Bug 225658
| Summary: | Merge Review: cpuspeed | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Nobody's working on this, feel free to take it <nobody> |
| Component: | Package Review | Assignee: | Nobody's working on this, feel free to take it <nobody> |
| Status: | CLOSED WONTFIX | QA Contact: | Fedora Package Reviews List <fedora-package-review> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | rawhide | CC: | jarod, ovasik, wtogami |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2011-08-18 12:10:53 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Nobody's working on this, feel free to take it
2007-01-31 17:52:17 UTC
http://www.redhat.com/archives/fedora-maintainers/2007-January/msg00339.html "make CFLAGS=..." is a no-op; the Makefile uses COPTS, thus the build ends up doing just "gcc -Wall -fno-exceptions -c -O2 cpuspeed.cc". Just changing CFLAGS to COPTS makes the build fail here on FC6 i386, though. Needs work then, assigning to owner before the review can go further. After you fix this issue, please set fedora-review back to BLANK and reassign back to the reviewer. Okay, I think I *almost* have this one sorted out... I've got the package building on rawhide x86_64 using most of our stock optflags, save -fexceptions, which is the one that was causing the build to fail when switching from CFLAGS= to COPTS= (note that that stock Makefile is set up to pass -fno- exceptions). I've tweaked things a bit further so the -z bits are only passed during linking, so as to suppress some warning messages. ----8<---- + make 'CFLAGS=-O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fno-exceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -fpie -pie' LDFLAGS=-Wl,-z,relro,-z,now gcc -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fno-exceptions -fstack-protector --param=ssp- buffer-size=4 -m64 -mtune=generic -fpie -pie -c -O2 cpuspeed.cc cpuspeed.cc:121: warning: non-local variable '<anonymous struct> speeds [16]' uses anonymous type cpuspeed.cc: In function 'int main(unsigned int, char**)': cpuspeed.cc:810: warning: ignoring return value of 'int daemon(int, int)', declared with attribute warn_unused_result gcc -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fno-exceptions -fstack-protector --param=ssp- buffer-size=4 -m64 -mtune=generic -fpie -pie -Wl,-z,relro,-z,now cpuspeed.o -o cpuspeed + exit 0 ----8<---- This results in built packages that, at a glance, seem to be functioning correctly. Unfortunately, I'm still winding up with no source in the debuginfo package for some reason that is beyond me at the moment... Turns out the lack of source files in my debuginfo is due to my rpm build root's $RPM_BUILD_DIR having a shorter path name than /usr/src/debug, which debugedit (called from find-debuginfo.sh) pukes on. With a suitably longer build root path, source files are indeed being properly included in -debuginfo now. as cpuspeed is dead package now, closing WONTFIX... |