Hello
Sorry if this is not the right newsgroup. If so, please direct me to the
right one.
I have some shared libs which link very slow. From about 350 libs, most of
them link in nearly no time, but for some ld needs about 30 Minutes to link
and uses about 260 MB of RAM (on Linux, FC6/32bit, on a
Hello,
for some years I've been running into problems with exorbitant link
times from time to time with different g++ and binutils releases.
Sometimes just switching either binutils or g++ to e.g. a self-built
one solved the problem. Now that I've been running into it again, I
decided to find the
Hello,
Stephan Kuhagen wrote:
> I have some shared libs which link very slow. From about 350 libs,
> most of them link in nearly no time, but for some ld needs about 30
> Minutes to link and uses about 260 MB of RAM (on Linux, FC6/32bit, on
> a AMD Athlon 64 X2 Dual Core Processor 4400+, 2GB RAM)
Bernd Strieder wrote:
Hello
> What a coincidence
Yes, the same came to my mind, when I saw your post... ;-)
> When using templates a lot, a lot of functions are generated multiple
> times and the duplicates have to be removed by the linker. There are
> optimizations to improve that, some gcc ma
Hello,
Stephan Kuhagen wrote:
> Bernd Strieder wrote:
>
>> When using templates a lot, a lot of functions are generated multiple
>> times and the duplicates have to be removed by the linker. There are
>> optimizations to improve that, some gcc make them possible, others
>> not, this is what I fo
I wrote this little program hello.c and compiled it on Fedora Core
6. It gives a Floating Point Error on Fedora Core 3. What is going on
here
My problem, obviously, lies with a larger app that I compile on FC6
and want to run on FC3, this example is just an illustration of the
problem. I a
At Thu, 08 Feb 2007 12:53:54 -0600 [EMAIL PROTECTED] wrote:
>
> I wrote this little program hello.c and compiled it on Fedora Core
> 6. It gives a Floating Point Error on Fedora Core 3. What is going on
> here
>
> My problem, obviously, lies with a larger app that I compile on FC6
> and
On Thu, 08 Feb 2007 21:35:37 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
> At Thu, 08 Feb 2007 12:53:54 -0600 [EMAIL PROTECTED] wrote:
>
>>
>> I wrote this little program hello.c and compiled it on Fedora Core
>> 6. It gives a Floating Point Error on Fedora Core 3. What is going on
>> here
At Thu, 08 Feb 2007 14:55:00 -0600 [EMAIL PROTECTED] wrote:
>
> On Thu, 08 Feb 2007 21:35:37 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
> > I'm guessing you did not re-compile for FC3?
>
> Correct.
>
> > Most likely the Glibc is different between FC6 and FC3 and the FC6
> > build is not (
On Thu, 08 Feb 2007 22:14:20 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
> At Thu, 08 Feb 2007 14:55:00 -0600 [EMAIL PROTECTED] wrote:
>
>>
>> On Thu, 08 Feb 2007 21:35:37 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
>> > I'm guessing you did not re-compile for FC3?
>>
>> Correct.
>>
>>
At Thu, 08 Feb 2007 15:32:49 -0600 [EMAIL PROTECTED] wrote:
>
> On Thu, 08 Feb 2007 22:14:20 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
> > At Thu, 08 Feb 2007 14:55:00 -0600 [EMAIL PROTECTED] wrote:
> >
> >>
> >> On Thu, 08 Feb 2007 21:35:37 +0100, Robert Heller <[EMAIL PROTECTED]>
> >> w
Robert Heller <[EMAIL PROTECTED]> writes:
> These would be the RHEL 4.x compatibility packages. They might do what
> you need though,
I seriously doubt they will -- OP will still be compiling against
newer libc and kernel headers, and linking against newer libc.so.
If it *does* work for non-tri
At Thu, 08 Feb 2007 15:02:00 -0800 Paul Pluzhnikov <[EMAIL PROTECTED]> wrote:
>
> Robert Heller <[EMAIL PROTECTED]> writes:
>
> > These would be the RHEL 4.x compatibility packages. They might do what
> > you need though,
>
> I seriously doubt they will -- OP will still be compiling against
>
Robert Heller <[EMAIL PROTECTED]> writes:
> *Usually* the libraries (the headers are not relevant once the
> executable has been built) will support a program built with an older
> version of the library. Often if the executable was built with a newer
> version of the library, it won't work prope
On Thu, 08 Feb 2007 23:42:06 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
> At Thu, 08 Feb 2007 15:32:49 -0600 [EMAIL PROTECTED] wrote:
>
>>
>> On Thu, 08 Feb 2007 22:14:20 +0100, Robert Heller <[EMAIL PROTECTED]> wrote:
>> > At Thu, 08 Feb 2007 14:55:00 -0600 [EMAIL PROTECTED] wrote:
>> >
>> >
On Thu, 08 Feb 2007 15:02:00 -0800, Paul Pluzhnikov <[EMAIL PROTECTED]> wrote:
> Robert Heller <[EMAIL PROTECTED]> writes:
>
>> These would be the RHEL 4.x compatibility packages. They might do what
>> you need though,
>
> I seriously doubt they will -- OP will still be compiling against
> newer l
Ignoramus25565 <[EMAIL PROTECTED]> writes:
> Seems to work.
I've reproduced the crash when compiling "int main() { return 0; }"
on FC6 and running it on FC4.
The reason dynamic linker crashes is that a.out doesn't have
DT_HASH dynamic tag at all, and FC4 ld-linux.so.2 expects it (and
divides by
After I built the same perl as on FC3 (i386-linux), and linked against
that, my app now runs seemingly fine. I will check it more in depth
tomorrow.
i
___
help-gplusplus mailing list
help-gplusplus@gnu.org
http://lists.gnu.org/mailman/listinfo/help-gplu
On Thu, 08 Feb 2007 19:04:58 -0800, Paul Pluzhnikov <[EMAIL PROTECTED]> wrote:
> Ignoramus25565 <[EMAIL PROTECTED]> writes:
>
>> Seems to work.
>
> I've reproduced the crash when compiling "int main() { return 0; }"
> on FC6 and running it on FC4.
>
> The reason dynamic linker crashes is that a.out
Ignoramus25565 <[EMAIL PROTECTED]> writes:
>> Still you are on dangerous ground ...
>
> Seems like using gcc34 is a better option than
> '-Wl,--hash-style=sysv', then?
Using gcc34 does *not* get you off the dangerous ground.
You still may have subtle bugs.
But if it passes all of your regressio
20 matches
Mail list logo
