[PATCH] powerpc/mm: Convert slb presence warning check to WARN_ON_ONCE

[Aneesh Kumar K.V]

We are hitting false positive in some case. Till we root cause
this, convert WARN_ON to WARN_ON_WONCE.

A sample stack dump looks like
NIP [c00000000007ac40] assert_slb_presence+0x90/0xa0
LR [c00000000007b270] slb_flush_and_restore_bolted+0x90/0xc0
Call Trace:
 arch_send_call_function_ipi_mask+0xcc/0x110 (unreliable)
 0xc000000f9f38f560
 slice_flush_segments+0x58/0xb0
 on_each_cpu+0x74/0xf0
 slice_get_unmapped_area+0x6d4/0x9e0
 hugetlb_get_unmapped_area+0x124/0x150
 get_unmapped_area+0xf0/0x1a0
 do_mmap+0x1a4/0x6b0
 vm_mmap_pgoff+0xbc/0x150
 ksys_mmap_pgoff+0x260/0x2f0
 sys_mmap+0x104/0x130
 system_call+0x5c/0x70

We are checking whether we were able to successfully insert
kernel stack SLB entries. If that is not case we will crash next.
So we are not losing much debug data.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar at linux.ibm.com>
---
 arch/powerpc/mm/slb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/mm/slb.c b/arch/powerpc/mm/slb.c
index bc3914d54e26..dca0cbd71b60 100644
--- a/arch/powerpc/mm/slb.c
+++ b/arch/powerpc/mm/slb.c
@@ -71,7 +71,7 @@ static void assert_slb_presence(bool present, unsigned long ea)
 
 	asm volatile(__PPC_SLBFEE_DOT(%0, %1) : "=r"(tmp) : "r"(ea) : "cr0");
 
-	WARN_ON(present == (tmp == 0));
+	WARN_ON_ONCE(present == (tmp == 0));
 #endif
 }
 
-- 
2.20.1