I have built rEFInd for ia32 from src using gnu_efi and when it is installed to the EFI partition it hangs with a black screen whenever a driver (any driver at all) is present in the drivers_ia32 directory.
Booting from a removeable medium (USB) works with the same build though.
I have also tested the 0.10.1 build which works as expected.
I have also tested with my build but drivers from 0.10.1 and that also works as expected so obviously it is something with the drivers.
I can do some debugging of this if given the pointers to what to look for, but for now I can use the pre-built drivers (as I only want to modify the binary at this point).
BR Gustaf
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've seen something similar on a 32-bit Mac Mini I own, but not under a 32-bit VirtualBox system. I banged my head against the wall for many hours trying to debug it, with little progress. It looked as if a function was being called in the middle, which of course meant things went downhill fast. There's a big fat comment about that in fsw_efi_read_block() in fsw_efi.c. If you can identify the cause, the help would be most welcome! If not, I'm sure I'll try again eventually. In the meantime, I recommend using the binaries built via TianoCore, which seem to work fine. (Mac users normally install my binary build, which provides TianoCore-built binaries, and since I only ran into this on a 32-bit Mac, using the TianoCore build seemed like an acceptable workaround.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I have built rEFInd for ia32 from src using gnu_efi and when it is installed to the EFI partition it hangs with a black screen whenever a driver (any driver at all) is present in the drivers_ia32 directory.
Booting from a removeable medium (USB) works with the same build though.
I have also tested the 0.10.1 build which works as expected.
I have also tested with my build but drivers from 0.10.1 and that also works as expected so obviously it is something with the drivers.
I can do some debugging of this if given the pointers to what to look for, but for now I can use the pre-built drivers (as I only want to modify the binary at this point).
BR Gustaf
I've seen something similar on a 32-bit Mac Mini I own, but not under a 32-bit VirtualBox system. I banged my head against the wall for many hours trying to debug it, with little progress. It looked as if a function was being called in the middle, which of course meant things went downhill fast. There's a big fat comment about that in fsw_efi_read_block() in fsw_efi.c. If you can identify the cause, the help would be most welcome! If not, I'm sure I'll try again eventually. In the meantime, I recommend using the binaries built via TianoCore, which seem to work fine. (Mac users normally install my binary build, which provides TianoCore-built binaries, and since I only ran into this on a 32-bit Mac, using the TianoCore build seemed like an acceptable workaround.)