From 9c4079d8c3b158369585d33f7b19fc7bd585401e Mon Sep 17 00:00:00 2001 From: Jeremy Allison Date: Thu, 25 May 2000 20:05:50 +0000 Subject: Fixed misunderstanding found during CIFS conference. Overlapping lock ranges (not just included lock ranges) should be handled correctly. UNIT test still needed. Jeremy. (This used to be commit 07872298e3ee8b4b50b69cb4e49b88635792128e) --- source3/locking/posix.c | 618 ++++++++++++++++++++++-------------------------- 1 file changed, 278 insertions(+), 340 deletions(-) (limited to 'source3/locking/posix.c') diff --git a/source3/locking/posix.c b/source3/locking/posix.c index d1edb1ef57..6ab4bef4cc 100644 --- a/source3/locking/posix.c +++ b/source3/locking/posix.c @@ -345,65 +345,22 @@ static BOOL delete_posix_lock_entry_by_index(files_struct *fsp, size_t entry) } /**************************************************************************** - Add an entry into the POSIX locking tdb. Returns the number of records that - completely overlap this request, or -1 on error. entry_num gets set to the - index number of the added lock (used in case we need to delete *exactly* - this entry). + Add an entry into the POSIX locking tdb. We return the index number of the + added lock (used in case we need to delete *exactly* this entry). Returns + False on fail, True on success. ****************************************************************************/ -static int add_posix_lock_entry(files_struct *fsp, SMB_OFF_T start, SMB_OFF_T size, int lock_type, size_t *entry_num) +static BOOL add_posix_lock_entry(files_struct *fsp, SMB_OFF_T start, SMB_OFF_T size, int lock_type, size_t *pentry_num) { TDB_DATA kbuf = locking_key_fsp(fsp); TDB_DATA dbuf; struct posix_lock pl; - struct posix_lock *entries; - size_t i, count; - int num_overlapping_records = 0; - *entry_num = 0; - - /* - * Windows is very strange. It allows read locks to be overlayed on - * a write lock, but leaves the write lock in force until the first - * unlock. It also reference counts the locks. This means the following sequence : - * - * process1 process2 - * ------------------------------------------------------------------------ - * WRITE LOCK : start = 0, len = 10 - * READ LOCK: start =0, len = 10 - FAIL - * READ LOCK : start = 5, len = 2 - * READ LOCK: start =0, len = 10 - FAIL - * UNLOCK : start = 0, len = 10 - * READ LOCK: start =0, len = 10 - OK - * - * Under POSIX, the same sequence in steps 1 and 2 would not be reference counted, but - * would leave a single read lock over the 0-10 region. In order to - * re-create Windows semantics mapped to POSIX locks, we create multiple TDB - * entries, one for each overlayed lock request. We are guarenteed by the brlock - * semantics that if a write lock is added, then it will be first in the array. - */ - dbuf.dptr = NULL; dbuf = tdb_fetch(posix_lock_tdb, kbuf); - count = (size_t)(dbuf.dsize / sizeof(pl)); - entries = (struct posix_lock *)dbuf.dptr; - - /* - * Ensure we look for overlapping entries *before* - * we add this entry. Count the number of entries - * that completely overlap this request. - */ - - for (i = 0; i < count; i++) { - struct posix_lock *entry = &entries[i]; - - if (fsp->fd == entry->fd && - start >= entry->start && - start + size <= entry->start + entry->size) - num_overlapping_records++; - } + *pentry_num = (size_t)(dbuf.dsize / sizeof(pl)); /* * Add new record. @@ -423,8 +380,6 @@ static int add_posix_lock_entry(files_struct *fsp, SMB_OFF_T start, SMB_OFF_T si memcpy(dbuf.dptr + dbuf.dsize, &pl, sizeof(pl)); dbuf.dsize += sizeof(pl); - *entry_num = count; - if (tdb_store(posix_lock_tdb, kbuf, dbuf, TDB_REPLACE) == -1) { DEBUG(0,("add_posix_lock: Failed to add lock entry on file %s\n", fsp->fsp_name)); goto fail; @@ -432,23 +387,36 @@ static int add_posix_lock_entry(files_struct *fsp, SMB_OFF_T start, SMB_OFF_T si free(dbuf.dptr); - DEBUG(10,("add_posix_lock: File %s: type = %s: start=%.0f size=%.0f: num_records = %d : dev=%.0f inode=%.0f\n", - fsp->fsp_name, posix_lock_type_name(lock_type), (double)start, (double)size, num_overlapping_records, + DEBUG(10,("add_posix_lock: File %s: type = %s: start=%.0f size=%.0f: dev=%.0f inode=%.0f\n", + fsp->fsp_name, posix_lock_type_name(lock_type), (double)start, (double)size, (double)fsp->dev, (double)fsp->inode )); - return num_overlapping_records; + return True; fail: if (dbuf.dptr) free(dbuf.dptr); - *entry_num = 0; - return -1; + return False; +} + +/**************************************************************************** + Calculate if locks have any overlap at all. +****************************************************************************/ + +static BOOL does_lock_overlap(SMB_OFF_T start1, SMB_OFF_T size1, SMB_OFF_T start2, SMB_OFF_T size2) +{ + if (start1 >= start2 && start1 <= start2 + size2) + return True; + + if (start1 < start2 && start1 + size1 > start2); + return True; + + return False; } /**************************************************************************** Delete an entry from the POSIX locking tdb. Returns a copy of the entry being - deleted and the number of records that are completely overlapped by this one, - or -1 on error. + deleted and the number of records that are overlapped by this one, or -1 on error. ****************************************************************************/ static int delete_posix_lock_entry(files_struct *fsp, SMB_OFF_T start, SMB_OFF_T size, struct posix_lock *pl) @@ -510,23 +478,20 @@ static int delete_posix_lock_entry(files_struct *fsp, SMB_OFF_T start, SMB_OFF_T /* * Count the number of entries that are - * overlapped completely by this unlock request. - * (Note that this is the reverse of the test in - * add_posix_lock). + * overlapped by this unlock request. */ for (i = 0; i < count; i++) { struct posix_lock *entry = &locks[i]; if (fsp->fd == entry->fd && - entry->start >= start && - entry->start + entry->size <= start + size) + does_lock_overlap( start, size, entry->start, entry->size)) num_overlapping_records++; } - DEBUG(10,("delete_posix_lock_entry: type = %s: start=%.0f size=%.0f, num_records = %d\n", - posix_lock_type_name(pl->lock_type), (double)pl->start, (double)pl->size, - (unsigned int)num_overlapping_records )); + DEBUG(10,("delete_posix_lock_entry: type = %s: start=%.0f size=%.0f, num_records = %d\n", + posix_lock_type_name(pl->lock_type), (double)pl->start, (double)pl->size, + (unsigned int)num_overlapping_records )); if (dbuf.dptr) free(dbuf.dptr); @@ -579,177 +544,81 @@ static int map_posix_lock_type( files_struct *fsp, enum brl_type lock_type) static BOOL posix_lock_in_range(SMB_OFF_T *offset_out, SMB_OFF_T *count_out, SMB_BIG_UINT u_offset, SMB_BIG_UINT u_count) { - SMB_OFF_T offset; - SMB_OFF_T count; - -#if defined(LARGE_SMB_OFF_T) && !defined(HAVE_BROKEN_FCNTL64_LOCKS) - - SMB_OFF_T mask2 = ((SMB_OFF_T)0x4) << (SMB_OFF_T_BITS-4); - SMB_OFF_T mask = (mask2<<1); - SMB_OFF_T neg_mask = ~mask; - - /* - * In this case SMB_OFF_T is 64 bits, - * and the underlying system can handle 64 bit signed locks. - * Cast to signed type. - */ - - offset = (SMB_OFF_T)u_offset; - count = (SMB_OFF_T)u_count; + SMB_OFF_T offset = (SMB_OFF_T)u_offset; + SMB_OFF_T count = (SMB_OFF_T)u_count; /* - * Deal with a very common case of count of all ones. - * (lock entire file). + * For the type of system we are, attempt to + * find the maximum positive lock offset as an SMB_OFF_T. */ - if(count == (SMB_OFF_T)-1) - count &= ~mask; +#if defined(LARGE_SMB_OFF_T) && !defined(HAVE_BROKEN_FCNTL64_LOCKS) /* - * POSIX lock ranges cannot be negative. - * Fail if any combination becomes negative. + * In this case SMB_OFF_T is 64 bits, + * and the underlying system can handle 64 bit signed locks. */ - if(offset < 0 || count < 0 || (offset + count < 0)) { - DEBUG(10,("posix_lock_in_range: negative range: offset = %.0f, count = %.0f. Ignoring lock.\n", - (double)offset, (double)count )); - return False; - } - - /* - * In this case SMB_OFF_T is 64 bits, the offset and count - * fit within the positive range, and the underlying - * system can handle 64 bit locks. Just return as the - * cast values are ok. - */ + SMB_OFF_T mask2 = ((SMB_OFF_T)0x4) << (SMB_OFF_T_BITS-4); + SMB_OFF_T mask = (mask2<<1); + SMB_OFF_T max_positive_lock_offset = ~mask; #else /* !LARGE_SMB_OFF_T || HAVE_BROKEN_FCNTL64_LOCKS */ /* * In this case either SMB_OFF_T is 32 bits, * or the underlying system cannot handle 64 bit signed locks. - * Either way we have to try and mangle to fit within 31 bits. - * This is difficult. + * All offsets & counts must be 2^31 or less. */ -#if defined(HAVE_BROKEN_FCNTL64_LOCKS) + SMB_OFF_T max_positive_lock_offset = 0x7FFFFFFF; - /* - * SMB_OFF_T is 64 bits, but we need to use 31 bits due to - * broken large locking. - */ +#endif /* !LARGE_SMB_OFF_T || HAVE_BROKEN_FCNTL64_LOCKS */ /* - * Deal with a very common case of count of all ones. - * (lock entire file). + * If the given offset was > max_positive_lock_offset then we cannot map this at all + * ignore this lock. */ - if(u_count == (SMB_BIG_UINT)-1) - count = 0x7FFFFFFF; - - if(((u_offset >> 32) & 0xFFFFFFFF) || ((u_count >> 32) & 0xFFFFFFFF)) { - DEBUG(10,("posix_lock_in_range: top 32 bits not zero. offset = %.0f, count = %.0f. Ignoring lock.\n", - (double)u_offset, (double)u_count )); - /* Top 32 bits of offset or count were not zero. */ + if (u_offset & ~((SMB_BIG_UINT)max_positive_lock_offset)) { + DEBUG(10,("posix_lock_in_range: (offset = %.0f) offset > %.0f and we cannot handle this. Ignoring lock.\n", + (double)u_offset, (double)((SMB_BIG_UINT)max_positive_lock_offset) )); return False; } - /* Cast from 64 bits unsigned to 64 bits signed. */ - offset = (SMB_OFF_T)u_offset; - count = (SMB_OFF_T)u_count; - - /* - * Check if we are within the 2^31 range. - */ - - { - int32 low_offset = (int32)offset; - int32 low_count = (int32)count; - - if(low_offset < 0 || low_count < 0 || (low_offset + low_count < 0)) { - DEBUG(10,("posix_lock_in_range: not within 2^31 range. low_offset = %d, low_count = %d. Ignoring lock.\n", - low_offset, low_count )); - return False; - } - } - - /* - * Ok - we can map from a 64 bit number to a 31 bit lock. - */ - -#else /* HAVE_BROKEN_FCNTL64_LOCKS */ - /* - * SMB_OFF_T is 32 bits. + * We must truncate the offset and count to less than max_positive_lock_offset. */ -#if defined(HAVE_LONGLONG) + offset &= max_positive_lock_offset; + count &= max_positive_lock_offset; - /* - * SMB_BIG_UINT is 64 bits, we can do a 32 bit shift. - */ /* * Deal with a very common case of count of all ones. * (lock entire file). */ - if(u_count == (SMB_BIG_UINT)-1) - count = 0x7FFFFFFF; - - if(((u_offset >> 32) & 0xFFFFFFFF) || ((u_count >> 32) & 0xFFFFFFFF)) { - DEBUG(10,("posix_lock_in_range: top 32 bits not zero. u_offset = %.0f, u_count = %.0f. Ignoring lock.\n", - (double)u_offset, (double)u_count )); - return False; - } - - /* Cast from 64 bits unsigned to 32 bits signed. */ - offset = (SMB_OFF_T)u_offset; - count = (SMB_OFF_T)u_count; - - /* - * Check if we are within the 2^31 range. - */ - - if(offset < 0 || count < 0 || (offset + count < 0)) { - DEBUG(10,("posix_lock_in_range: not within 2^31 range. offset = %d, count = %d. Ignoring lock.\n", - (int)offset, (int)count )); - return False; - } - -#else /* HAVE_LONGLONG */ - - /* - * SMB_BIG_UINT and SMB_OFF_T are both 32 bits, - * just cast. - */ + if(count == (SMB_OFF_T)-1) + count = max_positive_lock_offset; /* - * Deal with a very common case of count of all ones. - * (lock entire file). + * Truncate count to end at max lock offset. */ - if(u_count == (SMB_BIG_UINT)-1) - count = 0x7FFFFFFF; - - /* Cast from 32 bits unsigned to 32 bits signed. */ - offset = (SMB_OFF_T)u_offset; - count = (SMB_OFF_T)u_count; + if (offset + count < 0 || offset + count > max_positive_lock_offset) + count = max_positive_lock_offset - offset; /* - * Check if we are within the 2^31 range. + * If we ate all the count, ignore this lock. */ - if(offset < 0 || count < 0 || (offset + count < 0)) { - DEBUG(10,("posix_lock_in_range: not within 2^31 range. offset = %d, count = %d. Ignoring lock.\n", - (int)offset, (int)count )); + if (count == 0) { + DEBUG(10,("posix_lock_in_range: Count = 0. Ignoring lock u_offset = %.0f, u_count = %.0f\n", + (double)u_offset, (double)u_count )); return False; } -#endif /* HAVE_LONGLONG */ -#endif /* LARGE_SMB_OFF_T */ -#endif /* !LARGE_SMB_OFF_T || HAVE_BROKEN_FCNTL64_LOCKS */ - /* * The mapping was successful. */ @@ -914,89 +783,25 @@ BOOL is_posix_locked(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u_co return posix_fcntl_lock(fsp,SMB_F_GETLK,offset,count,posix_lock_type); } -/**************************************************************************** - POSIX function to acquire a lock. Returns True if the - lock could be granted, False if not. -****************************************************************************/ - -BOOL set_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u_count, enum brl_type lock_type) -{ - SMB_OFF_T offset; - SMB_OFF_T count; - BOOL ret = True; - size_t entry_num = 0; - int posix_lock_type = map_posix_lock_type(fsp,lock_type); - int num_overlapping_records; - - DEBUG(5,("set_posix_lock: File %s, offset = %.0f, count = %.0f, type = %s\n", - fsp->fsp_name, (double)u_offset, (double)u_count, posix_lock_type_name(lock_type) )); - - /* - * If the requested lock won't fit in the POSIX range, we will - * pretend it was successful. - */ - - if(!posix_lock_in_range(&offset, &count, u_offset, u_count)) - return True; - - /* - * Note that setting multiple overlapping locks on different - * file descriptors will not be held separately by the kernel (POSIX - * braindamage), but will be merged into one continuous lock - * range. We cope with this case in the release_posix_lock code - * below. We need to add the posix lock entry into the tdb before - * doing the real posix lock call to deal with the locking overlay - * case described above in add_posix_lock_entry(). - */ - - num_overlapping_records = add_posix_lock_entry(fsp,offset,count,posix_lock_type,&entry_num); - - if (num_overlapping_records == -1) { - DEBUG(0,("set_posix_lock: Unable to create posix lock entry !\n")); - return False; - } - - /* - * num_overlapping_records is the count of lock records that - * completely contain this request on the same fsp. Only bother - * with the real lock request if there are none, otherwise ignore. - */ - - if (num_overlapping_records == 0) { - /* - * First lock entry for this range created. Do a real POSIX lock. - */ - ret = posix_fcntl_lock(fsp,SMB_F_SETLK,offset,count,posix_lock_type); - - /* - * Oops, POSIX lock failed, delete the tdb entry we just added. - */ - if (!ret) - delete_posix_lock_entry_by_index(fsp,entry_num); - } - - return ret; -} - /* * Structure used when splitting a lock range * into a POSIX lock range. Doubly linked list. */ -struct unlock_list { - struct unlock_list *next; - struct unlock_list *prev; +struct lock_list { + struct lock_list *next; + struct lock_list *prev; SMB_OFF_T start; SMB_OFF_T size; }; /**************************************************************************** Create a list of lock ranges that don't overlap a given range. Used in calculating - POSIX lock unlocks. This is a difficult function that requires ASCII art to + POSIX locks and unlocks. This is a difficult function that requires ASCII art to understand it :-). ****************************************************************************/ -static struct unlock_list *posix_unlock_list(TALLOC_CTX *ctx, struct unlock_list *ulhead, files_struct *fsp) +static struct lock_list *posix_lock_list(TALLOC_CTX *ctx, struct lock_list *lhead, files_struct *fsp) { TDB_DATA kbuf = locking_key_fsp(fsp); TDB_DATA dbuf; @@ -1007,9 +812,8 @@ static struct unlock_list *posix_unlock_list(TALLOC_CTX *ctx, struct unlock_list dbuf = tdb_fetch(posix_lock_tdb, kbuf); - if (!dbuf.dptr) { - return ulhead; - } + if (!dbuf.dptr) + return lhead; locks = (struct posix_lock *)dbuf.dptr; num_locks = (size_t)(dbuf.dsize / sizeof(*locks)); @@ -1019,48 +823,48 @@ static struct unlock_list *posix_unlock_list(TALLOC_CTX *ctx, struct unlock_list * Quit if the list is deleted. */ - DEBUG(10,("posix_unlock_list: curr: start=%.0f,size=%.0f\n", - (double)ulhead->start, (double)ulhead->size )); + DEBUG(10,("posix_lock_list: curr: start=%.0f,size=%.0f\n", + (double)lhead->start, (double)lhead->size )); - for (i=0; ifd, + DEBUG(10,("posix_lock_list: lock: fd=%d: start=%.0f,size=%.0f:type=%s", lock->fd, (double)lock->start, (double)lock->size, posix_lock_type_name(lock->lock_type) )); - if ( (ul_curr->start >= (lock->start + lock->size)) || - (lock->start >= (ul_curr->start + ul_curr->size))) { + if ( (l_curr->start >= (lock->start + lock->size)) || + (lock->start >= (l_curr->start + l_curr->size))) { /* No overlap with this lock - leave this range alone. */ /********************************************* +---------+ - | ul_curr | + | l_curr | +---------+ +-------+ | lock | +-------+ OR.... +---------+ - | ul_curr | + | l_curr | +---------+ **********************************************/ DEBUG(10,("no overlap case.\n" )); - ul_curr = ul_curr->next; + l_curr = l_curr->next; - } else if ( (ul_curr->start >= lock->start) && - (ul_curr->start + ul_curr->size <= lock->start + lock->size) ) { + } else if ( (l_curr->start >= lock->start) && + (l_curr->start + l_curr->size <= lock->start + lock->size) ) { /* * This unlock is completely overlapped by this existing lock range @@ -1068,26 +872,26 @@ OR.... */ /********************************************* +---------+ - | ul_curr | + | l_curr | +---------+ +---------------------------+ | lock | +---------------------------+ **********************************************/ /* Save the next pointer */ - struct unlock_list *ul_next = ul_curr->next; + struct lock_list *ul_next = l_curr->next; DEBUG(10,("delete case.\n" )); - DLIST_REMOVE(ulhead, ul_curr); - if(ulhead == NULL) + DLIST_REMOVE(lhead, l_curr); + if(lhead == NULL) break; /* No more list... */ - ul_curr = ul_next; + l_curr = ul_next; - } else if ( (ul_curr->start >= lock->start) && - (ul_curr->start < lock->start + lock->size) && - (ul_curr->start + ul_curr->size > lock->start + lock->size) ) { + } else if ( (l_curr->start >= lock->start) && + (l_curr->start < lock->start + lock->size) && + (l_curr->start + l_curr->size > lock->start + lock->size) ) { /* * This unlock overlaps the existing lock range at the high end. @@ -1095,28 +899,28 @@ OR.... */ /********************************************* +---------------+ - | ul_curr | + | l_curr | +---------------+ +---------------+ | lock | +---------------+ BECOMES.... +-------+ - |ul_curr| + | l_curr| +-------+ **********************************************/ - ul_curr->size = (ul_curr->start + ul_curr->size) - (lock->start + lock->size); - ul_curr->start = lock->start + lock->size; + l_curr->size = (l_curr->start + l_curr->size) - (lock->start + lock->size); + l_curr->start = lock->start + lock->size; DEBUG(10,("truncate high case: start=%.0f,size=%.0f\n", - (double)ul_curr->start, (double)ul_curr->size )); + (double)l_curr->start, (double)l_curr->size )); - ul_curr = ul_curr->next; + l_curr = l_curr->next; - } else if ( (ul_curr->start < lock->start) && - (ul_curr->start + ul_curr->size > lock->start) && - (ul_curr->start + ul_curr->size <= lock->start + lock->size) ) { + } else if ( (l_curr->start < lock->start) && + (l_curr->start + l_curr->size > lock->start) && + (l_curr->start + l_curr->size <= lock->start + lock->size) ) { /* * This unlock overlaps the existing lock range at the low end. @@ -1124,26 +928,26 @@ BECOMES.... */ /********************************************* +---------------+ - | ul_curr | + | l_curr | +---------------+ +---------------+ | lock | +---------------+ BECOMES.... +-------+ - |ul_curr| + | l_curr| +-------+ **********************************************/ - ul_curr->size = lock->start - ul_curr->start; + l_curr->size = lock->start - l_curr->start; DEBUG(10,("truncate low case: start=%.0f,size=%.0f\n", - (double)ul_curr->start, (double)ul_curr->size )); + (double)l_curr->start, (double)l_curr->size )); - ul_curr = ul_curr->next; + l_curr = l_curr->next; - } else if ( (ul_curr->start < lock->start) && - (ul_curr->start + ul_curr->size > lock->start + lock->size) ) { + } else if ( (l_curr->start < lock->start) && + (l_curr->start + l_curr->size > lock->start + lock->size) ) { /* * Worst case scenario. Unlock request completely overlaps an existing * lock range. Split the request into two, push the new (upper) request @@ -1152,46 +956,46 @@ BECOMES.... */ /********************************************* +---------------------------+ - | ul_curr | + | l_curr | +---------------------------+ +---------+ | lock | +---------+ BECOMES..... +-------+ +---------+ - |ul_curr| |ul_new | + | l_curr| | l_new | +-------+ +---------+ **********************************************/ - struct unlock_list *ul_new = (struct unlock_list *)talloc(ctx, - sizeof(struct unlock_list)); + struct lock_list *l_new = (struct lock_list *)talloc(ctx, + sizeof(struct lock_list)); - if(ul_new == NULL) { - DEBUG(0,("posix_unlock_list: talloc fail.\n")); + if(l_new == NULL) { + DEBUG(0,("posix_lock_list: talloc fail.\n")); return NULL; /* The talloc_destroy takes care of cleanup. */ } - ZERO_STRUCTP(ul_new); - ul_new->start = lock->start + lock->size; - ul_new->size = ul_curr->start + ul_curr->size - ul_new->start; + ZERO_STRUCTP(l_new); + l_new->start = lock->start + lock->size; + l_new->size = l_curr->start + l_curr->size - l_new->start; - /* Truncate the ul_curr. */ - ul_curr->size = lock->start - ul_curr->start; + /* Truncate the l_curr. */ + l_curr->size = lock->start - l_curr->start; DEBUG(10,("split case: curr: start=%.0f,size=%.0f \ -new: start=%.0f,size=%.0f\n", (double)ul_curr->start, (double)ul_curr->size, - (double)ul_new->start, (double)ul_new->size )); +new: start=%.0f,size=%.0f\n", (double)l_curr->start, (double)l_curr->size, + (double)l_new->start, (double)l_new->size )); /* - * Add into the dlink list after the ul_curr point - NOT at ulhead. + * Add into the dlink list after the l_curr point - NOT at lhead. * Note we can't use DLINK_ADD here as this inserts at the head of the given list. */ - ul_new->prev = ul_curr; - ul_new->next = ul_curr->next; - ul_curr->next = ul_new; + l_new->prev = l_curr; + l_new->next = l_curr->next; + l_curr->next = l_new; /* And move after the link we added. */ - ul_curr = ul_new->next; + l_curr = l_new->next; } else { @@ -1201,18 +1005,159 @@ new: start=%.0f,size=%.0f\n", (double)ul_curr->start, (double)ul_curr->size, */ pstring msg; - slprintf(msg, sizeof(msg)-1, "logic flaw in cases: ul_curr: start = %.0f, size = %.0f : \ -lock: start = %.0f, size = %.0f\n", (double)ul_curr->start, (double)ul_curr->size, (double)lock->start, (double)lock->size ); + slprintf(msg, sizeof(msg)-1, "logic flaw in cases: l_curr: start = %.0f, size = %.0f : \ +lock: start = %.0f, size = %.0f\n", (double)l_curr->start, (double)l_curr->size, (double)lock->start, (double)lock->size ); smb_panic(msg); } - } /* end for ( ul_curr = ulhead; ul_curr;) */ + } /* end for ( l_curr = lhead; l_curr;) */ } /* end for (i=0; ifsp_name, (double)u_offset, (double)u_count, posix_lock_type_name(lock_type) )); + + /* + * If the requested lock won't fit in the POSIX range, we will + * pretend it was successful. + */ + + if(!posix_lock_in_range(&offset, &count, u_offset, u_count)) + return True; + + /* + * Windows is very strange. It allows read locks to be overlayed + * (even over a write lock), but leaves the write lock in force until the first + * unlock. It also reference counts the locks. This means the following sequence : + * + * process1 process2 + * ------------------------------------------------------------------------ + * WRITE LOCK : start = 2, len = 10 + * READ LOCK: start =0, len = 10 - FAIL + * READ LOCK : start = 0, len = 14 + * READ LOCK: start =0, len = 10 - FAIL + * UNLOCK : start = 2, len = 10 + * READ LOCK: start =0, len = 10 - OK + * + * Under POSIX, the same sequence in steps 1 and 2 would not be reference counted, but + * would leave a single read lock over the 0-14 region. In order to + * re-create Windows semantics mapped to POSIX locks, we create multiple TDB + * entries, one for each overlayed lock request. We are guarenteed by the brlock + * semantics that if a write lock is added, then it will be first in the array. + */ + + if ((l_ctx = talloc_init()) == NULL) { + DEBUG(0,("set_posix_lock: unable to init talloc context.\n")); + return True; /* Not a fatal error. */ + } + + if ((ll = (struct lock_list *)talloc(l_ctx, sizeof(struct lock_list))) == NULL) { + DEBUG(0,("set_posix_lock: unable to talloc unlock list.\n")); + talloc_destroy(l_ctx); + return True; /* Not a fatal error. */ + } + + /* + * Create the initial list entry containing the + * lock we want to add. + */ + + ZERO_STRUCTP(ll); + ll->start = offset; + ll->size = count; + + DLIST_ADD(llist, ll); + + /* + * The following call calculates if there are any + * overlapping locks held by this process on + * fd's open on the same file and splits this list + * into a list of lock ranges that do not overlap with existing + * POSIX locks. + */ + + llist = posix_lock_list(l_ctx, llist, fsp); + + /* + * Now we have the list of ranges to lock it is safe to add the + * entry into the POSIX lock tdb. We take note of the entry we + * added here in case we have to remove it on POSIX lock fail. + */ + + if (!add_posix_lock_entry(fsp,offset,count,posix_lock_type,&entry_num)) { + DEBUG(0,("set_posix_lock: Unable to create posix lock entry !\n")); + talloc_destroy(l_ctx); + return False; + } + + /* + * Add the POSIX locks on the list of ranges returned. + * As the lock is supposed to be added atomically, we need to + * back out all the locks if any one of these calls fail. + */ + + for (lock_count = 0, ll = llist; ll; ll = ll->next, lock_count++) { + offset = ll->start; + count = ll->size; + + DEBUG(5,("set_posix_lock: Real lock: Type = %s: offset = %.0f, count = %.0f\n", + posix_lock_type_name(posix_lock_type), (double)offset, (double)count )); + + if (!posix_fcntl_lock(fsp,SMB_F_SETLK,offset,count,posix_lock_type)) { + DEBUG(5,("set_posix_lock: Lock fail !: Type = %s: offset = %.0f, count = %.0f\n", + posix_lock_type_name(posix_lock_type), (double)offset, (double)count )); + ret = False; + break; + } + } + + if (!ret) { + + /* + * Back out all the POSIX locks we have on fail. + */ + + for (ll = llist; lock_count; ll = ll->next, lock_count--) { + offset = ll->start; + count = ll->size; + + DEBUG(5,("set_posix_lock: Backing out locks: Type = %s: offset = %.0f, count = %.0f\n", + posix_lock_type_name(posix_lock_type), (double)offset, (double)count )); + + posix_fcntl_lock(fsp,SMB_F_SETLK,offset,count,F_UNLCK); + } + + /* + * Remove the tdb entry for this lock. + */ + + delete_posix_lock_entry_by_index(fsp,entry_num); + } + + talloc_destroy(l_ctx); + return ret; } /**************************************************************************** @@ -1226,8 +1171,8 @@ BOOL release_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u SMB_OFF_T count; BOOL ret = True; TALLOC_CTX *ul_ctx = NULL; - struct unlock_list *ulist = NULL; - struct unlock_list *ul = NULL; + struct lock_list *ulist = NULL; + struct lock_list *ul = NULL; struct posix_lock deleted_lock; int num_overlapped_entries; @@ -1245,6 +1190,8 @@ BOOL release_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u /* * We treat this as one unlock request for POSIX accounting purposes even * if it may later be split into multiple smaller POSIX unlock ranges. + * num_overlapped_entries is the number of existing locks that have any + * overlap with this unlock request. */ num_overlapped_entries = delete_posix_lock_entry(fsp, offset, count, &deleted_lock); @@ -1256,7 +1203,8 @@ BOOL release_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u /* * If num_overlapped_entries is > 0, and the lock_type we just deleted from the tdb was * a POSIX write lock, then before doing the unlock we need to downgrade - * the POSIX lock to a read lock. + * the POSIX lock to a read lock. This allows any overlapping read locks + * to be atomically maintained. */ if (num_overlapped_entries > 0 && deleted_lock.lock_type == F_WRLCK) { @@ -1267,11 +1215,11 @@ BOOL release_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u } if ((ul_ctx = talloc_init()) == NULL) { - DEBUG(0,("release_posix_lock: unable to init talloc context.\n")); + DEBUG(0,("release_posix_lock: unable to init talloc context.\n")); return True; /* Not a fatal error. */ } - if ((ul = (struct unlock_list *)talloc(ul_ctx, sizeof(struct unlock_list))) == NULL) { + if ((ul = (struct lock_list *)talloc(ul_ctx, sizeof(struct lock_list))) == NULL) { DEBUG(0,("release_posix_lock: unable to talloc unlock list.\n")); talloc_destroy(ul_ctx); return True; /* Not a fatal error. */ @@ -1297,7 +1245,7 @@ BOOL release_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u * unlocks are performed. */ - ulist = posix_unlock_list(ul_ctx, ulist, fsp); + ulist = posix_lock_list(ul_ctx, ulist, fsp); /* * Release the POSIX locks on the list of ranges returned. @@ -1307,16 +1255,6 @@ BOOL release_posix_lock(files_struct *fsp, SMB_BIG_UINT u_offset, SMB_BIG_UINT u offset = ulist->start; count = ulist->size; - if(u_count == 0) { - - /* - * This lock must overlap with an existing lock. - * Don't do any POSIX call. - */ - - continue; - } - DEBUG(5,("release_posix_lock: Real unlock: offset = %.0f, count = %.0f\n", (double)offset, (double)count )); -- cgit