arch/tile: support 4KB page size as well as 64KB

The Tilera architecture traditionally supports 64KB page sizes
to improve TLB utilization and improve performance when the
hardware is being used primarily to run a single application.

For more generic server scenarios, it can be beneficial to run
with 4KB page sizes, so this commit allows that to be specified
(by modifying the arch/tile/include/hv/pagesize.h header).

As part of this change, we also re-worked the PTE management
slightly so that PTE writes all go through a __set_pte() function
where we can do some additional validation.  The set_pte_order()
function was eliminated since the "order" argument wasn't being used.

One bug uncovered was in the PCI DMA code, which wasn't properly
flushing the specified range.  This was benign with 64KB pages,
but with 4KB pages we were getting some larger flushes wrong.

The per-cpu memory reservation code also needed updating to
conform with the newer percpu stuff; before it always chose 64KB,
and that was always correct, but with 4KB granularity we now have
to pay closer attention and reserve the amount of memory that will
be requested when the percpu code starts allocating.

Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>

1 files changed, 13 insertions, 3 deletions

diff --git a/arch/tile/kernel/intvec_32.S b/arch/tile/kernel/intvec_32.S
index eabf1ef02cb2..fffcfa6b3a62 100644
--- a/arch/tile/kernel/intvec_32.S
+++ b/arch/tile/kernel/intvec_32.S
@@ -1556,7 +1556,10 @@ STD_ENTRY(_sys_clone)
 	.align 64
 	/* Align much later jump on the start of a cache line. */
 #if !ATOMIC_LOCKS_FOUND_VIA_TABLE()
-	nop; nop
+	nop
+#if PAGE_SIZE >= 0x10000
+	nop
+#endif
 #endif
 ENTRY(sys_cmpxchg)
 
@@ -1587,6 +1590,10 @@ ENTRY(sys_cmpxchg)
 	 * NOTE: this must match __atomic_hashed_lock() in lib/atomic_32.c.
 	 */
 
+#if (PAGE_OFFSET & 0xffff) != 0
+# error Code here assumes PAGE_OFFSET can be loaded with just hi16()
+#endif
+
 #if ATOMIC_LOCKS_FOUND_VIA_TABLE()
 	{
 	 /* Check for unaligned input. */
@@ -1679,11 +1686,14 @@ ENTRY(sys_cmpxchg)
 	 lw	r26, r0
 	}
 	{
-	 /* atomic_locks is page aligned so this suffices to get its addr. */
-	 auli	r21, zero, hi16(atomic_locks)
+	 auli	r21, zero, ha16(atomic_locks)
 
 	 bbns   r23, .Lcmpxchg_badaddr
 	}
+#if PAGE_SIZE < 0x10000
+	/* atomic_locks is page-aligned so for big pages we don't need this. */
+	addli   r21, r21, lo16(atomic_locks)
+#endif
 	{
 	 /*
 	  * Insert the hash bits into the page-aligned pointer.