From: Zwane Mwaikambo <zwane@arm.linux.org.uk>

It looks like when we do the WP test and trigger a (write) protection
fault, the 4G/4G page fault handling path doesn't expect this kind of fault
and instead results in recursive fault handling (or so it appears).



 arch/i386/mm/fault.c |    6 +++++-
 1 files changed, 5 insertions(+), 1 deletion(-)

diff -puN arch/i386/mm/fault.c~4g4g-wp-test-fix arch/i386/mm/fault.c
--- 25/arch/i386/mm/fault.c~4g4g-wp-test-fix	2003-10-16 00:31:51.000000000 -0700
+++ 25-akpm/arch/i386/mm/fault.c	2003-10-16 00:31:51.000000000 -0700
@@ -260,8 +260,12 @@ asmlinkage void do_page_fault(struct pt_
 	/*
 	 * On 4/4 all kernels faults are either bugs, vmalloc or prefetch
 	 */
-	if (unlikely((regs->xcs & 3) == 0))
+	if (unlikely((regs->xcs & 3) == 0)) {
+		if (error_code & 3)
+			goto bad_area_nosemaphore;
+
  		goto vmalloc_fault;
+	}
 #else
 	if (unlikely(address >= TASK_SIZE)) { 
 		if (!(error_code & 5))

_