Bug 358665 – stabilize net-analyzer/tcpreplay-3.4.4-r1

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Peter Volkov (RETIRED) 2011-03-13 09:15:56 UTC]

Arch teams, please, stabilize net-analyzer/tcpreplay-3.4.4-r1. Thank you in advance.

[Comment 1 Agostino Sarubbo 2011-03-13 14:51:27 UTC]

it does not respect CFLAGS:

example:

x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I../../src    -I.. -I../.. -I/usr/include -I/usr/include -D_U_="__attribute__((unused))" -march=native -O2 -Wall -O3 -std=gnu99 -Wextra -Wno-variadic-macros -Wfatal-errors -I/usr/include -MT cidr.o -MD -MP -MF .deps/cidr.Tpo -c -o cidr.o cidr.c

-O3 overwrites -O2

[Comment 2 Peter Volkov (RETIRED) 2011-03-13 18:44:55 UTC]

Thank you Agostino. Could you open new bug for this please and I'll check if this is intentional or not? In any case this is not a regression, so please, proceed with stabilization.

[Comment 3 Markos Chandras (RETIRED) 2011-03-14 19:17:40 UTC]

Current stable does not build for me so I marked this one stable anyway.

[Comment 4 Myckel Habets 2011-03-15 06:12:25 UTC]

Builds and runs tests fine on x86. Please mark stable for x86.

[Comment 5 Peter Volkov (RETIRED) 2011-03-15 08:39:39 UTC]

*** Bug 358533 has been marked as a duplicate of this bug. ***

[Comment 6 Peter Volkov (RETIRED) 2011-03-15 08:40:54 UTC]

And I've got answer from upstream developer:
 
in the past I did some performance testing and ended up settling on the flag
combination which worked the best.  Tcpreplay hasn't changed much
since then, but GCC is always evolving, so if you have some statistics
showing a different optimization flag works better with more recent
versions of GCC, then I'd be happy to accept them, but I would require
some stats to back up the patch.

IOW this optimization is upstream suggested and is not an overlook. I'll add a notion in ebuild about this.

[Comment 7 Thomas Kahle (RETIRED) 2011-03-21 08:14:05 UTC]

x86 stable. Closing.