Move the inline limit default variable to a per-arch place. For example,

the amd64 implementation of the pcpu macros is even more verbose than on
i386 and that causes gcc to way overestimate the complexity of this
2-instruction macro.  The other platforms can probably lower their
default values.

sys/conf/kern.mk

@@ -24,6 +24,7 @@ CWARNFLAGS?=	-Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes \
 #
 .if ${MACHINE_ARCH} == "i386"
 CFLAGS+=	-mno-align-long-strings -mpreferred-stack-boundary=2
+INLINE_LIMIT?=	15000
 .endif
 
 #
@@ -33,6 +34,7 @@ CFLAGS+=	-mno-align-long-strings -mpreferred-stack-boundary=2
 #
 .if ${MACHINE_ARCH} == "alpha"
 CFLAGS+=	-mno-fp-regs -ffixed-8 -Wa,-mev6
+INLINE_LIMIT?=	15000
 .endif
 
 #
@@ -41,6 +43,7 @@ CFLAGS+=	-mno-fp-regs -ffixed-8 -Wa,-mev6
 #
 .if ${MACHINE_ARCH} == "ia64"
 CFLAGS+=	-ffixed-r13 -mfixed-range=f32-f127 -mno-sdata
+INLINE_LIMIT?=	15000
 .endif
 
 #
@@ -50,6 +53,7 @@ CFLAGS+=	-ffixed-r13 -mfixed-range=f32-f127 -mno-sdata
 #
 .if ${MACHINE_ARCH} == "sparc64"
 CFLAGS+=	-mcmodel=medlow -msoft-float
+INLINE_LIMIT?=	15000
 .endif
 
 #
@@ -60,6 +64,7 @@ CFLAGS+=	-mcmodel=medlow -msoft-float
 CFLAGS+=	-mcmodel=kernel -mno-red-zone \
 		-mfpmath=387 -mno-sse -mno-sse2 -mno-mmx -mno-3dnow \
 		-msoft-float -fno-asynchronous-unwind-tables
+INLINE_LIMIT?=	20000
 .endif
 
 #

sys/conf/kern.pre.mk

@@ -56,7 +56,6 @@ PROF=	-pg
 .endif
 DEFINED_PROF=	${PROF}
 WERROR?=	-Werror
-INLINE_LIMIT?=	15000
 CFLAGS+=	-finline-limit=${INLINE_LIMIT} -fno-strict-aliasing
 
 

sys/conf/kmod.mk

@@ -99,7 +99,6 @@ CFLAGS+=	-I${DESTDIR}/usr/include
 CFLAGS+=	-I@/../include -I${DESTDIR}/usr/include
 .endif # @
 
-INLINE_LIMIT?=	15000
 CFLAGS+=	-finline-limit=${INLINE_LIMIT}
 
 # Disallow common variables, and if we end up with commons from