trace2: make SIDs more unique

Update SID component construction to use the current UTC datetime
and a portion of the SHA1 of the hostname.

Use an simplified date/time format to make it easier to use the
SID component as a logfile filename.

Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

3 files changed, 65 insertions, 4 deletions

diff --git a/trace2/tr2_sid.c b/trace2/tr2_sid.c
index 984524a43c..5047095478 100644
--- a/trace2/tr2_sid.c
+++ b/trace2/tr2_sid.c
@@ -1,4 +1,5 @@
 #include "cache.h"
+#include "trace2/tr2_tbuf.h"
 #include "trace2/tr2_sid.h"
 
 #define TR2_ENVVAR_PARENT_SID "GIT_TR2_PARENT_SID"
@@ -7,6 +8,53 @@ static struct strbuf tr2sid_buf = STRBUF_INIT;
 static int tr2sid_nr_git_parents;
 
 /*
+ * Compute the final component of the SID representing the current process.
+ * This should uniquely identify the process and be a valid filename (to
+ * allow writing trace2 data to per-process files).  It should also be fixed
+ * length for possible use as a database key.
+ *
+ * "<yyyymmdd>T<hhmmss>.<fraction>Z-<host>-<process>"
+ *
+ * where <host> is a 9 character string:
+ *    "H<first_8_chars_of_sha1_of_hostname>"
+ *    "Localhost" when no hostname.
+ *
+ * where <process> is a 9 character string containing the least signifcant
+ * 32 bits in the process-id.
+ *    "P<pid>"
+ * (This is an abribrary choice.  On most systems pid_t is a 32 bit value,
+ * so limit doesn't matter.  On larger systems, a truncated value is fine
+ * for our purposes here.)
+ */
+static void tr2_sid_append_my_sid_component(void)
+{
+	const struct git_hash_algo *algo = &hash_algos[GIT_HASH_SHA1];
+	struct tr2_tbuf tb_now;
+	git_hash_ctx ctx;
+	pid_t pid = getpid();
+	unsigned char hash[GIT_MAX_RAWSZ + 1];
+	char hex[GIT_MAX_HEXSZ + 1];
+	char hostname[HOST_NAME_MAX + 1];
+
+	tr2_tbuf_utc_datetime(&tb_now);
+	strbuf_addstr(&tr2sid_buf, tb_now.buf);
+
+	strbuf_addch(&tr2sid_buf, '-');
+	if (xgethostname(hostname, sizeof(hostname)))
+		strbuf_add(&tr2sid_buf, "Localhost", 9);
+	else {
+		algo->init_fn(&ctx);
+		algo->update_fn(&ctx, hostname, strlen(hostname));
+		algo->final_fn(hash, &ctx);
+		hash_to_hex_algop_r(hex, hash, algo);
+		strbuf_addch(&tr2sid_buf, 'H');
+		strbuf_add(&tr2sid_buf, hex, 8);
+	}
+
+	strbuf_addf(&tr2sid_buf, "-P%08"PRIx32, (uint32_t)pid);
+}
+
+/*
  * Compute a "unique" session id (SID) for the current process.  This allows
  * all events from this process to have a single label (much like a PID).
  *
@@ -20,7 +68,6 @@ static int tr2sid_nr_git_parents;
  */
 static void tr2_sid_compute(void)
 {
-	uint64_t us_now;
 	const char *parent_sid;
 
 	if (tr2sid_buf.len)
@@ -38,9 +85,7 @@ static void tr2_sid_compute(void)
 		tr2sid_nr_git_parents++;
 	}
 
-	us_now = getnanotime() / 1000;
-	strbuf_addf(&tr2sid_buf, "%" PRIuMAX "-%" PRIdMAX, (uintmax_t)us_now,
-		    (intmax_t)getpid());
+	tr2_sid_append_my_sid_component();
 
 	setenv(TR2_ENVVAR_PARENT_SID, tr2sid_buf.buf, 1);
 }
diff --git a/trace2/tr2_tbuf.c b/trace2/tr2_tbuf.c
index eb1b240d8a..2498482d9a 100644
--- a/trace2/tr2_tbuf.c
+++ b/trace2/tr2_tbuf.c
@@ -30,3 +30,18 @@ void tr2_tbuf_utc_datetime_extended(struct tr2_tbuf *tb)
 		  tm.tm_mon + 1, tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec,
 		  (long)tv.tv_usec);
 }
+
+void tr2_tbuf_utc_datetime(struct tr2_tbuf *tb)
+{
+	struct timeval tv;
+	struct tm tm;
+	time_t secs;
+
+	gettimeofday(&tv, NULL);
+	secs = tv.tv_sec;
+	gmtime_r(&secs, &tm);
+
+	xsnprintf(tb->buf, sizeof(tb->buf), "%4d%02d%02dT%02d%02d%02d.%06ldZ",
+		  tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, tm.tm_hour,
+		  tm.tm_min, tm.tm_sec, (long)tv.tv_usec);
+}
diff --git a/trace2/tr2_tbuf.h b/trace2/tr2_tbuf.h
index 2a1d173ff2..fa853d8f42 100644
--- a/trace2/tr2_tbuf.h
+++ b/trace2/tr2_tbuf.h
@@ -19,5 +19,6 @@ void tr2_tbuf_local_time(struct tr2_tbuf *tb);
  * Fill buffer with formatted UTC datatime string.
  */
 void tr2_tbuf_utc_datetime_extended(struct tr2_tbuf *tb);
+void tr2_tbuf_utc_datetime(struct tr2_tbuf *tb);
 
 #endif /* TR2_TBUF_H */