On Thu, Sep 10, 2020 at 04:24:20PM -0300, Sure Thing wrote: > Hi, I've been using dwm for a few months, excellent wm. Recently I > clicked a link (retro console review on youtube) and the session > closed (as if Meta-Shift-Q was pressed). > I compiled a version with symbols, attached gdb to the process, and > the last instruction from dwm is: > > 286 memcpy(buf, utf8str, len); > [...] > > the last jump is at: > 1888 in ../sysdeps/x86_64/multiarch/memcpy-ssse3.S > (gdb) s > 0x00007f30606195b3 in inflate () from /lib/x86_64-linux-gnu/libz.so.1 > > and the last output is: > __memcpy_ssse3 () at ../sysdeps/x86_64/multiarch/memcpy-ssse3.S:1888 > 1888 in ../sysdeps/x86_64/multiarch/memcpy-ssse3.S > (gdb) s > 0x00007f30606195b3 in inflate () from /lib/x86_64-linux-gnu/libz.so.1 > (gdb) s > Single stepping until exit from function inflate, > which has no line number information. > 0x00007f306063b737 in png_read_row () from > /lib/x86_64-linux-gnu/libpng16.so.16 > (gdb) s > Single stepping until exit from function png_read_row, > which has no line number information. > [Inferior 1 (process 53650) exited with code 01 > > Has this happened to someone else? Greets from south america! > > -- > Eronte. > ¡Fatwolf! >
Hi Eronte, Do you use the latest dwm version (without patches to reproduce the issue of course)? The last line seems to correspond to this: https://git.suckless.org/dwm/memfile/drw.c.html#l136 But I assume the trace is incorrect and it's the common Xft bug, but the latest git and release version has a workaround for this. The st FAQ has some information (maybe you see a similar error in the Xorg log)? https://git.suckless.org/st/file/FAQ.html#l238 in drw.c there are comments about this workaround also, like: https://git.suckless.org/dwm/file/drw.c.html#l136 Please let me know with more detailed information if it's not the Xft bug but some issue in drw_text() or so. Thanks, -- Kind regards, Hiltjo
