From 5a0e3ad6af8660be21ca98a971cd00f331318c05 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 24 Mar 2010 17:04:11 +0900 Subject: include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo Guess-its-ok-by: Christoph Lameter Cc: Ingo Molnar Cc: Lee Schermerhorn --- net/netlabel/netlabel_domainhash.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/netlabel/netlabel_domainhash.c') diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c index 0bfeaab88ef5..016ab9c75ebd 100644 --- a/net/netlabel/netlabel_domainhash.c +++ b/net/netlabel/netlabel_domainhash.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From b914f3a2a35812545f773645f340d7c075e5b64d Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Thu, 1 Apr 2010 10:43:57 +0000 Subject: netlabel: Fix several rcu_dereference() calls used without RCU read locks The recent changes to add RCU lock verification to rcu_dereference() calls caught out a problem with netlbl_unlhsh_hash(), see below. =================================================== [ INFO: suspicious rcu_dereference_check() usage. ] --------------------------------------------------- net/netlabel/netlabel_unlabeled.c:246 invoked rcu_dereference_check() without protection! This patch fixes this problem as well as others like it in the NetLabel code. Also included in this patch is the identification of future work to eliminate the RCU read lock in netlbl_domhsh_add(), but in the interest of getting this patch out quickly that work will happen in another patch to be finished later. Thanks to Eric Dumazet and Paul McKenney for their help in understanding the recent RCU changes. Signed-off-by: Paul Moore Reported-by: David Howells CC: Eric Dumazet CC: Paul E. McKenney Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/netlabel/netlabel_domainhash.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) (limited to 'net/netlabel/netlabel_domainhash.c') diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c index 0bfeaab88ef5..06ab41b6b57a 100644 --- a/net/netlabel/netlabel_domainhash.c +++ b/net/netlabel/netlabel_domainhash.c @@ -50,9 +50,12 @@ struct netlbl_domhsh_tbl { }; /* Domain hash table */ -/* XXX - updates should be so rare that having one spinlock for the entire - * hash table should be okay */ +/* updates should be so rare that having one spinlock for the entire hash table + * should be okay */ static DEFINE_SPINLOCK(netlbl_domhsh_lock); +#define netlbl_domhsh_rcu_deref(p) \ + rcu_dereference_check(p, rcu_read_lock_held() || \ + lockdep_is_held(&netlbl_domhsh_lock)) static struct netlbl_domhsh_tbl *netlbl_domhsh = NULL; static struct netlbl_dom_map *netlbl_domhsh_def = NULL; @@ -106,7 +109,8 @@ static void netlbl_domhsh_free_entry(struct rcu_head *entry) * Description: * This is the hashing function for the domain hash table, it returns the * correct bucket number for the domain. The caller is responsibile for - * calling the rcu_read_[un]lock() functions. + * ensuring that the hash table is protected with either a RCU read lock or the + * hash table lock. * */ static u32 netlbl_domhsh_hash(const char *key) @@ -120,7 +124,7 @@ static u32 netlbl_domhsh_hash(const char *key) for (iter = 0, val = 0, len = strlen(key); iter < len; iter++) val = (val << 4 | (val >> (8 * sizeof(u32) - 4))) ^ key[iter]; - return val & (rcu_dereference(netlbl_domhsh)->size - 1); + return val & (netlbl_domhsh_rcu_deref(netlbl_domhsh)->size - 1); } /** @@ -130,7 +134,8 @@ static u32 netlbl_domhsh_hash(const char *key) * Description: * Searches the domain hash table and returns a pointer to the hash table * entry if found, otherwise NULL is returned. The caller is responsibile for - * the rcu hash table locks (i.e. the caller much call rcu_read_[un]lock()). + * ensuring that the hash table is protected with either a RCU read lock or the + * hash table lock. * */ static struct netlbl_dom_map *netlbl_domhsh_search(const char *domain) @@ -141,7 +146,7 @@ static struct netlbl_dom_map *netlbl_domhsh_search(const char *domain) if (domain != NULL) { bkt = netlbl_domhsh_hash(domain); - bkt_list = &rcu_dereference(netlbl_domhsh)->tbl[bkt]; + bkt_list = &netlbl_domhsh_rcu_deref(netlbl_domhsh)->tbl[bkt]; list_for_each_entry_rcu(iter, bkt_list, list) if (iter->valid && strcmp(iter->domain, domain) == 0) return iter; @@ -159,8 +164,8 @@ static struct netlbl_dom_map *netlbl_domhsh_search(const char *domain) * Searches the domain hash table and returns a pointer to the hash table * entry if an exact match is found, if an exact match is not present in the * hash table then the default entry is returned if valid otherwise NULL is - * returned. The caller is responsibile for the rcu hash table locks - * (i.e. the caller much call rcu_read_[un]lock()). + * returned. The caller is responsibile ensuring that the hash table is + * protected with either a RCU read lock or the hash table lock. * */ static struct netlbl_dom_map *netlbl_domhsh_search_def(const char *domain) @@ -169,7 +174,7 @@ static struct netlbl_dom_map *netlbl_domhsh_search_def(const char *domain) entry = netlbl_domhsh_search(domain); if (entry == NULL) { - entry = rcu_dereference(netlbl_domhsh_def); + entry = netlbl_domhsh_rcu_deref(netlbl_domhsh_def); if (entry != NULL && !entry->valid) entry = NULL; } @@ -306,8 +311,11 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry, struct netlbl_af6list *tmp6; #endif /* IPv6 */ + /* XXX - we can remove this RCU read lock as the spinlock protects the + * entire function, but before we do we need to fixup the + * netlbl_af[4,6]list RCU functions to do "the right thing" with + * respect to rcu_dereference() when only a spinlock is held. */ rcu_read_lock(); - spin_lock(&netlbl_domhsh_lock); if (entry->domain != NULL) entry_old = netlbl_domhsh_search(entry->domain); -- cgit v1.2.3