All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Jeff Hostetler via GitGitGadget" <gitgitgadget@gmail.com>
To: git@vger.kernel.org
Cc: Jeff Hostetler <jeffhost@microsoft.com>,
	Jeff Hostetler <jeffhost@microsoft.com>
Subject: [PATCH 04/11] p7519: add trace logging during perf test
Date: Mon, 01 Feb 2021 22:02:13 +0000	[thread overview]
Message-ID: <8de9985a706675ff41a3549ce00dd42a9dc92c88.1612216941.git.gitgitgadget@gmail.com> (raw)
In-Reply-To: <pull.860.git.1612216941.gitgitgadget@gmail.com>

From: Jeff Hostetler <jeffhost@microsoft.com>

Add optional trace logging to allow us to better compare performance of
various fsmonitor providers and compare results with non-fsmonitor runs.

Currently, this includes Trace2 logging, but may be extended to include
other trace targets, such as GIT_TRACE_FSMONITOR if desired.

Using this logging helped me explain an odd behavior on MacOS where the
kernel was dropping events and causing the hook to Watchman to timeout.

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
---
 t/perf/.gitignore         |  1 +
 t/perf/Makefile           |  4 ++--
 t/perf/p7519-fsmonitor.sh | 32 ++++++++++++++++++++++++++++++++
 3 files changed, 35 insertions(+), 2 deletions(-)

diff --git a/t/perf/.gitignore b/t/perf/.gitignore
index 982eb8e3a94..72f5d0d3148 100644
--- a/t/perf/.gitignore
+++ b/t/perf/.gitignore
@@ -1,3 +1,4 @@
 /build/
 /test-results/
+/test-trace/
 /trash directory*/
diff --git a/t/perf/Makefile b/t/perf/Makefile
index fcb0e8865e4..2465770a782 100644
--- a/t/perf/Makefile
+++ b/t/perf/Makefile
@@ -7,10 +7,10 @@ perf: pre-clean
 	./run
 
 pre-clean:
-	rm -rf test-results
+	rm -rf test-results test-trace
 
 clean:
-	rm -rf build "trash directory".* test-results
+	rm -rf build "trash directory".* test-results test-trace
 
 test-lint:
 	$(MAKE) -C .. test-lint
diff --git a/t/perf/p7519-fsmonitor.sh b/t/perf/p7519-fsmonitor.sh
index 45bbba3c92f..e6724d3604b 100755
--- a/t/perf/p7519-fsmonitor.sh
+++ b/t/perf/p7519-fsmonitor.sh
@@ -32,6 +32,8 @@ test_description="Test core.fsmonitor"
 #
 # GIT_PERF_7519_DROP_CACHE: if set, the OS caches are dropped between tests
 #
+# GIT_PERF_7519_TRACE: if set, enable trace logging during the test.
+#   Trace logs will be grouped by fsmonitor provider.
 
 test_perf_large_repo
 test_checkout_worktree
@@ -70,6 +72,32 @@ then
 	fi
 fi
 
+trace_start() {
+	if test -n "$GIT_PERF_7519_TRACE"
+	then
+		name="$1"
+		TEST_TRACE_DIR="$TEST_OUTPUT_DIRECTORY/test-trace/p7519/"
+		echo "Writing trace logging to $TEST_TRACE_DIR"
+
+		mkdir -p "$TEST_TRACE_DIR"
+
+		# Start Trace2 logging and any other GIT_TRACE_* logs that you
+		# want for this named test case.
+
+		GIT_TRACE2_PERF="$TEST_TRACE_DIR/$name.trace2perf"
+		export GIT_TRACE2_PERF
+
+		>"$GIT_TRACE2_PERF"
+	fi
+}
+
+trace_stop() {
+	if test -n "$GIT_PERF_7519_TRACE"
+	then
+		unset GIT_TRACE2_PERF
+	fi
+}
+
 test_expect_success "one time repo setup" '
 	# set untrackedCache depending on the environment
 	if test -n "$GIT_PERF_7519_UNTRACKED_CACHE"
@@ -203,6 +231,7 @@ test_fsmonitor_suite() {
 # such as Watchman.
 #
 
+trace_start fsmonitor-watchman
 if test -n "$GIT_PERF_7519_FSMONITOR"; then
 	for INTEGRATION_PATH in $GIT_PERF_7519_FSMONITOR; do
 		test_expect_success "setup for fsmonitor $INTEGRATION_PATH" 'setup_for_fsmonitor'
@@ -221,11 +250,13 @@ then
 	# preventing the removal of the trash directory
 	watchman shutdown-server >/dev/null 2>&1
 fi
+trace_stop
 
 #
 # Run a full set of perf tests with the fsmonitor feature disabled.
 #
 
+trace_start fsmonitor-disabled
 test_expect_success "setup without fsmonitor" '
 	unset INTEGRATION_SCRIPT &&
 	git config --unset core.fsmonitor &&
@@ -233,5 +264,6 @@ test_expect_success "setup without fsmonitor" '
 '
 
 test_fsmonitor_suite
+trace_stop
 
 test_done
-- 
gitgitgadget


  parent reply	other threads:[~2021-02-01 22:03 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-01 22:02 [PATCH 00/11] FSMonitor Preliminary Commits Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 01/11] p7519: use xargs -0 rather than -d in test Jeff Hostetler via GitGitGadget
2021-02-01 23:25   ` Junio C Hamano
2021-02-02 18:16     ` Jeff Hostetler
2021-02-02 21:23       ` Junio C Hamano
2021-02-01 22:02 ` [PATCH 02/11] p7519: fix watchman watch-list test on Windows Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 03/11] p7519: move watchman cleanup earlier in the test Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` Jeff Hostetler via GitGitGadget [this message]
2021-02-01 22:02 ` [PATCH 05/11] preload-index: log the number of lstat calls to trace2 Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 06/11] read-cache: " Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 07/11] read-cache: log the number of scanned files " Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 08/11] fsmonitor: log invocation of FSMonitor hook " Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 09/11] fsmonitor: log FSMN token when reading and writing the index Jeff Hostetler via GitGitGadget
2021-02-01 22:02 ` [PATCH 10/11] fsmonitor: allow all entries for a folder to be invalidated Kevin Willford via GitGitGadget
2021-02-01 22:02 ` [PATCH 11/11] fsmonitor: refactor initialization of fsmonitor_last_update token Jeff Hostetler via GitGitGadget
2021-02-03 15:34 ` [PATCH v2 00/11] FSMonitor Preliminary Commits Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 01/11] p7519: do not rely on "xargs -d" in test Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 02/11] p7519: fix watchman watch-list test on Windows Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 03/11] p7519: move watchman cleanup earlier in the test Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 04/11] p7519: add trace logging during perf test Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 05/11] preload-index: log the number of lstat calls to trace2 Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 06/11] read-cache: " Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 07/11] read-cache: log the number of scanned files " Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 08/11] fsmonitor: log invocation of FSMonitor hook " Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 09/11] fsmonitor: log FSMN token when reading and writing the index Jeff Hostetler via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 10/11] fsmonitor: allow all entries for a folder to be invalidated Kevin Willford via GitGitGadget
2021-02-03 15:34   ` [PATCH v2 11/11] fsmonitor: refactor initialization of fsmonitor_last_update token Jeff Hostetler via GitGitGadget
2021-02-16 19:00   ` [PATCH v2 00/11] FSMonitor Preliminary Commits Jeff Hostetler
2021-02-17  1:54     ` Junio C Hamano
2021-02-03 21:19 ` [PATCH " Taylor Blau

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8de9985a706675ff41a3549ce00dd42a9dc92c88.1612216941.git.gitgitgadget@gmail.com \
    --to=gitgitgadget@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=jeffhost@microsoft.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.