s390: add support for virtually mapped kernel stacks

With virtually mapped kernel stacks the kernel stack overflow detection
is now fault based, every stack has a guard page in the vmalloc space.
The panic_stack is renamed to nodat_stack and is used for all function
that need to run without DAT, e.g. memcpy_real or do_start_kdump.

The main effect is a reduction in the kernel image size as with vmap
stacks the old style overflow checking that adds two instructions per
function is not needed anymore. Result from bloat-o-meter:

add/remove: 20/1 grow/shrink: 13/26854 up/down: 2198/-216240 (-214042)

In regard to performance the micro-benchmark for fork has a hit of a
few microseconds, allocating 4 pages in vmalloc space is more expensive
compare to an order-2 page allocation. But with real workload I could
not find a noticeable difference.

Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

1 files changed, 20 insertions, 5 deletions

diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c
index 7be06475809b..97b3ee53852b 100644
--- a/arch/s390/mm/maccess.c
+++ b/arch/s390/mm/maccess.c
@@ -89,10 +89,8 @@ static int __memcpy_real(void *dest, void *src, size_t count)
 	return rc;
 }
 
-/*
- * Copy memory in real mode (kernel to kernel)
- */
-int memcpy_real(void *dest, void *src, size_t count)
+static unsigned long _memcpy_real(unsigned long dest, unsigned long src,
+				  unsigned long count)
 {
 	int irqs_disabled, rc;
 	unsigned long flags;
@@ -103,7 +101,7 @@ int memcpy_real(void *dest, void *src, size_t count)
 	irqs_disabled = arch_irqs_disabled_flags(flags);
 	if (!irqs_disabled)
 		trace_hardirqs_off();
-	rc = __memcpy_real(dest, src, count);
+	rc = __memcpy_real((void *) dest, (void *) src, (size_t) count);
 	if (!irqs_disabled)
 		trace_hardirqs_on();
 	__arch_local_irq_ssm(flags);
@@ -111,6 +109,23 @@ int memcpy_real(void *dest, void *src, size_t count)
 }
 
 /*
+ * Copy memory in real mode (kernel to kernel)
+ */
+int memcpy_real(void *dest, void *src, size_t count)
+{
+	if (S390_lowcore.nodat_stack != 0)
+		return CALL_ON_STACK(_memcpy_real, S390_lowcore.nodat_stack,
+				     3, dest, src, count);
+	/*
+	 * This is a really early memcpy_real call, the stacks are
+	 * not set up yet. Just call _memcpy_real on the early boot
+	 * stack
+	 */
+	return _memcpy_real((unsigned long) dest,(unsigned long) src,
+			    (unsigned long) count);
+}
+
+/*
  * Copy memory in absolute mode (kernel to kernel)
  */
 void memcpy_absolute(void *dest, void *src, size_t count)