mod_mem_cache.c revision d7d551e53cdfb3288eb651447d7209599c40d17e
/* ====================================================================
* The Apache Software License, Version 1.1
*
* Copyright (c) 2000-2001 The Apache Software Foundation. All rights
* reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
*
* 3. The end-user documentation included with the redistribution,
* if any, must include the following acknowledgment:
* "This product includes software developed by the
* Apache Software Foundation (http://www.apache.org/)."
* Alternately, this acknowledgment may appear in the software itself,
* if and wherever such third-party acknowledgments normally appear.
*
* 4. The names "Apache" and "Apache Software Foundation" must
* not be used to endorse or promote products derived from this
* software without prior written permission. For written
* permission, please contact apache@apache.org.
*
* 5. Products derived from this software may not be called "Apache",
* nor may "Apache" appear in their name, without prior written
* permission of the Apache Software Foundation.
*
* THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
* ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
* USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
* ====================================================================
*
* This software consists of voluntary contributions made by many
* individuals on behalf of the Apache Software Foundation. For more
* information on the Apache Software Foundation, please see
* <http://www.apache.org/>.
*
* Portions of this software are based upon public domain software
* originally written at the National Center for Supercomputing Applications,
* University of Illinois, Urbana-Champaign.
*/
#define CORE_PRIVATE
#include "mod_cache.h"
#include "ap_mpm.h"
#include "apr_thread_mutex.h"
#if !APR_HAS_THREADS
#error This module does not currently compile unless you have a thread-capable APR. Sorry!
#endif
static apr_size_t max_cache_entry_size = 5000;
module AP_MODULE_DECLARE_DATA mem_cache_module;
/*
* XXX
* This cache uses apr_hash functions which leak storage when something is removed
* from the cache. This can be fixed in the apr_hash functions by making them use
* malloc/free rather than pools to manage their storage requirements.
*/
/*
* XXX Introduce a type field that identifies whether the cache obj
* references malloc'ed or mmap storage or a file descriptor
*/
typedef enum {
CACHE_TYPE_FILE = 1,
CACHE_TYPE_HEAP,
CACHE_TYPE_MMAP
} cache_type_e;
typedef struct {
char* hdr;
char* val;
} cache_header_tbl_t;
typedef struct mem_cache_object {
cache_type_e type;
apr_ssize_t num_header_out;
apr_ssize_t num_subprocess_env;
apr_ssize_t num_notes;
cache_header_tbl_t *header_out;
cache_header_tbl_t *subprocess_env;
cache_header_tbl_t *notes;
apr_size_t cleanup;
apr_size_t refcount;
apr_size_t m_len;
void *m;
} mem_cache_object_t;
typedef struct {
apr_thread_mutex_t *lock;
apr_hash_t *cacheht;
int space;
apr_time_t maxexpire;
apr_time_t defaultexpire;
} mem_cache_conf;
static mem_cache_conf *sconf;
#define DEFAULT_CACHE_SPACE 100*1024
#define CACHEFILE_LEN 20
/* Forward declarations */
static int remove_entity(cache_handle_t *h);
static int write_headers(cache_handle_t *h, request_rec *r, cache_info *i);
static int write_body(cache_handle_t *h, request_rec *r, apr_bucket_brigade *b);
static int read_headers(cache_handle_t *h, request_rec *r);
static int read_body(cache_handle_t *h, apr_pool_t *p, apr_bucket_brigade *bb);
static void cleanup_cache_object(cache_object_t *obj)
{
mem_cache_object_t *mobj = obj->vobj;
/* Cleanup the cache_object_t */
if (obj->key) {
free(obj->key);
}
free(obj);
/* Cleanup the mem_cache_object_t */
if (!mobj) {
return;
}
if (mobj->m) {
free(mobj->m);
}
/* XXX should freeing of the info be done here or in cache_storage ?
if (obj->info.content_type ) {
free((char*)obj->info.content_type );
obj->info.content_type =NULL;
}
if (obj->info.filename ) {
free( (char*)obj->info.filename );
obj->info.filename= NULL;
}
*/
/* XXX Cleanup the headers */
if (mobj->num_header_out) {
}
free(mobj);
}
static apr_status_t decrement_refcount(void *arg)
{
cache_object_t *obj = (cache_object_t *) arg;
mem_cache_object_t *mobj = (mem_cache_object_t*) obj->vobj;
if (sconf->lock) {
apr_thread_mutex_lock(sconf->lock);
}
mobj->refcount--;
/* If the object is marked for cleanup and the refcount
* has dropped to zero, cleanup the object
*/
if ((mobj->cleanup) && (!mobj->refcount)) {
cleanup_cache_object(obj);
}
if (sconf->lock) {
apr_thread_mutex_unlock(sconf->lock);
}
return APR_SUCCESS;
}
static apr_status_t cleanup_cache_mem(void *sconfv)
{
cache_object_t *obj;
mem_cache_object_t *mobj;
apr_hash_index_t *hi;
mem_cache_conf *co = (mem_cache_conf*) sconfv;
if (!co) {
return APR_SUCCESS;
}
/* Iterate over the frag hash table and clean up each entry */
if (sconf->lock) {
apr_thread_mutex_lock(sconf->lock);
}
for (hi = apr_hash_first(NULL, co->cacheht); hi; hi=apr_hash_next(hi)) {
apr_hash_this(hi, NULL, NULL, (void **)&obj);
if (obj) {
mobj = (mem_cache_object_t *) obj->vobj;
if (mobj->refcount) {
mobj->cleanup = 1;
}
else {
cleanup_cache_object(obj);
}
}
}
if (sconf->lock) {
apr_thread_mutex_unlock(sconf->lock);
}
return APR_SUCCESS;
}
static void *create_cache_config(apr_pool_t *p, server_rec *s)
{
int threaded_mpm;
sconf = apr_pcalloc(p, sizeof(mem_cache_conf));
sconf->space = DEFAULT_CACHE_SPACE;
ap_mpm_query(AP_MPMQ_IS_THREADED, &threaded_mpm);
if (threaded_mpm) {
apr_thread_mutex_create(&sconf->lock, APR_THREAD_MUTEX_DEFAULT, p);
}
sconf->cacheht = apr_hash_make(p);
apr_pool_cleanup_register(p, NULL, cleanup_cache_mem, apr_pool_cleanup_null);
return sconf;
}
static int create_entity(cache_handle_t *h, request_rec *r,
const char *type,
const char *key,
apr_size_t len)
{
cache_object_t *obj, *tmp_obj;
mem_cache_object_t *mobj;
if (strcasecmp(type, "mem")) {
return DECLINED;
}
/* XXX Check len to see if it is withing acceptable bounds
* max cache check should be configurable variable.
*/
if (len < 0 || len > max_cache_entry_size) {
return DECLINED;
}
/* XXX Check total cache size and number of entries. Are they within the
* configured limits? If not, kick off garbage collection thread.
*/
/* Allocate and initialize cache_object_t */
obj = malloc(sizeof(*obj));
if (!obj) {
return DECLINED;
}
memset(obj,'\0', sizeof(*obj));
obj->key = malloc(strlen(key) + 1);
if (!obj->key) {
free(obj);
return DECLINED;
}
strncpy(obj->key, key, strlen(key) + 1);
obj->info.len = len;
obj->complete = 0; /* Cache object is not complete */
/* Allocate and init mem_cache_object_t */
mobj = malloc(sizeof(*mobj));
if (!mobj) {
/* XXX: Cleanup */
cleanup_cache_object(obj);
}
memset(mobj,'\0', sizeof(*mobj));
obj->vobj = mobj; /* Reference the mem_cache_object_t out of
* cache_object_t
*/
mobj->refcount = 0;
mobj->cleanup = 0;
mobj->m_len = len; /* Duplicates info in cache_object_t info */
/* Place the cache_object_t into the hash table
* XXX Need a way to insert into the cache w/o such coarse grained locking
* XXX Need to enable caching multiple cache objects (representing different
* views of the same content) under a single search key
*/
if (sconf->lock) {
apr_thread_mutex_lock(sconf->lock);
}
tmp_obj = (cache_object_t *) apr_hash_get(sconf->cacheht,
key,
APR_HASH_KEY_STRING);
if (!tmp_obj) {
apr_hash_set(sconf->cacheht, obj->key, strlen(obj->key), obj);
}
if (sconf->lock) {
apr_thread_mutex_unlock(sconf->lock);
}
if (tmp_obj) {
/* This thread collided with another thread loading the same object
* into the cache at the same time. Defer to the other thread which
* is further along.
*/
cleanup_cache_object(obj);
return DECLINED;
}
/* Populate the cache handle */
h->cache_obj = obj;
h->read_body = &read_body;
h->read_headers = &read_headers;
h->write_body = &write_body;
h->write_headers = &write_headers;
h->remove_entity = &remove_entity;
return OK;
}
static int open_entity(cache_handle_t *h, request_rec *r, const char *type, const char *key)
{
cache_object_t *obj;
/* Look up entity keyed to 'url' */
if (strcasecmp(type, "mem")) {
return DECLINED;
}
if (sconf->lock) {
apr_thread_mutex_lock(sconf->lock);
}
obj = (cache_object_t *) apr_hash_get(sconf->cacheht, key,
APR_HASH_KEY_STRING);
if (obj) {
mem_cache_object_t *mobj = (mem_cache_object_t *) obj->vobj;
mobj->refcount++;
apr_pool_cleanup_register(r->pool, obj, decrement_refcount, apr_pool_cleanup_null);
}
if (sconf->lock) {
apr_thread_mutex_unlock(sconf->lock);
}
if (!obj || !(obj->complete)) {
return DECLINED;
}
/* Initialize the cache_handle */
h->read_body = &read_body;
h->read_headers = &read_headers;
h->write_body = &write_body;
h->write_headers = &write_headers;
h->remove_entity = &remove_entity;
h->cache_obj = obj;
return OK;
}
static int remove_entity(cache_handle_t *h)
{
cache_object_t *obj = h->cache_obj;
if (sconf->lock) {
apr_thread_mutex_lock(sconf->lock);
}
obj = (cache_object_t *) apr_hash_get(sconf->cacheht, obj->key,
APR_HASH_KEY_STRING);
if (obj) {
mem_cache_object_t *mobj = (mem_cache_object_t *) obj->vobj;
apr_hash_set(sconf->cacheht, obj->key, strlen(obj->key), NULL);
if (mobj->refcount) {
mobj->cleanup = 1;
}
else {
cleanup_cache_object(obj);
}
h->cache_obj = NULL;
}
if (sconf->lock) {
apr_thread_mutex_unlock(sconf->lock);
}
return OK;
}
static int serialize_table(cache_header_tbl_t **obj,
apr_ssize_t *nelts,
apr_table_t *table)
{
apr_table_entry_t *elts = (apr_table_entry_t *) table->a.elts;
apr_ssize_t i;
apr_size_t len = 0;
apr_size_t idx = 0;
char *buf;
*nelts = table->a.nelts;
if (*nelts ==0 ) {
*obj=NULL;
return OK;
}
*obj = malloc(sizeof(cache_header_tbl_t) * table->a.nelts);
if (NULL == *obj) {
/* cleanup_cache_obj(h->cache_obj); */
return DECLINED;
}
for (i = 0; i < table->a.nelts; ++i) {
len += strlen(elts[i].key);
len += strlen(elts[i].val);
len += 2; /* Extra space for NULL string terminator for key and val */
}
/* Transfer the headers into a contiguous memory block */
buf = malloc(len);
if (!buf) {
free(obj);
*obj = NULL;
/* cleanup_cache_obj(h->cache_obj); */
return DECLINED;
}
for (i = 0; i < *nelts; ++i) {
(*obj)[i].hdr = &buf[idx];
len = strlen(elts[i].key) + 1; /* Include NULL terminator */
strncpy(&buf[idx], elts[i].key, len);
idx+=len;
(*obj)[i].val = &buf[idx];
len = strlen(elts[i].val) + 1;
strncpy(&buf[idx], elts[i].val, len);
idx+=len;
}
return OK;
}
static int unserialize_table( cache_header_tbl_t *ctbl,
int num_headers,
apr_table_t *t )
{
int i;
for (i = 0; i < num_headers; ++i) {
apr_table_setn(t, ctbl[i].hdr, ctbl[i].val);
}
return OK;
}
/* Define request processing hook handlers */
static int remove_url(const char *type, const char *key)
{
cache_object_t *obj;
if (strcasecmp(type, "mem")) {
return DECLINED;
}
/* WIBNIF
* apr_hash_set(..,..,..,NULL) returned pointer to the object just removed.
* That way, we could free the object w/o doing another call to an
* apr_hash function.
*/
/* First, find the object in the cache */
if (sconf->lock) {
apr_thread_mutex_lock(sconf->lock);
}
obj = (cache_object_t *) apr_hash_get(sconf->cacheht, key,
APR_HASH_KEY_STRING);
if (obj) {
mem_cache_object_t *mobj = (mem_cache_object_t *) obj->vobj;
apr_hash_set(sconf->cacheht, key, APR_HASH_KEY_STRING, NULL);
if (mobj->refcount) {
mobj->cleanup = 1;
}
else {
cleanup_cache_object(obj);
}
}
if (sconf->lock) {
apr_thread_mutex_unlock(sconf->lock);
}
if (!obj) {
return DECLINED;
}
return OK;
}
static int read_headers(cache_handle_t *h, request_rec *r)
{
int rc;
mem_cache_object_t *mobj = (mem_cache_object_t*) h->cache_obj->vobj;
r->headers_out = apr_table_make(r->pool,mobj->num_header_out);
r->subprocess_env = apr_table_make(r->pool, mobj->num_subprocess_env);
r->notes = apr_table_make(r->pool, mobj->num_notes);
rc = unserialize_table( mobj->header_out,
mobj->num_header_out,
r->headers_out);
rc = unserialize_table( mobj->subprocess_env,
mobj->num_subprocess_env,
r->subprocess_env);
rc = unserialize_table( mobj->notes,
mobj->num_notes,
r->notes);
return rc;
}
static int read_body(cache_handle_t *h, apr_pool_t *p, apr_bucket_brigade *bb)
{
apr_bucket *b;
mem_cache_object_t *mobj = (mem_cache_object_t*) h->cache_obj->vobj;
b = apr_bucket_immortal_create(mobj->m, mobj->m_len);
APR_BRIGADE_INSERT_TAIL(bb, b);
b = apr_bucket_eos_create();
APR_BRIGADE_INSERT_TAIL(bb, b);
return OK;
}
static int write_headers(cache_handle_t *h, request_rec *r, cache_info *info)
{
cache_object_t *obj = h->cache_obj;
mem_cache_object_t *mobj = (mem_cache_object_t*) h->cache_obj->vobj;
int rc;
/* Precompute how much storage we need to hold the headers */
rc = serialize_table(&mobj->header_out,
&mobj->num_header_out,
r->headers_out);
if (rc != OK ) {
return rc;
}
rc = serialize_table(&mobj->subprocess_env,
&mobj->num_subprocess_env,
r->subprocess_env );
if (rc != OK ) {
return rc;
}
rc = serialize_table(&mobj->notes, &mobj->num_notes, r->notes);
if (rc != OK ) {
return rc;
}
/* Init the info struct */
if (info->date) {
obj->info.date = info->date;
}
if (info->lastmod) {
obj->info.lastmod = info->lastmod;
}
if (info->expire) {
obj->info.expire = info->expire;
}
if (info->content_type) {
obj->info.content_type = (char*) malloc(strlen(info->content_type) + 1);
if (!obj->info.content_type) {
/* cleanup the object? */
return DECLINED;
}
strcpy((char*) obj->info.content_type, info->content_type);
}
if ( info->filename) {
obj->info.filename = (char*) malloc(strlen(info->filename )+1);
if (!obj->info.filename ) {
free( (char*)obj->info.content_type );
obj->info.content_type =NULL;
return DECLINED;
}
strcpy((char*) obj->info.filename, info->filename );
}
return OK;
}
static int write_body(cache_handle_t *h, request_rec *r, apr_bucket_brigade *b)
{
apr_status_t rv;
mem_cache_object_t *mobj = (mem_cache_object_t*) h->cache_obj->vobj;
apr_read_type_e eblock = APR_BLOCK_READ;
apr_bucket *e;
char *cur;
/* XXX mmap, malloc or file?
* Enable this decision to be configured....
* XXX cache buckets...
*/
if (mobj->m == NULL) {
mobj->m = malloc(mobj->m_len);
if (mobj->m == NULL) {
/* Cleanup cache entry and return */
}
mobj->type = CACHE_TYPE_HEAP;
h->cache_obj->count = 0;
}
cur = (char*) mobj->m + h->cache_obj->count;
/* Iterate accross the brigade and populate the cache storage */
APR_BRIGADE_FOREACH(e, b) {
const char *s;
apr_size_t len;
if (APR_BUCKET_IS_EOS(e)) {
h->cache_obj->complete = 1;
break;
}
rv = apr_bucket_read(e, &s, &len, eblock);
if (rv != APR_SUCCESS) {
/* Big problem! Cleanup cache entry and return */
}
/* XXX Check for overflow */
if (len ) {
memcpy(cur, s, len);
cur+=len;
h->cache_obj->count+=len;
}
/* This should not happen, but if it does, we are in BIG trouble
* cause we just stomped all over the heap.
*/
AP_DEBUG_ASSERT(h->cache_obj->count > mobj->m_len);
}
return OK;
}
static const char
*set_cache_size(cmd_parms *parms, void *in_struct_ptr, const char *arg)
{
int val;
if (sscanf(arg, "%d", &val) != 1) {
return "CacheSize value must be an integer (kBytes)";
}
sconf->space = val;
return NULL;
}
static const char
*set_cache_entry_size(cmd_parms *parms, void *in_struct_ptr, const char *arg)
{
int val;
if (sscanf(arg, "%d", &val) != 1) {
return "CacheSize value must be an integer (bytes)";
}
max_cache_entry_size = val;
return NULL;
}
static const command_rec cache_cmds[] =
{
/* XXX
* What config directives does this module need?
* Should this module manage expire policy for its entries?
* Certainly cache limits like max number of entries,
* max entry size, and max size of the cache should
* be managed by this module.
*/
AP_INIT_TAKE1("CacheMemSize", set_cache_size, NULL, RSRC_CONF,
"The maximum space used by the cache in Kb"),
AP_INIT_TAKE1("CacheMemEntrySize", set_cache_entry_size, NULL, RSRC_CONF,
"The maximum size (in bytes) that a entry can take"),
{NULL}
};
static void register_hooks(apr_pool_t *p)
{
/* cache initializer */
/* cache_hook_cache_init(cache_init, NULL, NULL, AP_HOOK_FIRST); */
cache_hook_create_entity(create_entity, NULL, NULL, APR_HOOK_MIDDLE);
cache_hook_open_entity(open_entity, NULL, NULL, APR_HOOK_MIDDLE);
cache_hook_remove_url(remove_url, NULL, NULL, APR_HOOK_MIDDLE);
}
module AP_MODULE_DECLARE_DATA mem_cache_module =
{
STANDARD20_MODULE_STUFF,
NULL, /* create per-directory config structure */
NULL, /* merge per-directory config structures */
create_cache_config, /* create per-server config structure */
NULL, /* merge per-server config structures */
cache_cmds, /* command apr_table_t */
register_hooks
};