From 8ff7c917a8dbe541fb4bf797e21acc6d0edc7ceb Mon Sep 17 00:00:00 2001 From: Justus Winter <4winter@informatik.uni-hamburg.de> Date: Tue, 13 May 2014 20:09:05 +0200 Subject: new huge patch series --- .../0001-libihash-fix-type-of-max_load.patch | 31 ++ ...-use-an-integer-hash-function-on-the-keys.patch | 60 ++ ...e-linear-probing-and-fast-modulo-operatio.patch | 236 ++++++++ ...e-fast-binary-scaling-to-determine-the-lo.patch | 116 ++++ ...d-lock-less-reference-counting-primitives.patch | 213 +++++++ ...ock-less-reference-counting-for-peropen-o.patch | 143 +++++ ...ock-less-reference-counting-for-trivfs_pe.patch | 172 ++++++ .../0008-libports-use-a-single-hash-table.patch | 619 +++++++++++++++++++++ ...ck-less-reference-counting-for-port_info-.patch | 345 ++++++++++++ ...10-ext2fs-improve-enable-disable-_caching.patch | 38 ++ ...011-fatfs-improve-enable-disable-_caching.patch | 48 ++ ...s-use-a-seperate-lock-to-protect-nodehash.patch | 190 +++++++ ...s-use-a-seperate-lock-to-protect-nodehash.patch | 222 ++++++++ ...use-a-seperate-lock-to-protect-node_cache.patch | 222 ++++++++ ...-use-a-seperate-lock-to-protect-all_nodes.patch | 263 +++++++++ debian/patches/series | 15 + 16 files changed, 2933 insertions(+) create mode 100644 debian/patches/0001-libihash-fix-type-of-max_load.patch create mode 100644 debian/patches/0002-libihash-use-an-integer-hash-function-on-the-keys.patch create mode 100644 debian/patches/0003-libihash-use-linear-probing-and-fast-modulo-operatio.patch create mode 100644 debian/patches/0004-libihash-use-fast-binary-scaling-to-determine-the-lo.patch create mode 100644 debian/patches/0005-include-add-lock-less-reference-counting-primitives.patch create mode 100644 debian/patches/0006-libdiskfs-lock-less-reference-counting-for-peropen-o.patch create mode 100644 debian/patches/0007-libtrivfs-lock-less-reference-counting-for-trivfs_pe.patch create mode 100644 debian/patches/0008-libports-use-a-single-hash-table.patch create mode 100644 debian/patches/0009-libports-lock-less-reference-counting-for-port_info-.patch create mode 100644 debian/patches/0010-ext2fs-improve-enable-disable-_caching.patch create mode 100644 debian/patches/0011-fatfs-improve-enable-disable-_caching.patch create mode 100644 debian/patches/0012-ext2fs-use-a-seperate-lock-to-protect-nodehash.patch create mode 100644 debian/patches/0013-fatfs-use-a-seperate-lock-to-protect-nodehash.patch create mode 100644 debian/patches/0014-isofs-use-a-seperate-lock-to-protect-node_cache.patch create mode 100644 debian/patches/0015-tmpfs-use-a-seperate-lock-to-protect-all_nodes.patch diff --git a/debian/patches/0001-libihash-fix-type-of-max_load.patch b/debian/patches/0001-libihash-fix-type-of-max_load.patch new file mode 100644 index 00000000..557f0374 --- /dev/null +++ b/debian/patches/0001-libihash-fix-type-of-max_load.patch @@ -0,0 +1,31 @@ +From 6dcf53606fc7d46447176aab15954a897e19d6e5 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 13 May 2014 19:00:04 +0200 +Subject: [PATCH 01/15] libihash: fix type of max_load + +Previously, int was used for the field max_load of struct hurd_ihash. +There is no reason for this as far as I can tell. Furthermore, +hurd_ihash_set_max_load takes an unsigned int max_load. + +* libihash/ihash.h (struct hurd_ihash): Use unsigned int for field +max_load. +--- + libihash/ihash.h | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/libihash/ihash.h b/libihash/ihash.h +index 6bdc925..a24f384 100644 +--- a/libihash/ihash.h ++++ b/libihash/ihash.h +@@ -80,7 +80,7 @@ struct hurd_ihash + intptr_t locp_offset; + + /* The maximum load factor in percent. */ +- int max_load; ++ unsigned int max_load; + + /* When freeing or overwriting an element, this function is called + with the value as the first argument, and CLEANUP_DATA as the +-- +2.0.0.rc0 + diff --git a/debian/patches/0002-libihash-use-an-integer-hash-function-on-the-keys.patch b/debian/patches/0002-libihash-use-an-integer-hash-function-on-the-keys.patch new file mode 100644 index 00000000..820d0869 --- /dev/null +++ b/debian/patches/0002-libihash-use-an-integer-hash-function-on-the-keys.patch @@ -0,0 +1,60 @@ +From 2d898893815a980f1b821fcec267eb8e7ded678e Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Thu, 8 May 2014 15:45:00 +0200 +Subject: [PATCH 02/15] libihash: use an integer hash function on the keys + +Use an integer hash function to derive the index from the key. This +should reduce the number of collisions. + +* libihash/ihash.c (hash_int32): New function. +(find_index): Use hash_int32 on the key to derive the index. +(add_one): Likewise. +--- + libihash/ihash.c | 17 +++++++++++++++-- + 1 file changed, 15 insertions(+), 2 deletions(-) + +diff --git a/libihash/ihash.c b/libihash/ihash.c +index d670fee..71ddc26 100644 +--- a/libihash/ihash.c ++++ b/libihash/ihash.c +@@ -81,6 +81,19 @@ static const unsigned int ihash_nsizes = (sizeof ihash_sizes + / sizeof ihash_sizes[0]); + + ++/* This is the integer finalizer from MurmurHash3. */ ++static inline uint32_t ++murmur3_mix32 (uint32_t h, unsigned int bits) ++{ ++ h ^= h >> 16; ++ h *= 0x85ebca6b; ++ h ^= h >> 13; ++ h *= 0xc2b2ae35; ++ h ^= h >> 16; ++ ++ return h >> (32 - bits); ++} ++ + /* Return 1 if the slot with the index IDX in the hash table HT is + empty, and 0 otherwise. */ + static inline int +@@ -111,7 +124,7 @@ find_index (hurd_ihash_t ht, hurd_ihash_key_t key) + unsigned int up_idx; + unsigned int down_idx; + +- idx = key % ht->size; ++ idx = murmur3_mix32 (key, 32) % ht->size; + + if (ht->items[idx].value == _HURD_IHASH_EMPTY || ht->items[idx].key == key) + return idx; +@@ -264,7 +277,7 @@ add_one (hurd_ihash_t ht, hurd_ihash_key_t key, hurd_ihash_value_t value) + unsigned int idx; + unsigned int first_free; + +- idx = key % ht->size; ++ idx = murmur3_mix32 (key, 32) % ht->size; + first_free = idx; + + if (ht->items[idx].value != _HURD_IHASH_EMPTY && ht->items[idx].key != key) +-- +2.0.0.rc0 + diff --git a/debian/patches/0003-libihash-use-linear-probing-and-fast-modulo-operatio.patch b/debian/patches/0003-libihash-use-linear-probing-and-fast-modulo-operatio.patch new file mode 100644 index 00000000..e1cce7e5 --- /dev/null +++ b/debian/patches/0003-libihash-use-linear-probing-and-fast-modulo-operatio.patch @@ -0,0 +1,236 @@ +From aa4d444a81c32e2f1036041ad7474fb13b8bf693 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Thu, 8 May 2014 18:33:57 +0200 +Subject: [PATCH 03/15] libihash: use linear probing and fast modulo operation + +libihash uses open addressing. Previously, quadratic probing in both +directions was used to resolve collisions. Quadratic probing might +result in a less efficient use of caches. + +Also, prime numbers of the form 4 * i + 3 were used as array sizes. +This was used in combination with the integer modulo operation for +hashing. It has been known for some time that libihash suffers from +collisions, so a integer hash function is now applied to the keys to +derive the index. + +Use linear probing instead. Also, use powers of two for the array +sizes, so a bit mask can be used for the modulo operation. + +* libihash/ihash.c (ihash_sizes, ihash_nsizes): Remove. +(find_index): Use linear probing and fast modulo operation. +(add_one): Likewise. +* libihash/ihash.h (HURD_IHASH_MIN_SIZE): New macro. +--- + libihash/ihash.c | 125 +++++++------------------------------------------------ + libihash/ihash.h | 4 ++ + 2 files changed, 19 insertions(+), 110 deletions(-) + +diff --git a/libihash/ihash.c b/libihash/ihash.c +index 71ddc26..d628d75 100644 +--- a/libihash/ihash.c ++++ b/libihash/ihash.c +@@ -31,55 +31,6 @@ + #include + + #include "ihash.h" +- +- +-/* The prime numbers of the form 4 * i + 3 for some i, all greater +- than twice the previous one and smaller than 2^40 (for now). */ +-static const uint64_t ihash_sizes[] = +-{ +- 3, +- 7, +- 19, +- 43, +- 103, +- 211, +- 431, +- 863, +- 1747, +- 3499, +- 7019, +- 14051, +- 28111, +- 56239, +- 112507, +- 225023, +- 450067, +- 900139, +- 1800311, +- 3600659, +- 7201351, +- 14402743, +- 28805519, +- 57611039, +- 115222091, +- 230444239, +- 460888499, +- 921777067, +- 1843554151, +- UINT64_C (3687108307), +- UINT64_C (7374216631), +- UINT64_C (14748433279), +- UINT64_C (29496866579), +- UINT64_C (58993733159), +- UINT64_C (117987466379), +- UINT64_C (235974932759), +- UINT64_C (471949865531), +- UINT64_C (943899731087) +-}; +- +-static const unsigned int ihash_nsizes = (sizeof ihash_sizes +- / sizeof ihash_sizes[0]); +- + + /* This is the integer finalizer from MurmurHash3. */ + static inline uint32_t +@@ -120,40 +71,24 @@ static inline int + find_index (hurd_ihash_t ht, hurd_ihash_key_t key) + { + unsigned int idx; +- unsigned int i; + unsigned int up_idx; +- unsigned int down_idx; ++ unsigned int mask = ht->size - 1; + +- idx = murmur3_mix32 (key, 32) % ht->size; ++ idx = murmur3_mix32 (key, __builtin_ctz (ht->size)); + + if (ht->items[idx].value == _HURD_IHASH_EMPTY || ht->items[idx].key == key) + return idx; + +- /* Instead of calculating idx + 1, idx + 4, idx + 9, ..., idx + i^2, +- we add 1, 3, 5, 7, etc to the previous index. We do this in both +- directions separately. */ +- i = 1; + up_idx = idx; +- down_idx = idx; + + do + { +- up_idx = (up_idx + i) % ht->size; ++ up_idx = (up_idx + 1) & mask; + if (ht->items[up_idx].value == _HURD_IHASH_EMPTY + || ht->items[up_idx].key == key) + return up_idx; +- +- if (down_idx < i) +- down_idx += ht->size; +- down_idx = (down_idx - i) % ht->size; +- if (ht->items[down_idx].value == _HURD_IHASH_EMPTY +- || ht->items[down_idx].key == key) +- return down_idx; +- +- /* After (ht->size - 1) / 2 iterations, this will be 0. */ +- i = (i + 2) % ht->size; + } +- while (i); ++ while (up_idx != idx); + + /* If we end up here, the item could not be found. Return any + invalid index. */ +@@ -277,52 +212,25 @@ add_one (hurd_ihash_t ht, hurd_ihash_key_t key, hurd_ihash_value_t value) + unsigned int idx; + unsigned int first_free; + +- idx = murmur3_mix32 (key, 32) % ht->size; ++ idx = murmur3_mix32 (key, __builtin_ctz (ht->size)); + first_free = idx; + + if (ht->items[idx].value != _HURD_IHASH_EMPTY && ht->items[idx].key != key) + { +- /* Instead of calculating idx + 1, idx + 4, idx + 9, ..., idx + +- i^2, we add 1, 3, 5, 7, ... 2 * i - 1 to the previous index. +- We do this in both directions separately. */ +- unsigned int i = 1; ++ unsigned int mask = ht->size - 1; + unsigned int up_idx = idx; +- unsigned int down_idx = idx; +- ++ + do + { +- up_idx = (up_idx + i) % ht->size; ++ up_idx = (up_idx + 1) & mask; + if (ht->items[up_idx].value == _HURD_IHASH_EMPTY + || ht->items[up_idx].key == key) + { + idx = up_idx; + break; + } +- if (first_free == idx +- && ht->items[up_idx].value == _HURD_IHASH_DELETED) +- first_free = up_idx; +- +- if (down_idx < i) +- down_idx += ht->size; +- down_idx = (down_idx - i) % ht->size; +- if (down_idx < 0) +- down_idx += ht->size; +- else +- down_idx %= ht->size; +- if (ht->items[down_idx].value == _HURD_IHASH_EMPTY +- || ht->items[down_idx].key == key) +- { +- idx = down_idx; +- break; +- } +- if (first_free == idx +- && ht->items[down_idx].value == _HURD_IHASH_DELETED) +- first_free = down_idx; +- +- /* After (ht->size - 1) / 2 iterations, this will be 0. */ +- i = (i + 2) % ht->size; + } +- while (i); ++ while (up_idx != idx); + } + + /* Remove the old entry for this key if necessary. */ +@@ -365,21 +273,18 @@ hurd_ihash_add (hurd_ihash_t ht, hurd_ihash_key_t key, hurd_ihash_value_t item) + if (ht->size) + { + /* Only fill the hash table up to its maximum load factor. */ +- if (ht->nr_items * 100 / ht->size <= ht->max_load) ++ if ((ht->nr_items * 100) >> __builtin_ctz (ht->size) <= ht->max_load) + if (add_one (ht, key, item)) + return 0; + } + + /* The hash table is too small, and we have to increase it. */ +- for (i = 0; i < ihash_nsizes; i++) +- if (ihash_sizes[i] > old_ht.size) +- break; +- if (i == ihash_nsizes +- || ihash_sizes[i] > SIZE_MAX / sizeof (struct _hurd_ihash_item)) +- return ENOMEM; /* Surely will be true momentarily. */ +- + ht->nr_items = 0; +- ht->size = ihash_sizes[i]; ++ if (ht->size == 0) ++ ht->size = HURD_IHASH_MIN_SIZE; ++ else ++ ht->size <<= 1; ++ + /* calloc() will initialize all values to _HURD_IHASH_EMPTY implicitely. */ + ht->items = calloc (ht->size, sizeof (struct _hurd_ihash_item)); + +diff --git a/libihash/ihash.h b/libihash/ihash.h +index a24f384..8829e51 100644 +--- a/libihash/ihash.h ++++ b/libihash/ihash.h +@@ -93,6 +93,10 @@ typedef struct hurd_ihash *hurd_ihash_t; + + /* Construction and destruction of hash tables. */ + ++/* The size of the initial allocation in number of items. This must ++ be a power of two. */ ++#define HURD_IHASH_MIN_SIZE 32 ++ + /* The default value for the maximum load factor in percent. */ + #define HURD_IHASH_MAX_LOAD_DEFAULT 75 + +-- +2.0.0.rc0 + diff --git a/debian/patches/0004-libihash-use-fast-binary-scaling-to-determine-the-lo.patch b/debian/patches/0004-libihash-use-fast-binary-scaling-to-determine-the-lo.patch new file mode 100644 index 00000000..1a835a7f --- /dev/null +++ b/debian/patches/0004-libihash-use-fast-binary-scaling-to-determine-the-lo.patch @@ -0,0 +1,116 @@ +From 47ce159fbfca6e14f8ce442573c1fa10365113ff Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 13 May 2014 18:59:10 +0200 +Subject: [PATCH 04/15] libihash: use fast binary scaling to determine the load + +Expressing the maximum load in binary percent (where 128b% corresponds +to 100%) allows us to use fast binary scaling to determine if the +maximum load has been reached. + +* libihash/ihash.c (hurd_ihash_set_max_load): Update the comment. +(hurd_ihash_add): Use fast binary scaling to determine the current +load. +* libihash/ihash.h (hurd_ihash_set_max_load): Update the comment. +--- + libihash/ihash.c | 37 +++++++++++++++++++++++++++---------- + libihash/ihash.h | 22 ++++++++++++---------- + 2 files changed, 39 insertions(+), 20 deletions(-) + +diff --git a/libihash/ihash.c b/libihash/ihash.c +index d628d75..a4f92d1 100644 +--- a/libihash/ihash.c ++++ b/libihash/ihash.c +@@ -180,14 +180,15 @@ hurd_ihash_set_cleanup (hurd_ihash_t ht, hurd_ihash_cleanup_t cleanup, + } + + +-/* Set the maximum load factor in percent to MAX_LOAD, which should be +- between 1 and 100. The default is HURD_IHASH_MAX_LOAD_DEFAULT. +- New elements are only added to the hash table while the number of +- hashed elements is that much percent of the total size of the hash +- table. If more elements are added, the hash table is first +- expanded and reorganized. A MAX_LOAD of 100 will always fill the +- whole table before enlarging it, but note that this will increase +- the cost of operations significantly when the table is almost full. ++/* Set the maximum load factor in binary percent to MAX_LOAD, which ++ should be between 64 and 128. The default is ++ HURD_IHASH_MAX_LOAD_DEFAULT. New elements are only added to the ++ hash table while the number of hashed elements is that much percent ++ of the total size of the hash table. If more elements are added, ++ the hash table is first expanded and reorganized. A MAX_LOAD of ++ 128 will always fill the whole table before enlarging it, but note ++ that this will increase the cost of operations significantly when ++ the table is almost full. + + If the value is set to a smaller value than the current load + factor, the next reorganization will happen when a new item is +@@ -272,8 +273,24 @@ hurd_ihash_add (hurd_ihash_t ht, hurd_ihash_key_t key, hurd_ihash_value_t item) + + if (ht->size) + { +- /* Only fill the hash table up to its maximum load factor. */ +- if ((ht->nr_items * 100) >> __builtin_ctz (ht->size) <= ht->max_load) ++ /* Only fill the hash table up to its maximum load factor given ++ as "binary percent", where 128b% corresponds to 100%. As the ++ size is always a power of two, and 128 is also, the quotient ++ of both is also a power of two. Therefore, we can use bit ++ shifts to scale the number of items. ++ ++ load = nr_items * 128 / size ++ = nr_items^{log2 (128) - log2 (size)} ++ = nr_items >> (log2 (size) - log2 (128)) ++ -- if size >= 128 ++ = nr_items << (log2 (128) - log2 (size)) ++ -- otherwise ++ */ ++ int d = __builtin_ctz (ht->size) - 7; ++ unsigned int load = d >= 0 ++ ? ht->nr_items >> d ++ : ht->nr_items << -d; ++ if (load <= ht->max_load) + if (add_one (ht, key, item)) + return 0; + } +diff --git a/libihash/ihash.h b/libihash/ihash.h +index 8829e51..e245db4 100644 +--- a/libihash/ihash.h ++++ b/libihash/ihash.h +@@ -97,8 +97,9 @@ typedef struct hurd_ihash *hurd_ihash_t; + be a power of two. */ + #define HURD_IHASH_MIN_SIZE 32 + +-/* The default value for the maximum load factor in percent. */ +-#define HURD_IHASH_MAX_LOAD_DEFAULT 75 ++/* The default value for the maximum load factor in binary percent. ++ 96b% is equivalent to 75%, 128b% to 100%. */ ++#define HURD_IHASH_MAX_LOAD_DEFAULT 96 + + /* The LOCP_OFFS to use if no location pointer is available. */ + #define HURD_IHASH_NO_LOCP INTPTR_MIN +@@ -143,14 +144,15 @@ void hurd_ihash_free (hurd_ihash_t ht); + void hurd_ihash_set_cleanup (hurd_ihash_t ht, hurd_ihash_cleanup_t cleanup, + void *cleanup_data); + +-/* Set the maximum load factor in percent to MAX_LOAD, which should be +- between 50 and 100. The default is HURD_IHASH_MAX_LOAD_DEFAULT. +- New elements are only added to the hash table while the number of +- hashed elements is that much percent of the total size of the hash +- table. If more elements are added, the hash table is first +- expanded and reorganized. A MAX_LOAD of 100 will always fill the +- whole table before enlarging it, but note that this will increase +- the cost of operations significantly when the table is almost full. ++/* Set the maximum load factor in binary percent to MAX_LOAD, which ++ should be between 64 and 128. The default is ++ HURD_IHASH_MAX_LOAD_DEFAULT. New elements are only added to the ++ hash table while the number of hashed elements is that much percent ++ of the total size of the hash table. If more elements are added, ++ the hash table is first expanded and reorganized. A MAX_LOAD of ++ 128 will always fill the whole table before enlarging it, but note ++ that this will increase the cost of operations significantly when ++ the table is almost full. + + If the value is set to a smaller value than the current load + factor, the next reorganization will happen when a new item is +-- +2.0.0.rc0 + diff --git a/debian/patches/0005-include-add-lock-less-reference-counting-primitives.patch b/debian/patches/0005-include-add-lock-less-reference-counting-primitives.patch new file mode 100644 index 00000000..4a4dd9dc --- /dev/null +++ b/debian/patches/0005-include-add-lock-less-reference-counting-primitives.patch @@ -0,0 +1,213 @@ +From f6ed9ea2b93816dff52e33b71d69ed6787a57841 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 6 May 2014 19:52:04 +0200 +Subject: [PATCH 05/15] include: add lock-less reference counting primitives + +* include/refcount.h: New file. +--- + include/refcount.h | 193 +++++++++++++++++++++++++++++++++++++++++++++++++++++ + 1 file changed, 193 insertions(+) + create mode 100644 include/refcount.h + +diff --git a/include/refcount.h b/include/refcount.h +new file mode 100644 +index 0000000..0816220 +--- /dev/null ++++ b/include/refcount.h +@@ -0,0 +1,193 @@ ++/* Lock-less reference counting primitives ++ ++ Copyright (C) 2014 Free Software Foundation, Inc. ++ ++ Written by Justus Winter <4winter@informatik.uni-hamburg.de> ++ ++ This file is part of the GNU Hurd. ++ ++ The GNU Hurd is free software; you can redistribute it and/or ++ modify it under the terms of the GNU General Public License as ++ published by the Free Software Foundation; either version 2, or (at ++ your option) any later version. ++ ++ The GNU Hurd is distributed in the hope that it will be useful, but ++ WITHOUT ANY WARRANTY; without even the implied warranty of ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU ++ General Public License for more details. ++ ++ You should have received a copy of the GNU General Public License ++ along with the GNU Hurd. If not, see . */ ++ ++#ifndef _HURD_REFCOUNT_H_ ++#define _HURD_REFCOUNT_H_ ++ ++#include ++#include ++ ++/* Simple reference counting. */ ++ ++/* An opaque type. You must not access these values directly. */ ++typedef unsigned int refcount_t; ++ ++/* Initialize REF with REFERENCES. */ ++static inline void ++refcount_init (refcount_t *ref, unsigned int references) ++{ ++ *ref = references; ++} ++ ++/* Increment REF. Return the result of the operation. This function ++ uses atomic operations. It is not required to serialize calls to ++ this function. */ ++static inline unsigned int ++refcount_ref (refcount_t *ref) ++{ ++ unsigned int r; ++ r = __atomic_add_fetch (ref, 1, __ATOMIC_RELAXED); ++ assert (r != UINT_MAX); ++ return r; ++} ++ ++/* Decrement REF. Return the result of the operation. This function ++ uses atomic operations. It is not required to serialize calls to ++ this function. */ ++static inline unsigned int ++refcount_deref (refcount_t *ref) ++{ ++ unsigned int r; ++ r = __atomic_sub_fetch (ref, 1, __ATOMIC_RELAXED); ++ assert (r != UINT_MAX); ++ return r; ++} ++ ++/* Return REF. This function uses atomic operations. It is not ++ required to serialize calls to this function. */ ++static inline unsigned int ++refcount_references (refcount_t *ref) ++{ ++ return __atomic_load_n (ref, __ATOMIC_RELAXED); ++} ++ ++/* Reference counting with weak references. */ ++ ++/* An opaque type. You must not access these values directly. */ ++typedef union _references refcounts_t; ++ ++/* Instead, the functions manipulating refcounts_t values write the ++ results into this kind of objects. */ ++struct references { ++ uint32_t hard; ++ uint32_t weak; ++}; ++ ++/* We use a union to convert struct reference values to uint64_t which ++ we can manipulate atomically. While this behavior is not ++ guaranteed by the C standard, it is supported by all major ++ compilers. */ ++union _references { ++ struct references references; ++ uint64_t value; ++}; ++ ++/* Initialize REF with HARD and WEAK references. */ ++static inline void ++refcounts_init (refcounts_t *ref, uint32_t hard, uint32_t weak) ++{ ++ ref->references = (struct references) { .hard = hard, .weak = weak }; ++} ++ ++/* Increment the hard reference count of REF. If RESULT is not NULL, ++ the result of the operation is written there. This function uses ++ atomic operations. It is not required to serialize calls to this ++ function. */ ++static inline void ++refcounts_ref (refcounts_t *ref, struct references *result) ++{ ++ const union _references op = { .references = { .hard = 1 } }; ++ union _references r; ++ r.value = __atomic_add_fetch (&ref->value, op.value, __ATOMIC_RELAXED); ++ assert (r.references.hard != UINT32_MAX); ++ if (result) ++ *result = r.references; ++} ++ ++/* Decrement the hard reference count of REF. If RESULT is not NULL, ++ the result of the operation is written there. This function uses ++ atomic operations. It is not required to serialize calls to this ++ function. */ ++static inline void ++refcounts_deref (refcounts_t *ref, struct references *result) ++{ ++ const union _references op = { .references = { .hard = 1 } }; ++ union _references r; ++ r.value = __atomic_sub_fetch (&ref->value, op.value, __ATOMIC_RELAXED); ++ assert (r.references.hard != UINT32_MAX); ++ if (result) ++ *result = r.references; ++} ++ ++/* Increment the weak reference count of REF. If RESULT is not NULL, ++ the result of the operation is written there. This function uses ++ atomic operations. It is not required to serialize calls to this ++ function. */ ++static inline void ++refcounts_ref_weak (refcounts_t *ref, struct references *result) ++{ ++ const union _references op = { .references = { .weak = 1 } }; ++ union _references r; ++ r.value = __atomic_add_fetch (&ref->value, op.value, __ATOMIC_RELAXED); ++ assert (r.references.weak != UINT32_MAX); ++ if (result) ++ *result = r.references; ++} ++ ++/* Decrement the weak reference count of REF. If RESULT is not NULL, ++ the result of the operation is written there. This function uses ++ atomic operations. It is not required to serialize calls to this ++ function. */ ++static inline void ++refcounts_deref_weak (refcounts_t *ref, struct references *result) ++{ ++ const union _references op = { .references = { .weak = 1 } }; ++ union _references r; ++ r.value = __atomic_sub_fetch (&ref->value, op.value, __ATOMIC_RELAXED); ++ assert (r.references.weak != UINT32_MAX); ++ if (result) ++ *result = r.references; ++} ++ ++/* Store the current reference counts of REF in RESULT. This function ++ uses atomic operations. It is not required to serialize calls to ++ this function. */ ++static inline void ++refcounts_references (refcounts_t *ref, struct references *result) ++{ ++ union _references r; ++ r.value =__atomic_load_n (&ref->value, __ATOMIC_RELAXED); ++ *result = r.references; ++} ++ ++/* Return the hard reference count of REF. This function uses atomic ++ operations. It is not required to serialize calls to this ++ function. */ ++static inline uint32_t ++refcounts_hard_references (refcounts_t *ref) ++{ ++ struct references result; ++ refcounts_references (ref, &result); ++ return result.hard; ++} ++ ++/* Return the weak reference count of REF. This function uses atomic ++ operations. It is not required to serialize calls to this ++ function. */ ++static inline uint32_t ++refcounts_weak_references (refcounts_t *ref) ++{ ++ struct references result; ++ refcounts_references (ref, &result); ++ return result.weak; ++} ++ ++#endif /* _HURD_REFCOUNT_H_ */ +-- +2.0.0.rc0 + diff --git a/debian/patches/0006-libdiskfs-lock-less-reference-counting-for-peropen-o.patch b/debian/patches/0006-libdiskfs-lock-less-reference-counting-for-peropen-o.patch new file mode 100644 index 00000000..d688b3e5 --- /dev/null +++ b/debian/patches/0006-libdiskfs-lock-less-reference-counting-for-peropen-o.patch @@ -0,0 +1,143 @@ +From 78fbd3e318474ea61bc31783bc471809910b9244 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 6 May 2014 18:58:10 +0200 +Subject: [PATCH 06/15] libdiskfs: lock-less reference counting for peropen + objects + +* libdiskfs/diskfs.h (struct peropen): Use refcount_t for field refcnt. +* libdiskfs/peropen-make.c (diskfs_make_peropen): Initialize refcnt. +* libdiskfs/peropen-rele.c (diskfs_release_peropen): Adjust accordingly. +* libdiskfs/protid-make.c (diskfs_start_protid): Likewise. Also, the +node must no longer be locked, adjust comment accordingly. +(diskfs_create_protid): Likewise. +--- + libdiskfs/diskfs.h | 7 ++++--- + libdiskfs/peropen-make.c | 2 +- + libdiskfs/peropen-rele.c | 21 ++++++++++----------- + libdiskfs/protid-make.c | 8 ++++---- + 4 files changed, 19 insertions(+), 19 deletions(-) + +diff --git a/libdiskfs/diskfs.h b/libdiskfs/diskfs.h +index 8151ddc..ae1a150 100644 +--- a/libdiskfs/diskfs.h ++++ b/libdiskfs/diskfs.h +@@ -28,6 +28,7 @@ + #include + #include + #include ++#include + + #ifdef DISKFS_DEFINE_EXTERN_INLINE + #define DISKFS_EXTERN_INLINE +@@ -57,7 +58,7 @@ struct peropen + { + int filepointer; + int lock_status; +- int refcnt; ++ refcount_t refcnt; + int openstat; + + struct node *np; +@@ -792,12 +793,12 @@ diskfs_create_node (struct node *dir, const char *name, mode_t mode, + struct dirstat *ds); + + /* Create and return a protid for an existing peropen PO in CRED, +- referring to user USER. The node PO->np must be locked. */ ++ referring to user USER. */ + error_t diskfs_create_protid (struct peropen *po, struct iouser *user, + struct protid **cred); + + /* Build and return in CRED a protid which has no user identification, for +- peropen PO. The node PO->np must be locked. */ ++ peropen PO. */ + error_t diskfs_start_protid (struct peropen *po, struct protid **cred); + + /* Finish building protid CRED started with diskfs_start_protid; +diff --git a/libdiskfs/peropen-make.c b/libdiskfs/peropen-make.c +index eba037f..6d5ca01 100644 +--- a/libdiskfs/peropen-make.c ++++ b/libdiskfs/peropen-make.c +@@ -31,7 +31,7 @@ diskfs_make_peropen (struct node *np, int flags, struct peropen *context, + + po->filepointer = 0; + po->lock_status = LOCK_UN; +- po->refcnt = 0; ++ refcount_init (&po->refcnt, 0); + po->openstat = flags; + po->np = np; + po->path = NULL; +diff --git a/libdiskfs/peropen-rele.c b/libdiskfs/peropen-rele.c +index d3f7492..877137b 100644 +--- a/libdiskfs/peropen-rele.c ++++ b/libdiskfs/peropen-rele.c +@@ -22,13 +22,8 @@ + void + diskfs_release_peropen (struct peropen *po) + { +- pthread_mutex_lock (&po->np->lock); +- +- if (--po->refcnt) +- { +- pthread_mutex_unlock (&po->np->lock); +- return; +- } ++ if (refcount_deref (&po->refcnt) > 0) ++ return; + + if (po->root_parent) + mach_port_deallocate (mach_task_self (), po->root_parent); +@@ -40,10 +35,14 @@ diskfs_release_peropen (struct peropen *po) + mach_port_deallocate (mach_task_self (), po->shadow_root_parent); + + if (po->lock_status != LOCK_UN) +- fshelp_acquire_lock (&po->np->userlock, &po->lock_status, +- &po->np->lock, LOCK_UN); +- +- diskfs_nput (po->np); ++ { ++ pthread_mutex_lock (&po->np->lock); ++ fshelp_acquire_lock (&po->np->userlock, &po->lock_status, ++ &po->np->lock, LOCK_UN); ++ diskfs_nput (po->np); ++ } ++ else ++ diskfs_nrele (po->np); + + free (po->path); + free (po); +diff --git a/libdiskfs/protid-make.c b/libdiskfs/protid-make.c +index b39b92a..22aaa2e 100644 +--- a/libdiskfs/protid-make.c ++++ b/libdiskfs/protid-make.c +@@ -20,7 +20,7 @@ + #include + + /* Build and return in CRED a protid which has no user identification, for +- peropen PO. The node PO->np must be locked. */ ++ peropen PO. */ + error_t + diskfs_start_protid (struct peropen *po, struct protid **cred) + { +@@ -29,7 +29,7 @@ diskfs_start_protid (struct peropen *po, struct protid **cred) + sizeof (struct protid), cred); + if (! err) + { +- po->refcnt++; ++ refcount_ref (&po->refcnt); + (*cred)->po = po; + (*cred)->shared_object = MACH_PORT_NULL; + (*cred)->mapped = 0; +@@ -55,8 +55,8 @@ diskfs_finish_protid (struct protid *cred, struct iouser *user) + assert_perror (err); + } + +-/* Create and return a protid for an existing peropen PO in CRED for USER. +- The node PO->np must be locked. */ ++/* Create and return a protid for an existing peropen PO in CRED for ++ USER. */ + error_t + diskfs_create_protid (struct peropen *po, struct iouser *user, + struct protid **cred) +-- +2.0.0.rc0 + diff --git a/debian/patches/0007-libtrivfs-lock-less-reference-counting-for-trivfs_pe.patch b/debian/patches/0007-libtrivfs-lock-less-reference-counting-for-trivfs_pe.patch new file mode 100644 index 00000000..7e3bf100 --- /dev/null +++ b/debian/patches/0007-libtrivfs-lock-less-reference-counting-for-trivfs_pe.patch @@ -0,0 +1,172 @@ +From c4bd59b5b1bda56338215fd509e9414ec8c7f052 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 6 May 2014 19:07:13 +0200 +Subject: [PATCH 07/15] libtrivfs: lock-less reference counting for + trivfs_peropen objects + +* libtrivfs/trivfs.h (struct trivfs_peropen): Use refcount_t for field +refcnt. +(struct trivfs_control): Remove unused field lock. +* libtrivfs/cntl-create.c (trivfs_create_control): Drop the mutex +initialization. +* libtrivfs/io-reauthenticate.c (trivfs_S_io_reauthenticate): Adjust +accordingly. +* libtrivfs/io-restrict-auth.c (trivfs_S_io_restrict_auth): Likewise. +* libtrivfs/open.c (trivfs_open): Initialize refcnt. +* libtrivfs/protid-clean.c (trivfs_clean_protid): Likewise. +* libtrivfs/protid-dup.c (trivfs_protid_dup): Likewise. +--- + libtrivfs/cntl-create.c | 1 - + libtrivfs/io-reauthenticate.c | 5 +---- + libtrivfs/io-restrict-auth.c | 4 +--- + libtrivfs/open.c | 2 +- + libtrivfs/protid-clean.c | 26 +++++++++++++++----------- + libtrivfs/protid-dup.c | 5 +---- + libtrivfs/trivfs.h | 4 ++-- + 7 files changed, 21 insertions(+), 26 deletions(-) + +diff --git a/libtrivfs/cntl-create.c b/libtrivfs/cntl-create.c +index 910daf3..eb9a834 100644 +--- a/libtrivfs/cntl-create.c ++++ b/libtrivfs/cntl-create.c +@@ -85,7 +85,6 @@ trivfs_create_control (mach_port_t underlying, + } + + (*control)->hook = 0; +- pthread_mutex_init (&(*control)->lock, NULL); + } + + out: +diff --git a/libtrivfs/io-reauthenticate.c b/libtrivfs/io-reauthenticate.c +index 7677697..df0ed2e 100644 +--- a/libtrivfs/io-reauthenticate.c ++++ b/libtrivfs/io-reauthenticate.c +@@ -62,11 +62,8 @@ trivfs_S_io_reauthenticate (struct trivfs_protid *cred, + newcred->isroot = 1; + + newcred->hook = cred->hook; +- +- pthread_mutex_lock (&cred->po->cntl->lock); + newcred->po = cred->po; +- newcred->po->refcnt++; +- pthread_mutex_unlock (&cred->po->cntl->lock); ++ refcount_ref (&newcred->po->refcnt); + + do + err = io_restrict_auth (newcred->po->cntl->underlying, &newcred->realnode, +diff --git a/libtrivfs/io-restrict-auth.c b/libtrivfs/io-restrict-auth.c +index 65b4fd6..39670fe 100644 +--- a/libtrivfs/io-restrict-auth.c ++++ b/libtrivfs/io-restrict-auth.c +@@ -110,10 +110,8 @@ trivfs_S_io_restrict_auth (struct trivfs_protid *cred, + } + + newcred->isroot = 0; +- pthread_mutex_lock (&cred->po->cntl->lock); + newcred->po = cred->po; +- newcred->po->refcnt++; +- pthread_mutex_unlock (&cred->po->cntl->lock); ++ refcount_ref (&newcred->po->refcnt); + if (cred->isroot && idvec_contains (user->uids, 0)) + newcred->isroot = 1; + newcred->user = user; +diff --git a/libtrivfs/open.c b/libtrivfs/open.c +index f64d2ff..97e70a1 100644 +--- a/libtrivfs/open.c ++++ b/libtrivfs/open.c +@@ -40,7 +40,7 @@ trivfs_open (struct trivfs_control *cntl, + + ports_port_ref (cntl); + +- po->refcnt = 1; ++ refcount_init (&po->refcnt, 1); + po->cntl = cntl; + po->openmodes = flags; + po->hook = 0; +diff --git a/libtrivfs/protid-clean.c b/libtrivfs/protid-clean.c +index f98da6a..cce736d 100644 +--- a/libtrivfs/protid-clean.c ++++ b/libtrivfs/protid-clean.c +@@ -31,19 +31,23 @@ trivfs_clean_protid (void *arg) + (*trivfs_protid_destroy_hook) (cred); + + /* If we hold the only reference to the peropen, try to get rid of it. */ +- pthread_mutex_lock (&cntl->lock); +- if (cred->po->refcnt == 1 && trivfs_peropen_destroy_hook) ++ if (refcount_deref (&cred->po->refcnt) == 0) + { +- pthread_mutex_unlock (&cntl->lock); +- (*trivfs_peropen_destroy_hook) (cred->po); +- pthread_mutex_lock (&cntl->lock); ++ if (trivfs_peropen_destroy_hook) ++ { ++ /* Reaquire a reference while we call the hook. */ ++ refcount_ref (&cred->po->refcnt); ++ (*trivfs_peropen_destroy_hook) (cred->po); ++ if (refcount_deref (&cred->po->refcnt) == 0) ++ goto free_po; ++ } ++ else ++ { ++ free_po: ++ ports_port_deref (cntl); ++ free (cred->po); ++ } + } +- if (--cred->po->refcnt == 0) +- { +- ports_port_deref (cntl); +- free (cred->po); +- } +- pthread_mutex_unlock (&cntl->lock); + + iohelp_free_iouser (cred->user); + +diff --git a/libtrivfs/protid-dup.c b/libtrivfs/protid-dup.c +index 6169603..75f3ca8 100644 +--- a/libtrivfs/protid-dup.c ++++ b/libtrivfs/protid-dup.c +@@ -35,11 +35,8 @@ trivfs_protid_dup (struct trivfs_protid *cred, struct trivfs_protid **dup) + + if (! err) + { +- pthread_mutex_lock (&cred->po->cntl->lock); + new->po = cred->po; +- new->po->refcnt++; +- pthread_mutex_unlock (&cred->po->cntl->lock); +- ++ refcount_ref (&new->po->refcnt); + new->isroot = cred->isroot; + + err = iohelp_dup_iouser (&new->user, cred->user); +diff --git a/libtrivfs/trivfs.h b/libtrivfs/trivfs.h +index bb456ff..8902338 100644 +--- a/libtrivfs/trivfs.h ++++ b/libtrivfs/trivfs.h +@@ -24,6 +24,7 @@ + #include + #include + #include ++#include + + struct trivfs_protid + { +@@ -41,14 +42,13 @@ struct trivfs_peropen + { + void *hook; /* for user use */ + int openmodes; +- int refcnt; ++ refcount_t refcnt; + struct trivfs_control *cntl; + }; + + struct trivfs_control + { + struct port_info pi; +- pthread_mutex_t lock; + struct port_class *protid_class; + struct port_bucket *protid_bucket; + mach_port_t filesys_id; +-- +2.0.0.rc0 + diff --git a/debian/patches/0008-libports-use-a-single-hash-table.patch b/debian/patches/0008-libports-use-a-single-hash-table.patch new file mode 100644 index 00000000..b76029cd --- /dev/null +++ b/debian/patches/0008-libports-use-a-single-hash-table.patch @@ -0,0 +1,619 @@ +From 5d9f77f93862d2f404704071ccb79efdd506e2a7 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Sat, 3 May 2014 03:53:41 +0200 +Subject: [PATCH 08/15] libports: use a single hash table + +Previously, libports used a hash table per port bucket. This makes +looking up a port difficult if one does not know the port bucket, as +one has to iterate over all buckets and do a hash table lookup each. + +Having to iterate over the buckets makes it necessary to keep a list +of all buckets, which has to be updated and protected by a lock as +well. + +Also, the current code in _ports_bucket_class_iterate iterates over +the hash table associated with the bucket given. When +ports_class_iterate calls this common function, it obtains a reference +to the bucket from one of the ports in the given class. This will not +work if a class contains ports in different port buckets. This +limitation is not documented as far as I can see. Again, having to +maintain this list has its cost and requires serialization. + +Use a single hash table instead. Furthermore, serialize access to it +using a separate lock. Remove the linked lists of all buckets and all +ports in a class. + +* libports/bucket-iterate.c (ports_bucket_iterate): Acquire +_ports_htable_lock. Also, generalize ports_bucket_iterate to treat +the bucket argument the same way as the class argument. +* libports/class-iterate.c (ports_class_iterate): Just call the +generalized _ports_bucket_class_iterate with NULL as class. +* libports/ports.h (struct port_info): Remove the port class links. +(struct port_bucket): Remove the hash table, and the all buckets link. +(_ports_all_buckets): Remove declaration. +(_ports_htable): New global hash table. +(_ports_htable_lock): Protected by this lock. +* libports/claim-right.c: Adjust accordingly. +* libports/complete-deallocate.c: Likewise. +* libports/create-bucket.c: Likewise. +* libports/create-class.c: Likewise. +* libports/create-internal.c: Likewise. +* libports/destroy-right.c: Likewise. +* libports/import-port.c: Likewise. +* libports/lookup-port.c: Likewise. +* libports/reallocate-from-external.c: Likewise. +* libports/reallocate-port.c: Likewise. +* libports/transfer-right.c: Likewise. +* libports/inhibit-all-rpcs.c: Iterate over the hash table. +* libports/inhibit-bucket-rpcs.c: Likewise, but filter using bucket. +* libports/inhibit-class-rpcs.c: Likewise, but filter using class. +* libports/init.c (_ports_htable): Initialize. +(_ports_htable_lock): Likewise. +--- + libports/bucket-iterate.c | 21 +++++++++++++++------ + libports/claim-right.c | 6 ++++-- + libports/class-iterate.c | 10 +--------- + libports/complete-deallocate.c | 8 +++----- + libports/create-bucket.c | 7 ------- + libports/create-class.c | 1 - + libports/create-internal.c | 9 +++------ + libports/destroy-right.c | 6 +++--- + libports/import-port.c | 9 +++------ + libports/inhibit-all-rpcs.c | 27 +++++++++++++-------------- + libports/inhibit-bucket-rpcs.c | 7 +++++-- + libports/inhibit-class-rpcs.c | 27 ++++++++++++++++++--------- + libports/init.c | 7 ++++++- + libports/lookup-port.c | 21 ++++++++------------- + libports/ports.h | 11 ++++++----- + libports/reallocate-from-external.c | 16 ++++++++++------ + libports/reallocate-port.c | 10 +++++++--- + libports/transfer-right.c | 20 +++++++++++++------- + 18 files changed, 118 insertions(+), 105 deletions(-) + +diff --git a/libports/bucket-iterate.c b/libports/bucket-iterate.c +index babc204..d5f590e 100644 +--- a/libports/bucket-iterate.c ++++ b/libports/bucket-iterate.c +@@ -36,35 +36,44 @@ _ports_bucket_class_iterate (struct port_bucket *bucket, + error_t err; + + pthread_mutex_lock (&_ports_lock); ++ pthread_mutex_lock (&_ports_htable_lock); + +- if (bucket->htable.nr_items == 0) ++ if (_ports_htable.nr_items == 0) + { +- pthread_mutex_unlock (&_ports_lock); ++ pthread_mutex_unlock (&_ports_htable_lock); + return 0; + } + +- nr_items = bucket->htable.nr_items; ++ nr_items = _ports_htable.nr_items; + p = malloc (nr_items * sizeof *p); + if (p == NULL) + { +- pthread_mutex_unlock (&_ports_lock); ++ pthread_mutex_unlock (&_ports_htable_lock); + return ENOMEM; + } + + n = 0; +- HURD_IHASH_ITERATE (&bucket->htable, arg) ++ HURD_IHASH_ITERATE (&_ports_htable, arg) + { + struct port_info *const pi = arg; + +- if (class == 0 || pi->class == class) ++ if ((bucket == NULL || pi->bucket == bucket) ++ && (class == NULL || pi->class == class)) + { + pi->refcnt++; + p[n] = pi; + n++; + } + } ++ pthread_mutex_unlock (&_ports_htable_lock); + pthread_mutex_unlock (&_ports_lock); + ++ if (n == 0) ++ { ++ free (p); ++ return 0; ++ } ++ + if (n != nr_items) + { + /* We allocated too much. Release unused memory. */ +diff --git a/libports/claim-right.c b/libports/claim-right.c +index 4851ea3..f453a82 100644 +--- a/libports/claim-right.c ++++ b/libports/claim-right.c +@@ -34,10 +34,12 @@ ports_claim_right (void *portstruct) + if (ret == MACH_PORT_NULL) + return ret; + +- pthread_mutex_lock (&_ports_lock); +- hurd_ihash_locp_remove (&pi->bucket->htable, pi->hentry); ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, pi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); + err = mach_port_move_member (mach_task_self (), ret, MACH_PORT_NULL); + assert_perror (err); ++ pthread_mutex_lock (&_ports_lock); + pi->port_right = MACH_PORT_NULL; + if (pi->flags & PORT_HAS_SENDRIGHTS) + { +diff --git a/libports/class-iterate.c b/libports/class-iterate.c +index 1f8878a..5cfcfc8 100644 +--- a/libports/class-iterate.c ++++ b/libports/class-iterate.c +@@ -23,13 +23,5 @@ error_t + ports_class_iterate (struct port_class *class, + error_t (*fun)(void *)) + { +- pthread_mutex_lock (&_ports_lock); +- if (class->ports != 0) +- { +- struct port_bucket *bucket = class->ports->bucket; +- pthread_mutex_unlock (&_ports_lock); +- return _ports_bucket_class_iterate (bucket, class, fun); +- } +- pthread_mutex_unlock (&_ports_lock); +- return 0; ++ return _ports_bucket_class_iterate (NULL, class, fun); + } +diff --git a/libports/complete-deallocate.c b/libports/complete-deallocate.c +index 8ce095b..1371d92 100644 +--- a/libports/complete-deallocate.c ++++ b/libports/complete-deallocate.c +@@ -29,16 +29,14 @@ _ports_complete_deallocate (struct port_info *pi) + + if (pi->port_right) + { +- hurd_ihash_locp_remove (&pi->bucket->htable, pi->hentry); ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, pi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); + mach_port_mod_refs (mach_task_self (), pi->port_right, + MACH_PORT_RIGHT_RECEIVE, -1); + pi->port_right = MACH_PORT_NULL; + } + +- *pi->prevp = pi->next; +- if (pi->next) +- pi->next->prevp = pi->prevp; +- + pi->bucket->count--; + pi->class->count--; + +diff --git a/libports/create-bucket.c b/libports/create-bucket.c +index 52d50c3..9f6821d 100644 +--- a/libports/create-bucket.c ++++ b/libports/create-bucket.c +@@ -46,13 +46,6 @@ ports_create_bucket () + return NULL; + } + +- hurd_ihash_init (&ret->htable, offsetof (struct port_info, hentry)); + ret->rpcs = ret->flags = ret->count = 0; +- +- pthread_mutex_lock (&_ports_lock); +- ret->next = _ports_all_buckets; +- _ports_all_buckets = ret; +- pthread_mutex_unlock (&_ports_lock); +- + return ret; + } +diff --git a/libports/create-class.c b/libports/create-class.c +index 12c8add..782f52b 100644 +--- a/libports/create-class.c ++++ b/libports/create-class.c +@@ -39,7 +39,6 @@ ports_create_class (void (*clean_routine)(void *), + cl->dropweak_routine = dropweak_routine; + cl->flags = 0; + cl->rpcs = 0; +- cl->ports = NULL; + cl->count = 0; + cl->uninhibitable_rpcs = ports_default_uninhibitable_rpcs; + +diff --git a/libports/create-internal.c b/libports/create-internal.c +index 8551297..e773dd6 100644 +--- a/libports/create-internal.c ++++ b/libports/create-internal.c +@@ -81,15 +81,12 @@ _ports_create_port_internal (struct port_class *class, + goto loop; + } + +- err = hurd_ihash_add (&bucket->htable, port, pi); ++ pthread_mutex_lock (&_ports_htable_lock); ++ err = hurd_ihash_add (&_ports_htable, port, pi); ++ pthread_mutex_unlock (&_ports_htable_lock); + if (err) + goto lose; + +- pi->next = class->ports; +- pi->prevp = &class->ports; +- if (class->ports) +- class->ports->prevp = &pi->next; +- class->ports = pi; + bucket->count++; + class->count++; + pthread_mutex_unlock (&_ports_lock); +diff --git a/libports/destroy-right.c b/libports/destroy-right.c +index 65e19c7..6ba7302 100644 +--- a/libports/destroy-right.c ++++ b/libports/destroy-right.c +@@ -30,12 +30,12 @@ ports_destroy_right (void *portstruct) + + if (pi->port_right != MACH_PORT_NULL) + { +- pthread_mutex_lock (&_ports_lock); +- hurd_ihash_locp_remove (&pi->bucket->htable, pi->hentry); ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, pi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); + err = mach_port_mod_refs (mach_task_self (), pi->port_right, + MACH_PORT_RIGHT_RECEIVE, -1); + assert_perror (err); +- pthread_mutex_unlock (&_ports_lock); + + pi->port_right = MACH_PORT_NULL; + +diff --git a/libports/import-port.c b/libports/import-port.c +index 226f47e..d0b2ea4 100644 +--- a/libports/import-port.c ++++ b/libports/import-port.c +@@ -75,15 +75,12 @@ ports_import_port (struct port_class *class, struct port_bucket *bucket, + goto loop; + } + +- err = hurd_ihash_add (&bucket->htable, port, pi); ++ pthread_mutex_lock (&_ports_htable_lock); ++ err = hurd_ihash_add (&_ports_htable, port, pi); ++ pthread_mutex_unlock (&_ports_htable_lock); + if (err) + goto lose; + +- pi->next = class->ports; +- pi->prevp = &class->ports; +- if (class->ports) +- class->ports->prevp = &pi->next; +- class->ports = pi; + bucket->count++; + class->count++; + pthread_mutex_unlock (&_ports_lock); +diff --git a/libports/inhibit-all-rpcs.c b/libports/inhibit-all-rpcs.c +index d4a54ba..83c291f 100644 +--- a/libports/inhibit-all-rpcs.c ++++ b/libports/inhibit-all-rpcs.c +@@ -36,24 +36,23 @@ ports_inhibit_all_rpcs () + struct port_bucket *bucket; + int this_one = 0; + +- for (bucket = _ports_all_buckets; bucket; bucket = bucket->next) ++ pthread_mutex_lock (&_ports_htable_lock); ++ HURD_IHASH_ITERATE (&_ports_htable, portstruct) + { +- HURD_IHASH_ITERATE (&bucket->htable, portstruct) ++ struct rpc_info *rpc; ++ struct port_info *pi = portstruct; ++ ++ for (rpc = pi->current_rpcs; rpc; rpc = rpc->next) + { +- struct rpc_info *rpc; +- struct port_info *pi = portstruct; +- +- for (rpc = pi->current_rpcs; rpc; rpc = rpc->next) +- { +- /* Avoid cancelling the calling thread if it's currently +- handling a RPC. */ +- if (rpc->thread == hurd_thread_self ()) +- this_one = 1; +- else +- hurd_thread_cancel (rpc->thread); +- } ++ /* Avoid cancelling the calling thread if it's currently ++ handling a RPC. */ ++ if (rpc->thread == hurd_thread_self ()) ++ this_one = 1; ++ else ++ hurd_thread_cancel (rpc->thread); + } + } ++ pthread_mutex_unlock (&_ports_htable_lock); + + while (_ports_total_rpcs > this_one) + { +diff --git a/libports/inhibit-bucket-rpcs.c b/libports/inhibit-bucket-rpcs.c +index 965aa03..ed3e29d 100644 +--- a/libports/inhibit-bucket-rpcs.c ++++ b/libports/inhibit-bucket-rpcs.c +@@ -35,10 +35,13 @@ ports_inhibit_bucket_rpcs (struct port_bucket *bucket) + { + int this_one = 0; + +- HURD_IHASH_ITERATE (&bucket->htable, portstruct) ++ pthread_mutex_lock (&_ports_htable_lock); ++ HURD_IHASH_ITERATE (&_ports_htable, portstruct) + { + struct rpc_info *rpc; + struct port_info *pi = portstruct; ++ if (pi->bucket != bucket) ++ continue; + + for (rpc = pi->current_rpcs; rpc; rpc = rpc->next) + { +@@ -49,7 +52,7 @@ ports_inhibit_bucket_rpcs (struct port_bucket *bucket) + hurd_thread_cancel (rpc->thread); + } + } +- ++ pthread_mutex_unlock (&_ports_htable_lock); + + while (bucket->rpcs > this_one) + { +diff --git a/libports/inhibit-class-rpcs.c b/libports/inhibit-class-rpcs.c +index 7ee8653..1580bdb 100644 +--- a/libports/inhibit-class-rpcs.c ++++ b/libports/inhibit-class-rpcs.c +@@ -36,15 +36,24 @@ ports_inhibit_class_rpcs (struct port_class *class) + struct rpc_info *rpc; + int this_one = 0; + +- for (pi = class->ports; pi; pi = pi->next) +- for (rpc = pi->current_rpcs; rpc; rpc = rpc->next) +- { +- /* Avoid cancelling the calling thread. */ +- if (rpc->thread == hurd_thread_self ()) +- this_one = 1; +- else +- hurd_thread_cancel (rpc->thread); +- } ++ pthread_mutex_lock (&_ports_htable_lock); ++ HURD_IHASH_ITERATE (&_ports_htable, portstruct) ++ { ++ struct rpc_info *rpc; ++ struct port_info *pi = portstruct; ++ if (pi->class != class) ++ continue; ++ ++ for (rpc = pi->current_rpcs; rpc; rpc = rpc->next) ++ { ++ /* Avoid cancelling the calling thread. */ ++ if (rpc->thread == hurd_thread_self ()) ++ this_one = 1; ++ else ++ hurd_thread_cancel (rpc->thread); ++ } ++ } ++ pthread_mutex_unlock (&_ports_htable_lock); + + while (class->rpcs > this_one) + { +diff --git a/libports/init.c b/libports/init.c +index 3ef5388..8fbc7e8 100644 +--- a/libports/init.c ++++ b/libports/init.c +@@ -19,9 +19,14 @@ + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ + + #include "ports.h" ++#include + + pthread_mutex_t _ports_lock = PTHREAD_MUTEX_INITIALIZER; + pthread_cond_t _ports_block = PTHREAD_COND_INITIALIZER; +-struct port_bucket *_ports_all_buckets; ++ ++struct hurd_ihash _ports_htable = ++ HURD_IHASH_INITIALIZER (offsetof (struct port_info, hentry)); ++pthread_mutex_t _ports_htable_lock = PTHREAD_MUTEX_INITIALIZER; ++ + int _ports_total_rpcs; + int _ports_flags; +diff --git a/libports/lookup-port.c b/libports/lookup-port.c +index f79f6f0..fbb13ef 100644 +--- a/libports/lookup-port.c ++++ b/libports/lookup-port.c +@@ -26,26 +26,21 @@ ports_lookup_port (struct port_bucket *bucket, + mach_port_t port, + struct port_class *class) + { +- struct port_info *pi = 0; +- ++ struct port_info *pi; ++ + pthread_mutex_lock (&_ports_lock); ++ pthread_mutex_lock (&_ports_htable_lock); + +- if (bucket) +- pi = hurd_ihash_find (&bucket->htable, port); +- else +- for (bucket = _ports_all_buckets; bucket; bucket = bucket->next) +- { +- pi = hurd_ihash_find (&bucket->htable, port); +- if (pi) +- break; +- } +- +- if (pi && class && pi->class != class) ++ pi = hurd_ihash_find (&_ports_htable, port); ++ if (pi ++ && ((class && pi->class != class) ++ || (bucket && pi->bucket != bucket))) + pi = 0; + + if (pi) + pi->refcnt++; + ++ pthread_mutex_unlock (&_ports_htable_lock); + pthread_mutex_unlock (&_ports_lock); + + return pi; +diff --git a/libports/ports.h b/libports/ports.h +index 7f13124..9c5f14d 100644 +--- a/libports/ports.h ++++ b/libports/ports.h +@@ -48,7 +48,6 @@ struct port_info + struct rpc_info *current_rpcs; + struct port_bucket *bucket; + hurd_ihash_locp_t hentry; +- struct port_info *next, **prevp; /* links on port_class list */ + }; + typedef struct port_info *port_info_t; + +@@ -61,11 +60,9 @@ typedef struct port_info *port_info_t; + struct port_bucket + { + mach_port_t portset; +- struct hurd_ihash htable; + int rpcs; + int flags; + int count; +- struct port_bucket *next; + }; + /* FLAGS above are the following: */ + #define PORT_BUCKET_INHIBITED PORTS_INHIBITED +@@ -78,7 +75,6 @@ struct port_class + { + int flags; + int rpcs; +- struct port_info *ports; + int count; + void (*clean_routine) (void *); + void (*dropweak_routine) (void *); +@@ -402,7 +398,12 @@ extern kern_return_t + /* Private data */ + extern pthread_mutex_t _ports_lock; + extern pthread_cond_t _ports_block; +-extern struct port_bucket *_ports_all_buckets; ++ ++/* A hash table mapping port names to port_info objects. */ ++extern struct hurd_ihash _ports_htable; ++/* Access to the hash table is protected by this lock. */ ++extern pthread_mutex_t _ports_htable_lock; ++ + extern int _ports_total_rpcs; + extern int _ports_flags; + #define _PORTS_INHIBITED PORTS_INHIBITED +diff --git a/libports/reallocate-from-external.c b/libports/reallocate-from-external.c +index 8cccb2a..bbcf9ba 100644 +--- a/libports/reallocate-from-external.c ++++ b/libports/reallocate-from-external.c +@@ -43,8 +43,10 @@ ports_reallocate_from_external (void *portstruct, mach_port_t receive) + MACH_PORT_RIGHT_RECEIVE, -1); + assert_perror (err); + +- hurd_ihash_locp_remove (&pi->bucket->htable, pi->hentry); +- ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, pi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); ++ + if ((pi->flags & PORT_HAS_SENDRIGHTS) && !stat.mps_srights) + { + dropref = 1; +@@ -59,11 +61,13 @@ ports_reallocate_from_external (void *portstruct, mach_port_t receive) + pi->port_right = receive; + pi->cancel_threshold = 0; + pi->mscount = stat.mps_mscount; +- +- err = hurd_ihash_add (&pi->bucket->htable, receive, pi); +- assert_perror (err); ++ ++ pthread_mutex_lock (&_ports_htable_lock); ++ err = hurd_ihash_add (&_ports_htable, receive, pi); ++ pthread_mutex_unlock (&_ports_htable_lock); + pthread_mutex_unlock (&_ports_lock); +- ++ assert_perror (err); ++ + mach_port_move_member (mach_task_self (), receive, pi->bucket->portset); + + if (stat.mps_srights) +diff --git a/libports/reallocate-port.c b/libports/reallocate-port.c +index d2adaeb..99c4041 100644 +--- a/libports/reallocate-port.c ++++ b/libports/reallocate-port.c +@@ -36,7 +36,9 @@ ports_reallocate_port (void *portstruct) + MACH_PORT_RIGHT_RECEIVE, -1); + assert_perror (err); + +- hurd_ihash_locp_remove (&pi->bucket->htable, pi->hentry); ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, pi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); + + err = mach_port_allocate (mach_task_self (), MACH_PORT_RIGHT_RECEIVE, + &pi->port_right); +@@ -48,9 +50,11 @@ ports_reallocate_port (void *portstruct) + } + pi->cancel_threshold = 0; + pi->mscount = 0; +- err = hurd_ihash_add (&pi->bucket->htable, pi->port_right, pi); +- assert_perror (err); ++ pthread_mutex_lock (&_ports_htable_lock); ++ err = hurd_ihash_add (&_ports_htable, pi->port_right, pi); ++ pthread_mutex_unlock (&_ports_htable_lock); + pthread_mutex_unlock (&_ports_lock); ++ assert_perror (err); + + err = mach_port_move_member (mach_task_self (), pi->port_right, + pi->bucket->portset); +diff --git a/libports/transfer-right.c b/libports/transfer-right.c +index 72488a9..5a7653d 100644 +--- a/libports/transfer-right.c ++++ b/libports/transfer-right.c +@@ -41,7 +41,9 @@ ports_transfer_right (void *tostruct, + port = frompi->port_right; + if (port != MACH_PORT_NULL) + { +- hurd_ihash_locp_remove (&frompi->bucket->htable, frompi->hentry); ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, frompi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); + frompi->port_right = MACH_PORT_NULL; + if (frompi->flags & PORT_HAS_SENDRIGHTS) + { +@@ -54,7 +56,9 @@ ports_transfer_right (void *tostruct, + /* Destroy the existing right in TOPI. */ + if (topi->port_right != MACH_PORT_NULL) + { +- hurd_ihash_locp_remove (&topi->bucket->htable, topi->hentry); ++ pthread_mutex_lock (&_ports_htable_lock); ++ hurd_ihash_locp_remove (&_ports_htable, topi->hentry); ++ pthread_mutex_unlock (&_ports_htable_lock); + err = mach_port_mod_refs (mach_task_self (), topi->port_right, + MACH_PORT_RIGHT_RECEIVE, -1); + assert_perror (err); +@@ -74,10 +78,14 @@ ports_transfer_right (void *tostruct, + topi->port_right = port; + topi->cancel_threshold = frompi->cancel_threshold; + topi->mscount = frompi->mscount; +- ++ ++ pthread_mutex_unlock (&_ports_lock); ++ + if (port) + { +- err = hurd_ihash_add (&topi->bucket->htable, port, topi); ++ pthread_mutex_lock (&_ports_htable_lock); ++ err = hurd_ihash_add (&_ports_htable, port, topi); ++ pthread_mutex_unlock (&_ports_htable_lock); + assert_perror (err); + if (topi->bucket != frompi->bucket) + { +@@ -86,9 +94,7 @@ ports_transfer_right (void *tostruct, + assert_perror (err); + } + } +- +- pthread_mutex_unlock (&_ports_lock); +- ++ + /* Take care of any lowered reference counts. */ + if (dereffrompi) + ports_port_deref (frompi); +-- +2.0.0.rc0 + diff --git a/debian/patches/0009-libports-lock-less-reference-counting-for-port_info-.patch b/debian/patches/0009-libports-lock-less-reference-counting-for-port_info-.patch new file mode 100644 index 00000000..6c0f4eeb --- /dev/null +++ b/debian/patches/0009-libports-lock-less-reference-counting-for-port_info-.patch @@ -0,0 +1,345 @@ +From 6f5ee0a47ed6bade01f30ceda2be793bcfbb8661 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Sat, 3 May 2014 01:02:35 +0200 +Subject: [PATCH 09/15] libports: lock-less reference counting for port_info + objects + +* libports/ports.h (struct port_info): Use the new type. +* libports/lookup-port.c: No need to lock _ports_lock anymore. +* libports/bucket-iterate.c: Likewise. +* libports/complete-deallocate.c: Check if someone reacquired a +reference through a hash table lookup. +* libports/create-internal.c: Use the new reference counting primitives. +* libports/get-right.c: Likewise. +* libports/import-port.c: Likewise. +* libports/port-deref-weak.c: Likewise. +* libports/port-deref.c: Likewise. +* libports/port-ref-weak.c: Likewise. +* libports/port-ref.c: Likewise. +* libports/reallocate-from-external.c: Likewise. +* libports/transfer-right.c: Likewise. +* utils/rpctrace.c: Likewise. +--- + libports/bucket-iterate.c | 4 +--- + libports/complete-deallocate.c | 15 +++++++++++++++ + libports/create-internal.c | 3 +-- + libports/get-right.c | 2 +- + libports/import-port.c | 3 +-- + libports/lookup-port.c | 6 ++---- + libports/port-deref-weak.c | 10 +++------- + libports/port-deref.c | 34 ++++++++++++++++------------------ + libports/port-ref-weak.c | 8 +++----- + libports/port-ref.c | 8 +++----- + libports/ports.h | 4 ++-- + libports/reallocate-from-external.c | 2 +- + libports/transfer-right.c | 2 +- + utils/rpctrace.c | 10 ++++++++-- + 14 files changed, 58 insertions(+), 53 deletions(-) + +diff --git a/libports/bucket-iterate.c b/libports/bucket-iterate.c +index d5f590e..2a2b2b6 100644 +--- a/libports/bucket-iterate.c ++++ b/libports/bucket-iterate.c +@@ -35,7 +35,6 @@ _ports_bucket_class_iterate (struct port_bucket *bucket, + size_t i, n, nr_items; + error_t err; + +- pthread_mutex_lock (&_ports_lock); + pthread_mutex_lock (&_ports_htable_lock); + + if (_ports_htable.nr_items == 0) +@@ -60,13 +59,12 @@ _ports_bucket_class_iterate (struct port_bucket *bucket, + if ((bucket == NULL || pi->bucket == bucket) + && (class == NULL || pi->class == class)) + { +- pi->refcnt++; ++ refcounts_ref (&pi->refcounts, NULL); + p[n] = pi; + n++; + } + } + pthread_mutex_unlock (&_ports_htable_lock); +- pthread_mutex_unlock (&_ports_lock); + + if (n == 0) + { +diff --git a/libports/complete-deallocate.c b/libports/complete-deallocate.c +index 1371d92..6c9074a 100644 +--- a/libports/complete-deallocate.c ++++ b/libports/complete-deallocate.c +@@ -29,14 +29,29 @@ _ports_complete_deallocate (struct port_info *pi) + + if (pi->port_right) + { ++ struct references result; ++ + pthread_mutex_lock (&_ports_htable_lock); ++ ++ refcounts_references (&pi->refcounts, &result); ++ if (result.hard > 0 || result.weak > 0) ++ { ++ /* A reference was reacquired through a hash table lookup. ++ It's fine, we didn't touch anything yet. */ ++ pthread_mutex_unlock (&_ports_htable_lock); ++ return; ++ } ++ + hurd_ihash_locp_remove (&_ports_htable, pi->hentry); + pthread_mutex_unlock (&_ports_htable_lock); ++ + mach_port_mod_refs (mach_task_self (), pi->port_right, + MACH_PORT_RIGHT_RECEIVE, -1); + pi->port_right = MACH_PORT_NULL; + } + ++ pthread_mutex_lock (&_ports_lock); ++ + pi->bucket->count--; + pi->class->count--; + +diff --git a/libports/create-internal.c b/libports/create-internal.c +index e773dd6..9e3824d 100644 +--- a/libports/create-internal.c ++++ b/libports/create-internal.c +@@ -54,8 +54,7 @@ _ports_create_port_internal (struct port_class *class, + } + + pi->class = class; +- pi->refcnt = 1; +- pi->weakrefcnt = 0; ++ refcounts_init (&pi->refcounts, 1, 0); + pi->cancel_threshold = 0; + pi->mscount = 0; + pi->flags = 0; +diff --git a/libports/get-right.c b/libports/get-right.c +index 89050c6..8681f46 100644 +--- a/libports/get-right.c ++++ b/libports/get-right.c +@@ -41,7 +41,7 @@ ports_get_right (void *port) + if ((pi->flags & PORT_HAS_SENDRIGHTS) == 0) + { + pi->flags |= PORT_HAS_SENDRIGHTS; +- pi->refcnt++; ++ refcounts_ref (&pi->refcounts, NULL); + err = mach_port_request_notification (mach_task_self (), + pi->port_right, + MACH_NOTIFY_NO_SENDERS, +diff --git a/libports/import-port.c b/libports/import-port.c +index d0b2ea4..e250b0e 100644 +--- a/libports/import-port.c ++++ b/libports/import-port.c +@@ -48,8 +48,7 @@ ports_import_port (struct port_class *class, struct port_bucket *bucket, + return ENOMEM; + + pi->class = class; +- pi->refcnt = 1 + !!stat.mps_srights; +- pi->weakrefcnt = 0; ++ refcounts_init (&pi->refcounts, 1 + !!stat.mps_srights, 0); + pi->cancel_threshold = 0; + pi->mscount = stat.mps_mscount; + pi->flags = stat.mps_srights ? PORT_HAS_SENDRIGHTS : 0; +diff --git a/libports/lookup-port.c b/libports/lookup-port.c +index fbb13ef..1bf012f 100644 +--- a/libports/lookup-port.c ++++ b/libports/lookup-port.c +@@ -28,7 +28,6 @@ ports_lookup_port (struct port_bucket *bucket, + { + struct port_info *pi; + +- pthread_mutex_lock (&_ports_lock); + pthread_mutex_lock (&_ports_htable_lock); + + pi = hurd_ihash_find (&_ports_htable, port); +@@ -38,10 +37,9 @@ ports_lookup_port (struct port_bucket *bucket, + pi = 0; + + if (pi) +- pi->refcnt++; ++ ports_port_ref (pi); + + pthread_mutex_unlock (&_ports_htable_lock); +- pthread_mutex_unlock (&_ports_lock); +- ++ + return pi; + } +diff --git a/libports/port-deref-weak.c b/libports/port-deref-weak.c +index beb4842..8432660 100644 +--- a/libports/port-deref-weak.c ++++ b/libports/port-deref-weak.c +@@ -25,12 +25,8 @@ void + ports_port_deref_weak (void *portstruct) + { + struct port_info *pi = portstruct; +- +- pthread_mutex_lock (&_ports_lock); +- assert (pi->weakrefcnt); +- pi->weakrefcnt--; +- if (pi->refcnt == 0 && pi->weakrefcnt == 0) ++ struct references result; ++ refcounts_deref_weak (&pi->refcounts, &result); ++ if (result.hard == 0 && result.weak == 0) + _ports_complete_deallocate (pi); +- else +- pthread_mutex_unlock (&_ports_lock); + } +diff --git a/libports/port-deref.c b/libports/port-deref.c +index cf9b238..dd38f55 100644 +--- a/libports/port-deref.c ++++ b/libports/port-deref.c +@@ -25,26 +25,24 @@ void + ports_port_deref (void *portstruct) + { + struct port_info *pi = portstruct; +- int trieddroppingweakrefs = 0; +- +- retry: +- +- pthread_mutex_lock (&_ports_lock); +- +- if (pi->refcnt == 1 && pi->weakrefcnt +- && pi->class->dropweak_routine && !trieddroppingweakrefs) ++ struct references result; ++ ++ /* If we need to call the dropweak routine, we need to hold one ++ reference while doing so. We use a weak reference for this ++ purpose, which we acquire before we release our hard reference. ++ The order is important here to prevent a race. */ ++ if (pi->class->dropweak_routine) ++ refcounts_ref_weak (&pi->refcounts, NULL); ++ ++ refcounts_deref (&pi->refcounts, &result); ++ ++ if (pi->class->dropweak_routine) + { +- pthread_mutex_unlock (&_ports_lock); +- (*pi->class->dropweak_routine) (pi); +- trieddroppingweakrefs = 1; +- goto retry; ++ if (result.hard == 0 && result.weak > 1) ++ (*pi->class->dropweak_routine) (pi); ++ refcounts_deref_weak (&pi->refcounts, &result); + } +- +- assert (pi->refcnt); + +- pi->refcnt--; +- if (pi->refcnt == 0 && pi->weakrefcnt == 0) ++ if (result.hard == 0 && result.weak == 0) + _ports_complete_deallocate (pi); +- else +- pthread_mutex_unlock (&_ports_lock); + } +diff --git a/libports/port-ref-weak.c b/libports/port-ref-weak.c +index c7d3c69..e4b7fc8 100644 +--- a/libports/port-ref-weak.c ++++ b/libports/port-ref-weak.c +@@ -25,9 +25,7 @@ void + ports_port_ref_weak (void *portstruct) + { + struct port_info *pi = portstruct; +- +- pthread_mutex_lock (&_ports_lock); +- assert (pi->refcnt || pi->weakrefcnt); +- pi->weakrefcnt++; +- pthread_mutex_unlock (&_ports_lock); ++ struct references result; ++ refcounts_ref_weak (&pi->refcounts, &result); ++ assert (result.hard > 0 || result.weak > 1); + } +diff --git a/libports/port-ref.c b/libports/port-ref.c +index 92b7118..761c50f 100644 +--- a/libports/port-ref.c ++++ b/libports/port-ref.c +@@ -25,9 +25,7 @@ void + ports_port_ref (void *portstruct) + { + struct port_info *pi = portstruct; +- +- pthread_mutex_lock (&_ports_lock); +- assert (pi->refcnt || pi->weakrefcnt); +- pi->refcnt++; +- pthread_mutex_unlock (&_ports_lock); ++ struct references result; ++ refcounts_ref (&pi->refcounts, &result); ++ assert (result.hard > 1 || result.weak > 0); + } +diff --git a/libports/ports.h b/libports/ports.h +index 9c5f14d..116e920 100644 +--- a/libports/ports.h ++++ b/libports/ports.h +@@ -27,6 +27,7 @@ + #include + #include + #include ++#include + + /* These are global values for common flags used in the various structures. + Not all of these are meaningful in all flag fields. */ +@@ -39,8 +40,7 @@ + struct port_info + { + struct port_class *class; +- int refcnt; +- int weakrefcnt; ++ refcounts_t refcounts; + mach_port_mscount_t mscount; + mach_msg_seqno_t cancel_threshold; + int flags; +diff --git a/libports/reallocate-from-external.c b/libports/reallocate-from-external.c +index bbcf9ba..5ee579d 100644 +--- a/libports/reallocate-from-external.c ++++ b/libports/reallocate-from-external.c +@@ -55,7 +55,7 @@ ports_reallocate_from_external (void *portstruct, mach_port_t receive) + else if (((pi->flags & PORT_HAS_SENDRIGHTS) == 0) && stat.mps_srights) + { + pi->flags |= PORT_HAS_SENDRIGHTS; +- pi->refcnt++; ++ refcounts_ref (&pi->refcounts, NULL); + } + + pi->port_right = receive; +diff --git a/libports/transfer-right.c b/libports/transfer-right.c +index 5a7653d..d3ff0f4 100644 +--- a/libports/transfer-right.c ++++ b/libports/transfer-right.c +@@ -70,7 +70,7 @@ ports_transfer_right (void *tostruct, + else if (((topi->flags & PORT_HAS_SENDRIGHTS) == 0) && hassendrights) + { + topi->flags |= PORT_HAS_SENDRIGHTS; +- topi->refcnt++; ++ refcounts_ref (&topi->refcounts, NULL); + } + } + +diff --git a/utils/rpctrace.c b/utils/rpctrace.c +index fc913e3..b11fea4 100644 +--- a/utils/rpctrace.c ++++ b/utils/rpctrace.c +@@ -431,7 +431,9 @@ destroy_receiver_info (struct receiver_info *info) + while (send_wrapper) + { + struct sender_info *next = send_wrapper->next; +- assert (TRACED_INFO (send_wrapper)->pi.refcnt == 1); ++ assert ( ++ refcounts_hard_references (&TRACED_INFO (send_wrapper)->pi.refcounts) ++ == 1); + /* Reset the receive_right of the send wrapper in advance to avoid + * destroy_receiver_info is called when the port info is destroyed. */ + send_wrapper->receive_right = NULL; +@@ -848,7 +850,11 @@ rewrite_right (mach_port_t *right, mach_msg_type_name_t *type, + hurd_ihash_locp_remove (&traced_names, receiver_info->locp); + + send_wrapper2 = get_send_wrapper (receiver_info, dest, &rr); +- assert (TRACED_INFO (send_wrapper2)->pi.refcnt == 1); ++ assert ( ++ refcounts_hard_references ( ++ &TRACED_INFO (send_wrapper2)->pi.refcounts) ++ == 1); ++ + name = TRACED_INFO (send_wrapper2)->name; + TRACED_INFO (send_wrapper2)->name = NULL; + /* send_wrapper2 isn't destroyed normally, so we need to unlink +-- +2.0.0.rc0 + diff --git a/debian/patches/0010-ext2fs-improve-enable-disable-_caching.patch b/debian/patches/0010-ext2fs-improve-enable-disable-_caching.patch new file mode 100644 index 00000000..8cf1fabe --- /dev/null +++ b/debian/patches/0010-ext2fs-improve-enable-disable-_caching.patch @@ -0,0 +1,38 @@ +From 283c322efe55cf00f25cabaa22d48078b94c526e Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Fri, 9 May 2014 10:07:28 +0200 +Subject: [PATCH 10/15] ext2fs: improve {enable,disable}_caching + +* ext2fs/pager.c (enable_caching, disable_caching): Iterate over the +pager class instead of over both pager buckets. +--- + ext2fs/pager.c | 6 ++---- + 1 file changed, 2 insertions(+), 4 deletions(-) + +diff --git a/ext2fs/pager.c b/ext2fs/pager.c +index 017efcc..6328f3b 100644 +--- a/ext2fs/pager.c ++++ b/ext2fs/pager.c +@@ -1409,8 +1409,7 @@ disable_caching () + + /* Loop through the pagers and turn off caching one by one, + synchronously. That should cause termination of each pager. */ +- ports_bucket_iterate (disk_pager_bucket, block_cache); +- ports_bucket_iterate (file_pager_bucket, block_cache); ++ ports_class_iterate (_pager_class, block_cache); + } + + static void +@@ -1438,8 +1437,7 @@ enable_caching () + return 0; + } + +- ports_bucket_iterate (disk_pager_bucket, enable_cache); +- ports_bucket_iterate (file_pager_bucket, enable_cache); ++ ports_class_iterate (_pager_class, enable_cache); + } + + /* Tell diskfs if there are pagers exported, and if none, then +-- +2.0.0.rc0 + diff --git a/debian/patches/0011-fatfs-improve-enable-disable-_caching.patch b/debian/patches/0011-fatfs-improve-enable-disable-_caching.patch new file mode 100644 index 00000000..45b3f45a --- /dev/null +++ b/debian/patches/0011-fatfs-improve-enable-disable-_caching.patch @@ -0,0 +1,48 @@ +From 555a58f92cfcf75c5650ef1d96080c512a511f64 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Fri, 9 May 2014 10:11:45 +0200 +Subject: [PATCH 11/15] fatfs: improve {enable,disable}_caching + +* fatfs/pager.c (enable_caching, disable_caching): Iterate over the +pager class instead of over both pager buckets. +--- + fatfs/pager.c | 9 +++++---- + 1 file changed, 5 insertions(+), 4 deletions(-) + +diff --git a/fatfs/pager.c b/fatfs/pager.c +index f855ecf..7aa5c5e 100644 +--- a/fatfs/pager.c ++++ b/fatfs/pager.c +@@ -23,6 +23,9 @@ + #include + #include "fatfs.h" + ++/* XXX */ ++#include "../libpager/priv.h" ++ + /* A ports bucket to hold disk pager ports. */ + struct port_bucket *disk_pager_bucket; + +@@ -963,8 +966,7 @@ disable_caching () + + /* Loop through the pagers and turn off caching one by one, + synchronously. That should cause termination of each pager. */ +- ports_bucket_iterate (disk_pager_bucket, block_cache); +- ports_bucket_iterate (file_pager_bucket, block_cache); ++ ports_class_iterate (_pager_class, block_cache); + } + + static void +@@ -992,8 +994,7 @@ enable_caching () + return 0; + } + +- ports_bucket_iterate (disk_pager_bucket, enable_cache); +- ports_bucket_iterate (file_pager_bucket, enable_cache); ++ ports_class_iterate (_pager_class, enable_cache); + } + + /* Tell diskfs if there are pagers exported, and if none, then +-- +2.0.0.rc0 + diff --git a/debian/patches/0012-ext2fs-use-a-seperate-lock-to-protect-nodehash.patch b/debian/patches/0012-ext2fs-use-a-seperate-lock-to-protect-nodehash.patch new file mode 100644 index 00000000..20e9d5f3 --- /dev/null +++ b/debian/patches/0012-ext2fs-use-a-seperate-lock-to-protect-nodehash.patch @@ -0,0 +1,190 @@ +From 72937c396cc7a51e5aff5f79ee60e092f54f135c Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 13 May 2014 13:09:15 +0200 +Subject: [PATCH 12/15] ext2fs: use a seperate lock to protect nodehash + +Previously, ext2fs used diskfs_node_refcnt_lock to serialize access to +the nodehash. + +Use a separate lock to protect nodehash. Adjust the reference +counting accordingly. Every node in the nodehash carries a light +reference. When we are asked to give up that light reference, we +reacquire our lock momentarily to check whether someone else +reacquired a reference through the nodehash. + +* ext2fs/inode.c (nodehash_lock): New lock. +(diskfs_cached_lookup): Use a separate lock to protect nodehash. +Adjust the reference counting accordingly. +(ifind): Likewise. +(diskfs_node_iterate): Likewise. +(diskfs_node_norefs): Move the code removing the node from nodehash... +(diskfs_try_dropping_softrefs): ... here, where we check whether +someone reacquired a reference, and if so hold on to our light +reference. +--- + ext2fs/inode.c | 68 ++++++++++++++++++++++++++++++++++++++++++---------------- + 1 file changed, 50 insertions(+), 18 deletions(-) + +diff --git a/ext2fs/inode.c b/ext2fs/inode.c +index ed78265..aa070a2 100644 +--- a/ext2fs/inode.c ++++ b/ext2fs/inode.c +@@ -46,8 +46,18 @@ + #define INOHASH(ino) (((unsigned)(ino))%INOHSZ) + #endif + ++/* The nodehash is a cache of nodes. ++ ++ Access to nodehash and nodehash_nr_items is protected by ++ nodehash_lock. ++ ++ Every node in the nodehash carries a light reference. When we are ++ asked to give up that light reference, we reacquire our lock ++ momentarily to check whether someone else reacquired a reference ++ through the nodehash. */ + static struct node *nodehash[INOHSZ]; + static size_t nodehash_nr_items; ++static pthread_mutex_t nodehash_lock = PTHREAD_MUTEX_INITIALIZER; + + static error_t read_node (struct node *np); + +@@ -71,12 +81,12 @@ diskfs_cached_lookup (ino_t inum, struct node **npp) + struct node *np; + struct disknode *dn; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + for (np = nodehash[INOHASH(inum)]; np; np = np->dn->hnext) + if (np->cache_id == inum) + { +- np->references++; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref (np); ++ pthread_mutex_unlock (&nodehash_lock); + pthread_mutex_lock (&np->lock); + *npp = np; + return 0; +@@ -86,7 +96,7 @@ diskfs_cached_lookup (ino_t inum, struct node **npp) + dn = malloc (sizeof (struct disknode)); + if (! dn) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + return ENOMEM; + } + dn->dirents = 0; +@@ -107,9 +117,10 @@ diskfs_cached_lookup (ino_t inum, struct node **npp) + dn->hnext->dn->hprevp = &dn->hnext; + dn->hprevp = &nodehash[INOHASH(inum)]; + nodehash[INOHASH(inum)] = np; ++ diskfs_nref_light (np); + nodehash_nr_items += 1; + +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + + /* Get the contents of NP off disk. */ + err = read_node (np); +@@ -140,14 +151,13 @@ ifind (ino_t inum) + { + struct node *np; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + for (np = nodehash[INOHASH(inum)]; np; np = np->dn->hnext) + { + if (np->cache_id != inum) + continue; + +- assert (np->references); +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + return np; + } + assert (0); +@@ -158,11 +168,6 @@ ifind (ino_t inum) + void + diskfs_node_norefs (struct node *np) + { +- *np->dn->hprevp = np->dn->hnext; +- if (np->dn->hnext) +- np->dn->hnext->dn->hprevp = np->dn->hprevp; +- nodehash_nr_items -= 1; +- + if (np->dn->dirents) + free (np->dn->dirents); + assert (!np->dn->pager); +@@ -180,6 +185,33 @@ diskfs_node_norefs (struct node *np) + void + diskfs_try_dropping_softrefs (struct node *np) + { ++ pthread_mutex_lock (&nodehash_lock); ++ if (np->dn->hnext != NULL) ++ { ++ /* Check if someone reacquired a reference through the ++ nodehash. */ ++ unsigned int references; ++ pthread_spin_lock (&diskfs_node_refcnt_lock); ++ references = np->references; ++ pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ ++ if (references > 0) ++ { ++ /* A reference was reacquired through a hash table lookup. ++ It's fine, we didn't touch anything yet. */ ++ pthread_mutex_unlock (&nodehash_lock); ++ return; ++ } ++ ++ *np->dn->hprevp = np->dn->hnext; ++ if (np->dn->hnext) ++ np->dn->hnext->dn->hprevp = np->dn->hprevp; ++ np->dn->hnext = NULL; ++ nodehash_nr_items -= 1; ++ diskfs_nrele_light (np); ++ } ++ pthread_mutex_unlock (&nodehash_lock); ++ + drop_pager_softrefs (np); + } + +@@ -556,12 +588,12 @@ diskfs_node_iterate (error_t (*fun)(struct node *)) + size_t num_nodes; + struct node *node, **node_list, **p; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + + /* We must copy everything from the hash table into another data structure + to avoid running into any problems with the hash-table being modified + during processing (normally we delegate access to hash-table with +- diskfs_node_refcnt_lock, but we can't hold this while locking the ++ nodehash_lock, but we can't hold this while locking the + individual node locks). */ + num_nodes = nodehash_nr_items; + +@@ -570,7 +602,7 @@ diskfs_node_iterate (error_t (*fun)(struct node *)) + node_list = malloc (num_nodes * sizeof (struct node *)); + if (node_list == NULL) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + ext2_debug ("unable to allocate temporary node table"); + return ENOMEM; + } +@@ -580,10 +612,10 @@ diskfs_node_iterate (error_t (*fun)(struct node *)) + for (node = nodehash[n]; node; node = node->dn->hnext) + { + *p++ = node; +- node->references++; ++ diskfs_nref (node); + } + +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + + p = node_list; + while (num_nodes-- > 0) +-- +2.0.0.rc0 + diff --git a/debian/patches/0013-fatfs-use-a-seperate-lock-to-protect-nodehash.patch b/debian/patches/0013-fatfs-use-a-seperate-lock-to-protect-nodehash.patch new file mode 100644 index 00000000..a7918b0c --- /dev/null +++ b/debian/patches/0013-fatfs-use-a-seperate-lock-to-protect-nodehash.patch @@ -0,0 +1,222 @@ +From 664606b8d183455a6195ac954d78571af744362d Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 13 May 2014 15:14:53 +0200 +Subject: [PATCH 13/15] fatfs: use a seperate lock to protect nodehash + +Previously, fatfs used diskfs_node_refcnt_lock to serialize access to +the nodehash. + +Use a separate lock to protect nodehash. Adjust the reference +counting accordingly. Every node in the nodehash carries a light +reference. When we are asked to give up that light reference, we +reacquire our lock momentarily to check whether someone else +reacquired a reference through the nodehash. + +* fatfs/inode.c (nodehash_lock): New lock. +(diskfs_cached_lookup): Use a separate lock to protect nodehash. +Adjust the reference counting accordingly. +(ifind): Likewise. +(diskfs_node_iterate): Likewise. +(diskfs_node_norefs): Move the code removing the node from nodehash... +(diskfs_try_dropping_softrefs): ... here, where we check whether +someone reacquired a reference, and if so hold on to our light +reference. +--- + fatfs/inode.c | 81 +++++++++++++++++++++++++++++++++++++++++------------------ + 1 file changed, 57 insertions(+), 24 deletions(-) + +diff --git a/fatfs/inode.c b/fatfs/inode.c +index ed6f3f0..8b3385b 100644 +--- a/fatfs/inode.c ++++ b/fatfs/inode.c +@@ -44,8 +44,18 @@ + #define INOHASH(ino) (((unsigned)(ino))%INOHSZ) + #endif + ++/* The nodehash is a cache of nodes. ++ ++ Access to nodehash and nodehash_nr_items is protected by ++ nodehash_lock. ++ ++ Every node in the nodehash carries a light reference. When we are ++ asked to give up that light reference, we reacquire our lock ++ momentarily to check whether someone else reacquired a reference ++ through the nodehash. */ + static struct node *nodehash[INOHSZ]; + static size_t nodehash_nr_items; ++static pthread_mutex_t nodehash_lock = PTHREAD_MUTEX_INITIALIZER; + + static error_t read_node (struct node *np, vm_address_t buf); + +@@ -67,12 +77,12 @@ diskfs_cached_lookup (ino64_t inum, struct node **npp) + struct node *np; + struct disknode *dn; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + for (np = nodehash[INOHASH(inum)]; np; np = np->dn->hnext) + if (np->cache_id == inum) + { +- np->references++; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref (np); ++ pthread_mutex_unlock (&nodehash_lock); + pthread_mutex_lock (&np->lock); + *npp = np; + return 0; +@@ -82,7 +92,7 @@ diskfs_cached_lookup (ino64_t inum, struct node **npp) + dn = malloc (sizeof (struct disknode)); + if (! dn) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + return ENOMEM; + } + dn->pager = 0; +@@ -107,10 +117,11 @@ diskfs_cached_lookup (ino64_t inum, struct node **npp) + dn->hnext->dn->hprevp = &dn->hnext; + dn->hprevp = &nodehash[INOHASH(inum)]; + nodehash[INOHASH(inum)] = np; ++ diskfs_nref_light (np); + nodehash_nr_items += 1; + +- pthread_spin_unlock (&diskfs_node_refcnt_lock); +- ++ pthread_mutex_unlock (&nodehash_lock); ++ + /* Get the contents of NP off disk. */ + err = read_node (np, 0); + +@@ -133,12 +144,12 @@ diskfs_cached_lookup_in_dirbuf (int inum, struct node **npp, vm_address_t buf) + struct node *np; + struct disknode *dn; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + for (np = nodehash[INOHASH(inum)]; np; np = np->dn->hnext) + if (np->cache_id == inum) + { +- np->references++; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref (np); ++ pthread_mutex_unlock (&nodehash_lock); + pthread_mutex_lock (&np->lock); + *npp = np; + return 0; +@@ -148,7 +159,7 @@ diskfs_cached_lookup_in_dirbuf (int inum, struct node **npp, vm_address_t buf) + dn = malloc (sizeof (struct disknode)); + if (! dn) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + return ENOMEM; + } + dn->pager = 0; +@@ -173,10 +184,11 @@ diskfs_cached_lookup_in_dirbuf (int inum, struct node **npp, vm_address_t buf) + dn->hnext->dn->hprevp = &dn->hnext; + dn->hprevp = &nodehash[INOHASH(inum)]; + nodehash[INOHASH(inum)] = np; ++ diskfs_nref_light (np); + nodehash_nr_items += 1; + +- pthread_spin_unlock (&diskfs_node_refcnt_lock); +- ++ pthread_mutex_unlock (&nodehash_lock); ++ + /* Get the contents of NP off disk. */ + err = read_node (np, buf); + +@@ -196,14 +208,13 @@ ifind (ino_t inum) + { + struct node *np; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + for (np = nodehash[INOHASH(inum)]; np; np = np->dn->hnext) + { + if (np->cache_id != inum) + continue; + +- assert (np->references); +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + return np; + } + assert (0); +@@ -216,11 +227,6 @@ diskfs_node_norefs (struct node *np) + { + struct cluster_chain *last = np->dn->first; + +- *np->dn->hprevp = np->dn->hnext; +- if (np->dn->hnext) +- np->dn->hnext->dn->hprevp = np->dn->hprevp; +- nodehash_nr_items -= 1; +- + while (last) + { + struct cluster_chain *next = last->next; +@@ -251,6 +257,33 @@ diskfs_node_norefs (struct node *np) + void + diskfs_try_dropping_softrefs (struct node *np) + { ++ pthread_mutex_lock (&nodehash_lock); ++ if (np->dn->hnext != NULL) ++ { ++ /* Check if someone reacquired a reference through the ++ nodehash. */ ++ unsigned int references; ++ pthread_spin_lock (&diskfs_node_refcnt_lock); ++ references = np->references; ++ pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ ++ if (references > 0) ++ { ++ /* A reference was reacquired through a hash table lookup. ++ It's fine, we didn't touch anything yet. */ ++ pthread_mutex_unlock (&nodehash_lock); ++ return; ++ } ++ ++ *np->dn->hprevp = np->dn->hnext; ++ if (np->dn->hnext) ++ np->dn->hnext->dn->hprevp = np->dn->hprevp; ++ np->dn->hnext = NULL; ++ nodehash_nr_items -= 1; ++ diskfs_nrele_light (np); ++ } ++ pthread_mutex_unlock (&nodehash_lock); ++ + drop_pager_softrefs (np); + } + +@@ -554,12 +587,12 @@ diskfs_node_iterate (error_t (*fun)(struct node *)) + size_t num_nodes; + struct node *node, **node_list, **p; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&nodehash_lock); + + /* We must copy everything from the hash table into another data structure + to avoid running into any problems with the hash-table being modified + during processing (normally we delegate access to hash-table with +- diskfs_node_refcnt_lock, but we can't hold this while locking the ++ nodehash_lock, but we can't hold this while locking the + individual node locks). */ + + num_nodes = nodehash_nr_items; +@@ -570,10 +603,10 @@ diskfs_node_iterate (error_t (*fun)(struct node *)) + for (node = nodehash[n]; node; node = node->dn->hnext) + { + *p++ = node; +- node->references++; ++ diskfs_nref (node); + } + +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&nodehash_lock); + + p = node_list; + while (num_nodes-- > 0) +-- +2.0.0.rc0 + diff --git a/debian/patches/0014-isofs-use-a-seperate-lock-to-protect-node_cache.patch b/debian/patches/0014-isofs-use-a-seperate-lock-to-protect-node_cache.patch new file mode 100644 index 00000000..c6af3a4b --- /dev/null +++ b/debian/patches/0014-isofs-use-a-seperate-lock-to-protect-node_cache.patch @@ -0,0 +1,222 @@ +From 57da36c3f7f68de55bd2c360b441879902dc14f8 Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 13 May 2014 15:16:31 +0200 +Subject: [PATCH 14/15] isofs: use a seperate lock to protect node_cache + +Previously, isofs used diskfs_node_refcnt_lock to serialize access to +the node_cache. + +Use a separate lock to protect node_cache. Adjust the reference +counting accordingly. Every node in the node_cache carries a light +reference. When we are asked to give up that light reference, we +reacquire our lock momentarily to check whether someone else +reacquired a reference through the node_cache. + +* isofs/inode.c (node_cache_lock): New lock. +(inode_cache_find): Use a separate lock to protect node_cache. +Adjust the reference counting accordingly. +(diskfs_cached_lookup): Likewise. +(load_inode): Likewise. +(cache_inode): Update comment accordingly. +(diskfs_node_iterate): Likewise. +(diskfs_node_norefs): Move the code removing the node from node_cache... +(diskfs_try_dropping_softrefs): ... here, where we check whether +someone reacquired a reference, and if so hold on to our light +reference. +--- + isofs/inode.c | 72 +++++++++++++++++++++++++++++++++++++++++++---------------- + 1 file changed, 53 insertions(+), 19 deletions(-) + +diff --git a/isofs/inode.c b/isofs/inode.c +index cdc05ae..4f22086 100644 +--- a/isofs/inode.c ++++ b/isofs/inode.c +@@ -48,9 +48,19 @@ struct node_cache + struct node *np; /* if live */ + }; + ++/* The node_cache is a cache of nodes. ++ ++ Access to node_cache, node_cache_size, and node_cache_alloced is ++ protected by node_cache_lock. ++ ++ Every node in the node_cache carries a light reference. When we ++ are asked to give up that light reference, we reacquire our lock ++ momentarily to check whether someone else reacquired a reference ++ through the node_cache. */ + static int node_cache_size = 0; + static int node_cache_alloced = 0; + struct node_cache *node_cache = 0; ++static pthread_mutex_t node_cache_lock = PTHREAD_MUTEX_INITIALIZER; + + /* Forward */ + static error_t read_disknode (struct node *, +@@ -58,7 +68,7 @@ static error_t read_disknode (struct node *, + + + /* See if node with identifier ID is in the cache. If so, return it, +- with one additional reference. diskfs_node_refcnt_lock must be held ++ with one additional reference. node_cache_lock must be held + on entry to the call, and will be released iff the node was found + in the cache. */ + void +@@ -71,8 +81,8 @@ inode_cache_find (off_t id, struct node **npp) + && node_cache[i].np) + { + *npp = node_cache[i].np; +- (*npp)->references++; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref (*npp); ++ pthread_mutex_unlock (&node_cache_lock); + pthread_mutex_lock (&(*npp)->lock); + return; + } +@@ -92,7 +102,7 @@ use_file_start_id (struct dirrect *record, struct rrip_lookup *rr) + } + + /* Enter NP into the cache. The directory entry we used is DR, the +- cached Rock-Ridge info RR. diskfs_node_refcnt_lock must be held. */ ++ cached Rock-Ridge info RR. node_cache_lock must be held. */ + void + cache_inode (struct node *np, struct dirrect *record, + struct rrip_lookup *rr) +@@ -155,7 +165,7 @@ diskfs_cached_lookup (ino_t id, struct node **npp) + to avoid presenting zero cache ID's. */ + id--; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&node_cache_lock); + assert (id < node_cache_size); + + np = node_cache[id].np; +@@ -174,7 +184,7 @@ diskfs_cached_lookup (ino_t id, struct node **npp) + dn = malloc (sizeof (struct disknode)); + if (!dn) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&node_cache_lock); + release_rrip (&rr); + return ENOMEM; + } +@@ -185,16 +195,17 @@ diskfs_cached_lookup (ino_t id, struct node **npp) + if (!np) + { + free (dn); +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&node_cache_lock); + release_rrip (&rr); + return ENOMEM; + } + np->cache_id = id + 1; /* see above for rationale for increment */ + pthread_mutex_lock (&np->lock); + c->np = np; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref_light (np); ++ pthread_mutex_unlock (&node_cache_lock); + +- err = read_disknode (np, node_cache[id].dr, &rr); ++ err = read_disknode (np, dn->dr, &rr); + if (!err) + *npp = np; + +@@ -204,8 +215,8 @@ diskfs_cached_lookup (ino_t id, struct node **npp) + } + + +- np->references++; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref (np); ++ pthread_mutex_unlock (&node_cache_lock); + pthread_mutex_lock (&np->lock); + *npp = np; + return 0; +@@ -315,7 +326,7 @@ load_inode (struct node **npp, struct dirrect *record, + if (rr->valid & VALID_CL) + record = rr->realdirent; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&node_cache_lock); + + /* First check the cache */ + if (use_file_start_id (record, rr)) +@@ -325,7 +336,7 @@ load_inode (struct node **npp, struct dirrect *record, + + if (*npp) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&node_cache_lock); + return 0; + } + +@@ -333,7 +344,7 @@ load_inode (struct node **npp, struct dirrect *record, + dn = malloc (sizeof (struct disknode)); + if (!dn) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&node_cache_lock); + return ENOMEM; + } + dn->fileinfo = 0; +@@ -344,14 +355,14 @@ load_inode (struct node **npp, struct dirrect *record, + if (!np) + { + free (dn); +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&node_cache_lock); + return ENOMEM; + } + + pthread_mutex_lock (&np->lock); + + cache_inode (np, record, rr); +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&node_cache_lock); + + err = read_disknode (np, record, rr); + *npp = np; +@@ -505,9 +516,6 @@ error_t (*diskfs_read_symlink_hook) (struct node *, char *) + void + diskfs_node_norefs (struct node *np) + { +- assert (node_cache[np->cache_id - 1].np == np); +- node_cache[np->cache_id - 1].np = 0; +- + if (np->dn->translator) + free (np->dn->translator); + +@@ -521,6 +529,32 @@ diskfs_node_norefs (struct node *np) + void + diskfs_try_dropping_softrefs (struct node *np) + { ++ pthread_mutex_lock (&node_cache_lock); ++ if (np->cache_id != 0) ++ { ++ assert (node_cache[np->cache_id - 1].np == np); ++ ++ /* Check if someone reacquired a reference through the ++ node_cache. */ ++ unsigned int references; ++ pthread_spin_lock (&diskfs_node_refcnt_lock); ++ references = np->references; ++ pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ ++ if (references > 0) ++ { ++ /* A reference was reacquired through a hash table lookup. ++ It's fine, we didn't touch anything yet. */ ++ pthread_mutex_unlock (&node_cache_lock); ++ return; ++ } ++ ++ node_cache[np->cache_id - 1].np = 0; ++ np->cache_id = 0; ++ diskfs_nrele_light (np); ++ } ++ pthread_mutex_unlock (&node_cache_lock); ++ + drop_pager_softrefs (np); + } + +-- +2.0.0.rc0 + diff --git a/debian/patches/0015-tmpfs-use-a-seperate-lock-to-protect-all_nodes.patch b/debian/patches/0015-tmpfs-use-a-seperate-lock-to-protect-all_nodes.patch new file mode 100644 index 00000000..fafb113f --- /dev/null +++ b/debian/patches/0015-tmpfs-use-a-seperate-lock-to-protect-all_nodes.patch @@ -0,0 +1,263 @@ +From 7d53b7e08efefe1d6d24bf3a80b3391a9502c6aa Mon Sep 17 00:00:00 2001 +From: Justus Winter <4winter@informatik.uni-hamburg.de> +Date: Tue, 13 May 2014 15:35:42 +0200 +Subject: [PATCH 15/15] tmpfs: use a seperate lock to protect all_nodes + +Previously, tmpfs used diskfs_node_refcnt_lock to serialize access to +the all_nodes and some other related global state related to memory +consumption. + +Use a separate lock to protect all_nodes, and to the state related to +memory consumption as this is updated during insertion and removal +operations. Adjust the reference counting accordingly. Every node in +the all_nodes carries a light reference. When we are asked to give up +that light reference, we reacquire our lock momentarily to check +whether someone else reacquired a reference through the all_nodes. + +* tmpfs/node.c (all_nodes_lock): New lock. +(diskfs_alloc_node): Use a separate lock to protect all_nodes. +Adjust the reference counting accordingly. +(diskfs_free_node): Likewise. +(diskfs_cached_lookup):Likewise. +(diskfs_node_iterate): Likewise. +(diskfs_node_norefs): Move the code removing the node from all_nodes... +(diskfs_try_dropping_softrefs): ... here, where we check whether +someone reacquired a reference, and if so hold on to our light +reference. +* tmpfs/tmpfs.c (diskfs_set_statfs): Use all_nodes_lock. +* tmpfs/tmpfs.h (all_nodes_lock): New declaration. +(adjust_used): Use all_nodes_lock. +--- + tmpfs/node.c | 76 +++++++++++++++++++++++++++++++++++++++++++---------------- + tmpfs/tmpfs.c | 4 ++-- + tmpfs/tmpfs.h | 6 +++-- + 3 files changed, 62 insertions(+), 24 deletions(-) + +diff --git a/tmpfs/node.c b/tmpfs/node.c +index acc029a..74971ec 100644 +--- a/tmpfs/node.c ++++ b/tmpfs/node.c +@@ -29,8 +29,18 @@ the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */ + unsigned int num_files; + static unsigned int gen; + ++/* all_nodes is a cache of nodes. ++ ++ Access to all_nodes and all_nodes_nr_items is protected by ++ all_nodes_lock. ++ ++ Every node in the all_nodes carries a light reference. When we are ++ asked to give up that light reference, we reacquire our lock ++ momentarily to check whether someone else reacquired a reference ++ through the all_nodes. */ + struct node *all_nodes; + static size_t all_nodes_nr_items; ++pthread_mutex_t all_nodes_lock = PTHREAD_MUTEX_INITIALIZER; + + error_t + diskfs_alloc_node (struct node *dp, mode_t mode, struct node **npp) +@@ -40,18 +50,18 @@ diskfs_alloc_node (struct node *dp, mode_t mode, struct node **npp) + dn = calloc (1, sizeof *dn); + if (dn == 0) + return ENOSPC; +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&all_nodes_lock); + if (round_page (tmpfs_space_used + sizeof *dn) / vm_page_size + > tmpfs_page_limit) + { +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&all_nodes_lock); + free (dn); + return ENOSPC; + } + dn->gen = gen++; + ++num_files; + tmpfs_space_used += sizeof *dn; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&all_nodes_lock); + + dn->type = IFTODT (mode & S_IFMT); + return diskfs_cached_lookup ((ino_t) (uintptr_t) dn, npp); +@@ -75,15 +85,18 @@ diskfs_free_node (struct node *np, mode_t mode) + free (np->dn->u.lnk); + break; + } ++ ++ pthread_mutex_lock (&all_nodes_lock); + *np->dn->hprevp = np->dn->hnext; + if (np->dn->hnext != 0) + np->dn->hnext->dn->hprevp = np->dn->hprevp; + all_nodes_nr_items -= 1; +- free (np->dn); +- np->dn = 0; +- + --num_files; + tmpfs_space_used -= sizeof *np->dn; ++ pthread_mutex_unlock (&all_nodes_lock); ++ ++ free (np->dn); ++ np->dn = 0; + } + + void +@@ -117,14 +130,6 @@ diskfs_node_norefs (struct node *np) + np->dn->u.chr = np->dn_stat.st_rdev; + break; + } +- +- /* Remove this node from the cache list rooted at `all_nodes'. */ +- *np->dn->hprevp = np->dn->hnext; +- if (np->dn->hnext != 0) +- np->dn->hnext->dn->hprevp = np->dn->hprevp; +- all_nodes_nr_items -= 1; +- np->dn->hnext = 0; +- np->dn->hprevp = 0; + } + + free (np); +@@ -167,11 +172,14 @@ diskfs_cached_lookup (ino_t inum, struct node **npp) + + assert (npp); + ++ pthread_mutex_lock (&all_nodes_lock); ++ + if (dn->hprevp != 0) /* There is already a node. */ + { + np = *dn->hprevp; + assert (np->dn == dn); + assert (*dn->hprevp == np); ++ pthread_mutex_unlock (&all_nodes_lock); + + diskfs_nref (np); + } +@@ -183,14 +191,14 @@ diskfs_cached_lookup (ino_t inum, struct node **npp) + np = diskfs_make_node (dn); + np->cache_id = (ino_t) (uintptr_t) dn; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); + dn->hnext = all_nodes; + if (dn->hnext) + dn->hnext->dn->hprevp = &dn->hnext; + dn->hprevp = &all_nodes; + all_nodes = np; + all_nodes_nr_items += 1; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ diskfs_nref_light (np); ++ pthread_mutex_unlock (&all_nodes_lock); + + st = &np->dn_stat; + memset (st, 0, sizeof *st); +@@ -229,12 +237,12 @@ diskfs_node_iterate (error_t (*fun) (struct node *)) + size_t num_nodes; + struct node *node, **node_list, **p; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&all_nodes_lock); + + /* We must copy everything from the hash table into another data structure + to avoid running into any problems with the hash-table being modified + during processing (normally we delegate access to hash-table with +- diskfs_node_refcnt_lock, but we can't hold this while locking the ++ all_nodes_lock, but we can't hold this while locking the + individual node locks). */ + + num_nodes = all_nodes_nr_items; +@@ -243,10 +251,10 @@ diskfs_node_iterate (error_t (*fun) (struct node *)) + for (node = all_nodes; node != 0; node = node->dn->hnext) + { + *p++ = node; +- node->references++; ++ diskfs_nref (node); + } + +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&all_nodes_lock); + + p = node_list; + while (num_nodes-- > 0) +@@ -272,6 +280,34 @@ diskfs_node_iterate (error_t (*fun) (struct node *)) + void + diskfs_try_dropping_softrefs (struct node *np) + { ++ pthread_mutex_lock (&all_nodes_lock); ++ if (np->dn->hnext != NULL) ++ { ++ /* Check if someone reacquired a reference through the ++ all_nodes. */ ++ unsigned int references; ++ pthread_spin_lock (&diskfs_node_refcnt_lock); ++ references = np->references; ++ pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ ++ if (references > 0) ++ { ++ /* A reference was reacquired through a hash table lookup. ++ It's fine, we didn't touch anything yet. */ ++ pthread_mutex_unlock (&all_nodes_lock); ++ return; ++ } ++ ++ /* Remove this node from the cache list rooted at `all_nodes'. */ ++ *np->dn->hprevp = np->dn->hnext; ++ if (np->dn->hnext != 0) ++ np->dn->hnext->dn->hprevp = np->dn->hprevp; ++ all_nodes_nr_items -= 1; ++ np->dn->hnext = NULL; ++ np->dn->hprevp = NULL; ++ diskfs_nrele_light (np); ++ } ++ pthread_mutex_unlock (&all_nodes_lock); + } + + /* The user must define this funcction. Node NP has some light +diff --git a/tmpfs/tmpfs.c b/tmpfs/tmpfs.c +index a45d343..024b818 100644 +--- a/tmpfs/tmpfs.c ++++ b/tmpfs/tmpfs.c +@@ -67,10 +67,10 @@ diskfs_set_statfs (struct statfs *st) + st->f_bsize = vm_page_size; + st->f_blocks = tmpfs_page_limit; + +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&all_nodes_lock); + st->f_files = num_files; + pages = round_page (tmpfs_space_used) / vm_page_size; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&all_nodes_lock); + + st->f_bfree = pages < tmpfs_page_limit ? tmpfs_page_limit - pages : 0; + st->f_bavail = st->f_bfree; +diff --git a/tmpfs/tmpfs.h b/tmpfs/tmpfs.h +index b3c636d..187249e 100644 +--- a/tmpfs/tmpfs.h ++++ b/tmpfs/tmpfs.h +@@ -24,6 +24,7 @@ the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */ + #include + #include + #include ++#include + + struct disknode + { +@@ -71,15 +72,16 @@ struct tmpfs_dirent + + extern unsigned int num_files; + extern off_t tmpfs_page_limit, tmpfs_space_used; ++extern pthread_mutex_t all_nodes_lock; + + extern mach_port_t default_pager; + + static inline void + adjust_used (off_t change) + { +- pthread_spin_lock (&diskfs_node_refcnt_lock); ++ pthread_mutex_lock (&all_nodes_lock); + tmpfs_space_used += change; +- pthread_spin_unlock (&diskfs_node_refcnt_lock); ++ pthread_mutex_unlock (&all_nodes_lock); + } + + #endif +-- +2.0.0.rc0 + diff --git a/debian/patches/series b/debian/patches/series index c08b0f99..5bce22f8 100644 --- a/debian/patches/series +++ b/debian/patches/series @@ -40,3 +40,18 @@ xkb-compat.patch mach-defpager-protected-payload.patch +0001-libihash-fix-type-of-max_load.patch +0002-libihash-use-an-integer-hash-function-on-the-keys.patch +0003-libihash-use-linear-probing-and-fast-modulo-operatio.patch +0004-libihash-use-fast-binary-scaling-to-determine-the-lo.patch +0005-include-add-lock-less-reference-counting-primitives.patch +0006-libdiskfs-lock-less-reference-counting-for-peropen-o.patch +0007-libtrivfs-lock-less-reference-counting-for-trivfs_pe.patch +0008-libports-use-a-single-hash-table.patch +0009-libports-lock-less-reference-counting-for-port_info-.patch +0010-ext2fs-improve-enable-disable-_caching.patch +0011-fatfs-improve-enable-disable-_caching.patch +0012-ext2fs-use-a-seperate-lock-to-protect-nodehash.patch +0013-fatfs-use-a-seperate-lock-to-protect-nodehash.patch +0014-isofs-use-a-seperate-lock-to-protect-node_cache.patch +0015-tmpfs-use-a-seperate-lock-to-protect-all_nodes.patch -- cgit v1.2.3