]> git.saurik.com Git - apple/xnu.git/blobdiff - bsd/kern/ubc_subr.c
xnu-4903.221.2.tar.gz
[apple/xnu.git] / bsd / kern / ubc_subr.c
index 22dc9e5804ab8a174f8133099fc9d38a71edf681..6f09debf794c7940d9068cc7e85761ebfb1d6724 100644 (file)
@@ -1,31 +1,29 @@
 /*
- * Copyright (c) 1999-2004 Apple Computer, Inc. All rights reserved.
+ * Copyright (c) 1999-2014 Apple Inc. All rights reserved.
  *
- * @APPLE_LICENSE_OSREFERENCE_HEADER_START@
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
  * 
- * This file contains Original Code and/or Modifications of Original Code 
- * as defined in and that are subject to the Apple Public Source License 
- * Version 2.0 (the 'License'). You may not use this file except in 
- * compliance with the License.  The rights granted to you under the 
- * License may not be used to create, or enable the creation or 
- * redistribution of, unlawful or unlicensed copies of an Apple operating 
- * system, or to circumvent, violate, or enable the circumvention or 
- * violation of, any terms of an Apple operating system software license 
- * agreement.
- *
- * Please obtain a copy of the License at 
- * http://www.opensource.apple.com/apsl/ and read it before using this 
- * file.
- *
- * The Original Code and all software distributed under the License are 
- * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER 
- * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, 
- * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, 
- * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT. 
- * Please see the License for the specific language governing rights and 
+ * This file contains Original Code and/or Modifications of Original Code
+ * as defined in and that are subject to the Apple Public Source License
+ * Version 2.0 (the 'License'). You may not use this file except in
+ * compliance with the License. The rights granted to you under the License
+ * may not be used to create, or enable the creation or redistribution of,
+ * unlawful or unlicensed copies of an Apple operating system, or to
+ * circumvent, violate, or enable the circumvention or violation of, any
+ * terms of an Apple operating system software license agreement.
+ * 
+ * Please obtain a copy of the License at
+ * http://www.opensource.apple.com/apsl/ and read it before using this file.
+ * 
+ * The Original Code and all software distributed under the License are
+ * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
+ * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
+ * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
+ * Please see the License for the specific language governing rights and
  * limitations under the License.
- *
- * @APPLE_LICENSE_OSREFERENCE_HEADER_END@
+ * 
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  */
 /* 
  *     File:   ubc_subr.c
@@ -38,9 +36,6 @@
  *
  */ 
 
-#undef DIAGNOSTIC
-#define DIAGNOSTIC 1
-
 #include <sys/types.h>
 #include <sys/param.h>
 #include <sys/systm.h>
 #include <sys/kauth.h>
 #include <sys/buf.h>
 #include <sys/user.h>
+#include <sys/codesign.h>
+#include <sys/codedir_internal.h>
+#include <sys/fsevents.h>
+#include <sys/fcntl.h>
 
 #include <mach/mach_types.h>
 #include <mach/memory_object_types.h>
 #include <mach/memory_object_control.h>
 #include <mach/vm_map.h>
+#include <mach/mach_vm.h>
 #include <mach/upl.h>
 
 #include <kern/kern_types.h>
+#include <kern/kalloc.h>
 #include <kern/zalloc.h>
 #include <kern/thread.h>
+#include <vm/pmap.h>
 #include <vm/vm_kern.h>
 #include <vm/vm_protos.h> /* last */
 
+#include <libkern/crypto/sha1.h>
+#include <libkern/crypto/sha2.h>
+#include <libkern/libkern.h>
+
+#include <security/mac_framework.h>
+#include <stdbool.h>
+
+/* XXX These should be in a BSD accessible Mach header, but aren't. */
+extern kern_return_t memory_object_pages_resident(memory_object_control_t,
+                                                       boolean_t *);
+extern kern_return_t   memory_object_signed(memory_object_control_t control,
+                                            boolean_t is_signed);
+extern boolean_t       memory_object_is_signed(memory_object_control_t);
+
+/* XXX Same for those. */
+
+extern void Debugger(const char *message);
+
+
+/* XXX no one uses this interface! */
+kern_return_t ubc_page_op_with_control(
+       memory_object_control_t  control,
+       off_t                    f_offset,
+       int                      ops,
+       ppnum_t                  *phys_entryp,
+       int                      *flagsp);
+
+
 #if DIAGNOSTIC
 #if defined(assert)
-#undef assert()
+#undef assert
 #endif
 #define assert(cond)    \
-    ((void) ((cond) ? 0 : panic("%s:%d (%s)", __FILE__, __LINE__, # cond)))
+    ((void) ((cond) ? 0 : panic("Assert failed: %s", # cond)))
 #else
 #include <kern/assert.h>
 #endif /* DIAGNOSTIC */
 
-int ubc_info_init_internal(struct vnode *vp, int withfsize, off_t filesize);
-int ubc_umcallback(vnode_t, void *);
-int ubc_isinuse_locked(vnode_t, int, int);
-int ubc_msync_internal(vnode_t, off_t, off_t, off_t *, int, int *);
+static int ubc_info_init_internal(struct vnode *vp, int withfsize, off_t filesize);
+static int ubc_umcallback(vnode_t, void *);
+static int ubc_msync_internal(vnode_t, off_t, off_t, off_t *, int, int *);
+static void ubc_cs_free(struct ubc_info *uip);
+
+static boolean_t ubc_cs_supports_multilevel_hash(struct cs_blob *blob);
+static kern_return_t ubc_cs_convert_to_multilevel_hash(struct cs_blob *blob);
 
 struct zone    *ubc_info_zone;
+static uint32_t        cs_blob_generation_count = 1;
+
+/*
+ * CODESIGNING
+ * Routines to navigate code signing data structures in the kernel...
+ */
+
+extern int cs_debug;
+
+#define        PAGE_SHIFT_4K           (12)
+
+static boolean_t
+cs_valid_range(
+       const void *start,
+       const void *end,
+       const void *lower_bound,
+       const void *upper_bound)
+{
+       if (upper_bound < lower_bound ||
+           end < start) {
+               return FALSE;
+       }
+
+       if (start < lower_bound ||
+           end > upper_bound) {
+               return FALSE;
+       }
+
+       return TRUE;
+}
+
+typedef void (*cs_md_init)(void *ctx);
+typedef void (*cs_md_update)(void *ctx, const void *data, size_t size);
+typedef void (*cs_md_final)(void *hash, void *ctx);
+
+struct cs_hash {
+    uint8_t            cs_type;        /* type code as per code signing */
+    size_t             cs_size;        /* size of effective hash (may be truncated) */
+    size_t             cs_digest_size; /* size of native hash */
+    cs_md_init         cs_init;
+    cs_md_update       cs_update;
+    cs_md_final                cs_final;
+};
+
+uint8_t cs_hash_type(
+    struct cs_hash const * const cs_hash)
+{
+    return cs_hash->cs_type;
+}
+
+static const struct cs_hash cs_hash_sha1 = {
+    .cs_type = CS_HASHTYPE_SHA1,
+    .cs_size = CS_SHA1_LEN,
+    .cs_digest_size = SHA_DIGEST_LENGTH,
+    .cs_init = (cs_md_init)SHA1Init,
+    .cs_update = (cs_md_update)SHA1Update,
+    .cs_final = (cs_md_final)SHA1Final,
+};
+#if CRYPTO_SHA2
+static const struct cs_hash cs_hash_sha256 = {
+    .cs_type = CS_HASHTYPE_SHA256,
+    .cs_size = SHA256_DIGEST_LENGTH,
+    .cs_digest_size = SHA256_DIGEST_LENGTH,
+    .cs_init = (cs_md_init)SHA256_Init,
+    .cs_update = (cs_md_update)SHA256_Update,
+    .cs_final = (cs_md_final)SHA256_Final,
+};
+static const struct cs_hash cs_hash_sha256_truncate = {
+    .cs_type = CS_HASHTYPE_SHA256_TRUNCATED,
+    .cs_size = CS_SHA256_TRUNCATED_LEN,
+    .cs_digest_size = SHA256_DIGEST_LENGTH,
+    .cs_init = (cs_md_init)SHA256_Init,
+    .cs_update = (cs_md_update)SHA256_Update,
+    .cs_final = (cs_md_final)SHA256_Final,
+};
+static const struct cs_hash cs_hash_sha384 = {
+    .cs_type = CS_HASHTYPE_SHA384,
+    .cs_size = SHA384_DIGEST_LENGTH,
+    .cs_digest_size = SHA384_DIGEST_LENGTH,
+    .cs_init = (cs_md_init)SHA384_Init,
+    .cs_update = (cs_md_update)SHA384_Update,
+    .cs_final = (cs_md_final)SHA384_Final,
+};
+#endif
+
+static struct cs_hash const *
+cs_find_md(uint8_t type)
+{
+       if (type == CS_HASHTYPE_SHA1) {
+               return &cs_hash_sha1;
+#if CRYPTO_SHA2
+       } else if (type == CS_HASHTYPE_SHA256) {
+               return &cs_hash_sha256;
+       } else if (type == CS_HASHTYPE_SHA256_TRUNCATED) {
+               return &cs_hash_sha256_truncate;
+       } else if (type == CS_HASHTYPE_SHA384) {
+               return &cs_hash_sha384;
+#endif
+       }
+       return NULL;
+}
+
+union cs_hash_union {
+       SHA1_CTX                sha1ctxt;
+       SHA256_CTX              sha256ctx;
+       SHA384_CTX              sha384ctx;
+};
+
+
+/*
+ * Choose among different hash algorithms.
+ * Higher is better, 0 => don't use at all.
+ */
+static const uint32_t hashPriorities[] = {
+       CS_HASHTYPE_SHA1,
+       CS_HASHTYPE_SHA256_TRUNCATED,
+       CS_HASHTYPE_SHA256,
+       CS_HASHTYPE_SHA384,
+};
+
+static unsigned int
+hash_rank(const CS_CodeDirectory *cd)
+{
+       uint32_t type = cd->hashType;
+       unsigned int n;
+
+       for (n = 0; n < sizeof(hashPriorities) / sizeof(hashPriorities[0]); ++n)
+               if (hashPriorities[n] == type)
+                       return n + 1;
+       return 0;       /* not supported */
+}
+
+
+/*
+ * Locating a page hash
+ */
+static const unsigned char *
+hashes(
+       const CS_CodeDirectory *cd,
+       uint32_t page,
+       size_t hash_len,
+       const char *lower_bound,
+       const char *upper_bound)
+{
+       const unsigned char *base, *top, *hash;
+       uint32_t nCodeSlots = ntohl(cd->nCodeSlots);
+
+       assert(cs_valid_range(cd, cd + 1, lower_bound, upper_bound));
+
+       if((ntohl(cd->version) >= CS_SUPPORTSSCATTER) && (ntohl(cd->scatterOffset))) {
+               /* Get first scatter struct */
+               const SC_Scatter *scatter = (const SC_Scatter*)
+                       ((const char*)cd + ntohl(cd->scatterOffset));
+               uint32_t hashindex=0, scount, sbase=0;
+               /* iterate all scatter structs */
+               do {
+                       if((const char*)scatter > (const char*)cd + ntohl(cd->length)) {
+                               if(cs_debug) {
+                                       printf("CODE SIGNING: Scatter extends past Code Directory\n");
+                               }
+                               return NULL;
+                       }
+                       
+                       scount = ntohl(scatter->count);
+                       uint32_t new_base = ntohl(scatter->base);
+
+                       /* last scatter? */
+                       if (scount == 0) {
+                               return NULL;
+                       }
+                       
+                       if((hashindex > 0) && (new_base <= sbase)) {
+                               if(cs_debug) {
+                                       printf("CODE SIGNING: unordered Scatter, prev base %d, cur base %d\n",
+                                       sbase, new_base);
+                               }
+                               return NULL;    /* unordered scatter array */
+                       }
+                       sbase = new_base;
+
+                       /* this scatter beyond page we're looking for? */
+                       if (sbase > page) {
+                               return NULL;
+                       }
+                       
+                       if (sbase+scount >= page) {
+                               /* Found the scatter struct that is 
+                                * referencing our page */
+
+                               /* base = address of first hash covered by scatter */
+                               base = (const unsigned char *)cd + ntohl(cd->hashOffset) + 
+                                       hashindex * hash_len;
+                               /* top = address of first hash after this scatter */
+                               top = base + scount * hash_len;
+                               if (!cs_valid_range(base, top, lower_bound, 
+                                                   upper_bound) ||
+                                   hashindex > nCodeSlots) {
+                                       return NULL;
+                               }
+                               
+                               break;
+                       }
+                       
+                       /* this scatter struct is before the page we're looking 
+                        * for. Iterate. */
+                       hashindex+=scount;
+                       scatter++;
+               } while(1);
+               
+               hash = base + (page - sbase) * hash_len;
+       } else {
+               base = (const unsigned char *)cd + ntohl(cd->hashOffset);
+               top = base + nCodeSlots * hash_len;
+               if (!cs_valid_range(base, top, lower_bound, upper_bound) ||
+                   page > nCodeSlots) {
+                       return NULL;
+               }
+               assert(page < nCodeSlots);
+
+               hash = base + page * hash_len;
+       }
+       
+       if (!cs_valid_range(hash, hash + hash_len,
+                           lower_bound, upper_bound)) {
+               hash = NULL;
+       }
+
+       return hash;
+}
+
+/*
+ * cs_validate_codedirectory
+ *
+ * Validate that pointers inside the code directory to make sure that
+ * all offsets and lengths are constrained within the buffer.
+ *
+ * Parameters: cd                      Pointer to code directory buffer
+ *             length                  Length of buffer
+ *
+ * Returns:    0                       Success
+ *             EBADEXEC                Invalid code signature
+ */
+
+static int
+cs_validate_codedirectory(const CS_CodeDirectory *cd, size_t length)
+{
+       struct cs_hash const *hashtype;
+
+       if (length < sizeof(*cd))
+               return EBADEXEC;
+       if (ntohl(cd->magic) != CSMAGIC_CODEDIRECTORY)
+               return EBADEXEC;
+       if (cd->pageSize < PAGE_SHIFT_4K || cd->pageSize > PAGE_SHIFT)
+               return EBADEXEC;
+       hashtype = cs_find_md(cd->hashType);
+       if (hashtype == NULL)
+               return EBADEXEC;
+
+       if (cd->hashSize != hashtype->cs_size)
+               return EBADEXEC;
+
+       if (length < ntohl(cd->hashOffset))
+               return EBADEXEC;
+
+       /* check that nSpecialSlots fits in the buffer in front of hashOffset */
+       if (ntohl(cd->hashOffset) / hashtype->cs_size < ntohl(cd->nSpecialSlots))
+               return EBADEXEC;
+
+       /* check that codeslots fits in the buffer */
+       if ((length - ntohl(cd->hashOffset)) / hashtype->cs_size <  ntohl(cd->nCodeSlots))
+               return EBADEXEC;
+       
+       if (ntohl(cd->version) >= CS_SUPPORTSSCATTER && cd->scatterOffset) {
+
+               if (length < ntohl(cd->scatterOffset))
+                       return EBADEXEC;
+
+               const SC_Scatter *scatter = (const SC_Scatter *)
+                       (((const uint8_t *)cd) + ntohl(cd->scatterOffset));
+               uint32_t nPages = 0;
+
+               /*
+                * Check each scatter buffer, since we don't know the
+                * length of the scatter buffer array, we have to
+                * check each entry.
+                */
+               while(1) {
+                       /* check that the end of each scatter buffer in within the length */
+                       if (((const uint8_t *)scatter) + sizeof(scatter[0]) > (const uint8_t *)cd + length)
+                               return EBADEXEC;
+                       uint32_t scount = ntohl(scatter->count);
+                       if (scount == 0)
+                               break;
+                       if (nPages + scount < nPages)
+                               return EBADEXEC;
+                       nPages += scount;
+                       scatter++;
+
+                       /* XXX check that basees doesn't overlap */
+                       /* XXX check that targetOffset doesn't overlap */
+               }
+#if 0 /* rdar://12579439 */
+               if (nPages != ntohl(cd->nCodeSlots))
+                       return EBADEXEC;
+#endif
+       }
+
+       if (length < ntohl(cd->identOffset))
+               return EBADEXEC;
+
+       /* identifier is NUL terminated string */
+       if (cd->identOffset) {
+               const uint8_t *ptr = (const uint8_t *)cd + ntohl(cd->identOffset);
+               if (memchr(ptr, 0, length - ntohl(cd->identOffset)) == NULL)
+                       return EBADEXEC;
+       }
+
+       /* team identifier is NULL terminated string */
+       if (ntohl(cd->version) >= CS_SUPPORTSTEAMID && ntohl(cd->teamOffset)) {
+               if (length < ntohl(cd->teamOffset))
+                       return EBADEXEC;
+
+               const uint8_t *ptr = (const uint8_t *)cd + ntohl(cd->teamOffset);
+               if (memchr(ptr, 0, length - ntohl(cd->teamOffset)) == NULL)
+                       return EBADEXEC;
+       }
+
+       return 0;
+}
+
+/*
+ *
+ */
+
+static int
+cs_validate_blob(const CS_GenericBlob *blob, size_t length)
+{
+       if (length < sizeof(CS_GenericBlob) || length < ntohl(blob->length))
+               return EBADEXEC;
+       return 0;
+}
+
+/*
+ * cs_validate_csblob
+ *
+ * Validate that superblob/embedded code directory to make sure that
+ * all internal pointers are valid.
+ *
+ * Will validate both a superblob csblob and a "raw" code directory.
+ *
+ *
+ * Parameters: buffer                  Pointer to code signature
+ *             length                  Length of buffer
+ *             rcd                     returns pointer to code directory
+ *
+ * Returns:    0                       Success
+ *             EBADEXEC                Invalid code signature
+ */
+
+static int
+cs_validate_csblob(
+       const uint8_t *addr,
+       const size_t blob_size,
+       const CS_CodeDirectory **rcd,
+       const CS_GenericBlob **rentitlements)
+{
+       const CS_GenericBlob *blob;
+       int error;
+       size_t length;
+
+       *rcd = NULL;
+       *rentitlements = NULL;
+
+       blob = (const CS_GenericBlob *)(const void *)addr;
+
+       length = blob_size;
+       error = cs_validate_blob(blob, length);
+       if (error)
+               return error;
+       length = ntohl(blob->length);
+
+       if (ntohl(blob->magic) == CSMAGIC_EMBEDDED_SIGNATURE) {
+               const CS_SuperBlob *sb;
+               uint32_t n, count;
+               const CS_CodeDirectory *best_cd = NULL;
+               unsigned int best_rank = 0;
+#if PLATFORM_WatchOS
+               const CS_CodeDirectory *sha1_cd = NULL;
+#endif
+
+               if (length < sizeof(CS_SuperBlob))
+                       return EBADEXEC;
+
+               sb = (const CS_SuperBlob *)blob;
+               count = ntohl(sb->count);
+
+               /* check that the array of BlobIndex fits in the rest of the data */
+               if ((length - sizeof(CS_SuperBlob)) / sizeof(CS_BlobIndex) < count)
+                       return EBADEXEC;
+
+               /* now check each BlobIndex */
+               for (n = 0; n < count; n++) {
+                       const CS_BlobIndex *blobIndex = &sb->index[n];
+                       uint32_t type = ntohl(blobIndex->type);
+                       uint32_t offset = ntohl(blobIndex->offset);
+                       if (length < offset)
+                               return EBADEXEC;
+
+                       const CS_GenericBlob *subBlob =
+                               (const CS_GenericBlob *)(const void *)(addr + offset);
+
+                       size_t subLength = length - offset;
+
+                       if ((error = cs_validate_blob(subBlob, subLength)) != 0)
+                               return error;
+                       subLength = ntohl(subBlob->length);
+
+                       /* extra validation for CDs, that is also returned */
+                       if (type == CSSLOT_CODEDIRECTORY || (type >= CSSLOT_ALTERNATE_CODEDIRECTORIES && type < CSSLOT_ALTERNATE_CODEDIRECTORY_LIMIT)) {
+                               const CS_CodeDirectory *candidate = (const CS_CodeDirectory *)subBlob;
+                               if ((error = cs_validate_codedirectory(candidate, subLength)) != 0)
+                                       return error;
+                               unsigned int rank = hash_rank(candidate);
+                               if (cs_debug > 3)
+                                       printf("CodeDirectory type %d rank %d at slot 0x%x index %d\n", candidate->hashType, (int)rank, (int)type, (int)n);
+                               if (best_cd == NULL || rank > best_rank) {
+                                       best_cd = candidate;
+                                       best_rank = rank;
+
+                                       if (cs_debug > 2)
+                                               printf("using CodeDirectory type %d (rank %d)\n", (int)best_cd->hashType, best_rank);
+                                       *rcd = best_cd;
+                               } else if (best_cd != NULL && rank == best_rank) {
+                                       /* repeat of a hash type (1:1 mapped to ranks), illegal and suspicious */
+                                       printf("multiple hash=%d CodeDirectories in signature; rejecting\n", best_cd->hashType);
+                                       return EBADEXEC;
+                               }
+#if PLATFORM_WatchOS
+                               if (candidate->hashType == CS_HASHTYPE_SHA1) {
+                                       if (sha1_cd != NULL) {
+                                               printf("multiple sha1 CodeDirectories in signature; rejecting\n");
+                                               return EBADEXEC;
+                                       }
+                                       sha1_cd = candidate;
+                               }
+#endif
+                       } else if (type == CSSLOT_ENTITLEMENTS) {
+                               if (ntohl(subBlob->magic) != CSMAGIC_EMBEDDED_ENTITLEMENTS) {
+                                       return EBADEXEC;
+                               }
+                               if (*rentitlements != NULL) {
+                                       printf("multiple entitlements blobs\n");
+                                       return EBADEXEC;
+                               }
+                               *rentitlements = subBlob;
+                       }
+               }
+
+#if PLATFORM_WatchOS
+               /* To keep watchOS fast enough, we have to resort to sha1 for
+                * some code.
+                *
+                * At the time of writing this comment, known sha1 attacks are
+                * collision attacks (not preimage or second preimage
+                * attacks), which do not apply to platform binaries since
+                * they have a fixed hash in the trust cache.  Given this
+                * property, we only prefer sha1 code directories for adhoc
+                * signatures, which always have to be in a trust cache to be
+                * valid (can-load-cdhash does not exist for watchOS). Those
+                * are, incidentally, also the platform binaries, for which we
+                * care about the performance hit that sha256 would bring us.
+                *
+                * Platform binaries may still contain a (not chosen) sha256
+                * code directory, which keeps software updates that switch to
+                * sha256-only small.
+                */
+
+               if (*rcd != NULL && sha1_cd != NULL && (ntohl(sha1_cd->flags) & CS_ADHOC)) {
+                       if (sha1_cd->flags != (*rcd)->flags) {
+                               printf("mismatched flags between hash %d (flags: %#x) and sha1 (flags: %#x) cd.\n",
+                                          (int)(*rcd)->hashType, (*rcd)->flags, sha1_cd->flags);
+                               *rcd = NULL;
+                               return EBADEXEC;
+                       }
+
+                       *rcd = sha1_cd;
+               }
+#endif
+
+       } else if (ntohl(blob->magic) == CSMAGIC_CODEDIRECTORY) {
+
+               if ((error = cs_validate_codedirectory((const CS_CodeDirectory *)(const void *)addr, length)) != 0)
+                       return error;
+               *rcd = (const CS_CodeDirectory *)blob;
+       } else {
+               return EBADEXEC;
+       }
+
+       if (*rcd == NULL)
+               return EBADEXEC;
+
+       return 0;
+}
+
+/*
+ * cs_find_blob_bytes
+ *
+ * Find an blob from the superblob/code directory. The blob must have
+ * been been validated by cs_validate_csblob() before calling
+ * this. Use csblob_find_blob() instead.
+ * 
+ * Will also find a "raw" code directory if its stored as well as
+ * searching the superblob.
+ *
+ * Parameters: buffer                  Pointer to code signature
+ *             length                  Length of buffer
+ *             type                    type of blob to find
+ *             magic                   the magic number for that blob
+ *
+ * Returns:    pointer                 Success
+ *             NULL                    Buffer not found
+ */
+
+const CS_GenericBlob *
+csblob_find_blob_bytes(const uint8_t *addr, size_t length, uint32_t type, uint32_t magic)
+{
+       const CS_GenericBlob *blob = (const CS_GenericBlob *)(const void *)addr;
+
+       if (ntohl(blob->magic) == CSMAGIC_EMBEDDED_SIGNATURE) {
+               const CS_SuperBlob *sb = (const CS_SuperBlob *)blob;
+               size_t n, count = ntohl(sb->count);
+
+               for (n = 0; n < count; n++) {
+                       if (ntohl(sb->index[n].type) != type)
+                               continue;
+                       uint32_t offset = ntohl(sb->index[n].offset);
+                       if (length - sizeof(const CS_GenericBlob) < offset)
+                               return NULL;
+                       blob = (const CS_GenericBlob *)(const void *)(addr + offset);
+                       if (ntohl(blob->magic) != magic)
+                               continue;
+                       return blob;
+               }
+       } else if (type == CSSLOT_CODEDIRECTORY
+                  && ntohl(blob->magic) == CSMAGIC_CODEDIRECTORY
+                  && magic == CSMAGIC_CODEDIRECTORY)
+               return blob;
+       return NULL;
+}
+
+
+const CS_GenericBlob *
+csblob_find_blob(struct cs_blob *csblob, uint32_t type, uint32_t magic)
+{
+       if ((csblob->csb_flags & CS_VALID) == 0)
+               return NULL;
+       return csblob_find_blob_bytes((const uint8_t *)csblob->csb_mem_kaddr, csblob->csb_mem_size, type, magic);
+}
+
+static const uint8_t *
+find_special_slot(const CS_CodeDirectory *cd, size_t slotsize, uint32_t slot)
+{
+       /* there is no zero special slot since that is the first code slot */
+       if (ntohl(cd->nSpecialSlots) < slot || slot == 0)
+               return NULL;
+
+       return ((const uint8_t *)cd + ntohl(cd->hashOffset) - (slotsize * slot));
+}
+
+static uint8_t cshash_zero[CS_HASH_MAX_SIZE] = { 0 };
+
+int
+csblob_get_entitlements(struct cs_blob *csblob, void **out_start, size_t *out_length)
+{
+       uint8_t computed_hash[CS_HASH_MAX_SIZE];
+       const CS_GenericBlob *entitlements;
+       const CS_CodeDirectory *code_dir;
+       const uint8_t *embedded_hash;
+       union cs_hash_union context;
+
+       *out_start = NULL;
+       *out_length = 0;
+
+       if (csblob->csb_hashtype == NULL || csblob->csb_hashtype->cs_digest_size > sizeof(computed_hash))
+           return EBADEXEC;
+
+       code_dir = csblob->csb_cd;
+
+       if ((csblob->csb_flags & CS_VALID) == 0) {
+               entitlements = NULL;
+       } else {
+               entitlements = csblob->csb_entitlements_blob;
+       }
+       embedded_hash = find_special_slot(code_dir, csblob->csb_hashtype->cs_size, CSSLOT_ENTITLEMENTS);
+
+       if (embedded_hash == NULL) {
+               if (entitlements)
+                       return EBADEXEC;
+               return 0;
+       } else if (entitlements == NULL) {
+               if (memcmp(embedded_hash, cshash_zero, csblob->csb_hashtype->cs_size) != 0) {
+                       return EBADEXEC;
+               } else {
+                       return 0;
+               }
+       }
+
+       csblob->csb_hashtype->cs_init(&context);
+       csblob->csb_hashtype->cs_update(&context, entitlements, ntohl(entitlements->length));
+       csblob->csb_hashtype->cs_final(computed_hash, &context);
+
+       if (memcmp(computed_hash, embedded_hash, csblob->csb_hashtype->cs_size) != 0)
+               return EBADEXEC;
+
+       *out_start = __DECONST(void *, entitlements);
+       *out_length = ntohl(entitlements->length);
+
+       return 0;
+}
+
+/*
+ * CODESIGNING
+ * End of routines to navigate code signing data structures in the kernel.
+ */
+
+
 
 /*
- *     Initialization of the zone for Unified Buffer Cache.
+ * ubc_init
+ * 
+ * Initialization of the zone for Unified Buffer Cache.
+ *
+ * Parameters: (void)
+ *
+ * Returns:    (void)
+ *
+ * Implicit returns:
+ *             ubc_info_zone(global)   initialized for subsequent allocations
  */
 __private_extern__ void
-ubc_init()
+ubc_init(void)
 {
        int     i;
 
        i = (vm_size_t) sizeof (struct ubc_info);
-       /* XXX  the number of elements should be tied in to maxvnodes */
+
        ubc_info_zone = zinit (i, 10000*i, 8192, "ubc_info zone");
-       return;
+
+       zone_change(ubc_info_zone, Z_NOENCRYPT, TRUE);
 }
 
+
 /*
- *     Initialize a ubc_info structure for a vnode.
+ * ubc_info_init
+ *
+ * Allocate and attach an empty ubc_info structure to a vnode
+ *
+ * Parameters: vp                      Pointer to the vnode
+ *
+ * Returns:    0                       Success
+ *     vnode_size:ENOMEM               Not enough space
+ *     vnode_size:???                  Other error from vnode_getattr
+ *
  */
 int
 ubc_info_init(struct vnode *vp)
 {
        return(ubc_info_init_internal(vp, 0, 0));
 }
+
+
+/*
+ * ubc_info_init_withsize
+ *
+ * Allocate and attach a sized ubc_info structure to a vnode
+ *
+ * Parameters: vp                      Pointer to the vnode
+ *             filesize                The size of the file
+ *
+ * Returns:    0                       Success
+ *     vnode_size:ENOMEM               Not enough space
+ *     vnode_size:???                  Other error from vnode_getattr
+ */
 int
 ubc_info_init_withsize(struct vnode *vp, off_t filesize)
 {
        return(ubc_info_init_internal(vp, 1, filesize));
 }
 
-int
-ubc_info_init_internal(struct vnode *vp, int withfsize, off_t filesize)
+
+/*
+ * ubc_info_init_internal
+ *
+ * Allocate and attach a ubc_info structure to a vnode
+ *
+ * Parameters: vp                      Pointer to the vnode
+ *             withfsize{0,1}          Zero if the size should be obtained
+ *                                     from the vnode; otherwise, use filesize
+ *             filesize                The size of the file, if withfsize == 1
+ *
+ * Returns:    0                       Success
+ *     vnode_size:ENOMEM               Not enough space
+ *     vnode_size:???                  Other error from vnode_getattr
+ *
+ * Notes:      We call a blocking zalloc(), and the zone was created as an
+ *             expandable and collectable zone, so if no memory is available,
+ *             it is possible for zalloc() to block indefinitely.  zalloc()
+ *             may also panic if the zone of zones is exhausted, since it's
+ *             NOT expandable.
+ *
+ *             We unconditionally call vnode_pager_setup(), even if this is
+ *             a reuse of a ubc_info; in that case, we should probably assert
+ *             that it does not already have a pager association, but do not.
+ *
+ *             Since memory_object_create_named() can only fail from receiving
+ *             an invalid pager argument, the explicit check and panic is
+ *             merely precautionary.
+ */
+static int
+ubc_info_init_internal(vnode_t vp, int withfsize, off_t filesize)
 {
-       register struct ubc_info        *uip;
+       struct ubc_info *uip;
        void *  pager;
-       struct proc *p = current_proc();
        int error = 0;
        kern_return_t kret;
        memory_object_control_t control;
 
        uip = vp->v_ubcinfo;
 
+       /*
+        * If there is not already a ubc_info attached to the vnode, we
+        * attach one; otherwise, we will reuse the one that's there.
+        */
        if (uip == UBC_INFO_NULL) {
 
                uip = (struct ubc_info *) zalloc(ubc_info_zone);
@@ -133,28 +858,34 @@ ubc_info_init_internal(struct vnode *vp, int withfsize, off_t filesize)
                uip->ui_flags = UI_INITED;
                uip->ui_ucred = NOCRED;
        }
-#if DIAGNOSTIC
-       else
-               Debugger("ubc_info_init: already");
-#endif /* DIAGNOSTIC */
-       
        assert(uip->ui_flags != UI_NONE);
        assert(uip->ui_vnode == vp);
 
        /* now set this ubc_info in the vnode */
        vp->v_ubcinfo = uip;
 
+       /*
+        * Allocate a pager object for this vnode
+        *
+        * XXX The value of the pager parameter is currently ignored.
+        * XXX Presumably, this API changed to avoid the race between
+        * XXX setting the pager and the UI_HASPAGER flag.
+        */
        pager = (void *)vnode_pager_setup(vp, uip->ui_pager);
        assert(pager);
 
+       /*
+        * Explicitly set the pager into the ubc_info, after setting the
+        * UI_HASPAGER flag.
+        */
        SET(uip->ui_flags, UI_HASPAGER);
        uip->ui_pager = pager;
 
        /*
         * Note: We can not use VNOP_GETATTR() to get accurate
-        * value of ui_size. Thanks to NFS.
-        * nfs_getattr() can call vinvalbuf() and in this case
-        * ubc_info is not set up to deal with that.
+        * value of ui_size because this may be an NFS vnode, and
+        * nfs_getattr() can call vinvalbuf(); if this happens,
+        * ubc_info is not set up to deal with that event.
         * So use bogus size.
         */
 
@@ -175,137 +906,284 @@ ubc_info_init_internal(struct vnode *vp, int withfsize, off_t filesize)
        assert(control);
        uip->ui_control = control;      /* cache the value of the mo control */
        SET(uip->ui_flags, UI_HASOBJREF);       /* with a named reference */
-#if 0
-       /* create a pager reference on the vnode */
-       error = vnode_pager_vget(vp);
-       if (error)
-               panic("ubc_info_init: vnode_pager_vget error = %d", error);
-#endif
+
        if (withfsize == 0) {
-               struct vfs_context context;
                /* initialize the size */
-               context.vc_proc = p;
-               context.vc_ucred = kauth_cred_get();
-               error = vnode_size(vp, &uip->ui_size, &context);
+               error = vnode_size(vp, &uip->ui_size, vfs_context_current());
                if (error)
                        uip->ui_size = 0;
        } else {
                uip->ui_size = filesize;
        }
-       vp->v_lflag |= VNAMED_UBC;
+       vp->v_lflag |= VNAMED_UBC;      /* vnode has a named ubc reference */
 
        return (error);
 }
 
-/* Free the ubc_info */
+
+/*
+ * ubc_info_free
+ *
+ * Free a ubc_info structure
+ *
+ * Parameters: uip                     A pointer to the ubc_info to free
+ *
+ * Returns:    (void)
+ *
+ * Notes:      If there is a credential that has subsequently been associated
+ *             with the ubc_info via a call to ubc_setcred(), the reference
+ *             to the credential is dropped.
+ *
+ *             It's actually impossible for a ubc_info.ui_control to take the
+ *             value MEMORY_OBJECT_CONTROL_NULL.
+ */
 static void
 ubc_info_free(struct ubc_info *uip)
 {
-       kauth_cred_t credp;
-       
-       credp = uip->ui_ucred;
-       if (credp != NOCRED) {
-               uip->ui_ucred = NOCRED;
-               kauth_cred_rele(credp);
+       if (IS_VALID_CRED(uip->ui_ucred)) {
+               kauth_cred_unref(&uip->ui_ucred);
        }
 
        if (uip->ui_control != MEMORY_OBJECT_CONTROL_NULL)
                memory_object_control_deallocate(uip->ui_control);
        
        cluster_release(uip);
+       ubc_cs_free(uip);
 
-       zfree(ubc_info_zone, (vm_offset_t)uip);
+       zfree(ubc_info_zone, uip);
        return;
 }
 
+
 void
 ubc_info_deallocate(struct ubc_info *uip)
 {
         ubc_info_free(uip);
 }
 
+errno_t mach_to_bsd_errno(kern_return_t mach_err)
+{
+       switch (mach_err) {
+       case KERN_SUCCESS:
+               return 0;
+
+       case KERN_INVALID_ADDRESS:
+       case KERN_INVALID_ARGUMENT:
+       case KERN_NOT_IN_SET:
+       case KERN_INVALID_NAME:
+       case KERN_INVALID_TASK:
+       case KERN_INVALID_RIGHT:
+       case KERN_INVALID_VALUE:
+       case KERN_INVALID_CAPABILITY:
+       case KERN_INVALID_HOST:
+       case KERN_MEMORY_PRESENT:
+       case KERN_INVALID_PROCESSOR_SET:
+       case KERN_INVALID_POLICY:
+       case KERN_ALREADY_WAITING:
+       case KERN_DEFAULT_SET:
+       case KERN_EXCEPTION_PROTECTED:
+       case KERN_INVALID_LEDGER:
+       case KERN_INVALID_MEMORY_CONTROL:
+       case KERN_INVALID_SECURITY:
+       case KERN_NOT_DEPRESSED:
+       case KERN_LOCK_OWNED:
+       case KERN_LOCK_OWNED_SELF:
+               return EINVAL;
+
+       case KERN_PROTECTION_FAILURE:
+       case KERN_NOT_RECEIVER:
+       case KERN_NO_ACCESS:
+       case KERN_POLICY_STATIC:
+               return EACCES;
+
+       case KERN_NO_SPACE:
+       case KERN_RESOURCE_SHORTAGE:
+       case KERN_UREFS_OVERFLOW:
+       case KERN_INVALID_OBJECT:
+               return ENOMEM;
+
+       case KERN_FAILURE:
+               return EIO;
+
+       case KERN_MEMORY_FAILURE:
+       case KERN_POLICY_LIMIT:
+       case KERN_CODESIGN_ERROR:
+               return EPERM;
+
+       case KERN_MEMORY_ERROR:
+               return EBUSY;
+
+       case KERN_ALREADY_IN_SET:
+       case KERN_NAME_EXISTS:
+       case KERN_RIGHT_EXISTS:
+               return EEXIST;
+
+       case KERN_ABORTED:
+               return EINTR;
+
+       case KERN_TERMINATED:
+       case KERN_LOCK_SET_DESTROYED:
+       case KERN_LOCK_UNSTABLE:
+       case KERN_SEMAPHORE_DESTROYED:
+               return ENOENT;
+
+       case KERN_RPC_SERVER_TERMINATED:
+               return ECONNRESET;
+
+       case KERN_NOT_SUPPORTED:
+               return ENOTSUP;
+
+       case KERN_NODE_DOWN:
+               return ENETDOWN;
+
+       case KERN_NOT_WAITING:
+               return ENOENT;
+
+       case KERN_OPERATION_TIMED_OUT:
+               return ETIMEDOUT;
+
+       default:
+               return EIO;
+       }
+}
+
 /*
- * Communicate with VM the size change of the file
- * returns 1 on success, 0 on failure
+ * ubc_setsize_ex
+ *
+ * Tell the VM that the the size of the file represented by the vnode has
+ * changed
+ *
+ * Parameters: vp         The vp whose backing file size is
+ *                                        being changed
+ *                             nsize  The new size of the backing file
+ *                             opts   Options
+ *
+ * Returns:    EINVAL for new size < 0
+ *                     ENOENT if no UBC info exists
+ *          EAGAIN if UBC_SETSIZE_NO_FS_REENTRY option is set and new_size < old size
+ *          Other errors (mapped to errno_t) returned by VM functions
+ *
+ * Notes:   This function will indicate success if the new size is the
+ *                 same or larger than the old size (in this case, the
+ *                 remainder of the file will require modification or use of
+ *                 an existing upl to access successfully).
+ *
+ *                 This function will fail if the new file size is smaller,
+ *                 and the memory region being invalidated was unable to
+ *                 actually be invalidated and/or the last page could not be
+ *                 flushed, if the new size is not aligned to a page
+ *                 boundary.  This is usually indicative of an I/O error.
  */
-int
-ubc_setsize(struct vnode *vp, off_t nsize)
+errno_t ubc_setsize_ex(struct vnode *vp, off_t nsize, ubc_setsize_opts_t opts)
 {
        off_t osize;    /* ui_size before change */
        off_t lastpg, olastpgend, lastoff;
        struct ubc_info *uip;
        memory_object_control_t control;
-       kern_return_t kret;
+       kern_return_t kret = KERN_SUCCESS;
 
        if (nsize < (off_t)0)
-               return (0);
+               return EINVAL;
 
        if (!UBCINFOEXISTS(vp))
-               return (0);
+               return ENOENT;
 
        uip = vp->v_ubcinfo;
-       osize = uip->ui_size;   /* call ubc_getsize() ??? */
-       /* Update the size before flushing the VM */
+       osize = uip->ui_size;
+
+       if (ISSET(opts, UBC_SETSIZE_NO_FS_REENTRY) && nsize < osize)
+               return EAGAIN;
+
+       /*
+        * Update the size before flushing the VM
+        */
        uip->ui_size = nsize;
 
-       if (nsize >= osize)     /* Nothing more to do */
-               return (1);             /* return success */
+       if (nsize >= osize) {   /* Nothing more to do */
+               if (nsize > osize) {
+                       lock_vnode_and_post(vp, NOTE_EXTEND);
+               }
+
+               return 0;
+       }
 
        /*
         * When the file shrinks, invalidate the pages beyond the
         * new size. Also get rid of garbage beyond nsize on the
-        * last page. The ui_size already has the nsize. This
-        * insures that the pageout would not write beyond the new
-        * end of the file.
+        * last page. The ui_size already has the nsize, so any
+        * subsequent page-in will zero-fill the tail properly
         */
-
        lastpg = trunc_page_64(nsize);
        olastpgend = round_page_64(osize);
        control = uip->ui_control;
        assert(control);
        lastoff = (nsize & PAGE_MASK_64);
 
-       /*
-        * If length is multiple of page size, we should not flush
-        * invalidating is sufficient
-        */
-        if (!lastoff) {
-        /* invalidate last page and old contents beyond nsize */
-        kret = memory_object_lock_request(control,
-                    (memory_object_offset_t)lastpg,
-                   (memory_object_size_t)(olastpgend - lastpg), NULL, NULL,
-                    MEMORY_OBJECT_RETURN_NONE, MEMORY_OBJECT_DATA_FLUSH,
-                    VM_PROT_NO_CHANGE);
-        if (kret != KERN_SUCCESS)
-            printf("ubc_setsize: invalidate failed (error = %d)\n", kret);
-
-               return ((kret == KERN_SUCCESS) ? 1 : 0);
-        }
-
-       /* flush the last page */
-       kret = memory_object_lock_request(control,
-                               (memory_object_offset_t)lastpg,
-                               PAGE_SIZE_64, NULL, NULL,
-                               MEMORY_OBJECT_RETURN_DIRTY, FALSE,
-                               VM_PROT_NO_CHANGE);
-
-       if (kret == KERN_SUCCESS) {
-               /* invalidate last page and old contents beyond nsize */
-               kret = memory_object_lock_request(control,
-                                       (memory_object_offset_t)lastpg,
-                                       (memory_object_size_t)(olastpgend - lastpg), NULL, NULL,
-                                       MEMORY_OBJECT_RETURN_NONE, MEMORY_OBJECT_DATA_FLUSH,
-                                       VM_PROT_NO_CHANGE);
-               if (kret != KERN_SUCCESS)
-                       printf("ubc_setsize: invalidate failed (error = %d)\n", kret);
-       } else
-               printf("ubc_setsize: flush failed (error = %d)\n", kret);
+       if (lastoff) {
+               upl_t           upl;
+               upl_page_info_t *pl;
 
-       return ((kret == KERN_SUCCESS) ? 1 : 0);
+               /*
+                * new EOF ends up in the middle of a page
+                * zero the tail of this page if it's currently
+                * present in the cache
+                */
+               kret = ubc_create_upl_kernel(vp, lastpg, PAGE_SIZE, &upl, &pl, UPL_SET_LITE, VM_KERN_MEMORY_FILE);
+
+               if (kret != KERN_SUCCESS)
+                       panic("ubc_setsize: ubc_create_upl (error = %d)\n", kret);
+
+               if (upl_valid_page(pl, 0))
+                       cluster_zero(upl, (uint32_t)lastoff, PAGE_SIZE - (uint32_t)lastoff, NULL);
+
+               ubc_upl_abort_range(upl, 0, PAGE_SIZE, UPL_ABORT_FREE_ON_EMPTY);
+
+               lastpg += PAGE_SIZE_64;
+       }
+       if (olastpgend > lastpg) {
+               int     flags;
+
+               if (lastpg == 0)
+                       flags = MEMORY_OBJECT_DATA_FLUSH_ALL;
+               else
+                       flags = MEMORY_OBJECT_DATA_FLUSH;
+               /*
+                * invalidate the pages beyond the new EOF page
+                *
+                */
+               kret = memory_object_lock_request(control,
+                                                                                 (memory_object_offset_t)lastpg,
+                                                                                 (memory_object_size_t)(olastpgend - lastpg), NULL, NULL,
+                                                                                 MEMORY_OBJECT_RETURN_NONE, flags, VM_PROT_NO_CHANGE);
+               if (kret != KERN_SUCCESS)
+                       printf("ubc_setsize: invalidate failed (error = %d)\n", kret);
+       }
+       return mach_to_bsd_errno(kret);
+}
+
+// Returns true for success
+int ubc_setsize(vnode_t vp, off_t nsize)
+{
+       return ubc_setsize_ex(vp, nsize, 0) == 0;
 }
 
 /*
- * Get the size of the file
+ * ubc_getsize
+ *
+ * Get the size of the file assocated with the specified vnode
+ *
+ * Parameters: vp                      The vnode whose size is of interest
+ *
+ * Returns:    0                       There is no ubc_info associated with
+ *                                     this vnode, or the size is zero
+ *             !0                      The size of the file
+ *
+ * Notes:      Using this routine, it is not possible for a caller to
+ *             successfully distinguish between a vnode associate with a zero
+ *             length file, and a vnode with no associated ubc_info.  The
+ *             caller therefore needs to not care, or needs to ensure that
+ *             they have previously successfully called ubc_info_init() or
+ *             ubc_info_init_withsize().
  */
 off_t
 ubc_getsize(struct vnode *vp)
@@ -318,12 +1196,30 @@ ubc_getsize(struct vnode *vp)
        return (vp->v_ubcinfo->ui_size);
 }
 
+
 /*
- * call ubc_sync_range(vp, 0, EOF, UBC_PUSHALL) on all the vnodes
- * for this mount point.
- * returns 1 on success, 0 on failure
+ * ubc_umount
+ *
+ * Call ubc_msync(vp, 0, EOF, NULL, UBC_PUSHALL) on all the vnodes for this
+ * mount point
+ *
+ * Parameters: mp                      The mount point
+ *
+ * Returns:    0                       Success
+ *
+ * Notes:      There is no failure indication for this function.
+ *
+ *             This function is used in the unmount path; since it may block
+ *             I/O indefinitely, it should not be used in the forced unmount
+ *             path, since a device unavailability could also block that
+ *             indefinitely.
+ *
+ *             Because there is no device ejection interlock on USB, FireWire,
+ *             or similar devices, it's possible that an ejection that begins
+ *             subsequent to the vnode_iterate() completing, either on one of
+ *             those devices, or a network mount for which the server quits
+ *             responding, etc., may cause the caller to block indefinitely.
  */
-
 __private_extern__ int
 ubc_umount(struct mount *mp)
 {
@@ -331,22 +1227,40 @@ ubc_umount(struct mount *mp)
        return(0);
 }
 
+
+/*
+ * ubc_umcallback
+ *
+ * Used by ubc_umount() as an internal implementation detail; see ubc_umount()
+ * and vnode_iterate() for details of implementation.
+ */
 static int
 ubc_umcallback(vnode_t vp, __unused void * args)
 {
 
        if (UBCINFOEXISTS(vp)) {
 
-               cluster_push(vp, 0);
-
                (void) ubc_msync(vp, (off_t)0, ubc_getsize(vp), NULL, UBC_PUSHALL);
        }
        return (VNODE_RETURNED);
 }
 
 
-
-/* Get the credentials */
+/*
+ * ubc_getcred
+ *
+ * Get the credentials currently active for the ubc_info associated with the
+ * vnode.
+ *
+ * Parameters: vp                      The vnode whose ubc_info credentials
+ *                                     are to be retrieved
+ *
+ * Returns:    !NOCRED                 The credentials
+ *             NOCRED                  If there is no ubc_info for the vnode,
+ *                                     or if there is one, but it has not had
+ *                                     any credentials associated with it via
+ *                                     a call to ubc_setcred()
+ */
 kauth_cred_t
 ubc_getcred(struct vnode *vp)
 {
@@ -356,22 +1270,78 @@ ubc_getcred(struct vnode *vp)
        return (NOCRED);
 }
 
+
+/*
+ * ubc_setthreadcred
+ *
+ * If they are not already set, set the credentials of the ubc_info structure
+ * associated with the vnode to those of the supplied thread; otherwise leave
+ * them alone.
+ *
+ * Parameters: vp                      The vnode whose ubc_info creds are to
+ *                                     be set
+ *             p                       The process whose credentials are to
+ *                                     be used, if not running on an assumed
+ *                                     credential
+ *             thread                  The thread whose credentials are to
+ *                                     be used
+ *
+ * Returns:    1                       This vnode has no associated ubc_info
+ *             0                       Success
+ *
+ * Notes:      This function takes a proc parameter to account for bootstrap
+ *             issues where a task or thread may call this routine, either
+ *             before credentials have been initialized by bsd_init(), or if
+ *             there is no BSD info asscoiate with a mach thread yet.  This
+ *             is known to happen in both the initial swap and memory mapping
+ *             calls.
+ *
+ *             This function is generally used only in the following cases:
+ *
+ *             o       a memory mapped file via the mmap() system call
+ *             o       a swap store backing file
+ *             o       subsequent to a successful write via vn_write()
+ *
+ *             The information is then used by the NFS client in order to
+ *             cons up a wire message in either the page-in or page-out path.
+ *
+ *             There are two potential problems with the use of this API:
+ *
+ *             o       Because the write path only set it on a successful
+ *                     write, there is a race window between setting the
+ *                     credential and its use to evict the pages to the
+ *                     remote file server
+ *
+ *             o       Because a page-in may occur prior to a write, the
+ *                     credential may not be set at this time, if the page-in
+ *                     is not the result of a mapping established via mmap().
+ *
+ *             In both these cases, this will be triggered from the paging
+ *             path, which will instead use the credential of the current
+ *             process, which in this case is either the dynamic_pager or
+ *             the kernel task, both of which utilize "root" credentials.
+ *
+ *             This may potentially permit operations to occur which should
+ *             be denied, or it may cause to be denied operations which
+ *             should be permitted, depending on the configuration of the NFS
+ *             server.
+ */
 int
-ubc_setthreadcred(struct vnode *vp, struct proc *p, thread_t thread)
+ubc_setthreadcred(struct vnode *vp, proc_t p, thread_t thread)
 {
        struct ubc_info *uip;
        kauth_cred_t credp;
-       struct uthread *uthread = get_bsdthread_info(thread);
+       struct uthread  *uthread = get_bsdthread_info(thread);
 
        if (!UBCINFOEXISTS(vp))
-               return (1);
+               return (1); 
 
        vnode_lock(vp);
 
        uip = vp->v_ubcinfo;
        credp = uip->ui_ucred;
 
-       if (credp == NOCRED) {
+       if (!IS_VALID_CRED(credp)) {
                /* use per-thread cred, if assumed identity, else proc cred */
                if (uthread == NULL || (uthread->uu_flag & UT_SETUID) == 0) {
                        uip->ui_ucred = kauth_cred_proc_ref(p);
@@ -379,32 +1349,58 @@ ubc_setthreadcred(struct vnode *vp, struct proc *p, thread_t thread)
                        uip->ui_ucred = uthread->uu_ucred;
                        kauth_cred_ref(uip->ui_ucred);
                }
-       }
+       } 
        vnode_unlock(vp);
 
        return (0);
 }
 
+
 /*
- * Set the credentials
- * existing credentials are not changed
- * returns 1 on success and 0 on failure
+ * ubc_setcred
+ *
+ * If they are not already set, set the credentials of the ubc_info structure
+ * associated with the vnode to those of the process; otherwise leave them
+ * alone.
+ *
+ * Parameters: vp                      The vnode whose ubc_info creds are to
+ *                                     be set
+ *             p                       The process whose credentials are to
+ *                                     be used
+ *
+ * Returns:    0                       This vnode has no associated ubc_info
+ *             1                       Success
+ *
+ * Notes:      The return values for this function are inverted from nearly
+ *             all other uses in the kernel.
+ *
+ *             See also ubc_setthreadcred(), above.
+ *
+ *             This function is considered deprecated, and generally should
+ *             not be used, as it is incompatible with per-thread credentials;
+ *             it exists for legacy KPI reasons.
+ *
+ * DEPRECATION:        ubc_setcred() is being deprecated. Please use 
+ *             ubc_setthreadcred() instead.
  */
 int
-ubc_setcred(struct vnode *vp, struct proc *p)
+ubc_setcred(struct vnode *vp, proc_t p)
 {
        struct ubc_info *uip;
        kauth_cred_t credp;
 
-        if ( !UBCINFOEXISTS(vp))
+       /* If there is no ubc_info, deny the operation */
+       if ( !UBCINFOEXISTS(vp))
                return (0); 
 
+       /*
+        * Check to see if there is already a credential reference in the
+        * ubc_info; if there is not, take one on the supplied credential.
+        */
        vnode_lock(vp);
-
        uip = vp->v_ubcinfo;
        credp = uip->ui_ucred;
-
-       if (credp == NOCRED) {
+       if (!IS_VALID_CRED(credp)) {
                uip->ui_ucred = kauth_cred_proc_ref(p);
        } 
        vnode_unlock(vp);
@@ -412,7 +1408,21 @@ ubc_setcred(struct vnode *vp, struct proc *p)
        return (1);
 }
 
-/* Get the pager */
+/*
+ * ubc_getpager
+ *
+ * Get the pager associated with the ubc_info associated with the vnode.
+ *
+ * Parameters: vp                      The vnode to obtain the pager from
+ *
+ * Returns:    !VNODE_PAGER_NULL       The memory_object_t for the pager
+ *             VNODE_PAGER_NULL        There is no ubc_info for this vnode
+ *
+ * Notes:      For each vnode that has a ubc_info associated with it, that
+ *             ubc_info SHALL have a pager associated with it, so in the
+ *             normal case, it's impossible to return VNODE_PAGER_NULL for
+ *             a vnode with an associated ubc_info.
+ */
 __private_extern__ memory_object_t
 ubc_getpager(struct vnode *vp)
 {
@@ -422,65 +1432,149 @@ ubc_getpager(struct vnode *vp)
        return (0);
 }
 
+
 /*
- * Get the memory object associated with this vnode
- * If the vnode was reactivated, memory object would not exist.
- * Unless "do not rectivate" was specified, look it up using the pager.
- * If hold was requested create an object reference of one does not
- * exist already.
+ * ubc_getobject
+ *
+ * Get the memory object control associated with the ubc_info associated with
+ * the vnode
+ *
+ * Parameters: vp                      The vnode to obtain the memory object
+ *                                     from
+ *             flags                   DEPRECATED
+ *
+ * Returns:    !MEMORY_OBJECT_CONTROL_NULL
+ *             MEMORY_OBJECT_CONTROL_NULL
+ *
+ * Notes:      Historically, if the flags were not "do not reactivate", this
+ *             function would look up the memory object using the pager if
+ *             it did not exist (this could be the case if the vnode had
+ *             been previously reactivated).  The flags would also permit a
+ *             hold to be requested, which would have created an object
+ *             reference, if one had not already existed.  This usage is
+ *             deprecated, as it would permit a race between finding and
+ *             taking the reference vs. a single reference being dropped in
+ *             another thread.
  */
-
 memory_object_control_t
 ubc_getobject(struct vnode *vp, __unused int flags)
 {
         if (UBCINFOEXISTS(vp))
                return((vp->v_ubcinfo->ui_control));
 
-       return (0);
+       return (MEMORY_OBJECT_CONTROL_NULL);
 }
 
-
+/*
+ * ubc_blktooff
+ *
+ * Convert a given block number to a memory backing object (file) offset for a
+ * given vnode
+ *
+ * Parameters: vp                      The vnode in which the block is located
+ *             blkno                   The block number to convert
+ *
+ * Returns:    !-1                     The offset into the backing object
+ *             -1                      There is no ubc_info associated with
+ *                                     the vnode
+ *             -1                      An error occurred in the underlying VFS
+ *                                     while translating the block to an
+ *                                     offset; the most likely cause is that
+ *                                     the caller specified a block past the
+ *                                     end of the file, but this could also be
+ *                                     any other error from VNOP_BLKTOOFF().
+ *
+ * Note:       Representing the error in band loses some information, but does
+ *             not occlude a valid offset, since an off_t of -1 is normally
+ *             used to represent EOF.  If we had a more reliable constant in
+ *             our header files for it (i.e. explicitly cast to an off_t), we
+ *             would use it here instead.
+ */
 off_t
 ubc_blktooff(vnode_t vp, daddr64_t blkno)
 {
-       off_t file_offset;
+       off_t file_offset = -1;
        int error;
 
-       if (UBCINVALID(vp))
-               return ((off_t)-1);
-
-       error = VNOP_BLKTOOFF(vp, blkno, &file_offset);
-       if (error)
-               file_offset = -1;
+       if (UBCINFOEXISTS(vp)) {
+               error = VNOP_BLKTOOFF(vp, blkno, &file_offset);
+               if (error)
+                       file_offset = -1;
+       }
 
        return (file_offset);
 }
 
+
+/*
+ * ubc_offtoblk
+ *
+ * Convert a given offset in a memory backing object into a block number for a
+ * given vnode
+ *
+ * Parameters: vp                      The vnode in which the offset is
+ *                                     located
+ *             offset                  The offset into the backing object
+ *
+ * Returns:    !-1                     The returned block number
+ *             -1                      There is no ubc_info associated with
+ *                                     the vnode
+ *             -1                      An error occurred in the underlying VFS
+ *                                     while translating the block to an
+ *                                     offset; the most likely cause is that
+ *                                     the caller specified a block past the
+ *                                     end of the file, but this could also be
+ *                                     any other error from VNOP_OFFTOBLK().
+ *
+ * Note:       Representing the error in band loses some information, but does
+ *             not occlude a valid block number, since block numbers exceed
+ *             the valid range for offsets, due to their relative sizes.  If
+ *             we had a more reliable constant than -1 in our header files
+ *             for it (i.e. explicitly cast to an daddr64_t), we would use it
+ *             here instead.
+ */
 daddr64_t
 ubc_offtoblk(vnode_t vp, off_t offset)
 {
-       daddr64_t blkno;
+       daddr64_t blkno = -1;
        int error = 0;
 
-       if (UBCINVALID(vp))
-               return ((daddr64_t)-1);
-
-       error = VNOP_OFFTOBLK(vp, offset, &blkno);
-       if (error)
-               blkno = -1;
+       if (UBCINFOEXISTS(vp)) {
+               error = VNOP_OFFTOBLK(vp, offset, &blkno);
+               if (error)
+                       blkno = -1;
+       }
 
        return (blkno);
 }
 
+
+/*
+ * ubc_pages_resident
+ *
+ * Determine whether or not a given vnode has pages resident via the memory
+ * object control associated with the ubc_info associated with the vnode
+ *
+ * Parameters: vp                      The vnode we want to know about
+ *
+ * Returns:    1                       Yes
+ *             0                       No
+ */
 int
 ubc_pages_resident(vnode_t vp)
 {
        kern_return_t           kret;
        boolean_t                       has_pages_resident;
        
-       if ( !UBCINFOEXISTS(vp))
+       if (!UBCINFOEXISTS(vp))
                return (0);
                        
+       /*
+        * The following call may fail if an invalid ui_control is specified,
+        * or if there is no VM object associated with the control object.  In
+        * either case, reacting to it as if there were no pages resident will
+        * result in correct behavior.
+        */
        kret = memory_object_pages_resident(vp->v_ubcinfo->ui_control, &has_pages_resident);
        
        if (kret != KERN_SUCCESS)
@@ -492,29 +1586,38 @@ ubc_pages_resident(vnode_t vp)
        return (0);
 }
 
-
-
 /*
- * This interface will eventually be deprecated
+ * ubc_msync
  *
- * clean and/or invalidate  a range in the memory object that backs this
- * vnode. The start offset is truncated to the page boundary and the
- * size is adjusted to include the last page in the range.
+ * Clean and/or invalidate a range in the memory object that backs this vnode
+ *
+ * Parameters: vp                      The vnode whose associated ubc_info's
+ *                                     associated memory object is to have a
+ *                                     range invalidated within it
+ *             beg_off                 The start of the range, as an offset
+ *             end_off                 The end of the range, as an offset
+ *             resid_off               The address of an off_t supplied by the
+ *                                     caller; may be set to NULL to ignore
+ *             flags                   See ubc_msync_internal()
+ *
+ * Returns:    0                       Success
+ *             !0                      Failure; an errno is returned
+ *
+ * Implicit Returns:
+ *             *resid_off, modified    If non-NULL, the  contents are ALWAYS
+ *                                     modified; they are initialized to the
+ *                                     beg_off, and in case of an I/O error,
+ *                                     the difference between beg_off and the
+ *                                     current value will reflect what was
+ *                                     able to be written before the error
+ *                                     occurred.  If no error is returned, the
+ *                                     value of the resid_off is undefined; do
+ *                                     NOT use it in place of end_off if you
+ *                                     intend to increment from the end of the
+ *                                     last call and call iteratively.
+ *
+ * Notes:      see ubc_msync_internal() for more detailed information.
  *
- * returns 1 for success,  0 for failure
- */
-int
-ubc_sync_range(vnode_t vp, off_t beg_off, off_t end_off, int flags)
-{
-        return (ubc_msync_internal(vp, beg_off, end_off, NULL, flags, NULL));
-}
-
-
-/*
- * clean and/or invalidate  a range in the memory object that backs this
- * vnode. The start offset is truncated to the page boundary and the
- * size is adjusted to include the last page in the range.
- * if a
  */
 errno_t
 ubc_msync(vnode_t vp, off_t beg_off, off_t end_off, off_t *resid_off, int flags)
@@ -533,11 +1636,68 @@ ubc_msync(vnode_t vp, off_t beg_off, off_t end_off, off_t *resid_off, int flags)
 }
 
 
-
 /*
- * clean and/or invalidate  a range in the memory object that backs this
- * vnode. The start offset is truncated to the page boundary and the
- * size is adjusted to include the last page in the range.
+ * ubc_msync_internal
+ *
+ * Clean and/or invalidate a range in the memory object that backs this vnode
+ *
+ * Parameters: vp                      The vnode whose associated ubc_info's
+ *                                     associated memory object is to have a
+ *                                     range invalidated within it
+ *             beg_off                 The start of the range, as an offset
+ *             end_off                 The end of the range, as an offset
+ *             resid_off               The address of an off_t supplied by the
+ *                                     caller; may be set to NULL to ignore
+ *             flags                   MUST contain at least one of the flags
+ *                                     UBC_INVALIDATE, UBC_PUSHDIRTY, or
+ *                                     UBC_PUSHALL; if UBC_PUSHDIRTY is used,
+ *                                     UBC_SYNC may also be specified to cause
+ *                                     this function to block until the
+ *                                     operation is complete.  The behavior
+ *                                     of UBC_SYNC is otherwise undefined.
+ *             io_errno                The address of an int to contain the
+ *                                     errno from a failed I/O operation, if
+ *                                     one occurs; may be set to NULL to
+ *                                     ignore
+ *
+ * Returns:    1                       Success
+ *             0                       Failure
+ *
+ * Implicit Returns:
+ *             *resid_off, modified    The contents of this offset MAY be
+ *                                     modified; in case of an I/O error, the
+ *                                     difference between beg_off and the
+ *                                     current value will reflect what was
+ *                                     able to be written before the error
+ *                                     occurred.
+ *             *io_errno, modified     The contents of this offset are set to
+ *                                     an errno, if an error occurs; if the
+ *                                     caller supplies an io_errno parameter,
+ *                                     they should be careful to initialize it
+ *                                     to 0 before calling this function to
+ *                                     enable them to distinguish an error
+ *                                     with a valid *resid_off from an invalid
+ *                                     one, and to avoid potentially falsely
+ *                                     reporting an error, depending on use.
+ *
+ * Notes:      If there is no ubc_info associated with the vnode supplied,
+ *             this function immediately returns success.
+ *
+ *             If the value of end_off is less than or equal to beg_off, this
+ *             function immediately returns success; that is, end_off is NOT
+ *             inclusive.
+ *
+ *             IMPORTANT: one of the flags UBC_INVALIDATE, UBC_PUSHDIRTY, or
+ *             UBC_PUSHALL MUST be specified; that is, it is NOT possible to
+ *             attempt to block on in-progress I/O by calling this function
+ *             with UBC_PUSHDIRTY, and then later call it with just UBC_SYNC
+ *             in order to block pending on the I/O already in progress.
+ *
+ *             The start offset is truncated to the page boundary and the
+ *             size is adjusted to include the last page in the range; that
+ *             is, end_off on exactly a page boundary will not change if it
+ *             is rounded, and the range of bytes written will be from the
+ *             truncate beg_off to the rounded (end_off - 1).
  */
 static int
 ubc_msync_internal(vnode_t vp, off_t beg_off, off_t end_off, off_t *resid_off, int flags, int *io_errno)
@@ -549,10 +1709,10 @@ ubc_msync_internal(vnode_t vp, off_t beg_off, off_t end_off, off_t *resid_off, i
        
        if ( !UBCINFOEXISTS(vp))
                return (0);
-       if (end_off <= beg_off)
-               return (0);
        if ((flags & (UBC_INVALIDATE | UBC_PUSHDIRTY | UBC_PUSHALL)) == 0)
                return (0);
+       if (end_off <= beg_off)
+               return (1);
 
        if (flags & UBC_INVALIDATE)
                /*
@@ -574,8 +1734,8 @@ ubc_msync_internal(vnode_t vp, off_t beg_off, off_t end_off, off_t *resid_off, i
 
        if (flags & UBC_PUSHALL)
                /*
-                * then return all the interesting pages in the range (both dirty and precious)
-                * to the pager
+                * then return all the interesting pages in the range (both
+                * dirty and precious) to the pager
                 */
                flush_flags = MEMORY_OBJECT_RETURN_ALL;
 
@@ -585,15 +1745,64 @@ ubc_msync_internal(vnode_t vp, off_t beg_off, off_t end_off, off_t *resid_off, i
 
        /* flush and/or invalidate pages in the range requested */
        kret = memory_object_lock_request(vp->v_ubcinfo->ui_control,
-                                         beg_off, tsize, resid_off, io_errno,
-                                         flush_flags, request_flags, VM_PROT_NO_CHANGE);
+                                         beg_off, tsize,
+                                         (memory_object_offset_t *)resid_off,
+                                         io_errno, flush_flags, request_flags,
+                                         VM_PROT_NO_CHANGE);
        
        return ((kret == KERN_SUCCESS) ? 1 : 0);
 }
 
 
 /*
- * The vnode is mapped explicitly, mark it so.
+ * ubc_map
+ *
+ * Explicitly map a vnode that has an associate ubc_info, and add a reference
+ * to it for the ubc system, if there isn't one already, so it will not be
+ * recycled while it's in use, and set flags on the ubc_info to indicate that
+ * we have done this
+ *
+ * Parameters: vp                      The vnode to map
+ *             flags                   The mapping flags for the vnode; this
+ *                                     will be a combination of one or more of
+ *                                     PROT_READ, PROT_WRITE, and PROT_EXEC
+ *
+ * Returns:    0                       Success
+ *             EPERM                   Permission was denied
+ *
+ * Notes:      An I/O reference on the vnode must already be held on entry
+ *
+ *             If there is no ubc_info associated with the vnode, this function
+ *             will return success.
+ *
+ *             If a permission error occurs, this function will return
+ *             failure; all other failures will cause this function to return
+ *             success.
+ *
+ *             IMPORTANT: This is an internal use function, and its symbols
+ *             are not exported, hence its error checking is not very robust.
+ *             It is primarily used by:
+ *
+ *             o       mmap(), when mapping a file
+ *             o       When mapping a shared file (a shared library in the
+ *                     shared segment region)
+ *             o       When loading a program image during the exec process
+ *
+ *             ...all of these uses ignore the return code, and any fault that
+ *             results later because of a failure is handled in the fix-up path
+ *             of the fault handler.  The interface exists primarily as a
+ *             performance hint.
+ *
+ *             Given that third party implementation of the type of interfaces
+ *             that would use this function, such as alternative executable
+ *             formats, etc., are unsupported, this function is not exported
+ *             for general use.
+ *
+ *             The extra reference is held until the VM system unmaps the
+ *             vnode from its own context to maintain a vnode reference in
+ *             cases like open()/mmap()/close(), which leave the backing
+ *             object referenced by a mapped memory region in a process
+ *             address space.
  */
 __private_extern__ int
 ubc_map(vnode_t vp, int flags)
@@ -601,103 +1810,138 @@ ubc_map(vnode_t vp, int flags)
        struct ubc_info *uip;
        int error = 0;
        int need_ref = 0;
-       struct vfs_context context;
-
-       if (vnode_getwithref(vp))
-               return (0);
+       int need_wakeup = 0;
 
        if (UBCINFOEXISTS(vp)) {
-               context.vc_proc = current_proc();
-               context.vc_ucred = kauth_cred_get();
 
-               error = VNOP_MMAP(vp, flags, &context);
+               vnode_lock(vp);
+               uip = vp->v_ubcinfo;
 
-               if (error != EPERM)
-                       error = 0;
+               while (ISSET(uip->ui_flags, UI_MAPBUSY)) {
+                       SET(uip->ui_flags, UI_MAPWAITING);
+                       (void) msleep(&uip->ui_flags, &vp->v_lock,
+                                     PRIBIO, "ubc_map", NULL);
+               }
+               SET(uip->ui_flags, UI_MAPBUSY);
+               vnode_unlock(vp);
 
-               if (error == 0) {
-                       vnode_lock(vp);
-                       
-                       uip = vp->v_ubcinfo;
+               error = VNOP_MMAP(vp, flags, vfs_context_current());
+
+               /*
+                * rdar://problem/22587101 required that we stop propagating
+                * EPERM up the stack. Otherwise, we would have to funnel up 
+                * the error at all the call sites for memory_object_map().
+                * The risk is in having to undo the map/object/entry state at 
+                * all these call sites. It would also affect more than just mmap()
+                * e.g. vm_remap().
+                *
+                *      if (error != EPERM)
+                *              error = 0;
+                */
+
+               error = 0;
+
+               vnode_lock_spin(vp);
 
+               if (error == 0) {
                        if ( !ISSET(uip->ui_flags, UI_ISMAPPED))
                                need_ref = 1;
                        SET(uip->ui_flags, (UI_WASMAPPED | UI_ISMAPPED));
+                       if (flags & PROT_WRITE) {
+                               SET(uip->ui_flags, UI_MAPPEDWRITE);
+                       }
+               }
+               CLR(uip->ui_flags, UI_MAPBUSY);
 
-                       vnode_unlock(vp);
-                       
-                       if (need_ref)
-                               vnode_ref(vp);
+               if (ISSET(uip->ui_flags, UI_MAPWAITING)) {
+                       CLR(uip->ui_flags, UI_MAPWAITING);
+                       need_wakeup = 1;
                }
-       }
-       vnode_put(vp);
+               vnode_unlock(vp);
 
+               if (need_wakeup)
+                       wakeup(&uip->ui_flags);
+
+               if (need_ref) {
+                       /*
+                        * Make sure we get a ref as we can't unwind from here
+                        */
+                       if (vnode_ref_ext(vp, 0, VNODE_REF_FORCE))
+                               panic("%s : VNODE_REF_FORCE failed\n", __FUNCTION__);
+               }
+       }
        return (error);
 }
 
+
 /*
- * destroy the named reference for a given vnode
+ * ubc_destroy_named
+ *
+ * Destroy the named memory object associated with the ubc_info control object
+ * associated with the designated vnode, if there is a ubc_info associated
+ * with the vnode, and a control object is associated with it
+ *
+ * Parameters: vp                      The designated vnode
+ *
+ * Returns:    (void)
+ *
+ * Notes:      This function is called on vnode termination for all vnodes,
+ *             and must therefore not assume that there is a ubc_info that is
+ *             associated with the vnode, nor that there is a control object
+ *             associated with the ubc_info.
+ *
+ *             If all the conditions necessary are present, this function
+ *             calls memory_object_destory(), which will in turn end up
+ *             calling ubc_unmap() to release any vnode references that were
+ *             established via ubc_map().
+ *
+ *             IMPORTANT: This is an internal use function that is used
+ *             exclusively by the internal use function vclean().
  */
-__private_extern__ int
-ubc_destroy_named(struct vnode *vp)
+__private_extern__ void
+ubc_destroy_named(vnode_t vp)
 {
        memory_object_control_t control;
        struct ubc_info *uip;
        kern_return_t kret;
 
-       /*
-        * We may already have had the object terminated
-        * and the ubcinfo released as a side effect of
-        * some earlier processing.  If so, pretend we did
-        * it, because it probably was a result of our
-        * efforts.
-        */
-       if (!UBCINFOEXISTS(vp))
-               return (1);
-
-       uip = vp->v_ubcinfo;
-
-       /* 
-        * Terminate the memory object.
-        * memory_object_destroy() will result in
-        * vnode_pager_no_senders(). 
-        * That will release the pager reference
-        * and the vnode will move to the free list.
-        */
-       control = ubc_getobject(vp, UBC_HOLDOBJECT);
-       if (control != MEMORY_OBJECT_CONTROL_NULL) {
-
-         /*
-          * XXXXX - should we hold the vnode lock here?
-          */
-               if (ISSET(vp->v_flag, VTERMINATE))
-                       panic("ubc_destroy_named: already teminating");
-               SET(vp->v_flag, VTERMINATE);
-
-               kret = memory_object_destroy(control, 0);
-               if (kret != KERN_SUCCESS)
-                       return (0);
-
-               /* 
-                * memory_object_destroy() is asynchronous
-                * with respect to vnode_pager_no_senders().
-                * wait for vnode_pager_no_senders() to clear
-                * VTERMINATE
-                */
-               vnode_lock(vp);
-               while (ISSET(vp->v_lflag, VNAMED_UBC)) {
-                       (void)msleep((caddr_t)&vp->v_lflag, &vp->v_lock,
-                                                PINOD, "ubc_destroy_named", 0);
+       if (UBCINFOEXISTS(vp)) {
+               uip = vp->v_ubcinfo;
+
+               /* Terminate the memory object  */
+               control = ubc_getobject(vp, UBC_HOLDOBJECT);
+               if (control != MEMORY_OBJECT_CONTROL_NULL) {
+                       kret = memory_object_destroy(control, 0);
+                       if (kret != KERN_SUCCESS)
+                               panic("ubc_destroy_named: memory_object_destroy failed");
                }
-               vnode_unlock(vp);
        }
-       return (1);
 }
 
 
 /*
- * Find out whether a vnode is in use by UBC
- * Returns 1 if file is in use by UBC, 0 if not
+ * ubc_isinuse
+ *
+ * Determine whether or not a vnode is currently in use by ubc at a level in
+ * excess of the requested busycount
+ *
+ * Parameters: vp                      The vnode to check
+ *             busycount               The threshold busy count, used to bias
+ *                                     the count usually already held by the
+ *                                     caller to avoid races
+ *
+ * Returns:    1                       The vnode is in use over the threshold
+ *             0                       The vnode is not in use over the
+ *                                     threshold
+ *
+ * Notes:      Because the vnode is only held locked while actually asking
+ *             the use count, this function only represents a snapshot of the
+ *             current state of the vnode.  If more accurate information is
+ *             required, an additional busycount should be held by the caller
+ *             and a non-zero busycount used.
+ *
+ *             If there is no ubc_info associated with the vnode, this
+ *             function will report that the vnode is not in use by ubc.
  */
 int
 ubc_isinuse(struct vnode *vp, int busycount)
@@ -708,6 +1952,34 @@ ubc_isinuse(struct vnode *vp, int busycount)
 }
 
 
+/*
+ * ubc_isinuse_locked
+ *
+ * Determine whether or not a vnode is currently in use by ubc at a level in
+ * excess of the requested busycount
+ *
+ * Parameters: vp                      The vnode to check
+ *             busycount               The threshold busy count, used to bias
+ *                                     the count usually already held by the
+ *                                     caller to avoid races
+ *             locked                  True if the vnode is already locked by
+ *                                     the caller
+ *
+ * Returns:    1                       The vnode is in use over the threshold
+ *             0                       The vnode is not in use over the
+ *                                     threshold
+ *
+ * Notes:      If the vnode is not locked on entry, it is locked while
+ *             actually asking the use count.  If this is the case, this
+ *             function only represents a snapshot of the current state of
+ *             the vnode.  If more accurate information is required, the
+ *             vnode lock should be held by the caller, otherwise an
+ *             additional busycount should be held by the caller and a
+ *             non-zero busycount used.
+ *
+ *             If there is no ubc_info associated with the vnode, this
+ *             function will report that the vnode is not in use by ubc.
+ */
 int
 ubc_isinuse_locked(struct vnode *vp, int busycount, int locked)
 {
@@ -715,7 +1987,7 @@ ubc_isinuse_locked(struct vnode *vp, int busycount, int locked)
 
 
        if (!locked)
-               vnode_lock(vp);
+               vnode_lock_spin(vp);
 
        if ((vp->v_usecount - vp->v_kusecount) > busycount)
                retval = 1;
@@ -727,35 +1999,105 @@ ubc_isinuse_locked(struct vnode *vp, int busycount, int locked)
 
 
 /*
- * MUST only be called by the VM
+ * ubc_unmap
+ *
+ * Reverse the effects of a ubc_map() call for a given vnode
+ *
+ * Parameters: vp                      vnode to unmap from ubc
+ *
+ * Returns:    (void)
+ *
+ * Notes:      This is an internal use function used by vnode_pager_unmap().
+ *             It will attempt to obtain a reference on the supplied vnode,
+ *             and if it can do so, and there is an associated ubc_info, and
+ *             the flags indicate that it was mapped via ubc_map(), then the
+ *             flag is cleared, the mapping removed, and the reference taken
+ *             by ubc_map() is released.
+ *
+ *             IMPORTANT: This MUST only be called by the VM
+ *             to prevent race conditions.
  */
 __private_extern__ void
 ubc_unmap(struct vnode *vp)
 {
-       struct vfs_context context;
        struct ubc_info *uip;
        int     need_rele = 0;
+       int     need_wakeup = 0;
 
        if (vnode_getwithref(vp))
                return;
 
        if (UBCINFOEXISTS(vp)) {
-               vnode_lock(vp);
+               bool want_fsevent = false;
 
+               vnode_lock(vp);
                uip = vp->v_ubcinfo;
+
+               while (ISSET(uip->ui_flags, UI_MAPBUSY)) {
+                       SET(uip->ui_flags, UI_MAPWAITING);
+                       (void) msleep(&uip->ui_flags, &vp->v_lock,
+                                     PRIBIO, "ubc_unmap", NULL);
+               }
+               SET(uip->ui_flags, UI_MAPBUSY);
+
                if (ISSET(uip->ui_flags, UI_ISMAPPED)) {
-                       CLR(uip->ui_flags, UI_ISMAPPED);
+                       if (ISSET(uip->ui_flags, UI_MAPPEDWRITE))
+                               want_fsevent = true;
+
                        need_rele = 1;
+
+                       /*
+                        * We want to clear the mapped flags after we've called
+                        * VNOP_MNOMAP to avoid certain races and allow
+                        * VNOP_MNOMAP to call ubc_is_mapped_writable.
+                        */
                }
                vnode_unlock(vp);
-               
+
                if (need_rele) {
-                       context.vc_proc = current_proc();
-                       context.vc_ucred = kauth_cred_get();
-                       (void)VNOP_MNOMAP(vp, &context);
+                               vfs_context_t ctx = vfs_context_current();
+
+                       (void)VNOP_MNOMAP(vp, ctx);
+
+#if CONFIG_FSE
+                               /*
+                                * Why do we want an fsevent here?  Normally the
+                                * content modified fsevent is posted when a file is
+                                * closed and only if it's written to via conventional
+                                * means.  It's perfectly legal to close a file and
+                                * keep your mappings and we don't currently track
+                                * whether it was written to via a mapping.
+                                * Therefore, we need to post an fsevent here if the
+                                * file was mapped writable.  This may result in false
+                                * events, i.e. we post a notification when nothing
+                                * has really changed.
+                                */
+                               if (want_fsevent && need_fsevent(FSE_CONTENT_MODIFIED, vp)) {
+                                       add_fsevent(FSE_CONTENT_MODIFIED, ctx,
+                                                               FSE_ARG_VNODE, vp,
+                                                               FSE_ARG_DONE);
+                               }
+#endif
 
                        vnode_rele(vp);
                }
+
+               vnode_lock_spin(vp);
+
+               if (need_rele)
+                       CLR(uip->ui_flags, UI_ISMAPPED | UI_MAPPEDWRITE);
+
+               CLR(uip->ui_flags, UI_MAPBUSY);
+
+               if (ISSET(uip->ui_flags, UI_MAPWAITING)) {
+                       CLR(uip->ui_flags, UI_MAPWAITING);
+                       need_wakeup = 1;
+               }
+               vnode_unlock(vp);
+
+               if (need_wakeup)
+                       wakeup(&uip->ui_flags);
+
        }
        /*
         * the drop of the vnode ref will cleanup
@@ -763,6 +2105,82 @@ ubc_unmap(struct vnode *vp)
        vnode_put(vp);
 }
 
+
+/*
+ * ubc_page_op
+ *
+ * Manipulate individual page state for a vnode with an associated ubc_info
+ * with an associated memory object control.
+ *
+ * Parameters: vp                      The vnode backing the page
+ *             f_offset                A file offset interior to the page
+ *             ops                     The operations to perform, as a bitmap
+ *                                     (see below for more information)
+ *             phys_entryp             The address of a ppnum_t; may be NULL
+ *                                     to ignore
+ *             flagsp                  A pointer to an int to contain flags;
+ *                                     may be NULL to ignore
+ *
+ * Returns:    KERN_SUCCESS            Success
+ *             KERN_INVALID_ARGUMENT   If the memory object control has no VM
+ *                                     object associated
+ *             KERN_INVALID_OBJECT     If UPL_POP_PHYSICAL and the object is
+ *                                     not physically contiguous
+ *             KERN_INVALID_OBJECT     If !UPL_POP_PHYSICAL and the object is
+ *                                     physically contiguous
+ *             KERN_FAILURE            If the page cannot be looked up
+ *
+ * Implicit Returns:
+ *             *phys_entryp (modified) If phys_entryp is non-NULL and
+ *                                     UPL_POP_PHYSICAL
+ *             *flagsp (modified)      If flagsp is non-NULL and there was
+ *                                     !UPL_POP_PHYSICAL and a KERN_SUCCESS
+ *
+ * Notes:      For object boundaries, it is considerably more efficient to
+ *             ensure that f_offset is in fact on a page boundary, as this
+ *             will avoid internal use of the hash table to identify the
+ *             page, and would therefore skip a number of early optimizations.
+ *             Since this is a page operation anyway, the caller should try
+ *             to pass only a page aligned offset because of this.
+ *
+ *             *flagsp may be modified even if this function fails.  If it is
+ *             modified, it will contain the condition of the page before the
+ *             requested operation was attempted; these will only include the
+ *             bitmap flags, and not the PL_POP_PHYSICAL, UPL_POP_DUMP,
+ *             UPL_POP_SET, or UPL_POP_CLR bits.
+ *
+ *             The flags field may contain a specific operation, such as
+ *             UPL_POP_PHYSICAL or UPL_POP_DUMP:
+ *
+ *             o       UPL_POP_PHYSICAL        Fail if not contiguous; if
+ *                                             *phys_entryp and successful, set
+ *                                             *phys_entryp
+ *             o       UPL_POP_DUMP            Dump the specified page
+ *
+ *             Otherwise, it is treated as a bitmap of one or more page
+ *             operations to perform on the final memory object; allowable
+ *             bit values are:
+ *
+ *             o       UPL_POP_DIRTY           The page is dirty
+ *             o       UPL_POP_PAGEOUT         The page is paged out
+ *             o       UPL_POP_PRECIOUS        The page is precious
+ *             o       UPL_POP_ABSENT          The page is absent
+ *             o       UPL_POP_BUSY            The page is busy
+ *
+ *             If the page status is only being queried and not modified, then
+ *             not other bits should be specified.  However, if it is being
+ *             modified, exactly ONE of the following bits should be set:
+ *
+ *             o       UPL_POP_SET             Set the current bitmap bits
+ *             o       UPL_POP_CLR             Clear the current bitmap bits
+ *
+ *             Thus to effect a combination of setting an clearing, it may be
+ *             necessary to call this function twice.  If this is done, the
+ *             set should be used before the clear, since clearing may trigger
+ *             a wakeup on the destination page, and if the page is backed by
+ *             an encrypted swap file, setting will trigger the decryption
+ *             needed before the wakeup occurs.
+ */
 kern_return_t
 ubc_page_op(
        struct vnode    *vp,
@@ -783,22 +2201,66 @@ ubc_page_op(
                                      phys_entryp,
                                      flagsp));
 }
-                                     
-__private_extern__ kern_return_t
-ubc_page_op_with_control(
-       memory_object_control_t  control,
-       off_t                    f_offset,
-       int                      ops,
-       ppnum_t                  *phys_entryp,
-       int                      *flagsp)
-{
-       return (memory_object_page_op(control,
-                                     (memory_object_offset_t)f_offset,
-                                     ops,
-                                     phys_entryp,
-                                     flagsp));
-}
-                                     
+
+
+/*
+ * ubc_range_op
+ *
+ * Manipulate page state for a range of memory for a vnode with an associated
+ * ubc_info with an associated memory object control, when page level state is
+ * not required to be returned from the call (i.e. there are no phys_entryp or
+ * flagsp parameters to this call, and it takes a range which may contain
+ * multiple pages, rather than an offset interior to a single page).
+ *
+ * Parameters: vp                      The vnode backing the page
+ *             f_offset_beg            A file offset interior to the start page
+ *             f_offset_end            A file offset interior to the end page
+ *             ops                     The operations to perform, as a bitmap
+ *                                     (see below for more information)
+ *             range                   The address of an int; may be NULL to
+ *                                     ignore
+ *
+ * Returns:    KERN_SUCCESS            Success
+ *             KERN_INVALID_ARGUMENT   If the memory object control has no VM
+ *                                     object associated
+ *             KERN_INVALID_OBJECT     If the object is physically contiguous
+ *
+ * Implicit Returns:
+ *             *range (modified)       If range is non-NULL, its contents will
+ *                                     be modified to contain the number of
+ *                                     bytes successfully operated upon.
+ *
+ * Notes:      IMPORTANT: This function cannot be used on a range that
+ *             consists of physically contiguous pages.
+ *
+ *             For object boundaries, it is considerably more efficient to
+ *             ensure that f_offset_beg and f_offset_end are in fact on page
+ *             boundaries, as this will avoid internal use of the hash table
+ *             to identify the page, and would therefore skip a number of
+ *             early optimizations.  Since this is an operation on a set of
+ *             pages anyway, the caller should try to pass only a page aligned
+ *             offsets because of this.
+ *
+ *             *range will be modified only if this function succeeds.
+ *
+ *             The flags field MUST contain a specific operation; allowable
+ *             values are:
+ *
+ *             o       UPL_ROP_ABSENT  Returns the extent of the range
+ *                                     presented which is absent, starting
+ *                                     with the start address presented
+ *
+ *             o       UPL_ROP_PRESENT Returns the extent of the range
+ *                                     presented which is present (resident),
+ *                                     starting with the start address
+ *                                     presented
+ *             o       UPL_ROP_DUMP    Dump the pages which are found in the
+ *                                     target object for the target range.
+ *
+ *             IMPORTANT: For UPL_ROP_ABSENT and UPL_ROP_PRESENT; if there are
+ *             multiple regions in the range, only the first matching region
+ *             is returned.
+ */
 kern_return_t
 ubc_range_op(
        struct vnode    *vp,
@@ -819,48 +2281,166 @@ ubc_range_op(
                                      ops,
                                      range));
 }
-                                     
+
+
+/*
+ * ubc_create_upl
+ *
+ * Given a vnode, cause the population of a portion of the vm_object; based on
+ * the nature of the request, the pages returned may contain valid data, or
+ * they may be uninitialized.
+ *
+ * Parameters: vp                      The vnode from which to create the upl
+ *             f_offset                The start offset into the backing store
+ *                                     represented by the vnode
+ *             bufsize                 The size of the upl to create
+ *             uplp                    Pointer to the upl_t to receive the
+ *                                     created upl; MUST NOT be NULL
+ *             plp                     Pointer to receive the internal page
+ *                                     list for the created upl; MAY be NULL
+ *                                     to ignore
+ *
+ * Returns:    KERN_SUCCESS            The requested upl has been created
+ *             KERN_INVALID_ARGUMENT   The bufsize argument is not an even
+ *                                     multiple of the page size
+ *             KERN_INVALID_ARGUMENT   There is no ubc_info associated with
+ *                                     the vnode, or there is no memory object
+ *                                     control associated with the ubc_info 
+ *     memory_object_upl_request:KERN_INVALID_VALUE
+ *                                     The supplied upl_flags argument is
+ *                                     invalid
+ * Implicit Returns:
+ *             *uplp (modified)        
+ *             *plp (modified)         If non-NULL, the value of *plp will be
+ *                                     modified to point to the internal page
+ *                                     list; this modification may occur even
+ *                                     if this function is unsuccessful, in
+ *                                     which case the contents may be invalid
+ *
+ * Note:       If successful, the returned *uplp MUST subsequently be freed
+ *             via a call to ubc_upl_commit(), ubc_upl_commit_range(),
+ *             ubc_upl_abort(), or ubc_upl_abort_range().
+ */
+kern_return_t
+ubc_create_upl_external(
+       struct vnode    *vp,
+       off_t           f_offset,
+       int             bufsize,
+       upl_t           *uplp,
+       upl_page_info_t **plp,
+       int             uplflags)
+{
+    return (ubc_create_upl_kernel(vp, f_offset, bufsize, uplp, plp, uplflags, vm_tag_bt()));
+}
+
 kern_return_t
-ubc_create_upl(
+ubc_create_upl_kernel(
        struct vnode    *vp,
-       off_t                   f_offset,
-       long                    bufsize,
-       upl_t                   *uplp,
+       off_t           f_offset,
+       int             bufsize,
+       upl_t           *uplp,
        upl_page_info_t **plp,
-       int                             uplflags)
+       int             uplflags,
+       vm_tag_t tag)
 {
        memory_object_control_t         control;
-       int                             count;
-       int                             ubcflags;
        kern_return_t                   kr;
+
+       if (plp != NULL)
+               *plp = NULL;
+       *uplp = NULL;
        
        if (bufsize & 0xfff)
                return KERN_INVALID_ARGUMENT;
 
-       if (uplflags & UPL_FOR_PAGEOUT) {
+       if (bufsize > MAX_UPL_SIZE_BYTES)
+               return KERN_INVALID_ARGUMENT;
+
+       if (uplflags & (UPL_UBC_MSYNC | UPL_UBC_PAGEOUT | UPL_UBC_PAGEIN)) {
+
+               if (uplflags & UPL_UBC_MSYNC) {
+                       uplflags &= UPL_RET_ONLY_DIRTY;
+
+                       uplflags |= UPL_COPYOUT_FROM | UPL_CLEAN_IN_PLACE |
+                                   UPL_SET_INTERNAL | UPL_SET_LITE;
+
+               } else if (uplflags & UPL_UBC_PAGEOUT) {
+                       uplflags &= UPL_RET_ONLY_DIRTY;
+
+                       if (uplflags & UPL_RET_ONLY_DIRTY)
+                               uplflags |= UPL_NOBLOCK;
+
+                       uplflags |= UPL_FOR_PAGEOUT | UPL_CLEAN_IN_PLACE |
+                                    UPL_COPYOUT_FROM | UPL_SET_INTERNAL | UPL_SET_LITE;
+               } else {
+                       uplflags |= UPL_RET_ONLY_ABSENT |
+                                   UPL_NO_SYNC | UPL_CLEAN_IN_PLACE |
+                                   UPL_SET_INTERNAL | UPL_SET_LITE;
+
+                       /*
+                        * if the requested size == PAGE_SIZE, we don't want to set
+                        * the UPL_NOBLOCK since we may be trying to recover from a
+                        * previous partial pagein I/O that occurred because we were low
+                        * on memory and bailed early in order to honor the UPL_NOBLOCK...
+                        * since we're only asking for a single page, we can block w/o fear
+                        * of tying up pages while waiting for more to become available
+                        */
+                       if (bufsize > PAGE_SIZE)
+                               uplflags |= UPL_NOBLOCK;
+               }
+       } else {
                uplflags &= ~UPL_FOR_PAGEOUT;
-               ubcflags  =  UBC_FOR_PAGEOUT;
-       } else
-               ubcflags = UBC_FLAGS_NONE;
 
-       control = ubc_getobject(vp, ubcflags);
+               if (uplflags & UPL_WILL_BE_DUMPED) {
+                       uplflags &= ~UPL_WILL_BE_DUMPED;
+                       uplflags |= (UPL_NO_SYNC|UPL_SET_INTERNAL);
+               } else
+                       uplflags |= (UPL_NO_SYNC|UPL_CLEAN_IN_PLACE|UPL_SET_INTERNAL);
+       }
+       control = ubc_getobject(vp, UBC_FLAGS_NONE);
        if (control == MEMORY_OBJECT_CONTROL_NULL)
                return KERN_INVALID_ARGUMENT;
 
-       if (uplflags & UPL_WILL_BE_DUMPED) {
-               uplflags &= ~UPL_WILL_BE_DUMPED;
-               uplflags |= (UPL_NO_SYNC|UPL_SET_INTERNAL);
-       } else
-               uplflags |= (UPL_NO_SYNC|UPL_CLEAN_IN_PLACE|UPL_SET_INTERNAL);
-       count = 0;
-       kr = memory_object_upl_request(control, f_offset, bufsize,
-                                                                  uplp, NULL, &count, uplflags);
-       if (plp != NULL)
-                       *plp = UPL_GET_INTERNAL_PAGE_LIST(*uplp);
+       kr = memory_object_upl_request(control, f_offset, bufsize, uplp, NULL, NULL, uplflags, tag);
+       if (kr == KERN_SUCCESS && plp != NULL)
+               *plp = UPL_GET_INTERNAL_PAGE_LIST(*uplp);
        return kr;
 }
-                                     
+               
+                                                             
+/*
+ * ubc_upl_maxbufsize
+ *
+ * Return the maximum bufsize ubc_create_upl( ) will take.
+ *
+ * Parameters: none
+ *
+ * Returns:    maximum size buffer (in bytes) ubc_create_upl( ) will take.
+ */
+upl_size_t 
+ubc_upl_maxbufsize(
+       void)
+{
+       return(MAX_UPL_SIZE_BYTES);
+}
 
+/*
+ * ubc_upl_map
+ *
+ * Map the page list assocated with the supplied upl into the kernel virtual
+ * address space at the virtual address indicated by the dst_addr argument;
+ * the entire upl is mapped
+ *
+ * Parameters: upl                     The upl to map
+ *             dst_addr                The address at which to map the upl
+ *
+ * Returns:    KERN_SUCCESS            The upl has been mapped
+ *             KERN_INVALID_ARGUMENT   The upl is UPL_NULL
+ *             KERN_FAILURE            The upl is already mapped
+ *     vm_map_enter:KERN_INVALID_ARGUMENT
+ *                                     A failure code from vm_map_enter() due
+ *                                     to an invalid argument
+ */
 kern_return_t
 ubc_upl_map(
        upl_t           upl,
@@ -870,6 +2450,18 @@ ubc_upl_map(
 }
 
 
+/*
+ * ubc_upl_unmap
+ *
+ * Unmap the page list assocated with the supplied upl from the kernel virtual
+ * address space; the entire upl is unmapped.
+ *
+ * Parameters: upl                     The upl to unmap
+ *
+ * Returns:    KERN_SUCCESS            The upl has been unmapped
+ *             KERN_FAILURE            The upl is not currently mapped
+ *             KERN_INVALID_ARGUMENT   If the upl is UPL_NULL
+ */
 kern_return_t
 ubc_upl_unmap(
        upl_t   upl)
@@ -877,6 +2469,30 @@ ubc_upl_unmap(
        return(vm_upl_unmap(kernel_map, upl));
 }
 
+
+/*
+ * ubc_upl_commit
+ *
+ * Commit the contents of the upl to the backing store
+ *
+ * Parameters: upl                     The upl to commit
+ *
+ * Returns:    KERN_SUCCESS            The upl has been committed
+ *             KERN_INVALID_ARGUMENT   The supplied upl was UPL_NULL
+ *             KERN_FAILURE            The supplied upl does not represent
+ *                                     device memory, and the offset plus the
+ *                                     size would exceed the actual size of
+ *                                     the upl
+ *
+ * Notes:      In practice, the only return value for this function should be
+ *             KERN_SUCCESS, unless there has been data structure corruption;
+ *             since the upl is deallocated regardless of success or failure,
+ *             there's really nothing to do about this other than panic.
+ *
+ *             IMPORTANT: Use of this function should not be mixed with use of
+ *             ubc_upl_commit_range(), due to the unconditional deallocation
+ *             by this function.
+ */
 kern_return_t
 ubc_upl_commit(
        upl_t                   upl)
@@ -885,17 +2501,69 @@ ubc_upl_commit(
        kern_return_t   kr;
 
        pl = UPL_GET_INTERNAL_PAGE_LIST(upl);
-       kr = upl_commit(upl, pl, MAX_UPL_TRANSFER);
+       kr = upl_commit(upl, pl, MAX_UPL_SIZE_BYTES >> PAGE_SHIFT);
        upl_deallocate(upl);
        return kr;
 }
 
 
+/*
+ * ubc_upl_commit
+ *
+ * Commit the contents of the specified range of the upl to the backing store
+ *
+ * Parameters: upl                     The upl to commit
+ *             offset                  The offset into the upl
+ *             size                    The size of the region to be committed,
+ *                                     starting at the specified offset
+ *             flags                   commit type (see below)
+ *
+ * Returns:    KERN_SUCCESS            The range has been committed
+ *             KERN_INVALID_ARGUMENT   The supplied upl was UPL_NULL
+ *             KERN_FAILURE            The supplied upl does not represent
+ *                                     device memory, and the offset plus the
+ *                                     size would exceed the actual size of
+ *                                     the upl
+ *
+ * Notes:      IMPORTANT: If the commit is successful, and the object is now
+ *             empty, the upl will be deallocated.  Since the caller cannot
+ *             check that this is the case, the UPL_COMMIT_FREE_ON_EMPTY flag
+ *             should generally only be used when the offset is 0 and the size
+ *             is equal to the upl size.
+ *
+ *             The flags argument is a bitmap of flags on the rage of pages in
+ *             the upl to be committed; allowable flags are:
+ *
+ *             o       UPL_COMMIT_FREE_ON_EMPTY        Free the upl when it is
+ *                                                     both empty and has been
+ *                                                     successfully committed
+ *             o       UPL_COMMIT_CLEAR_DIRTY          Clear each pages dirty
+ *                                                     bit; will prevent a
+ *                                                     later pageout
+ *             o       UPL_COMMIT_SET_DIRTY            Set each pages dirty
+ *                                                     bit; will cause a later
+ *                                                     pageout
+ *             o       UPL_COMMIT_INACTIVATE           Clear each pages
+ *                                                     reference bit; the page
+ *                                                     will not be accessed
+ *             o       UPL_COMMIT_ALLOW_ACCESS         Unbusy each page; pages
+ *                                                     become busy when an
+ *                                                     IOMemoryDescriptor is
+ *                                                     mapped or redirected,
+ *                                                     and we have to wait for
+ *                                                     an IOKit driver
+ *
+ *             The flag UPL_COMMIT_NOTIFY_EMPTY is used internally, and should
+ *             not be specified by the caller.
+ *
+ *             The UPL_COMMIT_CLEAR_DIRTY and UPL_COMMIT_SET_DIRTY flags are
+ *             mutually exclusive, and should not be combined.
+ */
 kern_return_t
 ubc_upl_commit_range(
        upl_t                   upl,
-       vm_offset_t             offset,
-       vm_size_t               size,
+       upl_offset_t            offset,
+       upl_size_t              size,
        int                             flags)
 {
        upl_page_info_t *pl;
@@ -905,22 +2573,72 @@ ubc_upl_commit_range(
        if (flags & UPL_COMMIT_FREE_ON_EMPTY)
                flags |= UPL_COMMIT_NOTIFY_EMPTY;
 
+       if (flags & UPL_COMMIT_KERNEL_ONLY_FLAGS) {
+               return KERN_INVALID_ARGUMENT;
+       }
+
        pl = UPL_GET_INTERNAL_PAGE_LIST(upl);
 
        kr = upl_commit_range(upl, offset, size, flags,
-                                                 pl, MAX_UPL_TRANSFER, &empty);
+                             pl, MAX_UPL_SIZE_BYTES >> PAGE_SHIFT, &empty);
 
        if((flags & UPL_COMMIT_FREE_ON_EMPTY) && empty)
                upl_deallocate(upl);
 
        return kr;
 }
-       
+
+
+/*
+ * ubc_upl_abort_range
+ *
+ * Abort the contents of the specified range of the specified upl
+ *
+ * Parameters: upl                     The upl to abort
+ *             offset                  The offset into the upl
+ *             size                    The size of the region to be aborted,
+ *                                     starting at the specified offset
+ *             abort_flags             abort type (see below)
+ *
+ * Returns:    KERN_SUCCESS            The range has been aborted
+ *             KERN_INVALID_ARGUMENT   The supplied upl was UPL_NULL
+ *             KERN_FAILURE            The supplied upl does not represent
+ *                                     device memory, and the offset plus the
+ *                                     size would exceed the actual size of
+ *                                     the upl
+ *
+ * Notes:      IMPORTANT: If the abort is successful, and the object is now
+ *             empty, the upl will be deallocated.  Since the caller cannot
+ *             check that this is the case, the UPL_ABORT_FREE_ON_EMPTY flag
+ *             should generally only be used when the offset is 0 and the size
+ *             is equal to the upl size.
+ *
+ *             The abort_flags argument is a bitmap of flags on the range of
+ *             pages in the upl to be aborted; allowable flags are:
+ *
+ *             o       UPL_ABORT_FREE_ON_EMPTY Free the upl when it is both
+ *                                             empty and has been successfully
+ *                                             aborted
+ *             o       UPL_ABORT_RESTART       The operation must be restarted
+ *             o       UPL_ABORT_UNAVAILABLE   The pages are unavailable
+ *             o       UPL_ABORT_ERROR         An I/O error occurred
+ *             o       UPL_ABORT_DUMP_PAGES    Just free the pages
+ *             o       UPL_ABORT_NOTIFY_EMPTY  RESERVED
+ *             o       UPL_ABORT_ALLOW_ACCESS  RESERVED
+ *
+ *             The UPL_ABORT_NOTIFY_EMPTY is an internal use flag and should
+ *             not be specified by the caller.  It is intended to fulfill the
+ *             same role as UPL_COMMIT_NOTIFY_EMPTY does in the function
+ *             ubc_upl_commit_range(), but is never referenced internally.
+ *
+ *             The UPL_ABORT_ALLOW_ACCESS is defined, but neither set nor
+ *             referenced; do not use it.
+ */
 kern_return_t
 ubc_upl_abort_range(
        upl_t                   upl,
-       vm_offset_t             offset,
-       vm_size_t               size,
+       upl_offset_t            offset,
+       upl_size_t              size,
        int                             abort_flags)
 {
        kern_return_t   kr;
@@ -937,6 +2655,49 @@ ubc_upl_abort_range(
        return kr;
 }
 
+
+/*
+ * ubc_upl_abort
+ *
+ * Abort the contents of the specified upl
+ *
+ * Parameters: upl                     The upl to abort
+ *             abort_type              abort type (see below)
+ *
+ * Returns:    KERN_SUCCESS            The range has been aborted
+ *             KERN_INVALID_ARGUMENT   The supplied upl was UPL_NULL
+ *             KERN_FAILURE            The supplied upl does not represent
+ *                                     device memory, and the offset plus the
+ *                                     size would exceed the actual size of
+ *                                     the upl
+ *
+ * Notes:      IMPORTANT: If the abort is successful, and the object is now
+ *             empty, the upl will be deallocated.  Since the caller cannot
+ *             check that this is the case, the UPL_ABORT_FREE_ON_EMPTY flag
+ *             should generally only be used when the offset is 0 and the size
+ *             is equal to the upl size.
+ *
+ *             The abort_type is a bitmap of flags on the range of
+ *             pages in the upl to be aborted; allowable flags are:
+ *
+ *             o       UPL_ABORT_FREE_ON_EMPTY Free the upl when it is both
+ *                                             empty and has been successfully
+ *                                             aborted
+ *             o       UPL_ABORT_RESTART       The operation must be restarted
+ *             o       UPL_ABORT_UNAVAILABLE   The pages are unavailable
+ *             o       UPL_ABORT_ERROR         An I/O error occurred
+ *             o       UPL_ABORT_DUMP_PAGES    Just free the pages
+ *             o       UPL_ABORT_NOTIFY_EMPTY  RESERVED
+ *             o       UPL_ABORT_ALLOW_ACCESS  RESERVED
+ *
+ *             The UPL_ABORT_NOTIFY_EMPTY is an internal use flag and should
+ *             not be specified by the caller.  It is intended to fulfill the
+ *             same role as UPL_COMMIT_NOTIFY_EMPTY does in the function
+ *             ubc_upl_commit_range(), but is never referenced internally.
+ *
+ *             The UPL_ABORT_ALLOW_ACCESS is defined, but neither set nor
+ *             referenced; do not use it.
+ */
 kern_return_t
 ubc_upl_abort(
        upl_t                   upl,
@@ -949,6 +2710,25 @@ ubc_upl_abort(
        return kr;
 }
 
+
+/*
+ * ubc_upl_pageinfo
+ *
+ *  Retrieve the internal page list for the specified upl
+ *
+ * Parameters: upl                     The upl to obtain the page list from
+ *
+ * Returns:    !NULL                   The (upl_page_info_t *) for the page
+ *                                     list internal to the upl
+ *             NULL                    Error/no page list associated
+ *
+ * Notes:      IMPORTANT: The function is only valid on internal objects
+ *             where the list request was made with the UPL_INTERNAL flag.
+ *
+ *             This function is a utility helper function, since some callers
+ *             may not have direct access to the header defining the macro,
+ *             due to abstraction layering constraints.
+ */
 upl_page_info_t *
 ubc_upl_pageinfo(
        upl_t                   upl)
@@ -956,45 +2736,1614 @@ ubc_upl_pageinfo(
        return (UPL_GET_INTERNAL_PAGE_LIST(upl));
 }
 
-/************* UBC APIS **************/
 
 int 
-UBCINFOMISSING(struct vnode * vp)
+UBCINFOEXISTS(const struct vnode * vp)
 {
-       return((vp) && ((vp)->v_type == VREG) && ((vp)->v_ubcinfo == UBC_INFO_NULL));
+        return((vp) && ((vp)->v_type == VREG) && ((vp)->v_ubcinfo != UBC_INFO_NULL));
 }
 
-int 
-UBCINFORECLAIMED(struct vnode * vp)
+
+void
+ubc_upl_range_needed(
+       upl_t           upl,
+       int             index,
+       int             count)
 {
-       return((vp) && ((vp)->v_type == VREG) && ((vp)->v_ubcinfo == UBC_INFO_NULL));
+       upl_range_needed(upl, index, count);
 }
 
+boolean_t ubc_is_mapped(const struct vnode *vp, boolean_t *writable)
+{
+       if (!UBCINFOEXISTS(vp) || !ISSET(vp->v_ubcinfo->ui_flags, UI_ISMAPPED))
+               return FALSE;
+       if (writable)
+               *writable = ISSET(vp->v_ubcinfo->ui_flags, UI_MAPPEDWRITE);
+       return TRUE;
+}
 
-int 
-UBCINFOEXISTS(struct vnode * vp)
+boolean_t ubc_is_mapped_writable(const struct vnode *vp)
 {
-        return((vp) && ((vp)->v_type == VREG) && ((vp)->v_ubcinfo != UBC_INFO_NULL));
+       boolean_t writable;
+       return ubc_is_mapped(vp, &writable) && writable;
 }
-int 
-UBCISVALID(struct vnode * vp)
+
+
+/*
+ * CODE SIGNING
+ */
+static volatile SInt32 cs_blob_size = 0;
+static volatile SInt32 cs_blob_count = 0;
+static SInt32 cs_blob_size_peak = 0;
+static UInt32 cs_blob_size_max = 0;
+static SInt32 cs_blob_count_peak = 0;
+
+SYSCTL_INT(_vm, OID_AUTO, cs_blob_count, CTLFLAG_RD | CTLFLAG_LOCKED, (int *)(uintptr_t)&cs_blob_count, 0, "Current number of code signature blobs");
+SYSCTL_INT(_vm, OID_AUTO, cs_blob_size, CTLFLAG_RD | CTLFLAG_LOCKED, (int *)(uintptr_t)&cs_blob_size, 0, "Current size of all code signature blobs");
+SYSCTL_INT(_vm, OID_AUTO, cs_blob_count_peak, CTLFLAG_RD | CTLFLAG_LOCKED, &cs_blob_count_peak, 0, "Peak number of code signature blobs");
+SYSCTL_INT(_vm, OID_AUTO, cs_blob_size_peak, CTLFLAG_RD | CTLFLAG_LOCKED, &cs_blob_size_peak, 0, "Peak size of code signature blobs");
+SYSCTL_INT(_vm, OID_AUTO, cs_blob_size_max, CTLFLAG_RD | CTLFLAG_LOCKED, &cs_blob_size_max, 0, "Size of biggest code signature blob");
+
+/*
+ * Function: csblob_parse_teamid
+ *
+ * Description: This function returns a pointer to the team id
+               stored within the codedirectory of the csblob.
+               If the codedirectory predates team-ids, it returns
+               NULL.
+               This does not copy the name but returns a pointer to
+               it within the CD. Subsequently, the CD must be
+               available when this is used.
+*/
+
+static const char *
+csblob_parse_teamid(struct cs_blob *csblob)
 {
-       return((vp) && ((vp)->v_type == VREG) && !((vp)->v_flag & VSYSTEM));
+       const CS_CodeDirectory *cd;
+
+       cd = csblob->csb_cd;
+
+       if (ntohl(cd->version) < CS_SUPPORTSTEAMID)
+               return NULL;
+
+       if (cd->teamOffset == 0)
+               return NULL;
+
+       const char *name = ((const char *)cd) + ntohl(cd->teamOffset);
+       if (cs_debug > 1)
+               printf("found team-id %s in cdblob\n", name);
+
+       return name;
 }
-int 
-UBCINVALID(struct vnode * vp)
+
+
+kern_return_t
+ubc_cs_blob_allocate(
+       vm_offset_t     *blob_addr_p,
+       vm_size_t       *blob_size_p)
 {
-       return(((vp) == NULL) || ((vp) && ((vp)->v_type != VREG))
-               || ((vp) && ((vp)->v_flag & VSYSTEM)));
+       kern_return_t   kr = KERN_FAILURE;
+
+       {
+               *blob_addr_p = (vm_offset_t) kalloc_tag(*blob_size_p, VM_KERN_MEMORY_SECURITY);
+
+               if (*blob_addr_p == 0) {
+                       kr = KERN_NO_SPACE;
+               } else {
+                       kr = KERN_SUCCESS;
+               }
+       }
+
+       return kr;
 }
-int 
-UBCINFOCHECK(const char * fun, struct vnode * vp)
+
+void
+ubc_cs_blob_deallocate(
+       vm_offset_t     blob_addr,
+       vm_size_t       blob_size)
 {
-       if ((vp) && ((vp)->v_type == VREG) &&
-               ((vp)->v_ubcinfo == UBC_INFO_NULL)) {
-               panic("%s: lost ubc_info", (fun));
-               return(1);
+#if PMAP_CS
+       if (blob_size > pmap_cs_blob_limit) {
+               kmem_free(kernel_map, blob_addr, blob_size);
        } else
-               return(0);
+#endif
+       {
+               kfree((void *) blob_addr, blob_size);
+       }
+}
+
+/*
+ * Some codesigned files use a lowest common denominator page size of
+ * 4KiB, but can be used on systems that have a runtime page size of
+ * 16KiB. Since faults will only occur on 16KiB ranges in
+ * cs_validate_range(), we can convert the original Code Directory to
+ * a multi-level scheme where groups of 4 hashes are combined to form
+ * a new hash, which represents 16KiB in the on-disk file.  This can
+ * reduce the wired memory requirement for the Code Directory by
+ * 75%. Care must be taken for binaries that use the "fourk" VM pager
+ * for unaligned access, which may still attempt to validate on
+ * non-16KiB multiples for compatibility with 3rd party binaries.
+ */
+static boolean_t
+ubc_cs_supports_multilevel_hash(struct cs_blob *blob)
+{
+       const CS_CodeDirectory *cd;
+
+       
+       /*
+        * Only applies to binaries that ship as part of the OS,
+        * primarily the shared cache.
+        */
+       if (!blob->csb_platform_binary || blob->csb_teamid != NULL) {
+               return FALSE;
+       }
+
+       /*
+        * If the runtime page size matches the code signing page
+        * size, there is no work to do.
+        */
+       if (PAGE_SHIFT <= blob->csb_hash_pageshift) {
+               return FALSE;
+       }
+
+       cd = blob->csb_cd;
+
+       /*
+        * There must be a valid integral multiple of hashes
+        */
+       if (ntohl(cd->nCodeSlots) & (PAGE_MASK >> blob->csb_hash_pageshift)) {
+               return FALSE;
+       }
+
+       /*
+        * Scatter lists must also have ranges that have an integral number of hashes
+        */
+       if ((ntohl(cd->version) >= CS_SUPPORTSSCATTER) && (ntohl(cd->scatterOffset))) {
+
+               const SC_Scatter *scatter = (const SC_Scatter*)
+                       ((const char*)cd + ntohl(cd->scatterOffset));
+               /* iterate all scatter structs to make sure they are all aligned */
+               do {
+                       uint32_t sbase = ntohl(scatter->base);
+                       uint32_t scount = ntohl(scatter->count);
+
+                       /* last scatter? */
+                       if (scount == 0) {
+                               break;
+                       }
+
+                       if (sbase & (PAGE_MASK >> blob->csb_hash_pageshift)) {
+                               return FALSE;
+                       }
+
+                       if (scount & (PAGE_MASK >> blob->csb_hash_pageshift)) {
+                               return FALSE;
+                       }
+
+                       scatter++;
+               } while(1);
+       }
+
+       /* Covered range must be a multiple of the new page size */
+       if (ntohl(cd->codeLimit) & PAGE_MASK) {
+               return FALSE;
+       }
+
+       /* All checks pass */
+       return TRUE;
+}
+
+/*
+ * Given a cs_blob with an already chosen best code directory, this
+ * function allocates memory and copies into it only the blobs that
+ * will be needed by the kernel, namely the single chosen code
+ * directory (and not any of its alternatives) and the entitlement
+ * blob.
+ *
+ * This saves significant memory with agile signatures, and additional
+ * memory for 3rd Party Code because we also omit the CMS blob.
+ *
+ * To support multilevel and other potential code directory rewriting,
+ * the size of a new code directory can be specified. Since that code
+ * directory will replace the existing code directory,
+ * ubc_cs_reconstitute_code_signature does not copy the original code
+ * directory when a size is given, and the caller must fill it in.
+ */
+static int
+ubc_cs_reconstitute_code_signature(struct cs_blob const *blob, vm_size_t optional_new_cd_size,
+                                                                  vm_address_t *new_blob_addr_p, vm_size_t *new_blob_size_p,
+                                                                  CS_CodeDirectory **new_cd_p, CS_GenericBlob const **new_entitlements_p)
+{
+       const CS_CodeDirectory  *old_cd, *cd;
+       CS_CodeDirectory        *new_cd;
+       const CS_GenericBlob *entitlements;
+       vm_offset_t     new_blob_addr;
+       vm_size_t       new_blob_size;
+       vm_size_t       new_cdsize;
+       kern_return_t   kr;
+       int                             error;
+
+       old_cd = blob->csb_cd;
+
+       new_cdsize = optional_new_cd_size != 0 ? optional_new_cd_size : htonl(old_cd->length);
+
+       new_blob_size  = sizeof(CS_SuperBlob);
+       new_blob_size += sizeof(CS_BlobIndex);
+       new_blob_size += new_cdsize;
+
+       if (blob->csb_entitlements_blob) {
+               /* We need to add a slot for the entitlements */
+               new_blob_size += sizeof(CS_BlobIndex);
+               new_blob_size += ntohl(blob->csb_entitlements_blob->length);
+       }
+
+       kr = ubc_cs_blob_allocate(&new_blob_addr, &new_blob_size);
+       if (kr != KERN_SUCCESS) {
+               if (cs_debug > 1) {
+                       printf("CODE SIGNING: Failed to allocate memory for new Code Signing Blob: %d\n",
+                                  kr);
+               }
+               return ENOMEM;
+       }
+
+       CS_SuperBlob            *new_superblob;
+
+       new_superblob = (CS_SuperBlob *)new_blob_addr;
+       new_superblob->magic = htonl(CSMAGIC_EMBEDDED_SIGNATURE);
+       new_superblob->length = htonl((uint32_t)new_blob_size);
+       if (blob->csb_entitlements_blob) {
+               vm_size_t                       ent_offset, cd_offset;
+
+               cd_offset  = sizeof(CS_SuperBlob) + 2 * sizeof(CS_BlobIndex);
+               ent_offset = cd_offset +  new_cdsize;
+
+               new_superblob->count = htonl(2);
+               new_superblob->index[0].type = htonl(CSSLOT_CODEDIRECTORY);
+               new_superblob->index[0].offset = htonl((uint32_t)cd_offset);
+               new_superblob->index[1].type = htonl(CSSLOT_ENTITLEMENTS);
+               new_superblob->index[1].offset = htonl((uint32_t)ent_offset);
+
+               memcpy((void *)(new_blob_addr + ent_offset), blob->csb_entitlements_blob, ntohl(blob->csb_entitlements_blob->length));
+
+               new_cd = (CS_CodeDirectory *)(new_blob_addr + cd_offset);
+       } else {
+               // Blob is the code directory, directly.
+               new_cd = (CS_CodeDirectory *)new_blob_addr;
+       }
+
+       if (optional_new_cd_size == 0) {
+               // Copy code directory, and revalidate.
+               memcpy(new_cd, old_cd, new_cdsize);
+
+               vm_size_t length = new_blob_size;
+
+               error = cs_validate_csblob((const uint8_t *)new_blob_addr, length, &cd, &entitlements);
+
+               if (error) {
+                       printf("CODE SIGNING: Failed to validate new Code Signing Blob: %d\n",
+                                  error);
+
+                       ubc_cs_blob_deallocate(new_blob_addr, new_blob_size);
+                       return error;
+               }
+               *new_entitlements_p = entitlements;
+       } else {
+               // Caller will fill out and validate code directory.
+               memset(new_cd, 0, new_cdsize);
+               *new_entitlements_p = NULL;
+       }
+
+       *new_blob_addr_p = new_blob_addr;
+       *new_blob_size_p = new_blob_size;
+       *new_cd_p = new_cd;
+
+       return 0;
 }
 
+static int
+ubc_cs_convert_to_multilevel_hash(struct cs_blob *blob)
+{
+       const CS_CodeDirectory  *old_cd, *cd;
+       CS_CodeDirectory        *new_cd;
+       const CS_GenericBlob *entitlements;
+       vm_offset_t     new_blob_addr;
+       vm_size_t       new_blob_size;
+       vm_size_t       new_cdsize;
+       int                             error;
+
+       uint32_t                hashes_per_new_hash_shift = (uint32_t)(PAGE_SHIFT - blob->csb_hash_pageshift);
+
+       if (cs_debug > 1) {
+               printf("CODE SIGNING: Attempting to convert Code Directory for %lu -> %lu page shift\n",
+                          (unsigned long)blob->csb_hash_pageshift, (unsigned long)PAGE_SHIFT);
+       }
+
+       old_cd = blob->csb_cd;
+
+       /* Up to the hashes, we can copy all data */
+       new_cdsize  = ntohl(old_cd->hashOffset);
+       new_cdsize += (ntohl(old_cd->nCodeSlots) >> hashes_per_new_hash_shift) * old_cd->hashSize;
+
+       error = ubc_cs_reconstitute_code_signature(blob, new_cdsize,
+                                                                                       &new_blob_addr, &new_blob_size, &new_cd,
+                                                                                       &entitlements);
+       if (error != 0) {
+               printf("CODE SIGNING: Failed to reconsitute code signature: %d\n", error);
+               return error;
+       }
+
+       memcpy(new_cd, old_cd, ntohl(old_cd->hashOffset));
+
+       /* Update fields in the Code Directory structure */
+       new_cd->length = htonl((uint32_t)new_cdsize);
+
+       uint32_t nCodeSlots = ntohl(new_cd->nCodeSlots);
+       nCodeSlots >>= hashes_per_new_hash_shift;
+       new_cd->nCodeSlots = htonl(nCodeSlots);
+
+       new_cd->pageSize = PAGE_SHIFT; /* Not byte-swapped */
+
+       if ((ntohl(new_cd->version) >= CS_SUPPORTSSCATTER) && (ntohl(new_cd->scatterOffset))) {
+               SC_Scatter *scatter = (SC_Scatter*)
+                       ((char *)new_cd + ntohl(new_cd->scatterOffset));
+               /* iterate all scatter structs to scale their counts */
+               do {
+                       uint32_t scount = ntohl(scatter->count);
+                       uint32_t sbase  = ntohl(scatter->base);
+
+                       /* last scatter? */
+                       if (scount == 0) {
+                               break;
+                       }
+
+                       scount >>= hashes_per_new_hash_shift;
+                       scatter->count = htonl(scount);
+
+                       sbase >>= hashes_per_new_hash_shift;
+                       scatter->base = htonl(sbase);
+
+                       scatter++;
+               } while(1);
+       }
+
+       /* For each group of hashes, hash them together */
+       const unsigned char *src_base = (const unsigned char *)old_cd + ntohl(old_cd->hashOffset);
+       unsigned char *dst_base = (unsigned char *)new_cd + ntohl(new_cd->hashOffset);
+
+       uint32_t hash_index;
+       for (hash_index = 0; hash_index < nCodeSlots; hash_index++) {
+               union cs_hash_union     mdctx;
+
+               uint32_t source_hash_len = old_cd->hashSize << hashes_per_new_hash_shift;
+               const unsigned char *src = src_base + hash_index * source_hash_len;
+               unsigned char *dst = dst_base + hash_index * new_cd->hashSize;
+
+               blob->csb_hashtype->cs_init(&mdctx);
+               blob->csb_hashtype->cs_update(&mdctx, src, source_hash_len);
+               blob->csb_hashtype->cs_final(dst, &mdctx);
+       }
+
+       error = cs_validate_csblob((const uint8_t *)new_blob_addr, new_blob_size, &cd, &entitlements);
+       if (error != 0) {
+
+               printf("CODE SIGNING: Failed to validate new Code Signing Blob: %d\n",
+                          error);
+
+               ubc_cs_blob_deallocate(new_blob_addr, new_blob_size);
+               return error;
+       }
+
+    /* New Code Directory is ready for use, swap it out in the blob structure */
+       ubc_cs_blob_deallocate(blob->csb_mem_kaddr, blob->csb_mem_size);
+
+       blob->csb_mem_size = new_blob_size;
+       blob->csb_mem_kaddr = new_blob_addr;
+       blob->csb_cd = cd;
+       blob->csb_entitlements_blob = entitlements;
+
+       /* The blob has some cached attributes of the Code Directory, so update those */
+
+       blob->csb_hash_firstlevel_pagesize = blob->csb_hash_pagesize; /* Save the original page size */
+
+       blob->csb_hash_pagesize = PAGE_SIZE;
+       blob->csb_hash_pagemask = PAGE_MASK;
+       blob->csb_hash_pageshift = PAGE_SHIFT;
+       blob->csb_end_offset = ntohl(cd->codeLimit);
+       if((ntohl(cd->version) >= CS_SUPPORTSSCATTER) && (ntohl(cd->scatterOffset))) {
+               const SC_Scatter *scatter = (const SC_Scatter*)
+                       ((const char*)cd + ntohl(cd->scatterOffset));
+               blob->csb_start_offset = ((off_t)ntohl(scatter->base)) * PAGE_SIZE;
+       } else {
+               blob->csb_start_offset = 0;
+       }
+
+       return 0;
+}
+
+/*
+ * Validate the code signature blob, create a struct cs_blob wrapper
+ * and return it together with a pointer to the chosen code directory
+ * and entitlements blob.
+ *
+ * Note that this takes ownership of the memory as addr, mainly because
+ * this function can actually replace the passed in blob with another
+ * one, e.g. when performing multilevel hashing optimization.
+ */
+int
+cs_blob_create_validated(
+       vm_address_t * const            addr,
+       vm_size_t                       size,
+       struct cs_blob ** const         ret_blob,
+    CS_CodeDirectory const ** const    ret_cd)
+{
+       struct cs_blob          *blob;
+       int             error = EINVAL;
+       const CS_CodeDirectory *cd;
+       const CS_GenericBlob *entitlements;
+       union cs_hash_union     mdctx;
+       size_t                  length;
+
+       if (ret_blob)
+           *ret_blob = NULL;
+
+       blob = (struct cs_blob *) kalloc(sizeof (struct cs_blob));
+       if (blob == NULL) {
+               return ENOMEM;
+       }
+
+       /* fill in the new blob */
+       blob->csb_mem_size = size;
+       blob->csb_mem_offset = 0;
+       blob->csb_mem_kaddr = *addr;
+       blob->csb_flags = 0;
+       blob->csb_signer_type = CS_SIGNER_TYPE_UNKNOWN;
+       blob->csb_platform_binary = 0;
+       blob->csb_platform_path = 0;
+       blob->csb_teamid = NULL;
+       blob->csb_entitlements_blob = NULL;
+       blob->csb_entitlements = NULL;
+       blob->csb_reconstituted = false;
+
+       /* Transfer ownership. Even on error, this function will deallocate */
+       *addr = 0;
+
+       /*
+        * Validate the blob's contents
+        */
+       length = (size_t) size;
+       error = cs_validate_csblob((const uint8_t *)blob->csb_mem_kaddr,
+                                                          length, &cd, &entitlements);
+       if (error) {
+
+               if (cs_debug)
+                       printf("CODESIGNING: csblob invalid: %d\n", error);
+               /*
+                * The vnode checker can't make the rest of this function
+                * succeed if csblob validation failed, so bail */
+               goto out;
+
+       } else {
+               const unsigned char *md_base;
+               uint8_t hash[CS_HASH_MAX_SIZE];
+               int md_size;
+
+               blob->csb_cd = cd;
+               blob->csb_entitlements_blob = entitlements; /* may be NULL, not yet validated */
+               blob->csb_hashtype = cs_find_md(cd->hashType);
+               if (blob->csb_hashtype == NULL || blob->csb_hashtype->cs_digest_size > sizeof(hash))
+                       panic("validated CodeDirectory but unsupported type");
+
+               blob->csb_hash_pageshift = cd->pageSize;
+               blob->csb_hash_pagesize = (1U << cd->pageSize);
+               blob->csb_hash_pagemask = blob->csb_hash_pagesize - 1;
+               blob->csb_hash_firstlevel_pagesize = 0;
+               blob->csb_flags = (ntohl(cd->flags) & CS_ALLOWED_MACHO) | CS_VALID;
+               blob->csb_end_offset = (((vm_offset_t)ntohl(cd->codeLimit) + blob->csb_hash_pagemask) & ~((vm_offset_t)blob->csb_hash_pagemask));
+               if((ntohl(cd->version) >= CS_SUPPORTSSCATTER) && (ntohl(cd->scatterOffset))) {
+                       const SC_Scatter *scatter = (const SC_Scatter*)
+                               ((const char*)cd + ntohl(cd->scatterOffset));
+                       blob->csb_start_offset = ((off_t)ntohl(scatter->base)) * blob->csb_hash_pagesize;
+               } else {
+                       blob->csb_start_offset = 0;
+               }
+               /* compute the blob's cdhash */
+               md_base = (const unsigned char *) cd;
+               md_size = ntohl(cd->length);
+
+               blob->csb_hashtype->cs_init(&mdctx);
+               blob->csb_hashtype->cs_update(&mdctx, md_base, md_size);
+               blob->csb_hashtype->cs_final(hash, &mdctx);
+
+               memcpy(blob->csb_cdhash, hash, CS_CDHASH_LEN);
+       }
+
+    error = 0;
+
+out:
+    if (error != 0) {
+        cs_blob_free(blob);
+        blob = NULL;
+        cd = NULL;
+    }
+
+    if (ret_blob != NULL) {
+        *ret_blob = blob;
+    }
+    if (ret_cd != NULL) {
+        *ret_cd = cd;
+    }
+
+    return error;
+}
+
+/*
+ * Free a cs_blob previously created by cs_blob_create_validated.
+ */
+void
+cs_blob_free(
+    struct cs_blob * const blob)
+{
+    if (blob != NULL) {
+        if (blob->csb_mem_kaddr) {
+            ubc_cs_blob_deallocate(blob->csb_mem_kaddr, blob->csb_mem_size);
+            blob->csb_mem_kaddr = 0;
+        }
+        if (blob->csb_entitlements != NULL) {
+            osobject_release(blob->csb_entitlements);
+            blob->csb_entitlements = NULL;
+        }
+        kfree(blob, sizeof (*blob));
+    }
+}
+
+int
+ubc_cs_blob_add(
+       struct vnode    *vp,
+       cpu_type_t      cputype,
+       off_t           base_offset,
+       vm_address_t    *addr,
+       vm_size_t       size,
+       struct image_params *imgp,
+       __unused int    flags,
+       struct cs_blob  **ret_blob)
+{
+       kern_return_t           kr;
+       struct ubc_info         *uip;
+       struct cs_blob          *blob, *oblob;
+       int                     error;
+       CS_CodeDirectory const *cd;
+       off_t                   blob_start_offset, blob_end_offset;
+       boolean_t               record_mtime;
+
+       record_mtime = FALSE;
+       if (ret_blob)
+           *ret_blob = NULL;
+    /* Create the struct cs_blob wrapper that will be attached to the vnode.
+     * Validates the passed in blob in the process. */
+    error = cs_blob_create_validated(addr, size, &blob, &cd);
+
+    if (error != 0) {
+               printf("malform code signature blob: %d\n", error);
+        return error;
+    }
+
+    blob->csb_cpu_type = cputype;
+       blob->csb_base_offset = base_offset;
+
+       /*
+        * Let policy module check whether the blob's signature is accepted.
+        */
+#if CONFIG_MACF
+    unsigned int cs_flags = blob->csb_flags;
+       unsigned int signer_type = blob->csb_signer_type;
+       error = mac_vnode_check_signature(vp, blob, imgp, &cs_flags, &signer_type, flags);
+    blob->csb_flags = cs_flags;
+       blob->csb_signer_type = signer_type;
+
+       if (error) {
+               if (cs_debug) 
+                       printf("check_signature[pid: %d], error = %d\n", current_proc()->p_pid, error);
+               goto out;
+       }
+       if ((flags & MAC_VNODE_CHECK_DYLD_SIM) && !(blob->csb_flags & CS_PLATFORM_BINARY)) {
+               if (cs_debug)
+                       printf("check_signature[pid: %d], is not apple signed\n", current_proc()->p_pid);
+               error = EPERM;
+               goto out;
+       }
+#endif
+
+#if CONFIG_ENFORCE_SIGNED_CODE
+       /*
+        * Reconstitute code signature
+        */
+       {
+               vm_address_t new_mem_kaddr = 0;
+               vm_size_t new_mem_size = 0;
+
+               CS_CodeDirectory *new_cd = NULL;
+               CS_GenericBlob const *new_entitlements = NULL;
+
+               error = ubc_cs_reconstitute_code_signature(blob, 0,
+                                                                                                  &new_mem_kaddr, &new_mem_size,
+                                                                                                  &new_cd, &new_entitlements);
+
+               if (error != 0) {
+                       printf("failed code signature reconstitution: %d\n", error);
+                       goto out;
+               }
+
+               ubc_cs_blob_deallocate(blob->csb_mem_kaddr, blob->csb_mem_size);
+
+               blob->csb_mem_kaddr = new_mem_kaddr;
+               blob->csb_mem_size = new_mem_size;
+               blob->csb_cd = new_cd;
+               blob->csb_entitlements_blob = new_entitlements;
+               blob->csb_reconstituted = true;
+       }
+
+#endif
+
+
+       if (blob->csb_flags & CS_PLATFORM_BINARY) {
+               if (cs_debug > 1)
+                       printf("check_signature[pid: %d]: platform binary\n", current_proc()->p_pid);
+               blob->csb_platform_binary = 1;
+               blob->csb_platform_path = !!(blob->csb_flags & CS_PLATFORM_PATH);
+       } else {
+               blob->csb_platform_binary = 0;
+               blob->csb_platform_path = 0;
+               blob->csb_teamid = csblob_parse_teamid(blob);
+               if (cs_debug > 1) {
+                       if (blob->csb_teamid)
+                               printf("check_signature[pid: %d]: team-id is %s\n", current_proc()->p_pid, blob->csb_teamid);
+                       else
+                               printf("check_signature[pid: %d]: no team-id\n", current_proc()->p_pid);
+               }
+       }
+
+       /*
+        * Validate the blob's coverage
+        */
+       blob_start_offset = blob->csb_base_offset + blob->csb_start_offset;
+       blob_end_offset = blob->csb_base_offset + blob->csb_end_offset;
+
+       if (blob_start_offset >= blob_end_offset ||
+           blob_start_offset < 0 ||
+           blob_end_offset <= 0) {
+               /* reject empty or backwards blob */
+               error = EINVAL;
+               goto out;
+       }
+
+       if (ubc_cs_supports_multilevel_hash(blob)) {
+               error = ubc_cs_convert_to_multilevel_hash(blob);
+               if (error != 0) {
+                       printf("failed multilevel hash conversion: %d\n", error);
+                       goto out;
+               }
+               blob->csb_reconstituted = true;
+       }
+
+       vnode_lock(vp);
+       if (! UBCINFOEXISTS(vp)) {
+               vnode_unlock(vp);
+               error = ENOENT;
+               goto out;
+       }
+       uip = vp->v_ubcinfo;
+
+       /* check if this new blob overlaps with an existing blob */
+       for (oblob = uip->cs_blobs;
+            oblob != NULL;
+            oblob = oblob->csb_next) {
+                off_t oblob_start_offset, oblob_end_offset;
+
+                if (blob->csb_signer_type != oblob->csb_signer_type) { // signer type needs to be the same for slices
+                        vnode_unlock(vp);
+                        error = EALREADY;
+                        goto out;
+                } else if (blob->csb_platform_binary) { //platform binary needs to be the same for app slices
+                        if (!oblob->csb_platform_binary) {
+                                vnode_unlock(vp);
+                                error = EALREADY;
+                                goto out;
+                        }
+                } else if (blob->csb_teamid) { //teamid binary needs to be the same for app slices
+                       if (oblob->csb_platform_binary ||
+                           oblob->csb_teamid == NULL ||
+                           strcmp(oblob->csb_teamid, blob->csb_teamid) != 0) {
+                               vnode_unlock(vp);
+                               error = EALREADY;
+                               goto out;
+                       }
+                } else { // non teamid binary needs to be the same for app slices
+                       if (oblob->csb_platform_binary ||
+                               oblob->csb_teamid != NULL) {
+                               vnode_unlock(vp);
+                               error = EALREADY;
+                               goto out;
+                       }
+                }
+
+                oblob_start_offset = (oblob->csb_base_offset +
+                                      oblob->csb_start_offset);
+                oblob_end_offset = (oblob->csb_base_offset +
+                                    oblob->csb_end_offset);
+                if (blob_start_offset >= oblob_end_offset ||
+                    blob_end_offset <= oblob_start_offset) {
+                        /* no conflict with this existing blob */
+                } else {
+                        /* conflict ! */
+                        if (blob_start_offset == oblob_start_offset &&
+                            blob_end_offset == oblob_end_offset &&
+                            blob->csb_mem_size == oblob->csb_mem_size &&
+                            blob->csb_flags == oblob->csb_flags &&
+                            (blob->csb_cpu_type == CPU_TYPE_ANY ||
+                             oblob->csb_cpu_type == CPU_TYPE_ANY ||
+                             blob->csb_cpu_type == oblob->csb_cpu_type) &&
+                            !bcmp(blob->csb_cdhash,
+                                  oblob->csb_cdhash,
+                                  CS_CDHASH_LEN)) {
+                                /*
+                                 * We already have this blob:
+                                 * we'll return success but
+                                 * throw away the new blob.
+                                 */
+                                if (oblob->csb_cpu_type == CPU_TYPE_ANY) {
+                                        /*
+                                         * The old blob matches this one
+                                         * but doesn't have any CPU type.
+                                         * Update it with whatever the caller
+                                         * provided this time.
+                                         */
+                                        oblob->csb_cpu_type = cputype;
+                                }
+
+                                /* The signature is still accepted, so update the
+                                 * generation count. */
+                                uip->cs_add_gen = cs_blob_generation_count;
+
+                                vnode_unlock(vp);
+                                if (ret_blob)
+                                        *ret_blob = oblob;
+                                error = EAGAIN;
+                                goto out;
+                        } else {
+                                /* different blob: reject the new one */
+                                vnode_unlock(vp);
+                                error = EALREADY;
+                                goto out;
+                        }
+                }
+
+       }
+
+
+       /* mark this vnode's VM object as having "signed pages" */
+       kr = memory_object_signed(uip->ui_control, TRUE);
+       if (kr != KERN_SUCCESS) {
+               vnode_unlock(vp);
+               error = ENOENT;
+               goto out;
+       }
+
+       if (uip->cs_blobs == NULL) {
+               /* loading 1st blob: record the file's current "modify time" */
+               record_mtime = TRUE;
+       }
+
+       /* set the generation count for cs_blobs */
+       uip->cs_add_gen = cs_blob_generation_count;
+
+       /*
+        * Add this blob to the list of blobs for this vnode.
+        * We always add at the front of the list and we never remove a
+        * blob from the list, so ubc_cs_get_blobs() can return whatever
+        * the top of the list was and that list will remain valid
+        * while we validate a page, even after we release the vnode's lock.
+        */
+       blob->csb_next = uip->cs_blobs;
+       uip->cs_blobs = blob;
+
+       OSAddAtomic(+1, &cs_blob_count);
+       if (cs_blob_count > cs_blob_count_peak) {
+               cs_blob_count_peak = cs_blob_count; /* XXX atomic ? */
+       }
+       OSAddAtomic((SInt32) +blob->csb_mem_size, &cs_blob_size);
+       if ((SInt32) cs_blob_size > cs_blob_size_peak) {
+               cs_blob_size_peak = (SInt32) cs_blob_size; /* XXX atomic ? */
+       }
+       if ((UInt32) blob->csb_mem_size > cs_blob_size_max) {
+               cs_blob_size_max = (UInt32) blob->csb_mem_size;
+       }
+
+       if (cs_debug > 1) {
+               proc_t p;
+               const char *name = vnode_getname_printable(vp);
+               p = current_proc();
+               printf("CODE SIGNING: proc %d(%s) "
+                      "loaded %s signatures for file (%s) "
+                      "range 0x%llx:0x%llx flags 0x%x\n",
+                      p->p_pid, p->p_comm,
+                      blob->csb_cpu_type == -1 ? "detached" : "embedded",
+                      name,
+                      blob->csb_base_offset + blob->csb_start_offset,
+                      blob->csb_base_offset + blob->csb_end_offset,
+                      blob->csb_flags);
+               vnode_putname_printable(name);
+       }
+
+       vnode_unlock(vp);
+
+       if (record_mtime) {
+               vnode_mtime(vp, &uip->cs_mtime, vfs_context_current());
+       }
+
+       if (ret_blob)
+               *ret_blob = blob;
+
+       error = 0;      /* success ! */
+
+out:
+       if (error) {
+               if (cs_debug)
+                       printf("check_signature[pid: %d]: error = %d\n", current_proc()->p_pid, error);
+
+        cs_blob_free(blob);
+       }
+
+       if (error == EAGAIN) {
+               /*
+                * See above:  error is EAGAIN if we were asked 
+                * to add an existing blob again.  We cleaned the new
+                * blob and we want to return success.
+                */
+               error = 0;
+       }
+
+       return error;
+}
+
+void
+csvnode_print_debug(struct vnode *vp)
+{
+       const char      *name = NULL;
+       struct ubc_info *uip;
+       struct cs_blob *blob;
+
+       name = vnode_getname_printable(vp);
+       if (name) {
+               printf("csvnode: name: %s\n", name);
+               vnode_putname_printable(name);
+       }
+
+       vnode_lock_spin(vp);
+
+       if (! UBCINFOEXISTS(vp)) {
+               blob = NULL;
+               goto out;
+       }
+
+       uip = vp->v_ubcinfo;
+       for (blob = uip->cs_blobs; blob != NULL; blob = blob->csb_next) {
+               printf("csvnode: range: %lu -> %lu flags: 0x%08x platform: %s path: %s team: %s\n",
+                      (unsigned long)blob->csb_start_offset,
+                      (unsigned long)blob->csb_end_offset,
+                      blob->csb_flags,
+                      blob->csb_platform_binary ? "yes" : "no",
+                      blob->csb_platform_path ? "yes" : "no",
+                      blob->csb_teamid ? blob->csb_teamid : "<NO-TEAM>");
+       }
+
+out:
+       vnode_unlock(vp);
+
+}
+
+struct cs_blob *
+ubc_cs_blob_get(
+       struct vnode    *vp,
+       cpu_type_t      cputype,
+       off_t           offset)
+{
+       struct ubc_info *uip;
+       struct cs_blob  *blob;
+       off_t offset_in_blob;
+
+       vnode_lock_spin(vp);
+
+       if (! UBCINFOEXISTS(vp)) {
+               blob = NULL;
+               goto out;
+       }
+
+       uip = vp->v_ubcinfo;
+       for (blob = uip->cs_blobs;
+            blob != NULL;
+            blob = blob->csb_next) {
+               if (cputype != -1 && blob->csb_cpu_type == cputype) {
+                       break;
+               }
+               if (offset != -1) {
+                       offset_in_blob = offset - blob->csb_base_offset;
+                       if (offset_in_blob >= blob->csb_start_offset &&
+                           offset_in_blob < blob->csb_end_offset) {
+                               /* our offset is covered by this blob */
+                               break;
+                       }
+               }
+       }
+
+out:
+       vnode_unlock(vp);
+
+       return blob;
+}
+
+static void
+ubc_cs_free(
+       struct ubc_info *uip)
+{
+       struct cs_blob  *blob, *next_blob;
+
+       for (blob = uip->cs_blobs;
+            blob != NULL;
+            blob = next_blob) {
+               next_blob = blob->csb_next;
+               OSAddAtomic(-1, &cs_blob_count);
+               OSAddAtomic((SInt32) -blob->csb_mem_size, &cs_blob_size);
+               cs_blob_free(blob);
+       }
+#if CHECK_CS_VALIDATION_BITMAP
+       ubc_cs_validation_bitmap_deallocate( uip->ui_vnode );
+#endif
+       uip->cs_blobs = NULL;
+}
+
+/* check cs blob generation on vnode
+ * returns:
+ *    0         : Success, the cs_blob attached is current
+ *    ENEEDAUTH : Generation count mismatch. Needs authentication again.
+ */
+int
+ubc_cs_generation_check(
+       struct vnode    *vp)
+{
+       int retval = ENEEDAUTH;
+
+       vnode_lock_spin(vp);
+
+       if (UBCINFOEXISTS(vp) && vp->v_ubcinfo->cs_add_gen == cs_blob_generation_count) {
+               retval = 0;
+       }
+
+       vnode_unlock(vp);
+       return retval;
+}
+
+int
+ubc_cs_blob_revalidate(
+       struct vnode    *vp,
+       struct cs_blob *blob,
+       struct image_params *imgp,
+       int flags
+       )
+{
+       int error = 0;
+       const CS_CodeDirectory *cd = NULL;
+       const CS_GenericBlob *entitlements = NULL;
+       size_t size;
+       assert(vp != NULL);
+       assert(blob != NULL);
+
+       size = blob->csb_mem_size;
+       error = cs_validate_csblob((const uint8_t *)blob->csb_mem_kaddr,
+                                                          size, &cd, &entitlements);
+       if (error) {
+               if (cs_debug) {
+                       printf("CODESIGNING: csblob invalid: %d\n", error);
+               }
+               goto out;
+       }
+
+    unsigned int cs_flags = (ntohl(cd->flags) & CS_ALLOWED_MACHO) | CS_VALID;
+    unsigned int signer_type = CS_SIGNER_TYPE_UNKNOWN;
+
+       if (blob->csb_reconstituted) {
+               /*
+                * Code signatures that have been modified after validation
+                * cannot be revalidated inline from their in-memory blob.
+                *
+                * That's okay, though, because the only path left that relies
+                * on revalidation of existing in-memory blobs is the legacy
+                * detached signature database path, which only exists on macOS,
+                * which does not do reconstitution of any kind.
+                */
+               if (cs_debug) {
+                       printf("CODESIGNING: revalidate: not inline revalidating reconstituted signature.\n");
+               }
+
+               /*
+                * EAGAIN tells the caller that they may reread the code
+                * signature and try attaching it again, which is the same
+                * thing they would do if there was no cs_blob yet in the
+                * first place.
+                *
+                * Conveniently, after ubc_cs_blob_add did a successful
+                * validation, it will detect that a matching cs_blob (cdhash,
+                * offset, arch etc.) already exists, and return success
+                * without re-adding a cs_blob to the vnode.
+                */
+               return EAGAIN;
+       }
+
+       /* callout to mac_vnode_check_signature */
+#if CONFIG_MACF
+       error = mac_vnode_check_signature(vp, blob, imgp, &cs_flags, &signer_type, flags);
+       if (cs_debug && error) {
+                       printf("revalidate: check_signature[pid: %d], error = %d\n", current_proc()->p_pid, error);
+       }
+#else
+       (void)flags;
+       (void)signer_type;
+#endif
+
+       /* update generation number if success */
+       vnode_lock_spin(vp);
+    blob->csb_flags = cs_flags;
+       blob->csb_signer_type = signer_type;
+       if (UBCINFOEXISTS(vp)) {
+               if (error == 0)
+                       vp->v_ubcinfo->cs_add_gen = cs_blob_generation_count;
+               else
+                       vp->v_ubcinfo->cs_add_gen = 0;
+       }
+
+       vnode_unlock(vp);
+
+out:
+       return error;
+}
+
+void
+cs_blob_reset_cache()
+{
+       /* incrementing odd no by 2 makes sure '0' is never reached. */
+       OSAddAtomic(+2, &cs_blob_generation_count);
+       printf("Reseting cs_blob cache from all vnodes. \n");
+}
+
+struct cs_blob *
+ubc_get_cs_blobs(
+       struct vnode    *vp)
+{
+       struct ubc_info *uip;
+       struct cs_blob  *blobs;
+
+       /*
+        * No need to take the vnode lock here.  The caller must be holding
+        * a reference on the vnode (via a VM mapping or open file descriptor),
+        * so the vnode will not go away.  The ubc_info stays until the vnode
+        * goes away.  And we only modify "blobs" by adding to the head of the
+        * list.
+        * The ubc_info could go away entirely if the vnode gets reclaimed as
+        * part of a forced unmount.  In the case of a code-signature validation
+        * during a page fault, the "paging_in_progress" reference on the VM
+        * object guarantess that the vnode pager (and the ubc_info) won't go
+        * away during the fault.
+        * Other callers need to protect against vnode reclaim by holding the
+        * vnode lock, for example.
+        */
+
+       if (! UBCINFOEXISTS(vp)) {
+               blobs = NULL;
+               goto out;
+       }
+
+       uip = vp->v_ubcinfo;
+       blobs = uip->cs_blobs;
+
+out:
+       return blobs;
+}
+
+void
+ubc_get_cs_mtime(
+       struct vnode    *vp,
+       struct timespec *cs_mtime)
+{
+       struct ubc_info *uip;
+
+       if (! UBCINFOEXISTS(vp)) {
+               cs_mtime->tv_sec = 0;
+               cs_mtime->tv_nsec = 0;
+               return;
+       }
+
+       uip = vp->v_ubcinfo;
+       cs_mtime->tv_sec = uip->cs_mtime.tv_sec;
+       cs_mtime->tv_nsec = uip->cs_mtime.tv_nsec;
+}
+
+unsigned long cs_validate_page_no_hash = 0;
+unsigned long cs_validate_page_bad_hash = 0;
+static boolean_t
+cs_validate_hash(
+       struct cs_blob          *blobs,
+       memory_object_t         pager,
+       memory_object_offset_t  page_offset,
+       const void              *data,
+       vm_size_t               *bytes_processed,
+       unsigned                *tainted)
+{
+       union cs_hash_union     mdctx;
+       struct cs_hash const    *hashtype = NULL;
+       unsigned char           actual_hash[CS_HASH_MAX_SIZE];
+       unsigned char           expected_hash[CS_HASH_MAX_SIZE];
+       boolean_t               found_hash;
+       struct cs_blob          *blob;
+       const CS_CodeDirectory  *cd;
+       const unsigned char     *hash;
+       boolean_t               validated;
+       off_t                   offset; /* page offset in the file */
+       size_t                  size;
+       off_t                   codeLimit = 0;
+       const char              *lower_bound, *upper_bound;
+       vm_offset_t             kaddr, blob_addr;
+
+       /* retrieve the expected hash */
+       found_hash = FALSE;
+
+       for (blob = blobs;
+            blob != NULL;
+            blob = blob->csb_next) {
+               offset = page_offset - blob->csb_base_offset;
+               if (offset < blob->csb_start_offset ||
+                   offset >= blob->csb_end_offset) {
+                       /* our page is not covered by this blob */
+                       continue;
+               }
+
+               /* blob data has been released */
+               kaddr = blob->csb_mem_kaddr;
+               if (kaddr == 0) {
+                       continue;
+               }
+
+               blob_addr = kaddr + blob->csb_mem_offset;
+               lower_bound = CAST_DOWN(char *, blob_addr);
+               upper_bound = lower_bound + blob->csb_mem_size;
+               
+               cd = blob->csb_cd;
+               if (cd != NULL) {
+                       /* all CD's that have been injected is already validated */
+
+                       hashtype = blob->csb_hashtype;
+                       if (hashtype == NULL)
+                               panic("unknown hash type ?");
+                       if (hashtype->cs_digest_size > sizeof(actual_hash))
+                               panic("hash size too large");
+                       if (offset & blob->csb_hash_pagemask)
+                               panic("offset not aligned to cshash boundary");
+
+                       codeLimit = ntohl(cd->codeLimit);
+
+                       hash = hashes(cd, (uint32_t)(offset>>blob->csb_hash_pageshift),
+                                     hashtype->cs_size,
+                                     lower_bound, upper_bound);
+                       if (hash != NULL) {
+                               bcopy(hash, expected_hash, hashtype->cs_size);
+                               found_hash = TRUE;
+                       }
+
+                       break;
+               }
+       }
+
+       if (found_hash == FALSE) {
+               /*
+                * We can't verify this page because there is no signature
+                * for it (yet).  It's possible that this part of the object
+                * is not signed, or that signatures for that part have not
+                * been loaded yet.
+                * Report that the page has not been validated and let the
+                * caller decide if it wants to accept it or not.
+                */
+               cs_validate_page_no_hash++;
+               if (cs_debug > 1) {
+                       printf("CODE SIGNING: cs_validate_page: "
+                              "mobj %p off 0x%llx: no hash to validate !?\n",
+                              pager, page_offset);
+               }
+               validated = FALSE;
+               *tainted = 0;
+       } else {
+
+               *tainted = 0;
+
+               size = blob->csb_hash_pagesize;
+               *bytes_processed = size;
+
+               const uint32_t *asha1, *esha1;
+               if ((off_t)(offset + size) > codeLimit) {
+                       /* partial page at end of segment */
+                       assert(offset < codeLimit);
+                       size = (size_t) (codeLimit & blob->csb_hash_pagemask);
+                       *tainted |= CS_VALIDATE_NX;
+               }
+
+               hashtype->cs_init(&mdctx);
+
+               if (blob->csb_hash_firstlevel_pagesize) {
+                       const unsigned char *partial_data = (const unsigned char *)data;
+                       size_t i;
+                       for (i=0; i < size;) {
+                               union cs_hash_union     partialctx;
+                               unsigned char partial_digest[CS_HASH_MAX_SIZE];
+                               size_t partial_size = MIN(size-i, blob->csb_hash_firstlevel_pagesize);
+
+                               hashtype->cs_init(&partialctx);
+                               hashtype->cs_update(&partialctx, partial_data, partial_size);
+                               hashtype->cs_final(partial_digest, &partialctx);
+
+                               /* Update cumulative multi-level hash */
+                               hashtype->cs_update(&mdctx, partial_digest, hashtype->cs_size);
+                               partial_data = partial_data + partial_size;
+                               i += partial_size;
+                       }
+               } else {
+                       hashtype->cs_update(&mdctx, data, size);
+               }
+               hashtype->cs_final(actual_hash, &mdctx);
+
+               asha1 = (const uint32_t *) actual_hash;
+               esha1 = (const uint32_t *) expected_hash;
+
+               if (bcmp(expected_hash, actual_hash, hashtype->cs_size) != 0) {
+                       if (cs_debug) {
+                               printf("CODE SIGNING: cs_validate_page: "
+                                      "mobj %p off 0x%llx size 0x%lx: "
+                                      "actual [0x%x 0x%x 0x%x 0x%x 0x%x] != "
+                                      "expected [0x%x 0x%x 0x%x 0x%x 0x%x]\n",
+                                      pager, page_offset, size,
+                                      asha1[0], asha1[1], asha1[2],
+                                      asha1[3], asha1[4],
+                                      esha1[0], esha1[1], esha1[2],
+                                      esha1[3], esha1[4]);
+                       }
+                       cs_validate_page_bad_hash++;
+                       *tainted |= CS_VALIDATE_TAINTED;
+               } else {
+                       if (cs_debug > 10) {
+                               printf("CODE SIGNING: cs_validate_page: "
+                                      "mobj %p off 0x%llx size 0x%lx: "
+                                      "SHA1 OK\n",
+                                      pager, page_offset, size);
+                       }
+               }
+               validated = TRUE;
+       }
+       
+       return validated;
+}
+
+boolean_t
+cs_validate_range(
+       struct vnode    *vp,
+       memory_object_t         pager,
+       memory_object_offset_t  page_offset,
+       const void              *data,
+       vm_size_t               dsize,
+       unsigned                *tainted)
+{
+       vm_size_t offset_in_range;
+       boolean_t all_subranges_validated = TRUE; /* turn false if any subrange fails */
+
+       struct cs_blob *blobs = ubc_get_cs_blobs(vp);
+
+       *tainted = 0;
+
+       for (offset_in_range = 0;
+                offset_in_range < dsize;
+                /* offset_in_range updated based on bytes processed */) {
+               unsigned subrange_tainted = 0;
+               boolean_t subrange_validated;
+               vm_size_t bytes_processed = 0;
+
+               subrange_validated = cs_validate_hash(blobs,
+                                                                                         pager,
+                                                                                         page_offset + offset_in_range,
+                                                                                         (const void *)((const char *)data + offset_in_range),
+                                                                                         &bytes_processed,
+                                                                                         &subrange_tainted);
+
+               *tainted |= subrange_tainted;
+
+               if (bytes_processed == 0) {
+                       /* Cannote make forward progress, so return an error */
+                       all_subranges_validated = FALSE;
+                       break;
+               } else if (subrange_validated == FALSE) {
+                       all_subranges_validated = FALSE;
+                       /* Keep going to detect other types of failures in subranges */
+               }
+
+               offset_in_range += bytes_processed;
+       }
+
+       return all_subranges_validated;
+}
+
+int
+ubc_cs_getcdhash(
+       vnode_t         vp,
+       off_t           offset,
+       unsigned char   *cdhash)
+{
+       struct cs_blob  *blobs, *blob;
+       off_t           rel_offset;
+       int             ret;
+
+       vnode_lock(vp);
+
+       blobs = ubc_get_cs_blobs(vp);
+       for (blob = blobs;
+            blob != NULL;
+            blob = blob->csb_next) {
+               /* compute offset relative to this blob */
+               rel_offset = offset - blob->csb_base_offset;
+               if (rel_offset >= blob->csb_start_offset &&
+                   rel_offset < blob->csb_end_offset) {
+                       /* this blob does cover our "offset" ! */
+                       break;
+               }
+       }
+
+       if (blob == NULL) {
+               /* we didn't find a blob covering "offset" */
+               ret = EBADEXEC; /* XXX any better error ? */
+       } else {
+               /* get the SHA1 hash of that blob */
+               bcopy(blob->csb_cdhash, cdhash, sizeof (blob->csb_cdhash));
+               ret = 0;
+       }
+
+       vnode_unlock(vp);
+
+       return ret;
+}
+
+boolean_t
+ubc_cs_is_range_codesigned(
+       vnode_t                 vp,
+       mach_vm_offset_t        start,
+       mach_vm_size_t          size)
+{
+       struct cs_blob          *csblob;
+       mach_vm_offset_t        blob_start;
+       mach_vm_offset_t        blob_end;
+
+       if (vp == NULL) {
+               /* no file: no code signature */
+               return FALSE;
+       }
+       if (size == 0) {
+               /* no range: no code signature */
+               return FALSE;
+       }
+       if (start + size < start) {
+               /* overflow */
+               return FALSE;
+       }
+
+       csblob = ubc_cs_blob_get(vp, -1, start);
+       if (csblob == NULL) {
+               return FALSE;
+       }
+
+       /*
+        * We currently check if the range is covered by a single blob,
+        * which should always be the case for the dyld shared cache.
+        * If we ever want to make this routine handle other cases, we
+        * would have to iterate if the blob does not cover the full range.
+        */
+       blob_start = (mach_vm_offset_t) (csblob->csb_base_offset +
+                                        csblob->csb_start_offset);
+       blob_end = (mach_vm_offset_t) (csblob->csb_base_offset +
+                                      csblob->csb_end_offset);
+       if (blob_start > start || blob_end < (start + size)) {
+               /* range not fully covered by this code-signing blob */
+               return FALSE;
+       }
+
+       return TRUE;
+}
+
+#if CHECK_CS_VALIDATION_BITMAP
+#define stob(s)        (((atop_64(round_page_64(s))) + 07) >> 3)
+extern boolean_t       root_fs_upgrade_try;
+
+/*
+ * Should we use the code-sign bitmap to avoid repeated code-sign validation?
+ * Depends:
+ * a) Is the target vnode on the root filesystem?
+ * b) Has someone tried to mount the root filesystem read-write?
+ * If answers are (a) yes AND (b) no, then we can use the bitmap.
+ */
+#define USE_CODE_SIGN_BITMAP(vp)       ( (vp != NULL) && (vp->v_mount != NULL) && (vp->v_mount->mnt_flag & MNT_ROOTFS) && !root_fs_upgrade_try) 
+kern_return_t
+ubc_cs_validation_bitmap_allocate(
+       vnode_t         vp)
+{
+       kern_return_t   kr = KERN_SUCCESS;
+       struct ubc_info *uip;
+       char            *target_bitmap;
+       vm_object_size_t        bitmap_size;
+
+       if ( ! USE_CODE_SIGN_BITMAP(vp) || (! UBCINFOEXISTS(vp))) {
+               kr = KERN_INVALID_ARGUMENT;
+       } else {
+               uip = vp->v_ubcinfo;
+
+               if ( uip->cs_valid_bitmap == NULL ) {
+                       bitmap_size = stob(uip->ui_size);
+                       target_bitmap = (char*) kalloc( (vm_size_t)bitmap_size );
+                       if (target_bitmap == 0) {
+                               kr = KERN_NO_SPACE;
+                       } else {
+                               kr = KERN_SUCCESS;
+                       }
+                       if( kr == KERN_SUCCESS ) {
+                               memset( target_bitmap, 0, (size_t)bitmap_size);
+                               uip->cs_valid_bitmap = (void*)target_bitmap;
+                               uip->cs_valid_bitmap_size = bitmap_size;
+                       }
+               }
+       }
+       return kr;
+}
+
+kern_return_t
+ubc_cs_check_validation_bitmap (
+       vnode_t                 vp,
+       memory_object_offset_t          offset,
+       int                     optype)
+{
+       kern_return_t   kr = KERN_SUCCESS;
+
+       if ( ! USE_CODE_SIGN_BITMAP(vp) || ! UBCINFOEXISTS(vp)) {
+               kr = KERN_INVALID_ARGUMENT;
+       } else {
+               struct ubc_info *uip = vp->v_ubcinfo;
+               char            *target_bitmap = uip->cs_valid_bitmap;
+
+               if ( target_bitmap == NULL ) {
+                      kr = KERN_INVALID_ARGUMENT;
+               } else {
+                       uint64_t        bit, byte;
+                       bit = atop_64( offset );
+                       byte = bit >> 3;
+
+                       if ( byte > uip->cs_valid_bitmap_size ) {
+                              kr = KERN_INVALID_ARGUMENT;
+                       } else {
+
+                               if (optype == CS_BITMAP_SET) {
+                                       target_bitmap[byte] |= (1 << (bit & 07));
+                                       kr = KERN_SUCCESS;
+                               } else if (optype == CS_BITMAP_CLEAR) {
+                                       target_bitmap[byte] &= ~(1 << (bit & 07));
+                                       kr = KERN_SUCCESS;
+                               } else if (optype == CS_BITMAP_CHECK) {
+                                       if ( target_bitmap[byte] & (1 << (bit & 07))) {
+                                               kr = KERN_SUCCESS;
+                                       } else {
+                                               kr = KERN_FAILURE;
+                                       }
+                               }
+                       }
+               }
+       }
+       return kr;
+}
+
+void
+ubc_cs_validation_bitmap_deallocate(
+       vnode_t         vp)
+{
+       struct ubc_info *uip;
+       void            *target_bitmap;
+       vm_object_size_t        bitmap_size;
+
+       if ( UBCINFOEXISTS(vp)) {
+               uip = vp->v_ubcinfo;
+
+               if ( (target_bitmap = uip->cs_valid_bitmap) != NULL ) {
+                       bitmap_size = uip->cs_valid_bitmap_size;
+                       kfree( target_bitmap, (vm_size_t) bitmap_size );
+                       uip->cs_valid_bitmap = NULL;
+               }
+       }
+}
+#else
+kern_return_t  ubc_cs_validation_bitmap_allocate(__unused vnode_t vp){
+       return KERN_INVALID_ARGUMENT;
+}
+
+kern_return_t ubc_cs_check_validation_bitmap(
+       __unused struct vnode *vp, 
+       __unused memory_object_offset_t offset,
+       __unused int optype){
+
+       return KERN_INVALID_ARGUMENT;
+}
+
+void   ubc_cs_validation_bitmap_deallocate(__unused vnode_t vp){
+       return;
+}
+#endif /* CHECK_CS_VALIDATION_BITMAP */
+
+#if PMAP_CS
+kern_return_t
+cs_associate_blob_with_mapping(
+       void                    *pmap,
+       vm_map_offset_t         start,
+       vm_map_size_t           size,
+       vm_object_offset_t      offset,
+       void                    *blobs_p)
+{
+       off_t                   blob_start_offset, blob_end_offset;
+       kern_return_t           kr;
+       struct cs_blob          *blobs, *blob;
+       vm_offset_t             kaddr;
+       struct pmap_cs_code_directory *cd_entry = NULL;
+
+       if (!pmap_cs) {
+               return KERN_NOT_SUPPORTED;
+       }
+       
+       blobs = (struct cs_blob *)blobs_p;
+
+       for (blob = blobs;
+            blob != NULL;
+            blob = blob->csb_next) {
+               blob_start_offset = (blob->csb_base_offset +
+                                    blob->csb_start_offset);
+               blob_end_offset = (blob->csb_base_offset +
+                                  blob->csb_end_offset);
+               if ((off_t) offset < blob_start_offset ||
+                   (off_t) offset >= blob_end_offset ||
+                   (off_t) (offset + size) <= blob_start_offset ||
+                   (off_t) (offset + size) > blob_end_offset) {
+                       continue;
+               }
+               kaddr = blob->csb_mem_kaddr;
+               if (kaddr == 0) {
+                       /* blob data has been released */
+                       continue;
+               }
+               cd_entry = blob->csb_pmap_cs_entry;
+               if (cd_entry == NULL) {
+                       continue;
+               }
+
+               break;
+       }
+
+       if (cd_entry != NULL) {
+               kr = pmap_cs_associate(pmap,
+                                      cd_entry,
+                                      start,
+                                      size);
+       } else {
+               kr = KERN_CODESIGN_ERROR;
+       }
+#if 00
+       printf("FBDP %d[%s] pmap_cs_associate(%p,%p,0x%llx,0x%llx) -> kr=0x%x\n", proc_selfpid(), &(current_proc()->p_comm[0]), pmap, cd_entry, (uint64_t)start, (uint64_t)size, kr);
+       kr = KERN_SUCCESS;
+#endif
+       return kr;
+}
+#endif /* PMAP_CS */