mirror of
https://github.com/git-for-windows/git.git
synced 2025-12-12 04:41:35 -06:00
The threading model for fscache has been to have a single, global cache. This puts requirements on it to be thread safe so that callers like preload-index can call it from multiple threads. This was implemented with a single mutex and completion events which introduces contention between the calling threads. Simplify the threading model by making fscache thread specific. This allows us to remove the global mutex and synchronization events entirely and instead associate a fscache with every thread that requests one. This works well with the current multi-threading which divides the cache entries into blocks with a separate thread processing each block. At the end of each worker thread, if there is a fscache on the primary thread, merge the cached results from the worker into the primary thread cache. This enables us to reuse the cache later especially when scanning for untracked files. In testing, this reduced the time spent in preload_index() by about 25% and also reduced the CPU utilization significantly. On a repo with ~200K files, it reduced overall status times by ~12%. Signed-off-by: Ben Peart <benpeart@microsoft.com>
192 lines
4.5 KiB
C
192 lines
4.5 KiB
C
/*
|
|
* Copyright (C) 2008 Linus Torvalds
|
|
*/
|
|
|
|
#define USE_THE_REPOSITORY_VARIABLE
|
|
#define DISABLE_SIGN_COMPARE_WARNINGS
|
|
|
|
#include "git-compat-util.h"
|
|
#include "pathspec.h"
|
|
#include "dir.h"
|
|
#include "environment.h"
|
|
#include "fsmonitor.h"
|
|
#include "gettext.h"
|
|
#include "parse.h"
|
|
#include "preload-index.h"
|
|
#include "progress.h"
|
|
#include "read-cache.h"
|
|
#include "thread-utils.h"
|
|
#include "repository.h"
|
|
#include "symlinks.h"
|
|
#include "trace2.h"
|
|
|
|
static struct fscache *fscache;
|
|
|
|
/*
|
|
* Mostly randomly chosen maximum thread counts: we
|
|
* cap the parallelism to 20 threads, and we want
|
|
* to have at least 500 lstat's per thread for it to
|
|
* be worth starting a thread.
|
|
*/
|
|
#define MAX_PARALLEL (20)
|
|
#define THREAD_COST (500)
|
|
|
|
struct progress_data {
|
|
unsigned long n;
|
|
struct progress *progress;
|
|
pthread_mutex_t mutex;
|
|
};
|
|
|
|
struct thread_data {
|
|
pthread_t pthread;
|
|
struct index_state *index;
|
|
struct pathspec pathspec;
|
|
struct progress_data *progress;
|
|
int offset, nr;
|
|
int t2_nr_lstat;
|
|
};
|
|
|
|
static void *preload_thread(void *_data)
|
|
{
|
|
int nr, last_nr;
|
|
struct thread_data *p = _data;
|
|
struct index_state *index = p->index;
|
|
struct cache_entry **cep = index->cache + p->offset;
|
|
struct cache_def cache = CACHE_DEF_INIT;
|
|
|
|
nr = p->nr;
|
|
if (nr + p->offset > index->cache_nr)
|
|
nr = index->cache_nr - p->offset;
|
|
last_nr = nr;
|
|
|
|
enable_fscache(nr);
|
|
do {
|
|
struct cache_entry *ce = *cep++;
|
|
struct stat st;
|
|
|
|
if (ce_stage(ce))
|
|
continue;
|
|
if (S_ISGITLINK(ce->ce_mode))
|
|
continue;
|
|
if (ce_uptodate(ce))
|
|
continue;
|
|
if (ce_skip_worktree(ce))
|
|
continue;
|
|
if (ce->ce_flags & CE_FSMONITOR_VALID)
|
|
continue;
|
|
if (p->progress && !(nr & 31)) {
|
|
struct progress_data *pd = p->progress;
|
|
|
|
pthread_mutex_lock(&pd->mutex);
|
|
pd->n += last_nr - nr;
|
|
display_progress(pd->progress, pd->n);
|
|
pthread_mutex_unlock(&pd->mutex);
|
|
last_nr = nr;
|
|
}
|
|
if (!ce_path_match(index, ce, &p->pathspec, NULL))
|
|
continue;
|
|
if (threaded_has_symlink_leading_path(&cache, ce->name, ce_namelen(ce)))
|
|
continue;
|
|
p->t2_nr_lstat++;
|
|
if (lstat(ce->name, &st))
|
|
continue;
|
|
if (ie_match_stat(index, ce, &st, CE_MATCH_RACY_IS_DIRTY|CE_MATCH_IGNORE_FSMONITOR))
|
|
continue;
|
|
ce_mark_uptodate(ce);
|
|
mark_fsmonitor_valid(index, ce);
|
|
} while (--nr > 0);
|
|
if (p->progress) {
|
|
struct progress_data *pd = p->progress;
|
|
|
|
pthread_mutex_lock(&pd->mutex);
|
|
display_progress(pd->progress, pd->n + last_nr);
|
|
pthread_mutex_unlock(&pd->mutex);
|
|
}
|
|
cache_def_clear(&cache);
|
|
merge_fscache(fscache);
|
|
return NULL;
|
|
}
|
|
|
|
void preload_index(struct index_state *index,
|
|
const struct pathspec *pathspec,
|
|
unsigned int refresh_flags)
|
|
{
|
|
int threads, i, work, offset;
|
|
struct thread_data data[MAX_PARALLEL];
|
|
struct progress_data pd;
|
|
int t2_sum_lstat = 0;
|
|
|
|
if (!HAVE_THREADS || !core_preload_index)
|
|
return;
|
|
|
|
fscache = getcache_fscache();
|
|
threads = index->cache_nr / THREAD_COST;
|
|
if ((index->cache_nr > 1) && (threads < 2) && git_env_bool("GIT_TEST_PRELOAD_INDEX", 0))
|
|
threads = 2;
|
|
if (threads < 2)
|
|
return;
|
|
|
|
trace2_region_enter("index", "preload", NULL);
|
|
|
|
trace_performance_enter();
|
|
if (threads > MAX_PARALLEL)
|
|
threads = MAX_PARALLEL;
|
|
offset = 0;
|
|
work = DIV_ROUND_UP(index->cache_nr, threads);
|
|
memset(&data, 0, sizeof(data));
|
|
|
|
memset(&pd, 0, sizeof(pd));
|
|
if (refresh_flags & REFRESH_PROGRESS && isatty(2)) {
|
|
pd.progress = start_delayed_progress(the_repository,
|
|
_("Refreshing index"),
|
|
index->cache_nr);
|
|
pthread_mutex_init(&pd.mutex, NULL);
|
|
}
|
|
|
|
for (i = 0; i < threads; i++) {
|
|
struct thread_data *p = data+i;
|
|
int err;
|
|
|
|
p->index = index;
|
|
if (pathspec)
|
|
copy_pathspec(&p->pathspec, pathspec);
|
|
p->offset = offset;
|
|
p->nr = work;
|
|
if (pd.progress)
|
|
p->progress = &pd;
|
|
offset += work;
|
|
err = pthread_create(&p->pthread, NULL, preload_thread, p);
|
|
|
|
if (err)
|
|
die(_("unable to create threaded lstat: %s"), strerror(err));
|
|
}
|
|
for (i = 0; i < threads; i++) {
|
|
struct thread_data *p = data+i;
|
|
if (pthread_join(p->pthread, NULL))
|
|
die("unable to join threaded lstat");
|
|
t2_sum_lstat += p->t2_nr_lstat;
|
|
}
|
|
stop_progress(&pd.progress);
|
|
|
|
if (pathspec) {
|
|
/* earlier we made deep copies for each thread to work with */
|
|
for (i = 0; i < threads; i++)
|
|
clear_pathspec(&data[i].pathspec);
|
|
}
|
|
|
|
trace_performance_leave("preload index");
|
|
|
|
trace2_data_intmax("index", NULL, "preload/sum_lstat", t2_sum_lstat);
|
|
trace2_region_leave("index", "preload", NULL);
|
|
}
|
|
|
|
int repo_read_index_preload(struct repository *repo,
|
|
const struct pathspec *pathspec,
|
|
unsigned int refresh_flags)
|
|
{
|
|
int retval = repo_read_index(repo);
|
|
|
|
preload_index(repo->index, pathspec, refresh_flags);
|
|
return retval;
|
|
}
|