Re: [PATCH] fix build with gcc 4.8.5-4 and -Werror=inline option

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 17.05.2016 11:36, Maxim Nestratov wrote:
> fix numerous inlining errors like:
> 
> In file included from vircgroupmock.c:24:0:
> In function 'stat':
> virmock.h:259:1: error: inlining failed in call to 'callStat.part.1':
> call is unlikely and code size would grow [-Werror=inline]
>  callStat(int (*realStat)(const char *, struct stat *),
> ---
>  tests/Makefile.am |  8 +++++---
>  tests/virmock.c   | 43 +++++++++++++++++++++++++++++++++++++++++++
>  tests/virmock.h   | 16 ++--------------
>  3 files changed, 50 insertions(+), 17 deletions(-)
>  create mode 100644 tests/virmock.c
> 

I know about this issue and hunting it down for good two days now. While
this would resolve the inline issue, things are much more broken than
that. GNU glibc's implementation of stat() is probably the most
obfuscated code I've ever laid my eyes on. As a result of that, my
mocking makes some tests run in an endless loop on my arm machine (yeah,
raspberry, but still). The solution I'm developing for these two days
now, would completely drop this inline function. So stay tuned.

Michal

--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list



[Index of Archives]     [Virt Tools]     [Libvirt Users]     [Lib OS Info]     [Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]     [Fedora Tools]