I decided to complete the unification of piggyback.
This was inspired by comments from Julian Calaby.
The patchset documents my findings while reverse engineering
piggyback. And some limited refactoring also increased
readability a lot.
piggyback now takes an argument that tell if this is for
sparc or sparc64 and uses this info to adjust behaviour
as sparc and sparc64 differs in a few places.
With these additions we can now use the same piggyback for
both sparc and sparc64.
My testing turned out that piggyback_64 was broken on my x86 box,
because it was hardcoded to assume a big-endian host.
I also found a few code-spots that assuemd the host was big-endian.
All cases are fixed to be endian neutral.
Julian Calaby acknowledged that piggyback_64 had endian troubles.
This is boot tested on my sparc box.
I have only build tested the sparc64 bits!
Since v1 of the patchset I have included 4 additional patches,
the original 5 patches has not seen any updates.
And the patches are properly numbered this time.
Sam
Sam Ravnborg (9):
sparc: refactor piggy_32.c
sparc: use trapbase in setup_arch
sparc: use _start for the start entry (like 64 bit does)
sparc: additional comments to piggyback_32.c
sparc: remove obsolete ELF support in piggyback_32.c
sparc: add $BITS to piggyback arguments
sparc: use piggyback_32 for sparc64
sparc: rename piggyback_32 to piggyback
sparc: unify strip command in boot/Makefile
arch/sparc/boot/Makefile | 28 ++---
arch/sparc/boot/piggyback.c | 266 ++++++++++++++++++++++++++++++++++++++++
arch/sparc/boot/piggyback_32.c | 137 ---------------------
arch/sparc/boot/piggyback_64.c | 110 -----------------
arch/sparc/kernel/head_32.S | 3 +-
arch/sparc/kernel/setup_32.c | 3 +-
arch/sparc/mm/sun4c.c | 8 +-
7 files changed, 280 insertions(+), 275 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
