Hi Adrian,
I am with you,
I tried to apply the patches provided by solaris, I tried to compile but
on the Powermac itself it lasts three days to compile with the end of
did'nt successfully compiled. I do understand now the problem of not
having the time or not having the hardware or to be frustrated because
nothing works.
Meanwhile I tried to crosscompile that thing for powerpc64 on my macbook
pro m1 pro so it should go faster, but I always have problems with (I
think) big endian things.
Another option I thought could be to emulate a power8 system with a lot
of cores and to install also debian 12 ppc64 on this and compile there…
I also think about power consumption of the POwermac G5 when he compiles
3 days for nothing with both cores at 100%.
Til this day I never had success to compile Firefox successfully even once.
I know that even famous Rene Rebe is not able to compile because things
are broken.
Jan
Am 18.11.24 um 16:39 schrieb John Paul Adrian Glaubitz:
Hi Jan,
[…]
The problem is that Firefox is currently not fully compatible with big-endian
systems like your PowerMac G5. There are supposedly patches which address these
issues, notably on Solaris SPARC.
However, I simply did not have the time yet to go through Oracle's patches [1]
one by one and verify which of these patches fix the segfault. However, the
backtrace of the segfault might give us an idea which of the patches fixes the
crash since symbols (e.g. function names) of the problematic code should show
both in the backtrace as well as any of the patches from Oracle.
Please keep in mind that I'm all doing this in my free time and I'm not getting
paid for anything, so I really have to balance the time I'm spending on various
issues.
Adrian
[1]
https://github.com/oracle/solaris-userland/tree/master/components/desktop/firefox/patches
--
Herzlichst Jan Montag
via vintage computing
