Damn.
/me writes a patch
/me tests and finds an obvious typo
/me fixes and diffs fixed (and tested) version
/me sends the original one.
My apologies. Correct patch (taken between clean tree and
result of make distclean on the tree that gave a kernel
that passes all tests) follows. Please, apply
On Tue, 15 May 2001, Alexander Viro wrote:
>
>
> On Tue, 15 May 2001, Alexander Viro wrote:
>
> > ramdisks, etc.). Besides, it allows to start turning functions called
> > from device_init() into initcalls, thus getting rid of ifdef dungpiles
> > in them.
>
> ... and here's the next part. T
On Tue, 15 May 2001, Alexander Viro wrote:
> ramdisks, etc.). Besides, it allows to start turning functions called
> from device_init() into initcalls, thus getting rid of ifdef dungpiles
> in them.
... and here's the next part. Takes parport_init() out of device_init().
Since we have no initc
Patch below turns device_init() into initcall. Current tree
calls it from fs/partitions/check.c::partitions_setup() - definitely
odd place for that stuff.
Another thing done by partition_setup() is {init,}rd_load(). I.e.
setting the contents of /dev/ram0 from initrd or floppies. T
Patch below turns device_init() into initcall. Current tree
calls it from fs/partitions/check.c::partitions_setup() - definitely
odd place for that stuff.
Another thing done by partition_setup() is {init,}rd_load(). I.e.
setting the contents of /dev/ram0 from initrd or floppies. T
5 matches
Mail list logo
