Update pstat -s to handle both old and new swapper.

    Add pstat -ss to dump new swapper's radix tree.

2 files changed, 290 insertions, 76 deletions

diff --git a/usr.sbin/pstat/pstat.8 b/usr.sbin/pstat/pstat.8
index 792bbb5..d4e5dc6 100644
--- a/usr.sbin/pstat/pstat.8
+++ b/usr.sbin/pstat/pstat.8
@@ -30,7 +30,7 @@
 .\" SUCH DAMAGE.
 .\"
 .\"     @(#)pstat.8	8.5 (Berkeley) 5/13/94
-.\"	$Id: pstat.8,v 1.15 1997/10/09 07:22:05 charnier Exp $
+.\"	$Id: pstat.8,v 1.16 1998/01/17 16:58:55 bde Exp $
 .\"
 .Dd May 13, 1994
 .Dt PSTAT 8
@@ -131,6 +131,10 @@ reports the percentage of space used.
 .Pp
 If more than one partition is configured into the system, totals for all
 of the statistics will be reported in the final line of the report.
+.Pp
+If you supply the option again, as in
+.Fl ss ,
+the system will display a breakdown of the swap bitmap/radix-tree.
 .It Fl t
 Print table for terminals
 with these headings:
diff --git a/usr.sbin/pstat/pstat.c b/usr.sbin/pstat/pstat.c
index 0cc4339..157068d 100644
--- a/usr.sbin/pstat/pstat.c
+++ b/usr.sbin/pstat/pstat.c
@@ -42,7 +42,7 @@ static const char copyright[] =
 static char sccsid[] = "@(#)pstat.c	8.16 (Berkeley) 5/9/95";
 #endif
 static const char rcsid[] =
-	"$Id: pstat.c,v 1.36 1998/07/06 20:28:05 bde Exp $";
+	"$Id: pstat.c,v 1.37 1998/08/19 01:32:28 bde Exp $";
 #endif /* not lint */
 
 #include <sys/param.h>
@@ -68,6 +68,7 @@ static const char rcsid[] =
 #include <sys/tty.h>
 #include <sys/conf.h>
 #include <sys/rlist.h>
+#include <sys/blist.h>
 
 #include <sys/user.h>
 #include <sys/sysctl.h>
@@ -85,76 +86,77 @@ static const char rcsid[] =
 struct nlist nl[] = {
 #define VM_SWAPLIST	0
 	{ "_swaplist" },/* list of free swap areas */
+#define NLMANDATORYBEG	1
 #define VM_SWDEVT	1
 	{ "_swdevt" },	/* list of swap devices and sizes */
-#define VM_NSWAP	2
-	{ "_nswap" },	/* size of largest swap device */
-#define VM_NSWDEV	3
+#define VM_NSWDEV	2
 	{ "_nswdev" },	/* number of swap devices */
-#define VM_DMMAX	4
+#define VM_DMMAX	3
 	{ "_dmmax" },	/* maximum size of a swap block */
-#define	V_MOUNTLIST	5
+#define	V_MOUNTLIST	4
 	{ "_mountlist" },	/* address of head of mount list. */
-#define V_NUMV		6
+#define V_NUMV		5
 	{ "_numvnodes" },
-#define	FNL_NFILE	7
+#define	FNL_NFILE	6
 	{"_nfiles"},
-#define FNL_MAXFILE	8
+#define FNL_MAXFILE	7
 	{"_maxfiles"},
-#define NLMANDATORY FNL_MAXFILE	/* names up to here are mandatory */
-#define	SCONS		NLMANDATORY + 1
+#define NLMANDATORYEND FNL_MAXFILE	/* names up to here are mandatory */
+#define	SCONS		NLMANDATORYEND + 1
 	{ "_cons" },
-#define	SPTY		NLMANDATORY + 2
+#define	SPTY		NLMANDATORYEND + 2
 	{ "_pt_tty" },
-#define	SNPTY		NLMANDATORY + 3
+#define	SNPTY		NLMANDATORYEND + 3
 	{ "_npty" },
+#define	VM_SWAPBLIST	NLMANDATORYEND + 4
+	{ "_swapblist" },
 
 #ifdef hp300
-#define	SDCA	(SNPTY+1)
+#define	SDCA	(VM_SWAPBLIST+1)
 	{ "_dca_tty" },
-#define	SNDCA	(SNPTY+2)
+#define	SNDCA	(VM_SWAPBLIST+2)
 	{ "_ndca" },
-#define	SDCM	(SNPTY+3)
+#define	SDCM	(VM_SWAPBLIST+3)
 	{ "_dcm_tty" },
-#define	SNDCM	(SNPTY+4)
+#define	SNDCM	(VM_SWAPBLIST+4)
 	{ "_ndcm" },
-#define	SDCL	(SNPTY+5)
+#define	SDCL	(VM_SWAPBLIST+5)
 	{ "_dcl_tty" },
-#define	SNDCL	(SNPTY+6)
+#define	SNDCL	(VM_SWAPBLIST+6)
 	{ "_ndcl" },
-#define	SITE	(SNPTY+7)
+#define	SITE	(VM_SWAPBLIST+7)
 	{ "_ite_tty" },
-#define	SNITE	(SNPTY+8)
+#define	SNITE	(VM_SWAPBLIST+8)
 	{ "_nite" },
 #endif
 
 #ifdef mips
-#define SDC	(SNPTY+1)
+#define SDC	(VM_SWAPBLIST+1)
 	{ "_dc_tty" },
-#define SNDC	(SNPTY+2)
+#define SNDC	(VM_SWAPBLIST+2)
 	{ "_dc_cnt" },
 #endif
 
 #ifdef __FreeBSD__
-#define SCCONS	(SNPTY+1)
+#define SCCONS	(VM_SWAPBLIST+1)
 	{ "_sccons" },
-#define NSCCONS	(SNPTY+2)
+#define NSCCONS	(VM_SWAPBLIST+2)
 	{ "_nsccons" },
-#define SIO  (SNPTY+3)
+#define SIO  (VM_SWAPBLIST+3)
 	{ "_sio_tty" },
-#define NSIO (SNPTY+4)
+#define NSIO (VM_SWAPBLIST+4)
 	{ "_nsio_tty" },
-#define RC  (SNPTY+5)
+#define RC  (VM_SWAPBLIST+5)
 	{ "_rc_tty" },
-#define NRC (SNPTY+6)
+#define NRC (VM_SWAPBLIST+6)
 	{ "_nrc_tty" },
-#define CY  (SNPTY+7)
+#define CY  (VM_SWAPBLIST+7)
 	{ "_cy_tty" },
-#define NCY (SNPTY+8)
+#define NCY (VM_SWAPBLIST+8)
 	{ "_ncy_tty" },
-#define SI  (SNPTY+9)
+#define SI  (VM_SWAPBLIST+9)
 	{ "_si_tty" },
-#define NSI (SNPTY+10)
+#define NSI (VM_SWAPBLIST+10)
 	{ "_si_Nports" },
 #endif
 	{ "" }
@@ -162,6 +164,7 @@ struct nlist nl[] = {
 
 int	usenumflag;
 int	totalflag;
+int	swapflag;
 char	*nlistf	= NULL;
 char	*memf	= NULL;
 kvm_t	*kd;
@@ -212,6 +215,12 @@ struct {
 #define	KGET2(addr, p, s, msg)						\
 	if (kvm_read(kd, (u_long)(addr), p, s) != s)			\
 		warnx("cannot read %s: %s", msg, kvm_geterr(kd))
+#define	KGETN(idx, var)							\
+	KGET1N(idx, &var, sizeof(var), SVAR(var))
+#define	KGET1N(idx, p, s, msg)						\
+	KGET2N(nl[idx].n_value, p, s, msg)
+#define	KGET2N(addr, p, s, msg)						\
+	((kvm_read(kd, (u_long)(addr), p, s) == s) ? 1 : 0)
 #define	KGETRET(addr, p, s, msg)					\
 	if (kvm_read(kd, (u_long)(addr), p, s) != s) {			\
 		warnx("cannot read %s: %s", msg, kvm_geterr(kd));	\
@@ -248,7 +257,7 @@ main(argc, argv)
 	char *argv[];
 {
 	int ch, i, quit, ret;
-	int fileflag, swapflag, ttyflag, vnodeflag;
+	int fileflag, ttyflag, vnodeflag;
 	char buf[_POSIX2_LINE_MAX],*opts;
 
 	fileflag = swapflag = ttyflag = vnodeflag = 0;
@@ -286,7 +295,7 @@ main(argc, argv)
 			usenumflag = 1;
 			break;
 		case 's':
-			swapflag = 1;
+			++swapflag;
 			break;
 		case 'T':
 			totalflag = 1;
@@ -320,7 +329,7 @@ main(argc, argv)
 	if ((ret = kvm_nlist(kd, nl)) != 0) {
 		if (ret == -1)
 			errx(1, "kvm_nlist: %s", kvm_geterr(kd));
-		for (i = quit = 0; i <= NLMANDATORY; i++)
+		for (i = NLMANDATORYBEG, quit = 0; i <= NLMANDATORYEND; i++)
 			if (!nl[i].n_value) {
 				quit = 1;
 				warnx("undefined symbol: %s", nl[i].n_name);
@@ -1026,26 +1035,190 @@ getfiles(abuf, alen)
 }
 
 /*
+ * scanradix() - used by swapmode() to scan the swap device's new style 
+ * 		 radix tree 
+ */
+
+#define TABME	tab, tab, ""
+
+int
+scanradix(
+	blmeta_t *scan, 
+	daddr_t blk,
+	daddr_t radix,
+	daddr_t skip, 
+	daddr_t count,
+	int dmmax, 
+	int nswdev,
+	long *perdev,
+	int tab
+) {
+	blmeta_t meta;
+
+	KGET2(scan, &meta, sizeof(meta), "blmeta_t");
+
+	/*
+	 * Terminator
+	 */
+	if (meta.bm_bighint == (daddr_t)-1) {
+		if (swapflag > 1) {
+			printf("%*.*s(0x%06x,%d) Terminator\n", 
+			    TABME,
+			    blk, 
+			    radix
+			);
+		}
+		return(-1);
+	}
+
+	if (radix == BLIST_BMAP_RADIX) {
+		/*
+		 * Leaf bitmap
+		 */
+		int i;
+
+		if (swapflag > 1) {
+			printf("%*.*s(0x%06x,%d) Bitmap %08x big=%d\n", 
+			    TABME,
+			    blk, 
+			    radix,
+			    (int)meta.u.bmu_bitmap,
+			    meta.bm_bighint
+			);
+		}
+
+		/*
+		 * If not all allocated, count.
+		 */
+		if (meta.u.bmu_bitmap != 0) {
+			for (i = 0; i < BLIST_BMAP_RADIX && i < count; ++i) {
+				/*
+				 * A 0 bit means allocated
+				 */
+				if ((meta.u.bmu_bitmap & (1 << i))) {
+					int t = 0;
+
+					if (nswdev)
+						t = (blk + i) / dmmax % nswdev;
+					++perdev[t];
+				}
+			}
+		}
+	} else if (meta.u.bmu_avail == radix) {
+		/*
+		 * Meta node if all free
+		 */
+		if (swapflag > 1) {
+			printf("%*.*s(0x%06x,%d) Submap ALL-FREE {\n", 
+			    TABME,
+			    blk, 
+			    radix,
+			    (int)meta.u.bmu_avail,
+			    meta.bm_bighint
+			);
+		}
+		/*
+		 * Note: both dmmax and radix are powers of 2.  However, dmmax
+		 * may be larger then radix so use a smaller increment if
+		 * necessary.
+		 */
+		{
+			int t;
+			int tinc = dmmax;
+
+			while (tinc > radix)
+				tinc >>= 1;
+
+			for (t = blk; t < blk + radix; t += tinc) {
+				if (nswdev)
+					perdev[0] += tinc;
+				else
+					perdev[t / dmmax % nswdev] += tinc;
+			}
+		}
+	} else if (meta.u.bmu_avail == 0) {
+		/*
+		 * Meta node if all used
+		 */
+		if (swapflag > 1) {
+			printf("%*.*s(0x%06x,%d) Submap ALL-ALLOCATED\n", 
+			    TABME,
+			    blk, 
+			    radix,
+			    (int)meta.u.bmu_avail,
+			    meta.bm_bighint
+			);
+		}
+	} else {
+		/*
+		 * Meta node if not all free
+		 */
+		int i;
+		int next_skip;
+
+		radix >>= BLIST_META_RADIX_SHIFT;
+		next_skip = skip >> BLIST_META_RADIX_SHIFT;
+
+		if (swapflag > 1) {
+			printf("%*.*s(0x%06x,%d) Submap avail=%d big=%d {\n", 
+			    TABME,
+			    blk, 
+			    radix,
+			    (int)meta.u.bmu_avail,
+			    meta.bm_bighint
+			);
+		}
+
+		for (i = 1; i <= skip; i += next_skip) {
+			int r;
+			daddr_t vcount = (count > radix) ? radix : count;
+
+			r = scanradix(
+			    &scan[i],
+			    blk,
+			    radix,
+			    next_skip - 1,
+			    vcount,
+			    dmmax,
+			    nswdev,
+			    perdev,
+			    tab + 4
+			);
+			if (r < 0)
+				break;
+			blk += radix;
+		}
+		if (swapflag > 1) {
+			printf("%*.*s}\n", TABME);
+		}
+	}
+	return(0);
+}
+
+
+/*
  * swapmode is based on a program called swapinfo written
  * by Kevin Lahey <kml@rokkaku.atl.ga.us>.
  */
 void
-swapmode()
+swapmode(void)
 {
 	char *header, *p;
-	int hlen, nswap, nswdev, dmmax;
-	int i, div, avail, nfree, npfree, used;
+	int hlen, nswdev, dmmax;
+	int i, div, mul, avail, nfree, npfree, used;
 	struct swdevt *sw;
 	long blocksize, *perdev;
 	struct rlist head;
 	struct rlisthdr swaplist;
-	struct rlist *swapptr;
+	struct blist	*swapblist = NULL;
 	u_long ptr;
 
-	KGET(VM_NSWAP, nswap);
 	KGET(VM_NSWDEV, nswdev);
-	KGET(VM_DMMAX, dmmax);
-	KGET1(VM_SWAPLIST, &swaplist, sizeof swaplist, "swaplist");
+	KGET(VM_DMMAX, dmmax);	/* PAGE_BSIZE'd or PAGE_SIZE'd depending */
+	if (!KGET1N(VM_SWAPLIST, &swaplist, sizeof swaplist, "swaplist")) {
+	    KGET1(VM_SWAPBLIST, &swapblist, sizeof swapblist, "swapblist");
+	}
+
 	if ((sw = malloc(nswdev * sizeof(*sw))) == NULL ||
 	    (perdev = malloc(nswdev * sizeof(*perdev))) == NULL)
 		errx(1, "malloc");
@@ -1055,39 +1228,63 @@ swapmode()
 	/* Count up swap space. */
 	nfree = 0;
 	memset(perdev, 0, nswdev * sizeof(*perdev));
-	swapptr = swaplist.rlh_list;
-	while (swapptr) {
-		int	top, bottom, next_block;
 
-		KGET2(swapptr, &head, sizeof(struct rlist), "swapptr");
+	if (swapblist) {
+		/*
+		 * New swap radix tree (ugh!)
+		 */
+		struct blist blcopy = { 0 };
+		KGET2(swapblist, &blcopy, sizeof(blcopy), "*swapblist");
+		if (swapflag > 1) {
+			printf("radix tree: %d/%d/%d blocks, %dK wired\n",
+				blcopy.bl_free,
+				blcopy.bl_blocks,
+				blcopy.bl_radix,
+				(blcopy.bl_rootblks * sizeof(blmeta_t) + 1023)/
+				    1024
+			);
+		}
+		scanradix(blcopy.bl_root, 0, blcopy.bl_radix, blcopy.bl_skip, blcopy.bl_rootblks, dmmax, nswdev, perdev, 0);
+	} else {
+		/*
+		 * Old swap
+		 */
+		struct rlist *swapptr;
 
-		top = head.rl_end;
-		bottom = head.rl_start;
+		swapptr = swaplist.rlh_list;
+		while (swapptr) {
+			int	top, bottom, next_block;
 
-		nfree += top - bottom + 1;
+			KGET2(swapptr, &head, sizeof(struct rlist), "swapptr");
 
-		/*
-		 * Swap space is split up among the configured disks.
-		 *
-		 * For interleaved swap devices, the first dmmax blocks
-		 * of swap space some from the first disk, the next dmmax
-		 * blocks from the next, and so on up to nswap blocks.
-		 *
-		 * The list of free space joins adjacent free blocks,
-		 * ignoring device boundries.  If we want to keep track
-		 * of this information per device, we'll just have to
-		 * extract it ourselves.
-		 */
-		while (top / dmmax != bottom / dmmax) {
-			next_block = ((bottom + dmmax) / dmmax);
+			top = head.rl_end;
+			bottom = head.rl_start;
+
+			nfree += top - bottom + 1;
+
+			/*
+			 * Swap space is split up among the configured disks.
+			 *
+			 * For interleaved swap devices, the first dmmax blocks
+			 * of swap space some from the first disk, the next 
+			 * dmmax blocks from the next, and so.
+			 *
+			 * The list of free space joins adjacent free blocks,
+			 * ignoring device boundries.  If we want to keep track
+			 * of this information per device, we'll just have to
+			 * extract it ourselves.
+			 */
+			while (top / dmmax != bottom / dmmax) {
+				next_block = ((bottom + dmmax) / dmmax);
+				perdev[(bottom / dmmax) % nswdev] +=
+					next_block * dmmax - bottom;
+				bottom = next_block * dmmax;
+			}
 			perdev[(bottom / dmmax) % nswdev] +=
-				next_block * dmmax - bottom;
-			bottom = next_block * dmmax;
-		}
-		perdev[(bottom / dmmax) % nswdev] +=
-			top - bottom + 1;
+				top - bottom + 1;
 
-		swapptr = head.rl_next;
+			swapptr = head.rl_next;
+		}
 	}
 
 	header = getbsize(&hlen, &blocksize);
@@ -1095,7 +1292,20 @@ swapmode()
 		(void)printf("%-11s %*s %8s %8s %8s  %s\n",
 		    "Device", hlen, header,
 		    "Used", "Avail", "Capacity", "Type");
-	div = blocksize / 512;
+
+	mul = 1;
+	if (swapblist) {
+		if ((div = blocksize / PAGE_SIZE) == 0) {
+			mul = PAGE_SIZE / blocksize;
+			div = 1;
+		}
+	} else {
+		if ((div = blocksize / DEV_BSIZE) == 0) {
+			mul = DEV_BSIZE / blocksize;
+			div = 1;
+		}
+	}
+
 	avail = npfree = 0;
 	for (i = 0; i < nswdev; i++) {
 		int xsize, xfree;
@@ -1112,10 +1322,10 @@ swapmode()
 				p = devname(sw[i].sw_dev, S_IFBLK);
 				(void)printf("/dev/%-6s %*d ",
 					     p == NULL ? "??" : p,
-					     hlen, sw[i].sw_nblks / div);
+					     hlen, sw[i].sw_nblks * mul / div);
 			} else
 				(void)printf("[NFS swap]  %*d ",
-					     hlen, sw[i].sw_nblks / div);
+					     hlen, sw[i].sw_nblks * mul / div);
 
 		/* The first dmmax is never allocated to avoid trashing of
 		 * disklabels
@@ -1128,7 +1338,7 @@ swapmode()
 		if (totalflag)
 			continue;
 		(void)printf("%8d %8d %5.0f%%    %s\n",
-		    used / div, xfree / div,
+		    used * mul / div, xfree * mul / div,
 		    (double)used / (double)xsize * 100.0,
 		    (sw[i].sw_flags & SW_SEQUENTIAL) ?
 			     "Sequential" : "Interleaved");
@@ -1147,7 +1357,7 @@ swapmode()
 	}
 	if (npfree > 1) {
 		(void)printf("%-11s %*d %8d %8d %5.0f%%\n",
-		    "Total", hlen, avail / div, used / div, nfree / div,
+		    "Total", hlen, avail * mul / div, used * mul / div, nfree * mul / div,
 		    (double)used / (double)avail * 100.0);
 	}
 }