pmc(3)/hwpmc(4): update supported Intel processors to rely fully on the

vendor provided pmu-events tables and sundry cleanups.

The vendor pmu-events tables provide counter descriptions, default
sample rates, event, umask, and flag values for all the counter
configuration permutations. Using this gives us:

- much simpler kernel code for the MD component
- helpful long and short event descriptions
- simpler user code
- sample rates that won't overload the system

Update man page with newer sample types and remove unused sample type.

Squashed commit of the following:

commit 4459d43eff815bec08ccc5533dbe5de846f03128
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Sat May 26 00:06:31 2018 -0700

    libpmc: fix pmu function signatures for non amd64

commit a2cb8bbc586c65d41f9b291430a2261ec67b59fe
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 22:38:11 2018 -0700

    pmcstat: fix indentation of usage

commit f686954b15ff56a833ac80404898977cb80a265b
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 22:19:49 2018 -0700

    pmclog(3): add callchain and pmcallocatedyn, remove pcsample

commit 73e13a0d2e9498c81c150d14d022050cee7511bb
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 22:19:00 2018 -0700

    pmclog.h: GC pcsample field

commit 3e93ffd65da641fa657539dad3c48e281f8b5798
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 22:05:57 2018 -0700

    hwpmc: make Intel core CPUs use external event tables

commit 634f5fae1e1644ac324003136c66cd9c619d1c93
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 22:00:06 2018 -0700

    pmclog: update log record types, bump PMC_MAJOR
    - explicitly make log record types a multiple of 8 bytes
    - hook in pmu event types for pmc_allocate records
    - remove references to no longer PCSAMPLE record

commit 83d84fcd2d65bdf6ddcb2e155a22f0cfa2a9c225
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 21:52:10 2018 -0700

    libpmc: add support for having vendor table driven pmc_allocate

commit 9e6ad63c40c2fce8404847ace5078ca6cb33a736
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 19:11:33 2018 -0700

    hwpmc_core: add accessors for EVSEL & UMASK, make IAP_UMASK useful to user

commit 859dceb93daa6419a48c794db99b6758e5b041c9
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 19:09:45 2018 -0700

    pmcstat: update usage and man page as well as make -L consistent with pmccontrol

commit 79c7d8597e28c2eb13f5f9113e65ec2792ca57b1
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 18:07:03 2018 -0700

    pmu_util: add support for all current intel event keywords

commit d8089c7f6a6c8527f38324252b1ffb47004694c6
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 17:45:00 2018 -0700

    add description for new arguments

commit 058336740bab53c62ec88a3a026ea848cf3878c6
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 17:38:15 2018 -0700

    libpmc: move pmu_events table and pmu_utils out of libpmcstat so that they can be used by pmc_allocate

commit 049b66b382e2f833c3f47bc8df9e750cb265709f
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 16:12:41 2018 -0700

    pmcstat: hook pmu_events counter description utility routines in

commit f5e01e7b37a691dc045e1aa16b3ebdd162515de8
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 16:11:59 2018 -0700

    pmu_events: add utility routines for listing counters and their descriptions

commit cba4d4f8907f772279f86f18f915e0d74d33ac56
Author: Matt Macy <mmacy@mattmacy.io>
Date:   Fri May 25 16:09:50 2018 -0700

    pmu-events: expand out skylake regex to simplify string matches

3 files changed, 62 insertions, 29 deletions

diff --git a/usr.sbin/pmcstat/pmcstat.8 b/usr.sbin/pmcstat/pmcstat.8
index fa68a39a1483..f6ff4fbb32cb 100644
--- a/usr.sbin/pmcstat/pmcstat.8
+++ b/usr.sbin/pmcstat/pmcstat.8
@@ -39,7 +39,7 @@
 .Op Fl F Ar pathname
 .Op Fl G Ar pathname
 .Op Fl I
-.Op Fl L Ar lwp
+.Op Fl L
 .Op Fl M Ar mapfilename
 .Op Fl N
 .Op Fl O Ar logfilename
@@ -47,6 +47,7 @@
 .Op Fl R Ar logfilename
 .Op Fl S Ar event-spec
 .Op Fl T
+.Op Fl U Ar event-spec
 .Op Fl W
 .Op Fl a Ar pathname
 .Op Fl c Ar cpu-spec
@@ -54,6 +55,7 @@
 .Op Fl e
 .Op Fl f Ar pluginopt
 .Op Fl g
+.Op Fl i Ar lwp
 .Op Fl k Ar kerneldir
 .Op Fl l Ar secs
 .Op Fl m Ar pathname
@@ -64,6 +66,7 @@
 .Op Fl r Ar fsroot
 .Op Fl s Ar event-spec
 .Op Fl t Ar process-spec
+.Op Fl u Ar event-spec
 .Op Fl v
 .Op Fl w Ar secs
 .Op Fl z Ar graphdepth
@@ -159,13 +162,8 @@ this information is sent to the output file specified by the
 option.
 .It Fl I
 Skip symbol lookup and display address instead.
-.It Fl L Ar lwp
-Filter on thread ID
-.Ar lwp ,
-which you can get from
-.Xr ps 1
-.Fl o
-.Li lwp .
+.It Fl L
+List all event names.
 .It Fl M Ar mapfilename
 Write the mapping between executable objects encountered in the event
 log and the abbreviated pathnames used for
@@ -228,6 +226,8 @@ can be used: 'c+a' switch to accumulative mode, 'c+d' switch
 to delta mode, 'm' merge PMCs, 'n' change view, 'p' show next
 PMC, ' ' pause, 'q' quit. calltree only: 'f' cost under threshold
 is seen as a dot.
+.It Fl U Ar event-spec
+Provide long description of event.
 .It Fl W
 Toggle logging the incremental counts seen by the threads of a
 tracked process each time they are scheduled on a CPU.
@@ -291,6 +291,13 @@ A separate profile file is generated for each executable object
 encountered.
 Profile files are placed in sub-directories named by their PMC
 event name.
+.It Fl i Ar lwp
+Filter on thread ID
+.Ar lwp ,
+which you can get from
+.Xr ps 1
+.Fl o
+.Li lwp .
 .It Fl k Ar kerneldir
 Set the pathname of the kernel directory to argument
 .Ar kerneldir .
@@ -363,6 +370,8 @@ Argument
 .Ar process-spec
 may be a non-negative integer denoting a specific process id, or a
 regular expression for selecting processes based on their command names.
+.It Fl u Ar event-spec
+Provide short description of event.
 .It Fl v
 Increase verbosity.
 .It Fl w Ar secs
diff --git a/usr.sbin/pmcstat/pmcstat.c b/usr.sbin/pmcstat/pmcstat.c
index 63716e0e5c0d..45b272b60b59 100644
--- a/usr.sbin/pmcstat/pmcstat.c
+++ b/usr.sbin/pmcstat/pmcstat.c
@@ -365,8 +365,8 @@ pmcstat_show_usage(void)
 	    "\t -F file\t write a system-wide callgraph (Kcachegrind format)"
 		" to \"file\"\n"
 	    "\t -G file\t write a system-wide callgraph to \"file\"\n"
-	    "\t -I don't resolve leaf function name, show address instead\n"
-	    "\t -L lwp\t filter on thread id \"lwp\" in post-processing\n"
+	    "\t -I\t\t don't resolve leaf function name, show address instead\n"
+	    "\t -L\t\t list all counters available on this host\n"
 	    "\t -M file\t print executable/gmon file map to \"file\"\n"
 	    "\t -N\t\t (toggle) capture callchains\n"
 	    "\t -O file\t send log output to \"file\"\n"
@@ -374,6 +374,7 @@ pmcstat_show_usage(void)
 	    "\t -R file\t read events from \"file\"\n"
 	    "\t -S spec\t allocate a system-wide sampling PMC\n"
 	    "\t -T\t\t start in top mode\n"
+	    "\t -U spec \t provide long description of counters matching spec\n"
 	    "\t -W\t\t (toggle) show counts per context switch\n"
 	    "\t -a file\t print sampled PCs and callgraph to \"file\"\n"
 	    "\t -c cpu-list\t set cpus for subsequent system-wide PMCs\n"
@@ -381,6 +382,7 @@ pmcstat_show_usage(void)
 	    "\t -e\t\t use wide history counter for gprof(1) output\n"
 	    "\t -f spec\t pass \"spec\" to as plugin option\n"
 	    "\t -g\t\t produce gprof(1) compatible profiles\n"
+	    "\t -i lwp\t\t filter on thread id \"lwp\" in post-processing\n"
 	    "\t -k dir\t\t set the path to the kernel\n"
 	    "\t -l secs\t set duration time\n"
 	    "\t -m file\t print sampled PCs to \"file\"\n"
@@ -392,6 +394,7 @@ pmcstat_show_usage(void)
 	    "\t -s spec\t allocate a system-wide counting PMC\n"
 	    "\t -t process-spec attach to running processes matching "
 		"\"process-spec\"\n"
+	    "\t -u spec \t provide short description of counters matching spec\n"
 	    "\t -v\t\t increase verbosity\n"
 	    "\t -w secs\t set printing time interval\n"
 	    "\t -z depth\t limit callchain display depth"
@@ -429,14 +432,14 @@ main(int argc, char **argv)
 	int option, npmc;
 	int c, check_driver_stats; 
 	int do_callchain, do_descendants, do_logproccsw, do_logprocexit;
-	int do_print, do_read;
+	int do_print, do_read, do_listcounters, do_descr, do_long_descr;
 	size_t len;
 	int graphdepth;
 	int pipefd[2], rfd;
 	int use_cumulative_counts;
 	short cf, cb;
 	uint64_t current_sampling_count;
-	char *end, *tmp;
+	char *end, *tmp, *event;
 	const char *errmsg, *graphfilename;
 	enum pmcstat_state runstate;
 	struct pmc_driverstats ds_start, ds_end;
@@ -450,9 +453,12 @@ main(int argc, char **argv)
 	check_driver_stats      = 0;
 	current_sampling_count  = 0;
 	do_callchain		= 1;
+	do_descr                = 0;
 	do_descendants          = 0;
+	do_long_descr           = 0;
 	do_logproccsw           = 0;
 	do_logprocexit          = 0;
+	do_listcounters         = 0;
 	use_cumulative_counts   = 0;
 	graphfilename		= "-";
 	args.pa_required	= 0;
@@ -503,7 +509,7 @@ main(int argc, char **argv)
 	CPU_COPY(&rootmask, &cpumask);
 
 	while ((option = getopt(argc, argv,
-	    "CD:EF:G:IL:M:NO:P:R:S:TWa:c:def:gk:l:m:n:o:p:qr:s:t:vw:z:")) != -1)
+	    "CD:EF:G:ILM:NO:P:R:S:TU:WZa:c:def:gi:k:l:m:n:o:p:qr:s:t:u:vw:z:")) != -1)
 		switch (option) {
 		case 'a':	/* Annotate + callgraph */
 			args.pa_flags |= FLAG_DO_ANNOTATE;
@@ -544,6 +550,12 @@ main(int argc, char **argv)
 			args.pa_required |= FLAG_HAS_PROCESS_PMCS;
 			break;
 
+		case 'E':	/* log process exit */
+			do_logprocexit = !do_logprocexit;
+			args.pa_required |= (FLAG_HAS_PROCESS_PMCS |
+			    FLAG_HAS_COUNTING_PMCS | FLAG_HAS_OUTPUT_LOGFILE);
+			break;
+
 		case 'e':	/* wide gprof metrics */
 			args.pa_flags |= FLAG_DO_WIDE_GPROF_HC;
 			break;
@@ -575,6 +587,11 @@ main(int argc, char **argv)
 		case 'I':
 			args.pa_flags |= FLAG_SKIP_TOP_FN_RES;
 			break;
+		case 'i':
+			args.pa_flags |= FLAG_FILTER_THREAD_ID;
+			args.pa_tid = strtol(optarg, &end, 0);
+			break;
+
 		case 'k':	/* pathname to the kernel */
 			free(args.pa_kernel);
 			args.pa_kernel = strdup(optarg);
@@ -585,9 +602,9 @@ main(int argc, char **argv)
 			break;
 
 		case 'L':
-			args.pa_flags |= FLAG_FILTER_THREAD_ID;
-			args.pa_tid = strtol(optarg, &end, 0);
+			do_listcounters = 1;
 			break;
+
 		case 'l':	/* time duration in seconds */
 			duration = strtod(optarg, &end);
 			if (*end != '\0' || duration <= 0)
@@ -603,12 +620,6 @@ main(int argc, char **argv)
 			graphfilename  = optarg;
 			break;
 
-		case 'E':	/* log process exit */
-			do_logprocexit = !do_logprocexit;
-			args.pa_required |= (FLAG_HAS_PROCESS_PMCS |
-			    FLAG_HAS_COUNTING_PMCS | FLAG_HAS_OUTPUT_LOGFILE);
-			break;
-
 		case 'M':	/* mapfile */
 			args.pa_mapfilename = optarg;
 			break;
@@ -655,7 +666,7 @@ main(int argc, char **argv)
 				errx(EX_SOFTWARE, "ERROR: Out of memory.");
 
 			if (option == 'S' || option == 'P')
-				ev->ev_count = current_sampling_count ? current_sampling_count : pmcstat_pmu_sample_rate_get(ev->ev_spec);
+				ev->ev_count = current_sampling_count ? current_sampling_count : pmc_pmu_sample_rate_get(ev->ev_spec);
 			else
 				ev->ev_count = -1;
 
@@ -760,6 +771,14 @@ main(int argc, char **argv)
 				args.pa_printfile = stdout;
 			break;
 
+		case 'u':
+			do_descr = 1;
+			event = optarg;
+			break;
+		case 'U':
+			do_long_descr = 1;
+			event = optarg;
+			break;
 		case 'v':	/* verbose */
 			args.pa_verbosity++;
 			break;
@@ -796,6 +815,18 @@ main(int argc, char **argv)
 			break;
 
 		}
+	if ((do_listcounters | do_descr | do_long_descr) &&
+		pmc_pmu_enabled() == 0)
+			errx(EX_USAGE, "pmu features not supported on host or hwpmc not loaded");
+	if (do_listcounters) {
+		pmc_pmu_print_counters();
+	} else if (do_descr) {
+		pmc_pmu_print_counter_desc(event);
+	} else if (do_long_descr) {
+		pmc_pmu_print_counter_desc_long(event);
+	}
+	if (do_listcounters | do_descr | do_long_descr)
+		exit(0);
 
 	args.pa_argc = (argc -= optind);
 	args.pa_argv = (argv += optind);
diff --git a/usr.sbin/pmcstat/pmcstat_log.c b/usr.sbin/pmcstat/pmcstat_log.c
index cc3e9c49203f..5f88d274f393 100644
--- a/usr.sbin/pmcstat/pmcstat_log.c
+++ b/usr.sbin/pmcstat/pmcstat_log.c
@@ -423,13 +423,6 @@ pmcstat_print_log(void)
 			    (void *) ev.pl_u.pl_mo.pl_start,
 			    (void *) ev.pl_u.pl_mo.pl_end);
 			break;
-		case PMCLOG_TYPE_PCSAMPLE:
-			PMCSTAT_PRINT_ENTRY("sample","0x%x %d %p %c",
-			    ev.pl_u.pl_s.pl_pmcid,
-			    ev.pl_u.pl_s.pl_pid,
-			    (void *) ev.pl_u.pl_s.pl_pc,
-			    ev.pl_u.pl_s.pl_usermode ? 'u' : 's');
-			break;
 		case PMCLOG_TYPE_PMCALLOCATE:
 			PMCSTAT_PRINT_ENTRY("allocate","0x%x \"%s\" 0x%x",
 			    ev.pl_u.pl_a.pl_pmcid,