summaryrefslogtreecommitdiffstats
path: root/arch/m68k
diff options
context:
space:
mode:
authorGreg Ungerer <gerg@uclinux.org>2014-08-28 06:39:45 +0200
committerGreg Ungerer <gerg@uclinux.org>2014-09-29 01:18:36 +0200
commitecb6bdcde2474c8c954179b74fd3df3ad84fba75 (patch)
treeebd3ec082d8d319b95c2eb2a230b52a52cc08e4c /arch/m68k
parentm68k: fix crufty 68000 and 68360 intro comments (diff)
downloadlinux-ecb6bdcde2474c8c954179b74fd3df3ad84fba75.tar.xz
linux-ecb6bdcde2474c8c954179b74fd3df3ad84fba75.zip
m68knommu: fix size of address field for 5272 interrupt controller
Picked up by the 0-day buidler: All warnings: >> arch/m68k/coldfire/intc-5272.c:46:20: warning: large integer implicitly truncated to unsigned type [-Woverflow] /*MCF_IRQ_EINT1*/ { .icr = MCFSIM_ICR1, .index = 28, .ack = 1, }, ... The problem stems from the changes to make all ColdFire register addresses absolute, in commit d72a5abb ("make remaining ColdFire 5272 register definitions absolute"). That change did not take into account that the addresses were stored as offsets in the irqmap of the intc-5272.c code. Make the field that now stores register addresses big enough to hold addresses. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
Diffstat (limited to 'arch/m68k')
-rw-r--r--arch/m68k/coldfire/intc-5272.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/arch/m68k/coldfire/intc-5272.c b/arch/m68k/coldfire/intc-5272.c
index d7b695629a7e..d1e2fbad327c 100644
--- a/arch/m68k/coldfire/intc-5272.c
+++ b/arch/m68k/coldfire/intc-5272.c
@@ -36,7 +36,7 @@
* they also need acknowledging via acknowledge bits.
*/
struct irqmap {
- unsigned char icr;
+ unsigned int icr;
unsigned char index;
unsigned char ack;
};