June 2017 - openSUSE Bugs - openSUSE Mailing Lists

[Bug 1042090] TW 20170529 major breakage in Thunderbird
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.suse.com/show_bug.cgi?id=1042090 http://bugzilla.suse.com/show_bug.cgi?id=1042090#c7 --- Comment #7 from Marcus Meissner <meissner(a)suse.com> --- Thunderbird has a openQA test, but the needles were a bit unfortunately and did not include the UI corruption areas. I think the opensuse openqa guys will enhance them to avoid such a case in the future. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042012] Latest gcc 7-1.2 cause crash
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.novell.com/show_bug.cgi?id=1042012 http://bugzilla.novell.com/show_bug.cgi?id=1042012#c6 --- Comment #6 from Richard Biener <rguenther(a)suse.com> --- So as I thought GCC is taking advantage of #ifdef _MSC_VER __declspec(align(16)) #endif struct cdata { const struct ctype type #ifdef __GNUC__ __attribute__ ((aligned(16))) #endif ; }; when loading from cd->type but cd = (struct cdata*) lua_touserdata(L, idx); lua_touserdata doesn't honor that alignment. You can probably see that at the point of the crash %eax is not aligned to 16 bytes. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1039638] moc player gives segmentation fault
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.suse.com/show_bug.cgi?id=1039638 http://bugzilla.suse.com/show_bug.cgi?id=1039638#c1 Olaf Hering <ohering(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |i22a(a)yahoo.co.uk Flags| |needinfo?(i22a(a)yahoo.co.uk) --- Comment #1 from Olaf Hering <ohering(a)suse.com> --- I think this is fixed with a more recent Tumbleweed snapshot. There are changes in the way linking to ncurses is done. I tried it a few days ago on some system and moc played .mp3 files fine. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042012] Latest gcc 7-1.2 cause crash
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.suse.com/show_bug.cgi?id=1042012 http://bugzilla.suse.com/show_bug.cgi?id=1042012#c5 --- Comment #5 from Dr. Werner Fink <werner(a)suse.com> --- 252 *ct = cd->type; (gdb) print ct $1 = (struct ctype *) 0x7ffcaf207fc0 (gdb) print cd->type $2 = {base_size = 16, {{bit_size = 0, bit_offset = 0}, array_size = 0, variable_increment = 0}, offset = 0, align_mask = 0, pointers = 0, const_mask = 0, type = 6, is_reference = 0, is_array = 0, is_defined = 1, is_null = 0, has_member_name = 0, calling_convention = 0, has_var_arg = 0, is_variable_array = 0, is_variable_struct = 0, variable_size_known = 0, is_bitfield = 0, has_bitfield = 0, is_jitted = 0, is_packed = 0, is_unsigned = 0} (gdb) print *ct $3 = {base_size = 0, {{bit_size = 0, bit_offset = 0}, array_size = 0, variable_increment = 0}, offset = 0, align_mask = 0, pointers = 0, const_mask = 0, type = 0, is_reference = 0, is_array = 0, is_defined = 0, is_null = 0, has_member_name = 0, calling_convention = 0, has_var_arg = 0, is_variable_array = 0, is_variable_struct = 0, variable_size_known = 0, is_bitfield = 0, has_bitfield = 0, is_jitted = 0, is_packed = 0, is_unsigned = 0} -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1040546] LibreOffice using very high CPU doing nothing
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.opensuse.org/show_bug.cgi?id=1040546 http://bugzilla.opensuse.org/show_bug.cgi?id=1040546#c4 Kriton Kyrimis <kyrimis(a)alumni.princeton.edu> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |kyrimis(a)alumni.princeton.ed | |u --- Comment #4 from Kriton Kyrimis <kyrimis(a)alumni.princeton.edu> --- Same here, except that CPU usage is 100%. A workaround that I have found is to force LibreOffice to use the KDE instead of the gnome interface, by setting the OOO_FORCE_DESKTOP environment variable to kde4. I suspect that the reason that LibreOffice is always started using the gnome interface no longer exists (see bug #1042136 which I just submitted), so this may be the way to go, if you use KDE. -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042012] Latest gcc 7-1.2 cause crash
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.suse.com/show_bug.cgi?id=1042012 http://bugzilla.suse.com/show_bug.cgi?id=1042012#c4 --- Comment #4 from Dr. Werner Fink <werner(a)suse.com> --- Created attachment 727277 --> http://bugzilla.suse.com/attachment.cgi?id=727277&action=edit texk/web2c/luatexdir/luaffi/ffi.h -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042012] Latest gcc 7-1.2 cause crash
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.suse.com/show_bug.cgi?id=1042012 http://bugzilla.suse.com/show_bug.cgi?id=1042012#c3 --- Comment #3 from Dr. Werner Fink <werner(a)suse.com> --- The type of `struct cdata' is defined and declared in texk/web2c/luatexdir/luaffi/ffi.h with #ifdef _MSC_VER __declspec(align(16)) #endif struct cdata { const struct ctype type #ifdef __GNUC__ __attribute__ ((aligned(16))) #endif ; }; Whereas `struct ctype' is given in the same file at line 320 upto 366 with /* Note: if adding a new member that is associated with a struct/union * definition then it needs to be copied over in ctype.c:set_defined for when * we create types based off of the declaration alone. * * Since this is used as a header for every ctype and cdata, and we create a * ton of them on the stack, we try and minimise its size. */ struct ctype { size_t base_size; /* size of the base type in bytes */ union { /* valid if is_bitfield */ struct { /* size of bitfield in bits */ unsigned bit_size : 7; /* offset within the current byte between 0-63 */ unsigned bit_offset : 6; }; /* Valid if is_array */ size_t array_size; /* Valid for is_variable_struct or is_variable_array. If * variable_size_known (only used for is_variable_struct) then this is * the total increment otherwise this is the per element increment. */ size_t variable_increment; }; size_t offset; unsigned align_mask : 4; /* as (align bytes - 1) eg 7 gives 8 byte alignment */ unsigned pointers : POINTER_BITS; /* number of dereferences to get to the base type including +1 for arrays */ unsigned const_mask : POINTER_MAX + 1; /* const pointer mask, LSB is current pointer, +1 for the whether the base type is const */ unsigned type : 5; /* value given by type enum above */ unsigned is_reference : 1; unsigned is_array : 1; unsigned is_defined : 1; unsigned is_null : 1; unsigned has_member_name : 1; unsigned calling_convention : 2; unsigned has_var_arg : 1; unsigned is_variable_array : 1; /* set for variable array types where we don't know the variable size yet */ unsigned is_variable_struct : 1; unsigned variable_size_known : 1; /* used for variable structs after we know the variable size */ unsigned is_bitfield : 1; unsigned has_bitfield : 1; unsigned is_jitted : 1; unsigned is_packed : 1; unsigned is_unsigned : 1; }; with #define POINTER_BITS 2 #define POINTER_MAX ((1 << POINTER_BITS) - 1) at line 315 -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042118] Packaging bug : GTG should require python-notify.
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.opensuse.org/show_bug.cgi?id=1042118 Andreas Stieger <astieger(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Component|Basesystem |GNOME Assignee|bnc-team-screening(a)forge.pr |bnc-team-gnome(a)forge.provo. |ovo.novell.com |novell.com -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042134] New: vlc-3.0.0 can't play BluRays
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.novell.com/show_bug.cgi?id=1042134 Bug ID: 1042134 Summary: vlc-3.0.0 can't play BluRays Classification: openSUSE Product: openSUSE.org Version: unspecified Hardware: x86-64 OS: openSUSE 42.2 Status: NEW Severity: Normal Priority: P5 - None Component: 3rd party software Assignee: opensuse-communityscreening(a)forge.provo.novell.com Reporter: Winfrid.Tschiedel(a)t-online.de QA Contact: opensuse-communityscreening(a)forge.provo.novell.com Found By: --- Blocker: --- Created attachment 727275 --> http://bugzilla.novell.com/attachment.cgi?id=727275&action=edit compressed session-log for vlc -vvv After updating to the most recent version of vlc-beta you can't play BluRays. Type of the BluRay : with BDJ-menus vlc starts, but after showing an initial picture, vlc does not make any progress. Also the mouse disappears and there is no chance to kill vlc. Playing BluRays worked already on openSUSE Leap 42.2 - last time tested before : mid of February 2017. There is no problem on Windows 10 x64 with ./vlc --version VLC-Version 3.0.0-git Vetinari (2.2.0-git-12732-g27849047fc) Kompiliert von jenkins auf 5ac4d0d08b73 (May 27 2017 04:34:55) Works also on fedora 25 - but vlc-3.0.0 is from end of January 2015 -- You are receiving this mail because: You are on the CC list for the bug.

1 0

[Bug 1042012] Latest gcc 7-1.2 cause crash
by bugzilla_noreply＠novell.com 01 Jun '17

01 Jun '17

http://bugzilla.suse.com/show_bug.cgi?id=1042012 http://bugzilla.suse.com/show_bug.cgi?id=1042012#c2 --- Comment #2 from Dr. Werner Fink <werner(a)suse.com> --- With disassemble /s to_cdata in gdb I see 248 if (!cd) { 0x0000000000619314 <+116>: je 0x619375 <to_cdata+213> 249 lua_pushnil(L); 250 return NULL; 251 } 252 *ct = cd->type; => 0x0000000000619316 <+118>: movdqa (%rax),%xmm0 253 lua_getuservalue(L, idx); 0x000000000061931a <+122>: mov %r13d,%esi 0x000000000061931d <+125>: mov %rbx,%rdi that is that movdqa instruction does cause the segementation fault: Transcript written on luaimage.log. ../../../texk/web2c/luatexdir/luaimage.test: line 12: 31382 Segmentation fault (core dumped) ./luatex -ini luaimage FAIL luatexdir/luaimage.test (exit status: 1) On frame above the function to_cdata() its self is called with Dump of assembler code for function check_cdata: ../../../texk/web2c/luatexdir/luaffi/ctype.c: 269 { 0x0000000000619390 <+0>: push %r13 0x0000000000619392 <+2>: push %r12 0x0000000000619394 <+4>: mov %esi,%r12d 0x0000000000619397 <+7>: push %rbp 0x0000000000619398 <+8>: push %rbx 0x0000000000619399 <+9>: mov %rdx,%rbx 0x000000000061939c <+12>: mov %rdi,%rbp 0x000000000061939f <+15>: sub $0x8,%rsp 270 void* p = to_cdata(L, idx, ct); 0x00000000006193a3 <+19>: callq 0x6192a0 <to_cdata> 271 if (ct->type == INVALID_TYPE) { => 0x00000000006193a8 <+24>: testb $0x7c,0x19(%rbx) 270 void* p = to_cdata(L, idx, ct); 0x00000000006193ac <+28>: mov %rax,%r13 271 if (ct->type == INVALID_TYPE) { 0x00000000006193af <+31>: jne 0x6193c5 <check_cdata+53> -- You are receiving this mail because: You are on the CC list for the bug.

1 0