c * 0001-truetype-Better-protection-against-invalid-VF-data.patch + Upstream fix for bsc#1079601: Protection against invalid VF data - Add psaux-flex.patch to fix a regression in Type1 rendering - Update to version 2.9 * Advance width values of variation fonts were often wrong. * More fixes for variation font support; you should update to this version if you want to support them. * As a GSoC project, Ewald Hew extended the new (Adobe) CFF engine to handle Type 1 fonts also, thus greatly improving the rendering of this format. This is the new default. * A new function, `FT_Set_Named_Instance', can be used to set or change the current named instance. * Starting with this FreeType version, resetting variation coordinates will return to the currently selected named instance. Previously, FreeType returned to the base font (i.e., no instance set). * Some fuzzer fixes to better reject malformed fonts. - Update to version 2.8.1 * B/W hinting of TrueType fonts didn't work properly if interpreter version 38 or 40 was selected. * Some severe problems within the handling of TrueType Variation Fonts were found and fixed. * Function `FT_Set_Var_Design_Coordinates' didn't correctly handle the case with less input coordinates than axes. * By default, FreeType now offers high quality LCD-optimized output without resorting to ClearType techniques of resolution tripling and filtering. In this method, called Harmony, each color channel is generated separately after shifting the glyph outline, capitalizing on the fact that the color grids on LCD panels are shifted by a third of a pixel. This output is indistinguishable from ClearType with a light 3-tap filter. * Using the new function `FT_Get_Var_Axis_Flags', an application can access the `flags' field of a variation axis (introduced in OpenType version 1.8.2) * FreeType now synthesizes a missing Unicode cmap for (older) TrueType fonts also if glyph names are available. * The warping option has moved from `light' to `normal' hinting where it replaces the original hinting algorithm. The `light' mode is now always void of any hinting in x-direction. - Update to version 2.8 * Support for OpenType Variation Fonts is now complete. The last missing part was handling the `VVAR' and `MVAR' tables, which is available with this release. * A new function `FT_Face_Properties' allows the control of some module and library properties per font. Currently, the following properties can be handled: stem darkening, LCD filter weights, and the random seed for the `random' CFF operator. * The PCF change to show more `colourful' family names (introduced in version 2.7.1) was too radical; it can now be configured with PCF_CONFIG_OPTION_LONG_FAMILY_NAMES at compile time. If activated, it can be switched off at run time with the new pcf property `no-long-family-names'. If the `FREETYPE_PROPERTIES' environment variable is available, you can say FREETYPE_PROPERTIES=pcf:no-long-family-names=1 * Support for the following scripts has been added to the auto-hinter. Adlam, Avestan, Bamum, Buhid, Carian, Chakma, Coptic, Cypriot, Deseret, Glagolitic, Gothic, Kayah, Lisu, N'Ko, Ol Chiki, Old Turkic, Osage, Osmanya, Saurashtra, Shavian, Sundanese, Tai Viet, Tifinagh, Unified Canadian Syllabics, Vai * `Light' auto-hinting mode no longer uses TrueType metrics for TrueType fonts. This bug was introduced in version 2.4.6, causing horizontal scaling also. Almost all GNU/Linux distributions (with Fedora as a notable exception) disabled the corresponding patch for good reasons; chances are thus high that you won't notice a difference. * If a TrueType font gets loaded with FT_LOAD_NO_HINTING, FreeType now scales the font linearly again (bug introduced in version 2.4.6). * Fixed CVE-2017-8105, CVE-2017-8287: Older FreeType versions have out-of-bounds writes caused by heap-based buffer overflows related to Type 1 fonts. (boo#1035807, boo#1036457) - See https://sourceforge.net/projects/freetype/files/freetype2/2.8/ for the complete changelog. ==== ft2demos ==== Version update (2.7.1 -> 2.9) Subpackages: ftbench ftdiff ftdump ftgamma ftgrid ftinspect ftlint ftmulti ftstring ftvalid ftview - Update to version 2.9 * The `ftbench' demo program has a new test for opening a new face and loading some glyphs. * The `ftbench' demo program has a new option `-j' to specify the last glyph index to be used in the tests. * The `ftgrid' demo program has a new option `-n' to suppress display of named instances of variation fonts. * The `ttdebug' demo program can now show a stack trace (key `K') and switch between hexadecimal and decimal display of integers (key `I'). - Update to version 2.8.1 * The ftgrid demo program can now toggle the display of grid lines with the G key. * The ftgrid demo program can toggle a different set of colors (suitable to color-blind people) with the C key. * The ftgrid demo program now supports the -e command line option to select a cmap. * The ftdump demo program has a new command line option -t to output the SFNT table list. - A mode to display light auto-hinting with sub-pixel positioning has been added to `ftdiff'. ==== iproute2 ==== Version update (4.14.1 -> 4.15.0) - Update to new upstream release 4.15.0 * Add "show" subcommand to "ip fou" * Show 'external' link mode in output * bridge: vlan: support for per vlan tunnel info * gre6/tunnel: Output flowlabel after tclass * ila: support for configuring identifier and hook types * ila: support to configure checksum neutral-map-auto * ip6_gre: add support for ERSPAN tunnel * ip6tnl/tunnel: Output hoplimit before encapsulation limit * ip: add fastopen_no_cookie option to ip route * ip: bridge_slave: add neigh_suppress to the type help and * ip: bridge_slave: add support for per-port group_fwd_mask * ip: validate vlan value for vlan info * iplink: new option to set neigh suppression on a bridge port * link_iptnl: Print tunnel mode * qdisc: print offload indication * ss: Implement automatic column width calculation * ss: Introduce columns lightweight abstraction * ss: add AF_VSOCK support * ss: add missing path MTU parameter * ss: print tcpi_rcv_ssthresh * tc: Add support for the CBS qdisc * tc: bandwith limits can now be specified in %. * {f, m}_bpf: don't allow specifying multiple bpf programs - drop man-fix-documentation-references-in-manual-pages.patch * debian specific paths were removed altogether - partial cleanup with spec-cleaner - use https in urls ==== iptables ==== Version update (1.6.1 -> 1.6.2) Subpackages: libiptc0 libxtables12 xtables-plugins - Update to new upstream release 1.6.2 * add support for the "srh" match * add randomize-full for the "MASQUERADE" target * add rate match mode to the "hashlimit" match - Add iptables-batch-lock.patch: Fix a locking issue of iptables-batch which can cause it to spuriously fail when other programs modify the iptables rules in parallel (bnc#1045130). This can especially affect SuSEfirewall2 during startup. ==== kernel-source ==== Version update (4.15.1 -> 4.15.2) Subpackages: kernel-default kernel-default-devel kernel-devel kernel-docs kernel-macros kernel-syms - Linux 4.15.2 (bnc#1012628). - KVM: x86: Make indirect calls in emulator speculation safe (bnc#1012628). - KVM: VMX: Make indirect call speculation safe (bnc#1012628). - x86/alternative: Print unadorned pointers (bnc#1012628). - auxdisplay: img-ascii-lcd: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE (bnc#1012628). - iio: adc/accel: Fix up module licenses (bnc#1012628). - pinctrl: pxa: pxa2xx: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE (bnc#1012628). - ASoC: pcm512x: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE (bnc#1012628). - KVM: nVMX: Eliminate vmcs02 pool (bnc#1012628). - KVM: VMX: introduce alloc_loaded_vmcs (bnc#1012628). - objtool: Improve retpoline alternative handling (bnc#1012628). - objtool: Add support for alternatives at the end of a section (bnc#1012628). - objtool: Warn on stripped section symbol (bnc#1012628). - x86/mm: Fix overlap of i386 CPU_ENTRY_AREA with FIX_BTMAP (bnc#1012628). - x86/entry/64: Remove the SYSCALL64 fast path (bnc#1012628). - x86/entry/64: Push extra regs right away (bnc#1012628). - x86/asm: Move 'status' from thread_struct to thread_info (bnc#1012628). - x86/spectre: Fix spelling mistake: "vunerable"-> "vulnerable" (bnc#1012628). - x86/paravirt: Remove 'noreplace-paravirt' cmdline option (bnc#1012628). - KVM: VMX: make MSR bitmaps per-VCPU (bnc#1012628). - x86/kvm: Update spectre-v1 mitigation (bnc#1012628). - x86/retpoline: Avoid retpolines for built-in __init functions (bnc#1012628). - x86/speculation: Fix typo IBRS_ATT, which should be IBRS_ALL (bnc#1012628). - KVM/x86: Update the reverse_cpuid list to include CPUID_7_EDX (bnc#1012628). - KVM/x86: Add IBPB support (bnc#1012628). - KVM/VMX: Emulate MSR_IA32_ARCH_CAPABILITIES (bnc#1012628). - KVM/VMX: Allow direct access to MSR_IA32_SPEC_CTRL (bnc#1012628). - KVM/SVM: Allow direct access to MSR_IA32_SPEC_CTRL (bnc#1012628). - serial: core: mark port as initialized after successful IRQ change (bnc#1012628). - fpga: region: release of_parse_phandle nodes after use (bnc#1012628). - commit 54903ae - x86/pti: Mark constant arrays as __initconst (bsc#1068032 CVE-2017-5753). - x86/cpuid: Fix up "virtual" IBRS/IBPB/STIBP feature bits on Intel (bsc#1068032 CVE-2017-5753). - commit 7d4f257 - x86/speculation: Add basic IBRS support infrastructure (bsc#1068032 CVE-2017-5753). - x86/pti: Do not enable PTI on CPUs which are not vulnerable to Meltdown (bsc#1068032 CVE-2017-5753). - x86/cpufeature: Blacklist SPEC_CTRL/PRED_CMD on early Spectre v2 microcodes (bsc#1068032 CVE-2017-5753). - x86/nospec: Fix header guards names (bsc#1068032 CVE-2017-5753). - x86/bugs: Drop one "mitigation" from dmesg (bsc#1068032 CVE-2017-5753). - x86/cpu/bugs: Make retpoline module warning conditional (bsc#1068032 CVE-2017-5753). - x86/cpufeatures: Clean up Spectre v2 related CPUID flags (bsc#1068032 CVE-2017-5753). - x86/retpoline: Simplify vmexit_fill_RSB() (bsc#1068032 CVE-2017-5753). - x86/speculation: Simplify indirect_branch_prediction_barrier() (bsc#1068032 CVE-2017-5753). - module/retpoline: Warn about missing retpoline in module (bsc#1068032 CVE-2017-5753). - x86/spectre: Check CONFIG_RETPOLINE in command line parser (bsc#1068032 CVE-2017-5753). - x86/speculation: Use Indirect Branch Prediction Barrier in context switch (bsc#1068032 CVE-2017-5753). - Refresh patches.suse/0001-x86-cpufeatures-Add-CPUID_7_EDX-CPUID-leaf.patch. - Refresh patches.suse/0002-x86-cpufeatures-Add-Intel-feature-bits-for-Speculati.patch. - Refresh patches.suse/0003-x86-cpufeatures-Add-AMD-feature-bits-for-Speculation.patch. - Refresh patches.suse/0004-x86-msr-Add-definitions-for-new-speculation-control-.patch. - Refresh patches.suse/0007-x86-speculation-Add-basic-IBPB-Indirect-Branch-Predi.patch. - Refresh patches.suse/supported-flag. - Delete patches.suse/0005-x86-pti-Do-not-enable-PTI-on-processors-which-are-no.patch. - Delete patches.suse/0006-x86-cpufeature-Blacklist-SPEC_CTRL-on-early-Spectre-.patch. - Delete patches.suse/0008-x86-kvm-Add-IBPB-support.patch. - Delete patches.suse/0009-x86-speculation-Use-Indirect-Branch-Prediction-Barri.patch. - Delete patches.suse/0010-x86-mm-Only-flush-indirect-branches-when-switching-i.patch. - Delete patches.suse/0011-x86-speculation-Add-basic-IBRS-support-infrastructur.patch. - Delete patches.suse/0017-x86-ibrs-Add-new-helper-macros-to-save-restore-MSR_I-fix.patch. - Delete patches.suse/0017-x86-ibrs-Add-new-helper-macros-to-save-restore-MSR_I.patch. - Delete patches.suse/0018-x86-vmx-Direct-access-to-MSR_IA32_SPEC_CTRL.patch. Update to the patches from 4.16-rc1 and the updated IBRS patches from the dwmw's repo. - commit cd20d46 - Documentation: Document array_index_nospec (bsc#1068032 CVE-2017-5715). - array_index_nospec: Sanitize speculative array de-references (bsc#1068032 CVE-2017-5715). - x86: Implement array_index_mask_nospec (bsc#1068032 CVE-2017-5715). - x86: Introduce barrier_nospec (bsc#1068032 CVE-2017-5715). - x86: Introduce __uaccess_begin_nospec() and uaccess_try_nospec (bsc#1068032 CVE-2017-5715). - x86/usercopy: Replace open coded stac/clac with __uaccess_{begin, end} (bsc#1068032 CVE-2017-5715). - x86/syscall: Sanitize syscall table de-references under speculation (bsc#1068032 CVE-2017-5715). - nl80211: Sanitize array index in parse_txq_params (bsc#1068032 CVE-2017-5715). - x86/spectre: Report get_user mitigation for spectre_v1 (bsc#1068032 CVE-2017-5715). - Delete patches.suse/0001-Documentation-document-array_ptr.patch. - Delete patches.suse/0002-asm-nospec-array_ptr-sanitize-speculative-array-de-r.patch. - Delete patches.suse/0003-x86-implement-array_ptr_mask.patch. - Delete patches.suse/0004-x86-introduce-__uaccess_begin_nospec-and-ifence.patch. - Delete patches.suse/0007-x86-narrow-out-of-bounds-syscalls-to-sys_read-under-.patch. - Delete patches.suse/0009-kvm-x86-update-spectre-v1-mitigation.patch. - Delete patches.suse/0010-nl80211-sanitize-array-index-in-parse_txq_params.patch. Replace by the patches from 4.16-rc1. - commit 8343cab - scsi: aacraid: remove redundant setting of variable c (git-fixes). - commit 143e25c ==== libcue ==== Version update (2.1.0 -> 2.2.0) - Update to new upstream release 2.2.0 * Fix memory corruption in cue_scanner.l * Use MAXINDEX instead of MAXTRACK when indexing INDEX entries * Allow for 100 INDEX entries - Remove cmake.diff (applied upstream) ==== libisoburn ==== Version update (1.4.6 -> 1.4.8) Subpackages: libisoburn1 xorriso - Update to 1.4.8: * Bug fix: -as mkisofs -no-emul-boot without -boot-load-size defaulted to size 4, instead of full boot image size * Bug fix: -read_fs "norock" did not prevent reading of root Rock Ridge info * Bug fix: Mix of absolute and relative disk paths could cause SIGSEGV with pattern expansion * Bug fix: --mbr-force-bootable did not get into effect with - append_partition * Bug fix: Exit value of failed -mount command was reported as 0 * Bug fix: -boot_image actions "keep" and "patch" did not work any more. Regression by libisofs 1.4.4. * New -find tests -maxdepth and -mindepth * New commands -update_lxi and -update_li * New API calls isoburn_igopt_set_iso_mbr_part_type(), isoburn_igopt_get_iso_mbr_part_type() * New -boot_image bootspec iso_mbr_part_type= * New -as mkisofs option -iso_mbr_part_type * New -as mkisofs option -eltorito-platform * Properly refusing on Pseudo Overwritable formatted BD-R ==== libshout ==== - Fix typographic tidbits in description. Compact filelist. - Drop old Obsoletes. ==== lz4 ==== Version update (1.8.0 -> 1.8.1.2) Subpackages: liblz4-1 liblz4-1-32bit liblz4-devel - Update to new upstream release 1.8.1.2 * Most visible new feature is support for dictionary compression * Support is now more formally added on the API side within lib/lz4frame_static.h. It's early days, and this new API is tagged "experimental" for the time being. * Support is also added in the command line utility lz4, using the new command -D. The behavior of this command is identical to zstd, should you be already familiar. * lz4 doesn't specify how to build a dictionary. All it says is that it can be any file up to 64 KB. * This approach is compatible with zstd dictionary builder, which can be instructed to create a 64 KB dictionary. * Improved performance at ultra settings (levels 10+). * These levels receive a new code, called optimal parser, available in lib/lz4_opt.h. * New parser uses less memory (from 384KB to 256KB), performs faster, compresses a little bit better (not much, as it was already close to theoretical limit), and resists pathological patterns which could destroy performance. * New parser is directly compatible with lower compression levels, which brings additional benefits: + Compatibility with LZ4_*_destSize() variant, which reverses the logic by trying to fit as much data as possible into a predefined limited size buffer. + Compatibility with Dictionary compression, as it uses the same tables as regular HC mode * In the future, this compatibility will also allow dynamic change of compression level. * Small set of bug fixes and improvements: + perf : faster and stronger ultra modes (levels 10+) + perf : slightly faster compression and decompression speed + perf : fix bad degenerative case + fix : decompression failed when using a combination of extDict + low memory address + cli : support for dictionary compression (-D) + cli : fix : lz4 -d --rm preserves timestamp + cli : fix : do not modify /dev/null permission as root + api : _destSize() variant supported for all compression levels - partial cleanup with spec-cleaner - drop lz4-use-shlib.diff, breaks build - misc spec cleanups - remove Makefile fix - build and install separately - New URL ==== mono-core ==== Version update (5.4.1 -> 5.8.0) Subpackages: libmono-2_0-1 libmono-2_0-devel libmonosgen-2_0-1 libmonosgen-2_0-devel mono-data mono-data-sqlite mono-extras mono-mvc mono-wcf mono-web mono-winforms mono-winfxcore - Update to version 5.8.0.108 (5.8.0 Stable) - Minor packaging updates - Remove patch: bug-57918.patch - Remove patch: bug-59608.patch - Remove patch: mcs-compatibility-fix.patch - Add fix-64bit-portability-issue.patch: * Fix build error: E: mono-core 64bit-portability-issue threads.c:1866 - New features and changes: * WebAssembly: it should be possible to build both the runtime and class libraries targeting WebAssembly * Profiler: added a new "quiet" mode; added Mono.Profiler.Log.dll, which provides an API to the profiler * Verifier: the runtime verifier now supports C# 7.0 byref returns * Modes for the SGen GC: sgen garbage collector now has support for so called "modes", check out the mode parameter documentation for the MONO_GC_PARAMS environment variable in the man page * Other changes: improved compatibily with alternative languages that depend on implicit widening of locals * Class Libraries: more classes was replaced by CoreFX codebase * MCS changes: mcs now has better support for C# 7 syntax * csi: You are now able to use Roslyn?s csi (C# interactive) REPL tool from the command line * Roslyn: roslyn was updated to the 2.3.2 release * ilasm: ilasm gained support for the AggressiveInlining keyword - Bugfixes: * bxc#4786 - After failed call to SqlConnection.Open (due to invalid password), subsequent invalid password attempts result in wrong exception * bxc#5798 - Thread abort can lead to locks not being released * bxc#5804 - Thread.ResetAbort can prevent appdomain unloading * bxc#6339 - Headers not being populated in SoapExtension * bxc#10074 - Error while updating status of command: MonoDevelop.Ide.Commands.ViewCommands.LayoutList * bxc#11696 - Missing MethodImpl has a duplicate * bxc#12157 - Missing: Init state for this differs depending on path * bxc#17325 - RealProxy returns MarshalByRefObject for an Interface instead of the real interface type * bxc#25755 - HttpClient.Timeout does not throw ArgumentOutOfRangeException when setting Timeout property to 0 * bxc#31507 - ObjectDisposedException when canceling postAsync * bxc#36053 - Broken enum values conversion * bxc#36634 - IpcChannelTest.Bug609381 hangs on Linux * bxc#40556 - Output pad no longer displays unicode characters * bxc#41279 - Wrong syntax used for nested types in MethodAccessException (and similar) * bxc#43172 - Failure when alerting threads waiting on a socket that closes * bxc#43563 - Crash when struct Foo contains static field of type Foo[][] * bxc#44255 - Inconsistent results in the serialization of TimeZoneInfo. * bxc#44266 - Unable to change variable values whilst debugging * bxc#46683 - Recurring hang in System.Xaml on CI * bxc#52157 - SocketTest.ConnectedProperty test fails in FullAOT Linux runs * bxc#52621 - Typo in exception message in PKCS12.cs * bxc#55603 - Follow-up to bug 52845: Satellite assemblies not loaded by app when using ?Bundle assemblies into native code? even though they are now successfully mkbundled * bxc#55697 - .NET 2.0 style web services fail to build under msbuild in Release but work with xbuild * bxc#56111 - Hitting assertion after enabling AOT+LLVM on x86 devices in tramp-x86.c:124 * bxc#56242 - TypeTest.GetGenericMethodDefinitionOverInflatedMethodOnGTD throws NRE on Android in Release mode * bxc#56452 - Problem Stack too big * bxc#56653 - Zygote crashes * bxc#56684 - (mono-2.0-sgen.dll) in mono.exe: 0xC0000005: Access violation reading location 0x04FA3D9A. * bxc#56727 - Runtime crashes when disposing DeflateStream * bxc#56814 - Xamarin.Mac.dll Memory Leak starting with 3.2.0.175 Release * bxc#57002 - New bockbuild package is missing .dSYM directories * bxc#57069 - WaitHandle.SignalAndWait is broke * bxc#57222 - System.Reflection.AmbiguousMatchException for two fields with same name but different types * bxc#57232 - Compiler error when using ternary operator and no whitespace * bxc#57476 - AOT fails with large StructLayout size on Debug+ARMv7 * bxc#57488 - Debugger crashes when inspecting variable in a unit test * bxc#57496 - [Slovenian CultureInfo + others] .Parse throws exception when parsing negative values * bxc#57505 - Missing error CS0165 in presence of commented line * bxc#57528 - [THIS WEEK] Nested streams using SslStream cause ObjectDisposedException when disposed * bxc#57629 - [io-layer] File.Delete () throws System.UnauthorizedAccessException on non-existing files on a read-only path * bxc#57691 - InternalsVisibleTo doesn?t work with 2048-bit SNK files * bxc#57744 - ReflectionTypeLoadException.LoaderExceptions has null exceptions * bxc#57771 - F#.NET Core 2.0 projects do not compile * bxc#57796 - Compiler Crash instead of CS1621 * bxc#57850 - Mono assertion when ResolveEventHandler returns a reflection-only assembly * bxc#57851 - Assertion if using ResolveEventHandler to find custom attributes * bxc#57887 - Console.Error.WriteLine in System.Net.NetworkInformation * bxc#57918 - System.Net.Sockets.EndSend returns incorrect value * bxc#57930 - Building netstandard 2.0 project throws DllNotFoundException: hostfxr during msbuild * bxc#57936 - Race conditions of mempool.c * bxc#57959 - SIGSEGV in roslyn with corrupted stack * bxc#58100 - mono_w32socket_convert_error: no translation into winsock error for (6) ?Device not configured? * bxc#58114 - [llvm] csc (Rolsyn) produce fault clauses that the llvm backend does not support * bxc#58168 - pedump incorrectly reports mscorlib as invalid * bxc#58291 - Mono.CSharp Evaluator does not compile enums and throws an exception * bxc#58317 - Leaking threads: SGen worker and Finalizer * bxc#58344 - SIMD crash using System.Numerics.Vector * bxc#58361 - EnumBuilder can no longer be used to generate IL code * bxc#58379 - Mono jit gives different results that .net for simple il * bxc#58411 - System.Security.Cryptography.CryptographicException: Store root doesn?t exist * bxc#58421 - sre-save.c asserts in fixup_method () if TypeBuilder token is used across assemblies * bxc#58423 - Race condition of metadata.c (mono_metadata_get_canonical_generic_inst) * bxc#58454 - Cannot resolve method Subscription`1 Subscribe[TSelection](System.Func`2[TState,TSelection], System.Action`2[TSelection,TState]) because the declaring type of the method handle FSI_0002+Store`1[TState] is generic. Explicitly provide the declaring type to * bxc#58612 - When 2 breakpoints are hit at ?same time? 2nd thread frames count==0 * bxc#58621 - CS1703: An assembly `gactestlib? with the same identity has already been imported. Consider removing one of the references * bxc#58637 - Heap overflow when using GetAssemblyName * bxc#58738 - System.IO.MonoIO.Read - Native Crash when device file is gone * bxc#58770 - * Assertion at ..\mono\utils\mono-threads.c:707, condition `info? not met * bxc#58782 - Exception stack traces are deleted when reaching the native-to-managed wrapper in a reverse invoke * bxc#58793 - Fix pedump to use normal runtime init sequence * bxc#58809 - Return result of Type.IsAssignableFrom with the generic parameters differs from MS.NET one * bxc#58829 - Application Output Window being flooded with ?[Mono] worker parking, [Mono] worker unparking? messages when debugging agasint Android Emulator. * bxc#58877 - MonoTests.System.Threading.Tasks.TaskTests.Delay_Simple fails randomly after upgrade to Xcode 8.3 * bxc#58901 - Breaking ABI change on 32 bits with xcode 8.3 or newer * bxc#58911 - performance degradation with linq expressions * bxc#58965 - csi fails with ?type Object is defined in an assembly that is not referenced? * bxc#59039 - bug-18026.exe test times out on armel/armhf * bxc#59154 - runtime can?t load secondary-dependent library from newly unpacked temp directory * bxc#59180 - verifier does not handle TYPEDBYREF properties * bxc#59235 - [mono-2017-06] android.runtime.JavaProxyThrowable: System.ObjectDisposedException: Cannot access a disposed object. * bxc#59239 - [mono2017-06]FileTest.SymLinkLoop failed on tvOS device * bxc#59281 - System.Threading.AbandonedMutexException: The wait completed due to an abandoned mutex. * bxc#59334 - Expression: Can?t catch non-exception objects (mono wraps them with RuntimeWrappedException) * bxc#59364 - Assertion at dynamic-image.c:209, condition `prev == MONO_HANDLE_RAW (obj)? not met * bxc#59430 - api-diff not outputing out params correctly * bxc#59436 - MT3001 error * bxc#59664 - Type.GetType(??, true) throws ArgumentException on Mono but TypeLoadException on .NET * bxc#59824 - Casting an array of enums is inconsistent with .NET Framework behavior * bxc#59832 - [iOS]MT3001 error: Could not AOT the assembly error while building iOS/mobile samples * bxc#59909 - WebInvokeAttributeTest.RejectTwoParametersWhenNotWrapped() fails on Android * bxc#59916 - get_current_locale_name fails under the load with String conversion error: Illegal byte sequence encounted in the input * bxc#59953 - [llvm][bitcode] test_0_float_return_spill failed: got 1, expected 0 * bxc#60233 - Assertion in dynamic-image.c after 792b5367cd3a6ffa1a192c4d2d7ace3509cbb646 * bxc#60238 - Mono assertion hit when using csharp repl * bxc#60245 - FieldInfo.GetValue fails on ContextBoundObject * bxc#60317 - Stuck WebRequest * bxc#60340 - Csc msbuild task fails to find csc.exe * bxc#60480 - [mono-2017-10] ZipArchiveTests in System.iO Compression tests fail on device * bxc#60539 - mini-codegen.c:1265, condition `reginfo [sreg].born_in > 0? not met * bxc#60545 - Multiple argument generic with contravariant interface as an argument causes MissingMethodException. * bxc#60634 - Assertion failure ?align > 0? happening as of d15-5 * bxc#60771 - Attempting to JIT compile method ?System.Runtime.CompilerServices.Unsafe:Add (byte&,int)? while running in aot-only mode * bxc#61002 - Runtime exception: Cannot access a disposed object. Object name: ?MobileAuthenticatedStream?. * gh#mono/mono#6169 - Thread objects are being leaked * gh#mono/mono#6339 - System.TypeLoadException is thrown when run .net standard library with Mono * gh#mono/mono#6383 - System.Diagnostics.ProcessTest.Start1_FileName_NotFound failing ==== mpfr ==== Version update (4.0.0 -> 4.0.1) Subpackages: libmpfr6 mpfr-devel - update to 4.0.1: * Improved MPFR manual * bug fixes, including mpfr_div_ui, mpfr_div rounding issue ==== openblas_pthreads ==== - add openblas-s390.patch to build on s390 (bsc#1079513). ==== perl-DBI ==== Version update (1.639 -> 1.640) - updated to 1.640 see /usr/share/doc/packages/perl-DBI/Changes ==== perl-DateTime-TimeZone ==== Version update (2.15 -> 2.17) - updated to 2.17 see /usr/share/doc/packages/perl-DateTime-TimeZone/Changes 2.17 2018-01-23 - This release is based on version 2018b of the Olson database. This release reverts the changes for Ireland in the previous versions as these caused breakages in some systems that consumed the IANA time zone data. 2.16 2018-01-19 - This release is based on version 2018b of the Olson database. This release includes contemporary changes for S�o Tom� and Pr�ncipe, Brazil, and Ireland. The 2018a was skipped because it was missing a file, but there are no data changes from 2018a to 2018b. ==== perl-MailTools ==== Version update (2.19 -> 2.20) - updated to 2.20 see /usr/share/doc/packages/perl-MailTools/ChangeLog version 2.20: Mon 22 Jan 18:14:44 CET 2018 Improvements: - rewrite doc syntax to my current standard style. - text corrections rt.cpan.org#123823 [Ville Skytt�] - text corrections rt.cpan.org#123824 [Ville Skytt�] - convert to GIT - move to GitHUB ==== php7 ==== Version update (7.2.1 -> 7.2.2) Subpackages: apache2-mod_php7 php7-bcmath php7-bz2 php7-calendar php7-ctype php7-curl php7-dba php7-devel php7-dom php7-exif php7-fastcgi php7-ftp php7-gd php7-gettext php7-gmp php7-iconv php7-imap php7-json php7-ldap php7-mbstring php7-mysql php7-odbc php7-openssl php7-pdo php7-pear php7-pear-Archive_Tar php7-pgsql php7-shmop php7-snmp php7-sockets php7-sqlite php7-sysvsem php7-sysvshm php7-tidy php7-tokenizer php7-wddx php7-xmlreader php7-xmlwriter php7-xsl php7-zlib - updated to 7.2.2: This is a bugfix release, with several bug fixes included. http://php.net/ChangeLog-7.php#7.2.2 ==== postgresql10 ==== Version update (10.1 -> 10.2) Subpackages: libecpg6 libpq5 postgresql10-contrib postgresql10-devel postgresql10-docs postgresql10-server - Upate to PostgreSQL 10.2 * https://www.postgresql.org/docs/10/static/release-10-2.html * CVE-2018-1052, bsc#1080253: Fix processing of partition keys containing multiple expressions. * CVE-2018-1053, bsc#1077983: Ensure that all temporary files made by pg_upgrade are non-world-readable. * bsc#1079757: Rename pg_rewind's copy_file_range function to avoid conflict with new Linux system call of that name. - Use Python 3 for building PL/Python (boo#1067699). - Don't %config the symlinks to /etc/alternatives to avoid rpmlint warnings. ==== python-lxml ==== Version update (4.0.0 -> 4.1.1) Subpackages: python2-lxml python3-lxml - update to 4.1.1 - ElementPath supports text predicates for current node, like "[.='text']". - ElementPath allows spaces in predicates. - Custom Element classes and XPath functions can now be registered with a decorator rather than explicit dict assignments. - LP#1722776: Requesting non-Element objects like comments from a document with PythonElementClassLookup could fail with a TypeError. ==== smartmontools ==== Version update (6.5 -> 6.6) - Make possible to disable broken SAT support by -d scsi+cciss,N (bsc#1038271, smartmontools-scsiata-scsi_only.patch, https://www.smartmontools.org/ticket/871). - Update smartmontools.keyring. - update to 6.6 version - smartctl: * -i' and '--identify': ATA ACS-4 and SATA 3.3 enhancements. * Control ATA write cache through SCT Feature Control with '-s wcache-sct,ata|on|off[,p]' and '-g wcache-sct'. * Print ATA Pending Defects log with '-l defects'. * '-s wcreorder,on|off': New persistent flag ',p'. * '-s standby': Prevent temporary drive spinup. * '-n POWERMODE': New parameter to set exit status. * '-g security': ATA Security Level check fixed. * '-l scttemp*': Print minimum supported ERC Time Limit. * '-q noserial': Now also suppresses "SAS address" output. * '-i': Print IEEE EUI-64 of NVMe namespace. * '-c': Print NVMe 1.3 feature flags. * '-A': Print NVMe 1.3 thermal temperature transition statistic. * '-g/s dsn': Get/set ATA DSN. - smartd * Uses also device identify information to detect for duplicate devices. * '-e dsn' directive: Set ATA DSN. * Improved SCSI/SAS temperature logging. * Silence emails and log messages on open errors of '-d removable' devices. * Exit on device open error unless '-q never' or '-d removable' is specified (regression). - update-smart-drivedb: Now authenticates downloaded file with GnuPG. - update-smart-drivedb: New options '--trunk', '--no-verify' and '--export-key'. - Device type '-d intelliprop,N' for IntelliProp controllers. - SCSI: Default timeout increased to 1 minute. - configure: New options '--with-gnupg', '--with-scriptpath' and '--with-update-smart-drivedb=X.Y' - configure: Checks for C++11 support option and requires '--with[out]-cxx11-option' if option unknown or no C++11 support. - HDD, SSD and USB additions to drive database. - New smartmontools-* mailing list addresses. - Man page formatting reworked. - Linux: * Uses SG_IO V4 API if supported. * Devices behind hpsa driver are no longer detected as regular SCSI devices. - Darwin: Initial NVMe support based on undocumented API. - FreeBSD: * Fix panic on INVARIANTS enabled kernel. * Improve ATA SMART STATUS check for legacy controllers. * Compile fix for FreeBSD-11 and newer. - NetBSD: * NVMe support. * Full 28-bit ATA support. * Compile fix. * Use a raw disk device file. - OpenBSD: Compile fix. - OS/2: Support for the OS2AHCI driver, updating source code, adding autoscan support, adding self-test support. - Windows fixes: * Support for Windows 10 NVMe driver (stornvme.sys). * Fix CSMI access for IRST driver 15.2. * smartd: Ability to run PowerShell scripts with '-M exec'. * smartd: New PowerShell script to send smartd warning emails without external tools. * package now provides PDF man pages. - switch to new url - drop upstreamed patches: * smartmontools-scsi-suppress-no-temperature.patch * smartmontools-scsi-temperature-check.patch * smartmontools-log-scsi-temperature.patch - regen smartmontools-drivedb.h ==== unzip ==== Subpackages: unzip-doc - Add CVE-2018-1000035.patch: Fix a heap-based buffer overflow in password protected ZIP archives (CVE-2018-1000035 bsc#1080074) ==== vulkan ==== Version update (1.0.65.0 -> 1.0.68.0) Subpackages: libvulkan1 vulkan-devel - Update to new snapshot 1.0.68 * layers: add image layout validation to vkCmdBlitImage() * layers: Add VK_EXT_validation_cache implementation for shaders * layers: Add layer metadata for EXT_validation_cache extension * layers: Add descriptor limit checks to vkCreatePipelineLayout() * layers: Create instance & device extension lists - Add no-git.diff -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi, just did that update (from 0208) on my nvidia-based desktop. It boots to a black screen with cursor only :( No Mesa updates in this time (and I use the drivers from nvidia repo anyhow), so something else? I see in log Feb 13 14:25:29 lux kernel: NVRM: GPU at PCI:0000:01:00: GPU-12253b8f-a39d-51d4-7c47-40fcffb3fc22 Feb 13 14:25:29 lux kernel: NVRM: GPU Board Serial Number: Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception on GPC 0: 2D KIND Violation. Coordinates: (0x0, 0x0) Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ESR 0x500420=0x80000040 0x500434=0x0 0x500438=0x2a 0x50043c=0x10017 Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception on GPC 1: 2D KIND Violation. Coordinates: (0x8, 0x0) Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ESR 0x508420=0x80000040 0x508434=0x8 0x508438=0x2a 0x50843c=0x10017 Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ChID 0013, Class 0000902d, Offset 0000010c, Data 00000001 Kernel has been updated - maybe some incompatibility? I wanted to boot the old kernel, but am stuck there, as GRUB2 refuses to let me chose anything -> other thread. (First major issue with TW in 18 months....) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On úterý 13. února 2018 14:52:26 CET Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Someone reported bug with the same errors: https://bugzilla.opensuse.org/show_bug.cgi?id=1080787 No solution yet. Michal -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi Michal Thanks for this! I have overlooked the nvidia packages in the install list, allthough I (think I) always carefully check it before pressing ENTER :o Michal Srb wrote:
Someone reported bug with the same errors: https://bugzilla.opensuse.org/show_bug.cgi?id=1080787
No solution yet.
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi, Almost same here. kwin_x11 tries to start with OpenGL, but falls back to XRender, not every time. So I can use my KDE desktop, but without OpenGL effects. And I have to switch to VT by pressing ctrl-alt-F1, killall kwin_x11, switch back to X (cntra-alt-F7 or F8) and start openbox, before I realize what's wrong. On 2018-02-13 23:52, Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
No Mesa updates in this time (and I use the drivers from nvidia repo anyhow), so something else? I see in log
Feb 13 14:25:29 lux kernel: NVRM: GPU at PCI:0000:01:00: GPU-12253b8f-a39d-51d4-7c47-40fcffb3fc22 Feb 13 14:25:29 lux kernel: NVRM: GPU Board Serial Number: Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception on GPC 0: 2D KIND Violation. Coordinates: (0x0, 0x0) Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ESR 0x500420=0x80000040 0x500434=0x0 0x500438=0x2a 0x50043c=0x10017 Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception on GPC 1: 2D KIND Violation. Coordinates: (0x8, 0x0) Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ESR 0x508420=0x80000040 0x508434=0x8 0x508438=0x2a 0x50843c=0x10017 Feb 13 14:25:29 lux kernel: NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ChID 0013, Class 0000902d, Offset 0000010c, Data 00000001
Kernel has been updated - maybe some incompatibility? I wanted to boot the old kernel, but am stuck there, as GRUB2 refuses to let me chose anything -> other thread.
(First major issue with TW in 18 months....) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742 The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e., options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666 and rebuild the initrds. Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi, I can confirm the error and will try your fix as well. Hopefully it gets patched soon. Thanks to snapper+btrfs for a quick rollback :-) Kind regards, Erwin On Wed, Feb 14, 2018 at 6:17 AM, Peter Suetterlin <pit@astro.su.se> wrote:
Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742
The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values
A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e.,
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666
and rebuild the initrds.
Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
* Peter Suetterlin <pit@astro.su.se> [02-14-18 00:19]:
Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742
The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values
A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e.,
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666
and rebuild the initrds.
Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486)
must happen *only* with the rpm nvidia install as the packaged "run" files do not. -- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Registered Linux User #207535 @ http://linuxcounter.net Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet freenode -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Patrick Shanahan wrote:
* Peter Suetterlin <pit@astro.su.se> [02-14-18 00:19]:
Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742
The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values
A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e.,
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666
and rebuild the initrds.
Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486)
must happen *only* with the rpm nvidia install as the packaged "run" files do not.
Can you check the modprobe definitions the run file installs and/or the permissions of /dev/nvidia-modeset? The issue is that the new driver seems to enforce the values specified during load of the module and resets them on any access. If those are correct all is fine, but trying to tweak them using chown/chmod is futile... -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, Feb 14, 2018 at 02:30:16PM +0000, Peter Suetterlin wrote:
Patrick Shanahan wrote:
* Peter Suetterlin <pit@astro.su.se> [02-14-18 00:19]:
Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742
The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values
A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e.,
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666
and rebuild the initrds.
Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486)
must happen *only* with the rpm nvidia install as the packaged "run" files do not.
Can you check the modprobe definitions the run file installs and/or the permissions of /dev/nvidia-modeset?
The issue is that the new driver seems to enforce the values specified during load of the module and resets them on any access. If those are correct all is fine, but trying to tweak them using chown/chmod is futile...
Correct. Could anybody confirm my observations from comment #42 of this bugreport? https://bugzilla.suse.com/show_bug.cgi?id=1080742#c42 Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Stefan Dirsch wrote:
Correct. Could anybody confirm my observations from comment #42 of this bugreport?
Yes, works like a charm on my TW. MUCH nicer than world-writable devices... (guess I need to read a bit about this uaccess thing....) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, Feb 14, 2018 at 07:29:13PM +0000, Peter Suetterlin wrote:
Stefan Dirsch wrote:
Correct. Could anybody confirm my observations from comment #42 of this bugreport?
Yes, works like a charm on my TW.
MUCH nicer than world-writable devices...
(guess I need to read a bit about this uaccess thing....)
This is complete magic. Read https://bugzilla.opensuse.org/show_bug.cgi?id=1000625#c36 and the following comments. Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
* Peter Suetterlin <pit@astro.su.se> [02-14-18 09:48]:
Patrick Shanahan wrote:
* Peter Suetterlin <pit@astro.su.se> [02-14-18 00:19]:
Peter Suetterlin wrote:
Hi,
just did that update (from 0208) on my nvidia-based desktop.
It boots to a black screen with cursor only :(
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742
The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values
A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e.,
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666
and rebuild the initrds.
Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486)
must happen *only* with the rpm nvidia install as the packaged "run" files do not.
Can you check the modprobe definitions the run file installs and/or the permissions of /dev/nvidia-modeset?
crw-rw-rw- 1 root root 195, 254 Feb 12 16:46 /dev/nvidia-modeset
The issue is that the new driver seems to enforce the values specified during load of the module and resets them on any access. If those are correct all is fine, but trying to tweak them using chown/chmod is futile...
-- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Registered Linux User #207535 @ http://linuxcounter.net Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet freenode -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Peter Suetterlin wrote:
Summary from https://bugzilla.opensuse.org/show_bug.cgi?id=1080742
The permissions of /dev/nvidia-modeset are wrong, and something always sets them back to the default values
A (temporary) fix is to edit /etc/modprobe.d/50-nvidia-default.conf and, in the first line, set the default permission to 0666, i.e.,
options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0666
and rebuild the initrds.
Note that NVreg_DeviceFileGID should be the ID of the video group (on my TW system it is not 33, but 486) Personally I would recommend the solution from Stefan Dirsch, because his solution does not expand the permissions more then necessary. He recommends to add the line
L /run/udev/static_node-tags/uaccess/nvidia-modeset - - - - /dev/nvidia-modeset to the file /usr/lib/tmpfiles.d/nvidia-logind-acl-trick.conf and then reboot. This works for me. Source: https://bugzilla.opensuse.org/show_bug.cgi?id=1080742#c42 Greetings, Björn -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Bjoern Voigt wrote:
Personally I would recommend the solution from Stefan Dirsch, because his solution does not expand the permissions more then necessary. He recommends to add the line
I fully agree. That's why I had written 'sort of solved'. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (8)
-
Bjoern Voigt
-
Dominique Leuenberger
-
Erwin Van de Velde
-
Konstantin Voinov
-
Michal Srb
-
Patrick Shanahan
-
Peter Suetterlin
-
Stefan Dirsch