I copied over my working .config from 2.1.111 to 2.1.113 and
ran through the configuration, just to double check everything. I then
ran "make dep ; make clean ; make bzImage". It bombed in the i386/mm
area with the following errors:
] gcc -D__KERNEL__ -I/u/src/linux-2.1.113/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -pipe -fno-strength-reduce -m486 -malign-loops=2 -malign-jumps=2 -malign-functions=2 -DCPU=686 -c -o fault.o fault.c
] fault.c: In function `do_page_fault':
] fault.c:105: `local_bh_count' undeclared (first use this function)
] fault.c:105: (Each undeclared identifier is reported only once
] fault.c:105: for each function it appears in.)
] make[2]: *** [fault.o] Error 1
] make[2]: Leaving directory `/u/src/linux-2.1.113/arch/i386/mm'
] make[1]: *** [first_rule] Error 2
] make[1]: Leaving directory `/u/src/linux-2.1.113/arch/i386/mm'
] make: *** [linuxsubdirs] Error 2
Line 105 in arch/i386/mm/fault.c is here:
104] mm = tsk->mm;
105] if (in_interrupt())
106] die("page fault from irq handler",regs,error_code);
This appears to originate in include/asm-i386/hardirq.h with the
macro in_interrupt():
#define in_interrupt() ({ int __cpu = smp_processor_id(); \
(local_irq_count[__cpu] + local_bh_count[__cpu] != 0); })
It looks like local_bh_count is missing.
Here is the same macro from 2.1.111:
#define in_interrupt() (local_irq_count[smp_processor_id()] + local_bh_count[smp_processor_id()] != 0)
Ok... The local_bh_count array is still there, but was indexed
directly by the smp_processor_id function. Looks like a fix for some
sort of timing window between the two calls to smp_processor_id().
Where did local_bh_count disappear to?
Build information:
x86 Pentium Pro, Uniprocessor. RedHat 4.1 installation.
Thoughts on what to do about this animal?
Mike
-- Michael H. Warfield | (770) 985-6132 | mhw@WittsEnd.com (The Mad Wizard) | (770) 925-8248 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471 | possible worlds. A pessimist is sure of it!- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html