git/fsmonitor.c

819 lines
26 KiB
C
Raw Normal View History

#include "git-compat-util.h"
#include "config.h"
#include "dir.h"
#include "environment.h"
#include "ewah/ewok.h"
#include "fsmonitor.h"
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
#include "fsmonitor-ipc.h"
#include "name-hash.h"
#include "run-command.h"
#include "strbuf.h"
#include "trace2.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.
*/
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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;
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
if (fsm_settings__get_mode(r) != FSMONITOR_MODE_HOOK)
return -1;
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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;
}
/*
* Invalidate the FSM bit on this CE. This is like mark_fsmonitor_invalid()
* but we've already handled the untracked-cache, so let's not repeat that
* work. This also lets us have a different trace message so that we can
* see everything that was done as part of the refresh-callback.
*/
static void invalidate_ce_fsm(struct cache_entry *ce)
{
if (ce->ce_flags & CE_FSMONITOR_VALID) {
trace_printf_key(&trace_fsmonitor,
"fsmonitor_refresh_callback INV: '%s'",
ce->name);
ce->ce_flags &= ~CE_FSMONITOR_VALID;
}
}
static size_t handle_path_with_trailing_slash(
struct index_state *istate, const char *name, int pos);
/*
* Use the name-hash to do a case-insensitive cache-entry lookup with
* the pathname and invalidate the cache-entry.
*
* Returns the number of cache-entries that we invalidated.
*/
static size_t handle_using_name_hash_icase(
struct index_state *istate, const char *name)
{
struct cache_entry *ce = NULL;
ce = index_file_exists(istate, name, strlen(name), 1);
if (!ce)
return 0;
/*
* A case-insensitive search in the name-hash using the
* observed pathname found a cache-entry, so the observed path
* is case-incorrect. Invalidate the cache-entry and use the
* correct spelling from the cache-entry to invalidate the
* untracked-cache. Since we now have sparse-directories in
* the index, the observed pathname may represent a regular
* file or a sparse-index directory.
*
* Note that we should not have seen FSEvents for a
* sparse-index directory, but we handle it just in case.
*
* Either way, we know that there are not any cache-entries for
* children inside the cone of the directory, so we don't need to
* do the usual scan.
*/
trace_printf_key(&trace_fsmonitor,
"fsmonitor_refresh_callback MAP: '%s' '%s'",
name, ce->name);
/*
* NEEDSWORK: We used the name-hash to find the correct
* case-spelling of the pathname in the cache-entry[], so
* technically this is a tracked file or a sparse-directory.
* It should not have any entries in the untracked-cache, so
* we should not need to use the case-corrected spelling to
* invalidate the the untracked-cache. So we may not need to
* do this. For now, I'm going to be conservative and always
* do it; we can revisit this later.
*/
untracked_cache_invalidate_trimmed_path(istate, ce->name, 0);
invalidate_ce_fsm(ce);
return 1;
}
/*
* Use the dir-name-hash to find the correct-case spelling of the
* directory. Use the canonical spelling to invalidate all of the
* cache-entries within the matching cone.
*
* Returns the number of cache-entries that we invalidated.
*/
static size_t handle_using_dir_name_hash_icase(
struct index_state *istate, const char *name)
{
struct strbuf canonical_path = STRBUF_INIT;
int pos;
size_t len = strlen(name);
size_t nr_in_cone;
if (name[len - 1] == '/')
len--;
if (!index_dir_find(istate, name, len, &canonical_path))
return 0; /* name is untracked */
if (!memcmp(name, canonical_path.buf, canonical_path.len)) {
strbuf_release(&canonical_path);
/*
* NEEDSWORK: Our caller already tried an exact match
* and failed to find one. They called us to do an
* ICASE match, so we should never get an exact match,
* so we could promote this to a BUG() here if we
* wanted to. It doesn't hurt anything to just return
* 0 and go on because we should never get here. Or we
* could just get rid of the memcmp() and this "if"
* clause completely.
*/
BUG("handle_using_dir_name_hash_icase(%s) did not exact match",
name);
}
trace_printf_key(&trace_fsmonitor,
"fsmonitor_refresh_callback MAP: '%s' '%s'",
name, canonical_path.buf);
/*
* The dir-name-hash only tells us the corrected spelling of
* the prefix. We have to use this canonical path to do a
* lookup in the cache-entry array so that we repeat the
* original search using the case-corrected spelling.
*/
strbuf_addch(&canonical_path, '/');
pos = index_name_pos(istate, canonical_path.buf,
canonical_path.len);
nr_in_cone = handle_path_with_trailing_slash(
istate, canonical_path.buf, pos);
strbuf_release(&canonical_path);
return nr_in_cone;
}
/*
* The daemon sent an observed pathname without a trailing slash.
* (This is the normal case.) We do not know if it is a tracked or
* untracked file, a sparse-directory, or a populated directory (on a
* platform such as Windows where FSEvents are not qualified).
*
* The pathname contains the observed case reported by the FS. We
* do not know it is case-correct or -incorrect.
*
* Assume it is case-correct and try an exact match.
*
* Return the number of cache-entries that we invalidated.
*/
static size_t handle_path_without_trailing_slash(
struct index_state *istate, const char *name, int pos)
{
/*
* Mark the untracked cache dirty for this path (regardless of
* whether or not we find an exact match for it in the index).
* Since the path is unqualified (no trailing slash hint in the
* FSEvent), it may refer to a file or directory. So we should
* not assume one or the other and should always let the untracked
* cache decide what needs to invalidated.
*/
untracked_cache_invalidate_trimmed_path(istate, name, 0);
if (pos >= 0) {
/*
* An exact match on a tracked file. We assume that we
* do not need to scan forward for a sparse-directory
* cache-entry with the same pathname, nor for a cone
* at that directory. (That is, assume no D/F conflicts.)
*/
invalidate_ce_fsm(istate->cache[pos]);
return 1;
} else {
size_t nr_in_cone;
struct strbuf work_path = STRBUF_INIT;
/*
* The negative "pos" gives us the suggested insertion
* point for the pathname (without the trailing slash).
* We need to see if there is a directory with that
* prefix, but there can be lots of pathnames between
* "foo" and "foo/" like "foo-" or "foo-bar", so we
* don't want to do our own scan.
*/
strbuf_add(&work_path, name, strlen(name));
strbuf_addch(&work_path, '/');
pos = index_name_pos(istate, work_path.buf, work_path.len);
nr_in_cone = handle_path_with_trailing_slash(
istate, work_path.buf, pos);
strbuf_release(&work_path);
return nr_in_cone;
}
}
/*
* The daemon can decorate directory events, such as a move or rename,
* by adding a trailing slash to the observed name. Use this to
* explicitly invalidate the entire cone under that directory.
*
* The daemon can only reliably do that if the OS FSEvent contains
* sufficient information in the event.
*
* macOS FSEvents have enough information.
*
* Other platforms may or may not be able to do it (and it might
* depend on the type of event (for example, a daemon could lstat() an
* observed pathname after a rename, but not after a delete)).
*
* If we find an exact match in the index for a path with a trailing
* slash, it means that we matched a sparse-index directory in a
* cone-mode sparse-checkout (since that's the only time we have
* directories in the index). We should never see this in practice
* (because sparse directories should not be present and therefore
* not generating FS events). Either way, we can treat them in the
* same way and just invalidate the cache-entry and the untracked
* cache (and in this case, the forward cache-entry scan won't find
* anything and it doesn't hurt to let it run).
*
* Return the number of cache-entries that we invalidated. We will
* use this later to determine if we need to attempt a second
* case-insensitive search on case-insensitive file systems. That is,
* if the search using the observed-case in the FSEvent yields any
* results, we assume the prefix is case-correct. If there are no
* matches, we still don't know if the observed path is simply
* untracked or case-incorrect.
*/
static size_t handle_path_with_trailing_slash(
struct index_state *istate, const char *name, int pos)
{
int i;
size_t nr_in_cone = 0;
/*
* Mark the untracked cache dirty for this directory path
* (regardless of whether or not we find an exact match for it
* in the index or find it to be proper prefix of one or more
* files in the index), since the FSEvent is hinting that
* there may be changes on or within the directory.
*/
untracked_cache_invalidate_trimmed_path(istate, name, 0);
if (pos < 0)
pos = -pos - 1;
/* Mark all entries for the folder invalid */
for (i = pos; i < istate->cache_nr; i++) {
if (!starts_with(istate->cache[i]->name, name))
break;
invalidate_ce_fsm(istate->cache[i]);
nr_in_cone++;
}
return nr_in_cone;
}
static void fsmonitor_refresh_callback(struct index_state *istate, char *name)
{
int len = strlen(name);
int pos = index_name_pos(istate, name, len);
size_t nr_in_cone;
trace_printf_key(&trace_fsmonitor,
"fsmonitor_refresh_callback '%s' (pos %d)",
name, pos);
if (name[len - 1] == '/')
nr_in_cone = handle_path_with_trailing_slash(istate, name, pos);
else
nr_in_cone = handle_path_without_trailing_slash(istate, name, pos);
/*
* If we did not find an exact match for this pathname or any
* cache-entries with this directory prefix and we're on a
* case-insensitive file system, try again using the name-hash
* and dir-name-hash.
*/
if (!nr_in_cone && ignore_case) {
nr_in_cone = handle_using_name_hash_icase(istate, name);
if (!nr_in_cone)
nr_in_cone = handle_using_dir_name_hash_icase(
istate, name);
}
if (nr_in_cone)
trace_printf_key(&trace_fsmonitor,
"fsmonitor_refresh_callback CNT: %d",
(int)nr_in_cone);
}
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>
2022-03-26 02:03:12 +08:00
/*
* 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)
{
static int warn_once = 0;
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;
treewide: always have a valid "index_state.repo" member When the "repo" member was added to "the_index" in [1] the repo_read_index() was made to populate it, but the unpopulated "the_index" variable didn't get the same treatment. Let's do that in initialize_the_repository() when we set it up, and likewise for all of the current callers initialized an empty "struct index_state". This simplifies code that needs to deal with "the_index" or a custom "struct index_state", we no longer need to second-guess this part of the "index_state" deep in the stack. A recent example of such second-guessing is the "istate->repo ? istate->repo : the_repository" code in [2]. We can now simply use "istate->repo". We're doing this by making use of the INDEX_STATE_INIT() macro (and corresponding function) added in [3], which now have mandatory "repo" arguments. Because we now call index_state_init() in repository.c's initialize_the_repository() we don't need to handle the case where we have a "repo->index" whose "repo" member doesn't match the "repo" we're setting up, i.e. the "Complete the double-reference" code in repo_read_index() being altered here. That logic was originally added in [1], and was working around the lack of what we now have in initialize_the_repository(). For "fsmonitor-settings.c" we can remove the initialization of a NULL "r" argument to "the_repository". This was added back in [4], and was needed at the time for callers that would pass us the "r" from an "istate->repo". Before this change such a change to "fsmonitor-settings.c" would segfault all over the test suite (e.g. in t0002-gitfile.sh). This change has wider eventual implications for "fsmonitor-settings.c". The reason the other lazy loading behavior in it is required (starting with "if (!r->settings.fsmonitor) ..." is because of the previously passed "r" being "NULL". I have other local changes on top of this which move its configuration reading to "prepare_repo_settings()" in "repo-settings.c", as we could now start to rely on it being called for our "r". But let's leave all of that for now, and narrowly remove this particular part of the lazy-loading. 1. 1fd9ae517c4 (repository: add repo reference to index_state, 2021-01-23) 2. ee1f0c242ef (read-cache: add index.skipHash config option, 2023-01-06) 3. 2f6b1eb794e (cache API: add a "INDEX_STATE_INIT" macro/function, add release_index(), 2023-01-12) 4. 1e0ea5c4316 (fsmonitor: config settings are repository-specific, 2022-03-25) Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Acked-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2023-01-17 21:57:00 +08:00
struct repository *r = istate->repo;
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
enum fsmonitor_mode fsm_mode = fsm_settings__get_mode(r);
enum fsmonitor_reason reason = fsm_settings__get_reason(r);
if (!warn_once && reason > FSMONITOR_REASON_OK) {
char *msg = fsm_settings__get_incompatible_msg(r, reason);
warn_once = 1;
warning("%s", msg);
free(msg);
}
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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");
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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;
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
}
assert(fsm_mode == FSMONITOR_MODE_HOOK);
hook_version = fsmonitor_hook_version();
/*
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
* 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);
/*
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
* 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) {
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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) {
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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);
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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.
*/
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>
2022-03-26 02:03:12 +08:00
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.
*/
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>
2022-03-26 02:03:12 +08:00
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;
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>
2022-03-26 02:03:12 +08:00
count++;
}
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>
2022-03-26 02:03:12 +08:00
if (bol < query_result.len) {
fsmonitor_refresh_callback(istate, buf + bol);
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>
2022-03-26 02:03:12 +08:00
count++;
}
/* Now mark the untracked cache for fsmonitor usage */
if (istate->untracked)
istate->untracked->use_fsmonitor = 1;
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>
2022-03-26 02:03:12 +08:00
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;
}
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>
2022-03-26 02:03:12 +08:00
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;
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
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++) {
fsmonitor: never set CE_FSMONITOR_VALID on submodules Never set CE_FSMONITOR_VALID on the cache-entry of submodule directories. During a client command like 'git status', we may need to recurse into each submodule to compute a status summary for the submodule. Since the purpose of the ce_flag is to let Git avoid scanning a cache-entry, setting the flag causes the recursive call to be avoided and we report incorrect (no status) for the submodule. We created an OS watch on the root directory of our working directory and we receive events for everything in the cone under it. When submodules are present inside our working directory, we receive events for both our repo (the super) and any subs within it. Since our index doesn't have any information for items within the submodules, we can't use those events. We could try to truncate the paths of those events back to the submodule boundary and mark the GITLINK as dirty, but that feels expensive since we would have to prefix compare every FS event that we receive against a list of submodule roots. And it still wouldn't be sufficient to correctly report status on the submodule, since we don't have any space in the cache-entry to cache the submodule's status (the 'SCMU' bits in porcelain V2 speak). That is, the CE_FSMONITOR_VALID bit just says that we don't need to scan/inspect it because we already know the answer -- it doesn't say that the item is clean -- and we don't have space in the cache-entry to store those answers. So we should always do the recursive scan. Therefore, we should never set the flag on GITLINK cache-entries. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-05-27 05:47:17 +08:00
if (S_ISGITLINK(istate->cache[i]->ce_mode))
continue;
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;
}
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
if (fsmonitor_enabled)
add_fsmonitor(istate);
fsmonitor: config settings are repository-specific Move fsmonitor config settings to a new and opaque `struct fsmonitor_settings` structure. Add a lazily-loaded pointer to this into `struct repo_settings` Create an `enum fsmonitor_mode` type in `struct fsmonitor_settings` to represent the state of fsmonitor. This lets us represent which, if any, fsmonitor provider (hook or IPC) is enabled. Create `fsm_settings__get_*()` getters to lazily look up fsmonitor- related config settings. Get rid of the `core_fsmonitor` global variable. Move the code to lookup the existing `core.fsmonitor` config value into the fsmonitor settings. Create a hook pathname variable in `struct fsmonitor-settings` and only set it when in hook mode. Extend the definition of `core.fsmonitor` to be either a boolean or a hook pathname. When true, the builtin FSMonitor is used. When false or unset, no FSMonitor (neither builtin nor hook) is used. The existing `core_fsmonitor` global variable was used to store the pathname to the fsmonitor hook *and* it was used as a boolean to see if fsmonitor was enabled. This dual usage and global visibility leads to confusion when we add the IPC-based provider. So lets hide the details in fsmonitor-settings.c and let it decide which provider to use in the case of multiple settings. This avoids cluttering up repo-settings.c with these private details. A future commit in builtin-fsmonitor series will add the ability to disqualify worktrees for various reasons, such as being mounted from a remote volume, where fsmonitor should not be started. Having the config settings hidden in fsmonitor-settings.c allows such worktree restrictions to override the config values used. Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2022-03-26 02:02:46 +08:00
else
remove_fsmonitor(istate);
}