Illumos 5027 - zfs large block support

5027 zfs large block support
Reviewed by: Alek Pinchuk <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Josef 'Jeff' Sipek <[email protected]>
Reviewed by: Richard Elling <[email protected]>
Reviewed by: Saso Kiselkov <[email protected]>
Reviewed by: Brian Behlendorf <[email protected]>
Approved by: Dan McDonald <[email protected]>

References:
  https://www.illumos.org/issues/5027
  https://github.com/illumos/illumos-gate/commit/b515258

Porting Notes:

* Included in this patch is a tiny ISP2() cleanup in zio_init() from
Illumos 5255.

* Unlike the upstream Illumos commit this patch does not impose an
arbitrary 128K block size limit on volumes.  Volumes, like filesystems,
are limited by the zfs_max_recordsize=1M module option.

* By default the maximum record size is limited to 1M by the module
option zfs_max_recordsize.  This value may be safely increased up to
16M which is the largest block size supported by the on-disk format.
At the moment, 1M blocks clearly offer a significant performance
improvement but the benefits of going beyond this for the majority
of workloads are less clear.

* The illumos version of this patch increased DMU_MAX_ACCESS to 32M.
This was determined not to be large enough when using 16M blocks
because the zfs_make_xattrdir() function will fail (EFBIG) when
assigning a TX.  This was immediately observed under Linux because
all newly created files must have a security xattr created and
that was failing.  Therefore, we've set DMU_MAX_ACCESS to 64M.

* On 32-bit platforms a hard limit of 1M is set for blocks due
to the limited virtual address space.  We should be able to relax
this one the ABD patches are merged.

Ported-by: Brian Behlendorf <[email protected]>
Closes #354

1 files changed, 22 insertions, 7 deletions

diff --git a/module/zfs/zio.c b/module/zfs/zio.c
index 9204df2b2..2b338f2a7 100644
--- a/module/zfs/zio.c
+++ b/module/zfs/zio.c
@@ -24,6 +24,7 @@
  * Copyright (c) 2011 Nexenta Systems, Inc. All rights reserved.
  */
 
+#include <sys/sysmacros.h>
 #include <sys/zfs_context.h>
 #include <sys/fm/fs/zfs.h>
 #include <sys/spa.h>
@@ -107,9 +108,8 @@ zio_init(void)
 
 	/*
 	 * For small buffers, we want a cache for each multiple of
-	 * SPA_MINBLOCKSIZE.  For medium-size buffers, we want a cache
-	 * for each quarter-power of 2.  For large buffers, we want
-	 * a cache for each multiple of PAGESIZE.
+	 * SPA_MINBLOCKSIZE.  For larger buffers, we want a cache
+	 * for each quarter-power of 2.
 	 */
 	for (c = 0; c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT; c++) {
 		size_t size = (c + 1) << SPA_MINBLOCKSHIFT;
@@ -117,7 +117,16 @@ zio_init(void)
 		size_t align = 0;
 		size_t cflags = (size > zio_buf_debug_limit) ? KMC_NODEBUG : 0;
 
-		while (p2 & (p2 - 1))
+#ifdef _ILP32
+		/*
+		 * Cache size limited to 1M on 32-bit platforms until ARC
+		 * buffers no longer require virtual address space.
+		 */
+		if (size > zfs_max_recordsize)
+			break;
+#endif
+
+		while (!ISP2(p2))
 			p2 &= p2 - 1;
 
 #ifndef _KERNEL
@@ -132,10 +141,8 @@ zio_init(void)
 #endif
 		if (size <= 4 * SPA_MINBLOCKSIZE) {
 			align = SPA_MINBLOCKSIZE;
-		} else if (IS_P2ALIGNED(size, PAGESIZE)) {
-			align = PAGESIZE;
 		} else if (IS_P2ALIGNED(size, p2 >> 2)) {
-			align = p2 >> 2;
+			align = MIN(p2 >> 2, PAGESIZE);
 		}
 
 		if (align != 0) {
@@ -174,6 +181,14 @@ zio_fini(void)
 	kmem_cache_t *last_data_cache = NULL;
 
 	for (c = 0; c < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT; c++) {
+#ifdef _ILP32
+		/*
+		 * Cache size limited to 1M on 32-bit platforms until ARC
+		 * buffers no longer require virtual address space.
+		 */
+		if (((c + 1) << SPA_MINBLOCKSHIFT) > zfs_max_recordsize)
+			break;
+#endif
 		if (zio_buf_cache[c] != last_cache) {
 			last_cache = zio_buf_cache[c];
 			kmem_cache_destroy(zio_buf_cache[c]);