Matti Aarnio writes:
> As Linus hasn't put it into i386 (nor Alpha), he has
> some good reason. Ask him for guidance.
That is probably because libgcc contained some stuff which was unsuitable
for the kernel on x86.
> Mainly I think we should NEVER use slow 64/* division routines
> in fast paths.
Ah, are you saying that the only thing that libgcc.a is used for is
supposidly slow division routines? Some architectures need it for much
much more than that.
> As an end-result you will get your system free of need for
> libgcc.a, and possible dangers lurking in it.
No. libgcc.a is dependent on the compiler. I do know that code generated
with one ARM gcc compiler does not work with the libgcc.a of a later
compiler, suggesting that there is a difference in the API between the
two versions of GCC. Hence, the only real way to get the correct behaviour
in this case is to use libgcc.a for ARM.
_____
|_____| ------------------------------------------------- ---+---+-
| | Russell King rmk@arm.linux.org.uk --- ---
| | | | http://www.arm.linux.org.uk/~rmk/aboutme.html / / |
| +-+-+ --- -+-
/ | THE developer of ARM Linux |+| /|\
/ | | | --- |
+-+-+ ------------------------------------------------- /\\\ |
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:19 EST