Commits · 5f19f14fed7786652b9617c633db101d26a42251 · Kirill Smelkov / linux

22 May, 2012 1 commit

sh: intc: Kill off special reservation interface. · 5f19f14f

Paul Mundt authored May 22, 2012

At present reserving the IRLs in the IRQ bitmap in addition to the
dropping of the legacy IRQ pre-allocation prevent IRL IRQs from being
allocated for the x3proto board.

The only reason to permit reservations was to lock down possible hardware
vectors prior to dynamic IRQ scanning, but this doesn't matter much given
that the hardware controller configuration is sorted before we get around
to doing any dynamic IRQ allocation anyways. Beyond that, all of the
tables are __init annotated, so quite a bit more work would need to be
done to support reconfiguring things like IRL controllers on the fly,
much more than would ever make it worth the hassle.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

5f19f14f

21 May, 2012 2 commits

sh: Enable PIO API for hp6xx and se770x. · c30b9c43

Paul Mundt authored May 21, 2012

Both of these boards are still using the PIO API sans PCI, and will need
to be carefully converted. As conversion is out of scope for this merge
window, simply make sure NO_IOPORT doesn't get accidentally set for
these, joining a long list of other crap platforms.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

c30b9c43

sh: Kill off machvec IRQ hinting. · 63dc02bd

Paul Mundt authored May 21, 2012

Everything is using sparseirq these days, so we have no need to
arbitrarily size nr_irqs ahead of time. The legacy IRQ pre-allocation
likewise has no meaning for us, so that's killed off too. We now depend
on nr_irqs expansion by the generic hardirq layer instead.

It's also worth noting that the majority of boards had completely bogus
values for their nr_irqs relative to their CPU and configurations, so
this ends up correcting behaviour for quite a few platforms.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

63dc02bd

19 May, 2012 4 commits

sh: dma: More legacy cpu dma chainsawing. · 7f47c718

Paul Mundt authored May 19, 2012

Attempt to "tidy" up some of the multi IRQ handling and base + IRQ
management. This should keep it limping along without too much hassle,
and no new parts should ever be enabling or using this API anyways.

It doesn't get any closer to lipstick on a pig as this.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

7f47c718

sh: Kill off MAX_DMA_ADDRESS leftovers. · e24cca19

Paul Mundt authored May 19, 2012

We don't support the ISA DMA API, so this is only ever misused. The
dma-sh case inadvertently broke the dreamcast case by testing the wrong
variable for the total number of channels, so this fixes that up too.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

e24cca19

sh: Tidy up some of the cpu legacy dma header mess. · 2ff9f317

Paul Mundt authored May 19, 2012

This has turned in to quite a mess, and with CPUs that care using
dmaengine now it's about time to start cleaning up after the legacy DMA
code. For starters, kill off the stubs for the CPUs that don't do
anything, as well as all of the unused definitions. This leaves us with a
set of IRQs and base addresses we can deal with later.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

2ff9f317

sh: Move sh4a dma header from cpu-sh4 to cpu-sh4a. · 0c601231

Paul Mundt authored May 19, 2012

cpu-sh4a headers take priority over cpu-sh4 ones by virtue of the build
system, there's no need to try and mingle sh4a stuff in cpu-sh4.
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

0c601231

18 May, 2012 33 commits

sh64: Fix up vmalloc fault range check. · d8fd35fc