René Rebe wrote:
Hi,
Adam Stirk wrote:
Hi,
Is there a reason why when compiling GCC compiler cache is not used?
If I would like to test it where would I need to start looking?
Thanks
Adam
In theory it should indeed just work. We disabled ccache during gcc
explicitly,
because we experienced some rather hard to track cache miss-use building
multilib gcc variants (those supporting more than one target architecture,
such as i386/x86-64, ppc(32)/ppc64, or sparc(32)/sparc64 at a time).
If you want to investigate this, remove the conditional from:
package/develop/ccache/parse-config
if [ "$pkg" == gcc -a "$SDECFG_MULTILIB" = 1 ]; then
: # in GCC multilib builds the ccache hash currently collides
(or so)
...
And give it a spin with a fresh x86-64 build, if you are lucky it
even just works these days :-)
PS: "svn blame" and "svn log" are usually a great help to get an idea
why some specific code is there.
Yours,
I ran into problem on 8.0 head building a i486 target on my x86_64 building
libtool and file. The complaint was something about relocation of a local
symbol and
the suggestion to resolve the error was compile with -fpic. After I made a
patch to
the configure script to do this, both packages then compiled.
this is the error:
relocation R_X86_64_32S against `a local symbol' can not be used when making a
shared object; recompile with -fPIC
The patch system is very awesome.
Steve
-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
[EMAIL PROTECTED] with a subject of: unsubscribe t2
