Account for the fact that contigmalloc(9) can return a NULL pointer.

Fix the flags argument: M_WAITOK is not a valid flag. Its presence
leaves the indication that contigmalloc(9) will not return a NULL
pointer.

The use of contigmalloc(9) in this place is probably not a good idea
given the constraints. It's probably better to lift the constraints
and instead add a permanent mapping to the ITR. It's possible that
the first 256MB of memory is exhausted when we get here.

This fixes a kernel panic on a 16GB rx3600.

1 files changed, 6 insertions, 3 deletions

diff --git a/sys/ia64/ia64/mca.c b/sys/ia64/ia64/mca.c
index dbec087..bba0bfb 100644
--- a/sys/ia64/ia64/mca.c
+++ b/sys/ia64/ia64/mca.c
@@ -92,6 +92,9 @@ ia64_mca_save_state(int type)
 	if (mca_info_size[type] == -1)
 		return;
 
+	if (mca_info_block == 0)
+		return;
+
 	while (1) {
 		mtx_lock_spin(&mca_info_block_lock);
 
@@ -198,9 +201,9 @@ ia64_mca_init(void)
 	}
 	max_size = round_page(max_size);
 
-	if (max_size) {
-		p = contigmalloc(max_size, M_TEMP, M_WAITOK, 0ul,
-		    256*1024*1024 - 1, PAGE_SIZE, 256*1024*1024);
+	p = (max_size) ? contigmalloc(max_size, M_TEMP, 0, 0ul,
+	    256*1024*1024 - 1, PAGE_SIZE, 256*1024*1024) : NULL;
+	if (p != NULL) {
 		mca_info_block = IA64_PHYS_TO_RR7(ia64_tpa((u_int64_t)p));
 
 		if (bootverbose)