Klemens Nanni <k...@openbsd.org> wrote: > Are there more reasons or benefits to this approach than a) less intrusive > chmod() and b) more shared, overall less code? > > I obviously don't see the full picture (yet).
A normal bootloader only reads from disk. No write occur. The chmod hack is a write. In the libsa case, it narrowly modifies just the block that is required, because libsa's filesystem code is not a buffering filesystem implimentation. But these octeon / powerpc64 bootloaders are real kernels, running the real ffs. When the "allow a block write" operation is activated, the root filesystem is *REALLY MOUNTED*, which means all the ffs read, read-ahead, inode update, and other write operations come into play. There is no complete "only do what libsa does" scheme. So the worry is that doing mount rw, chmod, mount ro, can create (or make worse) existing filesystem damage, to the point where maybe you cannot boot at all anymore.
