I have been carrying out some preliminary tests to see how feasible it is to compile the Linux Kernel, and NVIDIA drivers, and VMware, using the ‘alternative’ to GCC – CLANG.. The kernel compile is relatively easy, and well documented: # make CC=clang HOSTCC=clang -j8 (for a 4-core system) The compile seems to be slower than GCC, … Continue reading →
Recently acquired one of the Khadas VIM3-Pro arm64 SoCs, and was impressed with the level of documentation available.. Unfortunately – as is so often the case – this was all relentlessly Ubuntu-focussed, with a semi-proprietary kernel based on the 4.X kernel tree.. At least the U-Boot version was fairly recent, and appeared to support EFI etc.. The … Continue reading →
Fedora 33 will be using ZRAM as standard for swap, rather than a partition or swapfile on disk. I had no problem with this on a new F33 install (Odroid H2Plus) but after upgrading an existing Fedora 32 system it refused to initialise, and ‘/dev/zram0‘ did not exist. Research revealed that the documentation for much of this … Continue reading →
I have been testing Fedora 33 pre-release, and on x86_64 it seems relatively stable (usual caveats…) but on arm64 systems (Odroid N2 and Pi4) I found that after compiling a new kernel (5.8.3), the system booted, but login was impossible, and screen output was missing (uart/console was OK). Kernels compiled with Fedora 32 all booted OK, and … Continue reading →
I have been investigating alternatives, as VMware is currently a non-starter (literally..) with Kernel 5.8.. I installed the latest VirtualBox (6.1.97 r139689) from SVN (another blast from the past…) and managed – after some messing around – to get it to compile OK on my Fedora 32 test system with Kernel 5.8 (from kernel.org..) I had … Continue reading →
My 8GB Raspbery Pi 4 finally arrived, and I was looking forward to trying Fedora 32 and KDE/Plasma on it, and comparing it to the Odroid N2 from a few months back.. ( http://rglinuxtech.com/?p=2689 ) The Fedora 32 aarch64 install is supposed to support the Pi 4 now, but the resulting sdcard image failed to boot, with … Continue reading →
Kernel 5.8-rc2 has been released, and brief details are here: http://lkml.iu.edu/hypermail/linux/kernel/2006.2/08747.html There is no change to the problem with VMware runtime, and vmware-modconfig still segfaults, and the system still reboots immediately, when a VM open is attempted. To check that this was not due to a kernel misconfiguration, I installed the Fedora (Rawhide) version of Kernel 5.8-rc1 … Continue reading →
Kernel 5.8-rc1 has been released, and details are here: http://lkml.iu.edu/hypermail/linux/kernel/2006.1/07293.html NVIDIA 440.82 with the 5.7 and 5.8 patches seems to be OK, but I have encountered a serious problem with VMware, and have now confirmed the same fault on two different systems (both Fedora 32 x86_64) The VMware 15.5.6 vmmon/vmnet modules compile/run OK with the kernel 5.7 … Continue reading →
There is an updated patch for NVIDIA 440-59 with Kernel 5.3, which includes the Y2038 fixes, at https://gitlab.com/snippets/1937313 This compiles with GCC10 now, and I have briefly tested it on my Fedora32 system, with KDE/Plasma and VMware, and it seems to be working OK. Robert Gadsdon. February 25th 2020. … Continue reading →
Thanks to Isaak I. Aleksandrov, there is now an additional patch for NVIDIA driver 440.59 to accommodate the Y2038 changes in Kernel 5.6-rc3. This should be applied after the original 5.6-rc1 patch ( https://gitlab.com/snippets/1937313 ), and can be found here: https://gitlab.com/snippets/1944237 I have tested this, and it compiles OK with GCC9 (current distros) but fails with GCC10 … Continue reading →