1
0
Fork 0
mirror of https://github.com/git/git.git synced 2024-05-08 01:36:08 +02:00
git/fsmonitor.c
Jeff Hostetler 26b9f34ab3 fsmonitor: force update index after large responses
Measure the time taken to apply the FSMonitor query result
to the index and the untracked-cache.

Set the `FSMONITOR_CHANGED` bit on `istate->cache_changed` when
FSMonitor returns a very large repsonse to ensure that the index is
written to disk.

Normally, when the FSMonitor response includes a tracked file, the
index is always updated.  Similarly, the index might be updated when
the response alters the untracked-cache (when enabled).  However, in
cases where neither of those cause the index to be considered changed,
the FSMonitor response is wasted.  Subsequent Git commands will make
requests with the same token and receive the same response.

If that response is very large, performance may suffer.  It would be
more efficient to force update the index now (and the token in the
index extension) in order to reduce the size of the response received
by future commands.

This was observed on Windows after a large checkout.  On Windows, the
kernel emits events for the files that are changed as they are
changed.  However, it might delay events for the containing
directories until the system is more idle (or someone scans the
directory (so it seems)).  The first status following a checkout would
get the list of files.  The subsequent status commands would get the
list of directories as the events trickled out.  But they would never
catch up because the token was not advanced because the index wasn't
updated.

This list of directories caused `wt_status_collect_untracked()` to
unnecessarily spend time actually scanning them during each command.

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-25 16:04:17 -07:00

565 lines
17 KiB
C

#include "cache.h"
#include "config.h"
#include "dir.h"
#include "ewah/ewok.h"
#include "fsmonitor.h"
#include "fsmonitor-ipc.h"
#include "run-command.h"
#include "strbuf.h"
#define INDEX_EXTENSION_VERSION1 (1)
#define INDEX_EXTENSION_VERSION2 (2)
#define HOOK_INTERFACE_VERSION1 (1)
#define HOOK_INTERFACE_VERSION2 (2)
struct trace_key trace_fsmonitor = TRACE_KEY_INIT(FSMONITOR);
static void assert_index_minimum(struct index_state *istate, size_t pos)
{
if (pos > istate->cache_nr)
BUG("fsmonitor_dirty has more entries than the index (%"PRIuMAX" > %u)",
(uintmax_t)pos, istate->cache_nr);
}
static void fsmonitor_ewah_callback(size_t pos, void *is)
{
struct index_state *istate = (struct index_state *)is;
struct cache_entry *ce;
assert_index_minimum(istate, pos + 1);
ce = istate->cache[pos];
ce->ce_flags &= ~CE_FSMONITOR_VALID;
}
static int fsmonitor_hook_version(void)
{
int hook_version;
if (git_config_get_int("core.fsmonitorhookversion", &hook_version))
return -1;
if (hook_version == HOOK_INTERFACE_VERSION1 ||
hook_version == HOOK_INTERFACE_VERSION2)
return hook_version;
warning("Invalid hook version '%i' in core.fsmonitorhookversion. "
"Must be 1 or 2.", hook_version);
return -1;
}
int read_fsmonitor_extension(struct index_state *istate, const void *data,
unsigned long sz)
{
const char *index = data;
uint32_t hdr_version;
uint32_t ewah_size;
struct ewah_bitmap *fsmonitor_dirty;
int ret;
uint64_t timestamp;
struct strbuf last_update = STRBUF_INIT;
if (sz < sizeof(uint32_t) + 1 + sizeof(uint32_t))
return error("corrupt fsmonitor extension (too short)");
hdr_version = get_be32(index);
index += sizeof(uint32_t);
if (hdr_version == INDEX_EXTENSION_VERSION1) {
timestamp = get_be64(index);
strbuf_addf(&last_update, "%"PRIu64"", timestamp);
index += sizeof(uint64_t);
} else if (hdr_version == INDEX_EXTENSION_VERSION2) {
strbuf_addstr(&last_update, index);
index += last_update.len + 1;
} else {
return error("bad fsmonitor version %d", hdr_version);
}
istate->fsmonitor_last_update = strbuf_detach(&last_update, NULL);
ewah_size = get_be32(index);
index += sizeof(uint32_t);
fsmonitor_dirty = ewah_new();
ret = ewah_read_mmap(fsmonitor_dirty, index, ewah_size);
if (ret != ewah_size) {
ewah_free(fsmonitor_dirty);
return error("failed to parse ewah bitmap reading fsmonitor index extension");
}
istate->fsmonitor_dirty = fsmonitor_dirty;
if (!istate->split_index)
assert_index_minimum(istate, istate->fsmonitor_dirty->bit_size);
trace2_data_string("index", NULL, "extension/fsmn/read/token",
istate->fsmonitor_last_update);
trace_printf_key(&trace_fsmonitor,
"read fsmonitor extension successful '%s'",
istate->fsmonitor_last_update);
return 0;
}
void fill_fsmonitor_bitmap(struct index_state *istate)
{
unsigned int i, skipped = 0;
istate->fsmonitor_dirty = ewah_new();
for (i = 0; i < istate->cache_nr; i++) {
if (istate->cache[i]->ce_flags & CE_REMOVE)
skipped++;
else if (!(istate->cache[i]->ce_flags & CE_FSMONITOR_VALID))
ewah_set(istate->fsmonitor_dirty, i - skipped);
}
}
void write_fsmonitor_extension(struct strbuf *sb, struct index_state *istate)
{
uint32_t hdr_version;
uint32_t ewah_start;
uint32_t ewah_size = 0;
int fixup = 0;
if (!istate->split_index)
assert_index_minimum(istate, istate->fsmonitor_dirty->bit_size);
put_be32(&hdr_version, INDEX_EXTENSION_VERSION2);
strbuf_add(sb, &hdr_version, sizeof(uint32_t));
strbuf_addstr(sb, istate->fsmonitor_last_update);
strbuf_addch(sb, 0); /* Want to keep a NUL */
fixup = sb->len;
strbuf_add(sb, &ewah_size, sizeof(uint32_t)); /* we'll fix this up later */
ewah_start = sb->len;
ewah_serialize_strbuf(istate->fsmonitor_dirty, sb);
ewah_free(istate->fsmonitor_dirty);
istate->fsmonitor_dirty = NULL;
/* fix up size field */
put_be32(&ewah_size, sb->len - ewah_start);
memcpy(sb->buf + fixup, &ewah_size, sizeof(uint32_t));
trace2_data_string("index", NULL, "extension/fsmn/write/token",
istate->fsmonitor_last_update);
trace_printf_key(&trace_fsmonitor,
"write fsmonitor extension successful '%s'",
istate->fsmonitor_last_update);
}
/*
* Call the query-fsmonitor hook passing the last update token of the saved results.
*/
static int query_fsmonitor_hook(struct repository *r,
int version,
const char *last_update,
struct strbuf *query_result)
{
struct child_process cp = CHILD_PROCESS_INIT;
int result;
if (fsm_settings__get_mode(r) != FSMONITOR_MODE_HOOK)
return -1;
strvec_push(&cp.args, fsm_settings__get_hook_path(r));
strvec_pushf(&cp.args, "%d", version);
strvec_pushf(&cp.args, "%s", last_update);
cp.use_shell = 1;
cp.dir = get_git_work_tree();
trace2_region_enter("fsm_hook", "query", NULL);
result = capture_command(&cp, query_result, 1024);
if (result)
trace2_data_intmax("fsm_hook", NULL, "query/failed", result);
else
trace2_data_intmax("fsm_hook", NULL, "query/response-length",
query_result->len);
trace2_region_leave("fsm_hook", "query", NULL);
return result;
}
static void fsmonitor_refresh_callback(struct index_state *istate, char *name)
{
int i, len = strlen(name);
if (name[len - 1] == '/') {
/*
* TODO We should binary search to find the first path with
* TODO this directory prefix. Then linearly update entries
* TODO while the prefix matches. Taking care to search without
* TODO the trailing slash -- because '/' sorts after a few
* TODO interesting special chars, like '.' and ' '.
*/
/* Mark all entries for the folder invalid */
for (i = 0; i < istate->cache_nr; i++) {
if (istate->cache[i]->ce_flags & CE_FSMONITOR_VALID &&
starts_with(istate->cache[i]->name, name))
istate->cache[i]->ce_flags &= ~CE_FSMONITOR_VALID;
}
/* Need to remove the / from the path for the untracked cache */
name[len - 1] = '\0';
} else {
int pos = index_name_pos(istate, name, strlen(name));
if (pos >= 0) {
struct cache_entry *ce = istate->cache[pos];
ce->ce_flags &= ~CE_FSMONITOR_VALID;
}
}
/*
* Mark the untracked cache dirty even if it wasn't found in the index
* as it could be a new untracked file.
*/
trace_printf_key(&trace_fsmonitor, "fsmonitor_refresh_callback '%s'", name);
untracked_cache_invalidate_path(istate, name, 0);
}
/*
* The number of pathnames that we need to receive from FSMonitor
* before we force the index to be updated.
*
* Note that any pathname within the set of received paths MAY cause
* cache-entry or istate flag bits to be updated and thus cause the
* index to be updated on disk.
*
* However, the response may contain many paths (such as ignored
* paths) that will not update any flag bits. And thus not force the
* index to be updated. (This is fine and normal.) It also means
* that the token will not be updated in the FSMonitor index
* extension. So the next Git command will find the same token in the
* index, make the same token-relative request, and receive the same
* response (plus any newly changed paths). If this response is large
* (and continues to grow), performance could be impacted.
*
* For example, if the user runs a build and it writes 100K object
* files but doesn't modify any source files, the index would not need
* to be updated. The FSMonitor response (after the build and
* relative to a pre-build token) might be 5MB. Each subsequent Git
* command will receive that same 100K/5MB response until something
* causes the index to be updated. And `refresh_fsmonitor()` will
* have to iterate over those 100K paths each time.
*
* Performance could be improved if we optionally force update the
* index after a very large response and get an updated token into
* the FSMonitor index extension. This should allow subsequent
* commands to get smaller and more current responses.
*
* The value chosen here does not need to be precise. The index
* will be updated automatically the first time the user touches
* a tracked file and causes a command like `git status` to
* update an mtime to be updated and/or set a flag bit.
*/
static int fsmonitor_force_update_threshold = 100;
void refresh_fsmonitor(struct index_state *istate)
{
struct strbuf query_result = STRBUF_INIT;
int query_success = 0, hook_version = -1;
size_t bol = 0; /* beginning of line */
uint64_t last_update;
struct strbuf last_update_token = STRBUF_INIT;
char *buf;
unsigned int i;
int is_trivial = 0;
struct repository *r = istate->repo ? istate->repo : the_repository;
enum fsmonitor_mode fsm_mode = fsm_settings__get_mode(r);
if (fsm_mode <= FSMONITOR_MODE_DISABLED ||
istate->fsmonitor_has_run_once)
return;
istate->fsmonitor_has_run_once = 1;
trace_printf_key(&trace_fsmonitor, "refresh fsmonitor");
if (fsm_mode == FSMONITOR_MODE_IPC) {
query_success = !fsmonitor_ipc__send_query(
istate->fsmonitor_last_update ?
istate->fsmonitor_last_update : "builtin:fake",
&query_result);
if (query_success) {
/*
* The response contains a series of nul terminated
* strings. The first is the new token.
*
* Use `char *buf` as an interlude to trick the CI
* static analysis to let us use `strbuf_addstr()`
* here (and only copy the token) rather than
* `strbuf_addbuf()`.
*/
buf = query_result.buf;
strbuf_addstr(&last_update_token, buf);
bol = last_update_token.len + 1;
is_trivial = query_result.buf[bol] == '/';
if (is_trivial)
trace2_data_intmax("fsm_client", NULL,
"query/trivial-response", 1);
} else {
/*
* The builtin daemon is not available on this
* platform -OR- we failed to get a response.
*
* Generate a fake token (rather than a V1
* timestamp) for the index extension. (If
* they switch back to the hook API, we don't
* want ambiguous state.)
*/
strbuf_addstr(&last_update_token, "builtin:fake");
}
goto apply_results;
}
assert(fsm_mode == FSMONITOR_MODE_HOOK);
hook_version = fsmonitor_hook_version();
/*
* This could be racy so save the date/time now and query_fsmonitor_hook
* should be inclusive to ensure we don't miss potential changes.
*/
last_update = getnanotime();
if (hook_version == HOOK_INTERFACE_VERSION1)
strbuf_addf(&last_update_token, "%"PRIu64"", last_update);
/*
* If we have a last update token, call query_fsmonitor_hook for the set of
* changes since that token, else assume everything is possibly dirty
* and check it all.
*/
if (istate->fsmonitor_last_update) {
if (hook_version == -1 || hook_version == HOOK_INTERFACE_VERSION2) {
query_success = !query_fsmonitor_hook(
r, HOOK_INTERFACE_VERSION2,
istate->fsmonitor_last_update, &query_result);
if (query_success) {
if (hook_version < 0)
hook_version = HOOK_INTERFACE_VERSION2;
/*
* First entry will be the last update token
* Need to use a char * variable because static
* analysis was suggesting to use strbuf_addbuf
* but we don't want to copy the entire strbuf
* only the chars up to the first NUL
*/
buf = query_result.buf;
strbuf_addstr(&last_update_token, buf);
if (!last_update_token.len) {
warning("Empty last update token.");
query_success = 0;
} else {
bol = last_update_token.len + 1;
is_trivial = query_result.buf[bol] == '/';
}
} else if (hook_version < 0) {
hook_version = HOOK_INTERFACE_VERSION1;
if (!last_update_token.len)
strbuf_addf(&last_update_token, "%"PRIu64"", last_update);
}
}
if (hook_version == HOOK_INTERFACE_VERSION1) {
query_success = !query_fsmonitor_hook(
r, HOOK_INTERFACE_VERSION1,
istate->fsmonitor_last_update, &query_result);
if (query_success)
is_trivial = query_result.buf[0] == '/';
}
if (is_trivial)
trace2_data_intmax("fsm_hook", NULL,
"query/trivial-response", 1);
trace_performance_since(last_update, "fsmonitor process '%s'",
fsm_settings__get_hook_path(r));
trace_printf_key(&trace_fsmonitor,
"fsmonitor process '%s' returned %s",
fsm_settings__get_hook_path(r),
query_success ? "success" : "failure");
}
apply_results:
/*
* The response from FSMonitor (excluding the header token) is
* either:
*
* [a] a (possibly empty) list of NUL delimited relative
* pathnames of changed paths. This list can contain
* files and directories. Directories have a trailing
* slash.
*
* [b] a single '/' to indicate the provider had no
* information and that we should consider everything
* invalid. We call this a trivial response.
*/
trace2_region_enter("fsmonitor", "apply_results", istate->repo);
if (query_success && !is_trivial) {
/*
* Mark all pathnames returned by the monitor as dirty.
*
* This updates both the cache-entries and the untracked-cache.
*/
int count = 0;
buf = query_result.buf;
for (i = bol; i < query_result.len; i++) {
if (buf[i] != '\0')
continue;
fsmonitor_refresh_callback(istate, buf + bol);
bol = i + 1;
count++;
}
if (bol < query_result.len) {
fsmonitor_refresh_callback(istate, buf + bol);
count++;
}
/* Now mark the untracked cache for fsmonitor usage */
if (istate->untracked)
istate->untracked->use_fsmonitor = 1;
if (count > fsmonitor_force_update_threshold)
istate->cache_changed |= FSMONITOR_CHANGED;
trace2_data_intmax("fsmonitor", istate->repo, "apply_count",
count);
} else {
/*
* We failed to get a response or received a trivial response,
* so invalidate everything.
*
* We only want to run the post index changed hook if
* we've actually changed entries, so keep track if we
* actually changed entries or not.
*/
int is_cache_changed = 0;
for (i = 0; i < istate->cache_nr; i++) {
if (istate->cache[i]->ce_flags & CE_FSMONITOR_VALID) {
is_cache_changed = 1;
istate->cache[i]->ce_flags &= ~CE_FSMONITOR_VALID;
}
}
/*
* If we're going to check every file, ensure we save
* the results.
*/
if (is_cache_changed)
istate->cache_changed |= FSMONITOR_CHANGED;
if (istate->untracked)
istate->untracked->use_fsmonitor = 0;
}
trace2_region_leave("fsmonitor", "apply_results", istate->repo);
strbuf_release(&query_result);
/* Now that we've updated istate, save the last_update_token */
FREE_AND_NULL(istate->fsmonitor_last_update);
istate->fsmonitor_last_update = strbuf_detach(&last_update_token, NULL);
}
/*
* The caller wants to turn on FSMonitor. And when the caller writes
* the index to disk, a FSMonitor extension should be included. This
* requires that `istate->fsmonitor_last_update` not be NULL. But we
* have not actually talked to a FSMonitor process yet, so we don't
* have an initial value for this field.
*
* For a protocol V1 FSMonitor process, this field is a formatted
* "nanoseconds since epoch" field. However, for a protocol V2
* FSMonitor process, this field is an opaque token.
*
* Historically, `add_fsmonitor()` has initialized this field to the
* current time for protocol V1 processes. There are lots of race
* conditions here, but that code has shipped...
*
* The only true solution is to use a V2 FSMonitor and get a current
* or default token value (that it understands), but we cannot do that
* until we have actually talked to an instance of the FSMonitor process
* (but the protocol requires that we send a token first...).
*
* For simplicity, just initialize like we have a V1 process and require
* that V2 processes adapt.
*/
static void initialize_fsmonitor_last_update(struct index_state *istate)
{
struct strbuf last_update = STRBUF_INIT;
strbuf_addf(&last_update, "%"PRIu64"", getnanotime());
istate->fsmonitor_last_update = strbuf_detach(&last_update, NULL);
}
void add_fsmonitor(struct index_state *istate)
{
unsigned int i;
if (!istate->fsmonitor_last_update) {
trace_printf_key(&trace_fsmonitor, "add fsmonitor");
istate->cache_changed |= FSMONITOR_CHANGED;
initialize_fsmonitor_last_update(istate);
/* reset the fsmonitor state */
for (i = 0; i < istate->cache_nr; i++)
istate->cache[i]->ce_flags &= ~CE_FSMONITOR_VALID;
/* reset the untracked cache */
if (istate->untracked) {
add_untracked_cache(istate);
istate->untracked->use_fsmonitor = 1;
}
/* Update the fsmonitor state */
refresh_fsmonitor(istate);
}
}
void remove_fsmonitor(struct index_state *istate)
{
if (istate->fsmonitor_last_update) {
trace_printf_key(&trace_fsmonitor, "remove fsmonitor");
istate->cache_changed |= FSMONITOR_CHANGED;
FREE_AND_NULL(istate->fsmonitor_last_update);
}
}
void tweak_fsmonitor(struct index_state *istate)
{
unsigned int i;
int fsmonitor_enabled = (fsm_settings__get_mode(istate->repo)
> FSMONITOR_MODE_DISABLED);
if (istate->fsmonitor_dirty) {
if (fsmonitor_enabled) {
/* Mark all entries valid */
for (i = 0; i < istate->cache_nr; i++) {
istate->cache[i]->ce_flags |= CE_FSMONITOR_VALID;
}
/* Mark all previously saved entries as dirty */
assert_index_minimum(istate, istate->fsmonitor_dirty->bit_size);
ewah_each_bit(istate->fsmonitor_dirty, fsmonitor_ewah_callback, istate);
refresh_fsmonitor(istate);
}
ewah_free(istate->fsmonitor_dirty);
istate->fsmonitor_dirty = NULL;
}
if (fsmonitor_enabled)
add_fsmonitor(istate);
else
remove_fsmonitor(istate);
}