1
Applications and Porting / Re: [GAMES] dosbox-staging -- one of the best choice for DOS gaming
« Last post by ClassicHasClass on Today at 01:09:17 pm »tle has a fork of it with the JIT I wrote just before the deletion landed.
Recent Posts
2
Applications and Porting / Re: [GAMES] dosbox-staging -- one of the best choice for DOS gaming
« Last post by Hasturtium on July 09, 2026, 10:37:48 am »So what’s the best option now? Is there a specific version I should download, compile, and use moving forward?
3
Applications and Porting / Re: [DEV] JIT support for Firefox
« Last post by MPC7500 on July 04, 2026, 09:17:42 pm »tle, are you already using Fable, or which model do you use?
4
Operating Systems and Porting / Follow-up: Void Linux on ppc64le - Firefox Native Build vs Fedora 42
« Last post by DKnoto on July 03, 2026, 12:55:44 pm »Following up on my recent project of bootstrapping Void Linux on ppc64le, I decided to quantify the exact performance penalty we pay when using generic distribution binaries on our Talos II workstations.
I compiled Firefox 152 natively using both GCC and Clang (with and without LTO/PGO) and ran a rigorous Speedometer 3.1 benchmark against the standard Firefox 151 package provided by Fedora 42.
To eliminate DVFS noise, the CPU was strictly pinned to 2.8 GHz, and tests were run in an isolated E27 environment using --new-instance --private-window to avoid the unaccelerated paths of the standard safe mode.
The Build Times & Toolchain Limits
Clang: 43m 30.668s
Clang + LTO: 45m 32.136s
Gcc: 67m 05.885s
Gcc + LTO: 84m 28.716s
Note on PGO: I attempted PGO builds, but the current toolchain state on ppc64le (specifically regarding VSX) makes it a dead end. GCC instrumentation resulted in an Internal Compiler Error, and Clang took over 3 hours (185m real time) only to produce a crashing binary.
Performance Results (Speedometer 3.1)
FF151 / F42: 1.647; stdev = 0.018;
FF152 / GCC: 5.449; stdev = 0.086
FF152 / GCC+LTO: 5.518; stdev = 0.063;
FF152 / Clang: 5.478; stdev = 0.065;
FF152 / Clang+LTO: 5.779l stdev = 0.050.
Key Architectural Takeaways:
The full raw data, generated charts, XBPS build templates, and the critical patch fixing VSX alignment issues in the Mozilla codebase are available in my public GitHub repository here: https://github.com/DKnoto/Firefox-Benchmarks
I compiled Firefox 152 natively using both GCC and Clang (with and without LTO/PGO) and ran a rigorous Speedometer 3.1 benchmark against the standard Firefox 151 package provided by Fedora 42.
To eliminate DVFS noise, the CPU was strictly pinned to 2.8 GHz, and tests were run in an isolated E27 environment using --new-instance --private-window to avoid the unaccelerated paths of the standard safe mode.
The Build Times & Toolchain Limits
Clang: 43m 30.668s
Clang + LTO: 45m 32.136s
Gcc: 67m 05.885s
Gcc + LTO: 84m 28.716s
Note on PGO: I attempted PGO builds, but the current toolchain state on ppc64le (specifically regarding VSX) makes it a dead end. GCC instrumentation resulted in an Internal Compiler Error, and Clang took over 3 hours (185m real time) only to produce a crashing binary.
Performance Results (Speedometer 3.1)
FF151 / F42: 1.647; stdev = 0.018;
FF152 / GCC: 5.449; stdev = 0.086
FF152 / GCC+LTO: 5.518; stdev = 0.063;
FF152 / Clang: 5.478; stdev = 0.065;
FF152 / Clang+LTO: 5.779l stdev = 0.050.
Key Architectural Takeaways:
- The VSX/JIT Effect: The native Clang+LTO build delivers a ~250% performance increase (a 3.5x multiplier) over the default Fedora 42 package. This heavily implies that standard distro builds are either running scalar code or heavily crippling the JS JIT compiler for stability.
- LLVM Dominance: Clang not only builds significantly faster than GCC, but it also scales much better with LTO.
- GCC UI Anomaly trimmed: During the tests, I noticed that regardless of LTO, the GCC-compiled binaries exhibit significant UI desynchronization (the "burger menu" lags about three cursor positions behind). The SpiderMonkey engine runs fine, but the interaction with system libraries (GTK/Cairo) is visibly degraded compared to the Clang build.
The full raw data, generated charts, XBPS build templates, and the critical patch fixing VSX alignment issues in the Mozilla codebase are available in my public GitHub repository here: https://github.com/DKnoto/Firefox-Benchmarks
5
Operating Systems and Porting / [HELP] Install grub boot loader for Chimera installation [SOLVED]
« Last post by tle on June 23, 2026, 07:26:00 pm »Good morning everyone
I would like to ask if anyone has managed to install grub boot-loader successfully for their Chimera installation. This is how I did it but upon restarting the boot-loader failed to load correct firmware:
Great thanks in advance
EDIT: I managed to fix the issue myself, I forgot to install linux kernel, that is `apk add linux-stable` step. I documented the full installation at https://gist.github.com/runlevel5/ca4d531c8e90c133fc31c47f823c073d
I would like to ask if anyone has managed to install grub boot-loader successfully for their Chimera installation. This is how I did it but upon restarting the boot-loader failed to load correct firmware:
Code: [Select]
# NOTES: This installation is done in libvirt VM (so no petitboot obviously)
# Download the ISO then boot it up under VM
$ cfdisk /dev/vda
choose `gpt`
select Free space -> choose [New] -> 16MB
select /dev/vda1 -> [Type] -> PowerPC PReP boot
select Free space -> [New] -> 1GB
select /dev/vda2 -> [Type] -> Linux filesystem
select Free space -> [New] -> 4GB
select /dev/vda3 -> [Type] -> Linux swap
select Free space -> [New] -> 45GB
select /dev/vda4 -> [Type] -> Linux filesystem
choose [Write] --> yes
choose [quit]
$ mkfs.ext4 /dev/vda2
$ mkswap /dev/vda3
$ mkfs.ext4 /dev/vda4
$ mount /dev/vda4 /mnt
$ mount -m /dev/vda2 /mnt/boot/
$ swapon /dev/vda3
$ chimera-bootstrap /mnt
$ chimra-chroot /mnt
$ apk add grub-powerpc-ieee1275
$ mkdir -p /boot/grub/
$ grub-install --target=powerpc-ieee1275 /dev/vda1
$ update-grub
$ exit
$ umount -R /mnt
$ reboot
Great thanks in advance
EDIT: I managed to fix the issue myself, I forgot to install linux kernel, that is `apk add linux-stable` step. I documented the full installation at https://gist.github.com/runlevel5/ca4d531c8e90c133fc31c47f823c073d
6
Applications and Porting / Re: [DEV] JIT support for Firefox
« Last post by carlosgonz on June 21, 2026, 02:32:37 pm »Thank you Hero, yes it is better to i compile it, however the heat manager is not good on my Raptor Blackbird, but i guess i could compile with 1 core to reduce heat.
Guys i like so much powerpc be.
sent from motorola powerpc,altivec,32bits and netsurf!!!
debian
Guys i like so much powerpc be.
sent from motorola powerpc,altivec,32bits and netsurf!!!
debian
7
Applications and Porting / Re: [DEV] JIT support for Firefox
« Last post by tle on June 17, 2026, 08:12:59 pm »I can not boot tle-hero ppc-firefox build on trisquel ecne.
Can you bump it firefox newer and recompile on glibc 2.39 <=
please, please tle
~/Downloads/firefox$ ./firefox
XPCOMGlueLoad error for file /Downloads/firefox/libxul.so:
/lib/powerpc64le-linux-gnu/libm.so.6: version `GLIBC_2.43' not found (required by /Downloads/firefox/libxul.so)
Couldn't load XPCOM.
~/Downloads/firefox$ ldd --version
ldd (trisquel GLIBC 2.39-0trisquel8.7trisquel1) 2.39
Copyright (C) 2024 Free Software Foundation, Inc.
Esto es software libre; vea el código fuente para las condiciones de copia.
No hay NINGUNA garantía; ni siquiera de COMERCIABILIDAD o IDONEIDAD PARA UN
FIN DETERMINADO.
Escrito por Roland McGrath y Ulrich Drepper.
I mean you have to compile it on Trisquel because that distribution does not have the same glibc version. The build instructions is kinda like this:
git clone --tags https://github.com/mozilla-firefox/firefox.git
cd firefox
git checkout FIREFOX_152_0_1_RELEASE
# download 3 patches from https://github.com/runlevel5/firefox/releases/tag/FIREFOX_152_0_1_RELEASE_PPC64
git am 000*.patch
mkdir -p obj-ff-ppc64
export CC=gcc
export CXX=g++
ac_add_options --enable-application=browser
ac_add_options --enable-optimize
ac_add_options --disable-debug
ac_add_options --enable-release
ac_add_options --disable-tests
ac_add_options --enable-jit
ac_add_options --with-ccache
ac_add_options --with-libclang-path=/usr/lib64
ac_add_options --without-wasm-sandboxed-libraries
ac_add_options --enable-linker=lld
mk_add_options MOZ_OBJDIR=@TOPSRCDIR@/obj-ff-ppc64
./mach build
./mach package
Then let me know how you go, I highly encourage you ask the Trisque packagers to adopt my downstream patches so you do not have to build it yourself
8
Applications and Porting / Re: [GAME] Minecraft Java Edition (LWJGL being ported)
« Last post by tle on June 17, 2026, 08:06:26 pm »26.2 is also out
Check https://github.com/runlevel5/PrismLauncher-rpmspec/releases/tag/26.2-ppc64le-64kpages
Check https://github.com/runlevel5/PrismLauncher-rpmspec/releases/tag/26.2-ppc64le-64kpages
9
Operating Systems and Porting / [NEWS] Linux 7.1
« Last post by tle on June 14, 2026, 06:15:40 pm »7.1 is out! https://lore.kernel.org/lkml/CAHk-=wi4BF4bMhZNZ1tqs+FFV4OuZRe3ZqdWB+LxRLmRweUzQw@mail.gmail.com/T/#u
No changes related to ppc64 however the new NTFS driver is a welcoming change for users who still interact with Windows filesystem.
No changes related to ppc64 however the new NTFS driver is a welcoming change for users who still interact with Windows filesystem.
10
Operating Systems and Porting / Re: [NEWS] Bootstrapping native Void Linux (ppc64le) on Talos II
« Last post by DKnoto on June 14, 2026, 03:07:18 pm »I've finished restoring a minimal version of Void Linux on ppc64 with DE E25/X11: https://www.dropbox.com/scl/fi/iok4f2j0l2mxfxbhgxpee/Void-Linux-Talos2-ppc64le-03-Minimal-with-E25.jpg?rlkey=o31u971una54ajj95t252kvy4&st=n5pu0fxt&dl=0
