sh: default to sparseirq.

As SH has a very sparse IRQ map by default, all new CPUs and boards
benefit from using sparseirq by default. Despite this, there are still a
few stragglers (mostly due to using a fixed IRQ range for their FPGA
IRQ mappings), and these still need to be converted over one by one. As
these are now in the minority, and we do not want to encourage this sort
of brain-damage in newer board ports, we force sparseirq on.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 2121fbb2ff4c313c5f0891b783ea4cbf3a571fe0..ae6c73689036ced70c2fa8740a9ced72426cb84d 100644 (file)
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -726,8 +726,9 @@ config GUSA_RB
          disabling interrupts around the atomic sequence.
 
 config SPARSE_IRQ
-       bool "Support sparse irq numbering"
-       depends on EXPERIMENTAL
+       def_bool y
+       depends on SUPERH32 && !SH_DREAMCAST && !SH_HIGHLANDER && \
+                  !SH_RTS7751R2D && !HD64461 && !SH_7724_SOLUTION_ENGINE
        help
          This enables support for sparse irqs. This is useful in general
          as most CPUs have a fairly sparse array of IRQ vectors, which