Red Hat Bugzilla – Bug 1469861
Powerline is very slow on F26 even with powerline-daemon running
Last modified: 2017-07-24 15:20:26 EDT
Description of problem: The execution of a new prompt (by say pressing enter) is very slow after upgrading to F26. Version-Release number of selected component (if applicable): powerline-2.6-1.fc26.x86_64 How reproducible: Always, on all systems I upgraded to F26 Steps to Reproduce: 1. Use powerline on a F26 system 2. Hold enter at the command prompt 3. Observe slow behavior compared with F25 Actual results: Prompt is generated slowly, you can see a MUCH more pronounced effect on older slower systems. You can also see it hang slightly with set -x; <enter> Expected results: Since this is a regression, I would expect that it worked as quickly as before. Additional info: I'd love to work with troubleshooting this with you, but I'm not really sure where to start. I'm pretty confident that you're going to see this behavior on any desktop environment and in any terminal, as I also tested it in a normal tty and it is also slow there.
See upstream bug here: https://github.com/powerline/powerline/issues/1800
Developer closed both bugs because it is apparently our fault for packaging the python version of the script as /usr/bin/powerline. See: F25: file $(which powerline) /usr/bin/powerline: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=04e4042f5650ff2b23f858862231a59472aeaddd, stripped F26: file $(which powerline) /usr/bin/powerline: Python script, ASCII text executable I suspect this unintentional and is some install time weirdness based on the way the developer decided to automatically choose which one would be build based on available dependencies, but I'll have to dive into the SPEC to say more. Get back when I can!
client/powerline.c: In function ‘main’: client/powerline.c:125:36: warning: ‘__builtin___snprintf_chk’ output may be truncated before the last format character snprintf(num_args, NUM_ARGS_SIZE, "%x", argc - 1); ^~~~ In file included from /usr/include/stdio.h:939:0, from client/powerline.c:4: /usr/include/bits/stdio2.h:64:10: note: ‘__builtin___snprintf_chk’ output between 2 and 9 bytes into a destination of si return __builtin___snprintf_chk (__s, __n, __USE_FORTIFY_LEVEL - 1, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ __bos (__s), __fmt, __va_arg_pack ()); ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /tmp/ccKwNdgL.o: In function `main': /builddir/build/BUILD/powerline-2.6/client/powerline.c:114: undefined reference to `true_sun_len' collect2: error: ld returned 1 exit status ERROR:root:Command '['cc', '-O2', '-g', '-pipe', '-Wall', '-Werror=format-security', '-Wp,-D_FORTIFY_SOURCE=2', '-fexcep Traceback (most recent call last): File "setup.py", line 38, in <module> compile_client() File "setup.py", line 35, in compile_client subprocess.check_call(compiler + shlex.split(cflags) + ['client/powerline.c', '-o', 'scripts/powerline']) File "/usr/lib64/python3.6/subprocess.py", line 291, in check_call raise CalledProcessError(retcode, cmd) subprocess.CalledProcessError: Command '['cc', '-O2', '-g', '-pipe', '-Wall', '-Werror=format-security', '-Wp,-D_FORTIFY Compiling C version of powerline-client failed Using powerline.py script instead of C version This is why F25 and F26 differ. It seems that you found that out the hard way given that as I'm looking at the SPEC differences between F25 and F26, you added this: +# As we only have python files do not create a debuginfo package +%global debug_package %{nil} And switch it to the `noarch` BuildArch. The way the comment is written, with the absence of info about this in the changelog leads me to believe you didn't realize that this was an automatic fallback because the C program failed to build, so that's why I'm documenting that here. Again, more info happening at github, too.
powerline-2.6-2.fc26 has been submitted as an update to Fedora 26. https://bodhi.fedoraproject.org/updates/FEDORA-2017-222bbba94a
powerline-2.6-2.fc26 has been pushed to the Fedora 26 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2017-222bbba94a
powerline-2.6-2.fc26 has been pushed to the Fedora 26 stable repository. If problems still persist, please make note of it in this bug report.