[PATCH] 2.5.8 IDE 35

- Expand configure help options a bit
- Fix xconfig bug
- Decrease queue depth if a command takes too long to complete
- Test master/slave stuff. It works, but one device can heavily starve
   another. This is the simple approach right now, means that one device
   will wait until the other is completely idle before starting any
   commands This is not necessary since we can have queued commands on
   both devices at the same time. TODO.
- Add proc output for oldest command, just for testing.
- pci_dev compile fixes.
- Make sure ide-disk doesn't BUG if TCQ is not used, basically this was
   fixed by off-loading the using_tcq setting to ide-tcq.
- Remove warning about 'queued feature set not supported'
- Abstract ide_tcq_wait_dataphase() into a function

drivers/ide/Config.help

@@ -753,9 +753,17 @@ CONFIG_BLK_DEV_IDE_TCQ
 
   If you have such a drive, say Y here.
 
-CONFIG_BLK_DEV_IDE_TCQ_DEFAULT
-  Enabled tagged command queueing unconditionally on drives that report
-  support for it.
+CONFIG_BLK_DEV_IDE_TCQ_FULL
+  When a command completes from the drive, the SERVICE bit is checked to
+  see if other queued commands are ready to be started. Doing this
+  immediately after a command completes has a tendency to 'starve' the
+  device hardware queue, since we risk emptying the queue completely
+  before starting any new commands. This shows up during stressing the
+  drive as a /\/\/\/\ queue size balance, where we could instead try and
+  maintain a minimum queue size and get a /---------\ graph instead.
+
+  Saying Y here will attempt to always keep the queue full when possible
+  at the cost of possibly increasing command turn-around latency.
 
   Generally say Y here.
 
@@ -766,6 +774,18 @@ CONFIG_BLK_DEV_IDE_TCQ_DEPTH
   You probably just want the default of 32 here. If you enter an invalid
   number, the default value will be used.
 
+CONFIG_BLK_DEV_IDE_TCQ_DEFAULT
+  Enabled tagged command queueing unconditionally on drives that report
+  support for it. Regardless of the chosen value here, tagging can be
+  controlled at run time:
+
+  echo "using_tcq:32" > /proc/ide/hdX/settings
+
+  where any value between 1-32 selects chosen queue depth and enables
+  TCQ, and 0 disables it.
+
+  Generally say Y here.
+
 CONFIG_BLK_DEV_IT8172
   Say Y here to support the on-board IDE controller on the Integrated
   Technology Express, Inc. ITE8172 SBC.  Vendor page at

drivers/ide/Config.in

@@ -48,10 +48,11 @@ if [ "$CONFIG_BLK_DEV_IDE" != "n" ]; then
          dep_bool '    Enable DMA only for disks ' CONFIG_IDEDMA_ONLYDISK $CONFIG_IDEDMA_PCI_AUTO
 	 define_bool CONFIG_BLK_DEV_IDEDMA $CONFIG_BLK_DEV_IDEDMA_PCI
 	 dep_bool '    ATA tagged command queueing' CONFIG_BLK_DEV_IDE_TCQ $CONFIG_BLK_DEV_IDEDMA_PCI
-	   dep_bool '      TCQ on by default' CONFIG_BLK_DEV_IDE_TCQ_DEFAULT $CONFIG_BLK_DEV_IDE_TCQ
-	   if [ $CONFIG_BLK_DEV_IDE_TCQ_DEFAULT != "n" ]; then
-		int '      Default queue depth' CONFIG_BLK_DEV_IDE_TCQ_DEPTH 32
-	   fi
+	 dep_bool '      Attempt to keep queue full' CONFIG_BLK_DEV_IDE_TCQ_FULL $CONFIG_BLK_DEV_IDE_TCQ
+	 dep_bool '      TCQ on by default' CONFIG_BLK_DEV_IDE_TCQ_DEFAULT $CONFIG_BLK_DEV_IDE_TCQ
+	 if [ "$CONFIG_BLK_DEV_IDE_TCQ" != "n" ]; then
+	    int '      Default queue depth' CONFIG_BLK_DEV_IDE_TCQ_DEPTH 8
+	 fi
 	 dep_bool '    ATA Work(s) In Progress (EXPERIMENTAL)' CONFIG_IDEDMA_PCI_WIP $CONFIG_BLK_DEV_IDEDMA_PCI $CONFIG_EXPERIMENTAL
 	 dep_bool '    Good-Bad DMA Model-Firmware (WIP)' CONFIG_IDEDMA_NEW_DRIVE_LISTINGS $CONFIG_IDEDMA_PCI_WIP
 	 dep_bool '    AEC62XX chipset support' CONFIG_BLK_DEV_AEC62XX $CONFIG_BLK_DEV_IDEDMA_PCI

drivers/ide/ide-disk.c

@@ -107,10 +107,7 @@ static u8 get_command(ide_drive_t *drive, int cmd)
 	 * 48-bit commands are pretty sanely laid out
 	 */
 	if (lba48bit) {
-		if (cmd == READ)
-			command = WIN_READ_EXT;
-		else
-			command = WIN_WRITE_EXT;
+		command = cmd == READ ? WIN_READ_EXT : WIN_WRITE_EXT;
 
 		if (drive->using_dma) {
 			command++;		/* WIN_*DMA_EXT */
@@ -118,31 +115,33 @@ static u8 get_command(ide_drive_t *drive, int cmd)
 				command++;	/* WIN_*DMA_QUEUED_EXT */
 		} else if (drive->mult_count)
 			command += 5;		/* WIN_MULT*_EXT */
-	} else {
-		/*
-		 * 28-bit commands seem not to be, though...
-		 */
-		if (cmd == READ) {
-			if (drive->using_dma) {
-				if (drive->using_tcq)
-					command = WIN_READDMA_QUEUED;
-				else
-					command = WIN_READDMA;
-			} else if (drive->mult_count)
-				command = WIN_MULTREAD;
+
+		return command;
+	}
+
+	/*
+	 * 28-bit commands seem not to be, though...
+	 */
+	if (cmd == READ) {
+		if (drive->using_dma) {
+			if (drive->using_tcq)
+				command = WIN_READDMA_QUEUED;
 			else
-				command = WIN_READ;
-		} else {
-			if (drive->using_dma) {
-				if (drive->using_tcq)
-					command = WIN_WRITEDMA_QUEUED;
-				else
-					command = WIN_WRITEDMA;
-			} else if (drive->mult_count)
-				command = WIN_MULTWRITE;
+				command = WIN_READDMA;
+		} else if (drive->mult_count)
+			command = WIN_MULTREAD;
+		else
+			command = WIN_READ;
+	} else {
+		if (drive->using_dma) {
+			if (drive->using_tcq)
+				command = WIN_WRITEDMA_QUEUED;
 			else
-				command = WIN_WRITE;
-		}
+				command = WIN_WRITEDMA;
+		} else if (drive->mult_count)
+			command = WIN_MULTWRITE;
+		else
+			command = WIN_WRITE;
 	}
 
 	return command;
@@ -895,24 +894,24 @@ static int proc_idedisk_read_tcq
 
 		__set_bit(i, &tag_mask);
 		len += sprintf(out+len, "%d, ", i);
-		if (ar->ar_time > max_jif)
-			max_jif = ar->ar_time;
+		if (cur_jif - ar->ar_time > max_jif)
+			max_jif = cur_jif - ar->ar_time;
 		cmds++;
 	}
 	len += sprintf(out+len, "]\n");
 
+	len += sprintf(out+len, "Queue:\t\t\treleased [ %d ] - started [ %d ]\n", drive->tcq->immed_rel, drive->tcq->immed_comp);
+
 	if (drive->tcq->queued != cmds)
-		len += sprintf(out+len, "pending request and queue count mismatch (%d)\n", cmds);
+		len += sprintf(out+len, "pending request and queue count mismatch (counted: %d)\n", cmds);
 
 	if (tag_mask != drive->tcq->tag_mask)
 		len += sprintf(out+len, "tag masks differ (counted %lx != %lx\n", tag_mask, drive->tcq->tag_mask);
 
 	len += sprintf(out+len, "DMA status:\t\t%srunning\n", test_bit(IDE_DMA, &HWGROUP(drive)->flags) ? "" : "not ");
 
-	if (max_jif)
-		len += sprintf(out+len, "Oldest command:\t\t%lu\n", cur_jif - max_jif);
-
-	len += sprintf(out+len, "immed rel %d, immed comp %d\n", drive->tcq->immed_rel, drive->tcq->immed_comp);
+	len += sprintf(out+len, "Oldest command:\t\t%lu jiffies\n", max_jif);
+	len += sprintf(out+len, "Oldest command ever:\t%lu\n", drive->tcq->oldest_command);
 
 	drive->tcq->max_last_depth = 0;
 
@@ -1017,8 +1016,10 @@ static int set_using_tcq(ide_drive_t *drive, int arg)
 		return -EPERM;
 	if (!drive->channel->dmaproc)
 		return -EPERM;
+	if (arg == drive->queue_depth && drive->using_tcq)
+		return 0;
 
-	drive->using_tcq = arg;
+	drive->queue_depth = arg ? arg : 1;
 	if (drive->channel->dmaproc(arg ? ide_dma_queued_on : ide_dma_queued_off, drive))
 		return -EIO;
 

drivers/ide/ide-dma.c

@@ -610,9 +610,6 @@ int ide_dmaproc (ide_dma_action_t func, ide_drive_t *drive)
 		case ide_dma_check:
 			return config_drive_for_dma (drive);
 		case ide_dma_begin:
-#ifdef DEBUG
-			printk("ide_dma_begin: from %p\n", __builtin_return_address(0));
-#endif
 			if (test_and_set_bit(IDE_DMA, &HWGROUP(drive)->flags))
 				BUG();
 			/* Note that this is done *after* the cmd has
@@ -634,14 +631,13 @@ int ide_dmaproc (ide_dma_action_t func, ide_drive_t *drive)
 		case ide_dma_read:
 			reading = 1 << 3;
 		case ide_dma_write:
-			ar = HWGROUP(drive)->rq->special;
+			ar = IDE_CUR_AR(drive);
 
 			if (ide_start_dma(hwif, drive, func))
 				return 1;
 
 			if (drive->type != ATA_DISK)
 				return 0;
-
 			BUG_ON(HWGROUP(drive)->handler);
 			ide_set_handler(drive, &ide_dma_intr, WAIT_CMD, dma_timer_expiry);	/* issue cmd to drive */
 			if ((ar->ar_rq->flags & REQ_DRIVE_TASKFILE) &&
@@ -655,20 +651,13 @@ int ide_dmaproc (ide_dma_action_t func, ide_drive_t *drive)
 			}
 			return hwif->dmaproc(ide_dma_begin, drive);
 		case ide_dma_end: /* returns 1 on error, 0 otherwise */
-#ifdef DEBUG
-			printk("ide_dma_end: from %p\n", __builtin_return_address(0));
-#endif
-			if (!test_and_clear_bit(IDE_DMA, &HWGROUP(drive)->flags)) {
-				printk("ide_dma_end: dma not going? %p\n", __builtin_return_address(0));
-				return 1;
-			}
+			if (!test_and_clear_bit(IDE_DMA, &HWGROUP(drive)->flags))
+				BUG();
 			drive->waiting_for_dma = 0;
 			outb(inb(dma_base)&~1, dma_base);	/* stop DMA */
 			dma_stat = inb(dma_base+2);	/* get DMA status */
 			outb(dma_stat|6, dma_base+2);	/* clear the INTR & ERROR bits */
 			ide_destroy_dmatable(drive);	/* purge DMA mappings */
-			if (drive->tcq)
-				IDE_SET_CUR_TAG(drive, -1);
 			return (dma_stat & 7) != 4 ? (0x10 | dma_stat) : 0;	/* verify good DMA status */
 		case ide_dma_test_irq: /* returns 1 if dma irq issued, 0 otherwise */
 			dma_stat = inb(dma_base+2);

drivers/ide/ide-probe.c

@@ -192,19 +192,16 @@ static inline void do_identify (ide_drive_t *drive, byte cmd)
 	/*
 	 * it's an ata drive, build command list
 	 */
-#ifndef CONFIG_BLK_DEV_IDE_TCQ
 	drive->queue_depth = 1;
+#ifdef CONFIG_BLK_DEV_IDE_TCQ_DEPTH
+	drive->queue_depth = CONFIG_BLK_DEV_IDE_TCQ_DEPTH;
 #else
-# ifndef CONFIG_BLK_DEV_IDE_TCQ_DEPTH
-#  define CONFIG_BLK_DEV_IDE_TCQ_DEPTH 1
-# endif
 	drive->queue_depth = drive->id->queue_depth + 1;
-	if (drive->queue_depth > CONFIG_BLK_DEV_IDE_TCQ_DEPTH)
-		drive->queue_depth = CONFIG_BLK_DEV_IDE_TCQ_DEPTH;
+#endif
 	if (drive->queue_depth < 1 || drive->queue_depth > IDE_MAX_TAG)
 		drive->queue_depth = IDE_MAX_TAG;
-#endif
-	if (ide_build_commandlist(drive))
+
+	if (ide_init_commandlist(drive))
 		goto err_misc;
 
 	return;

drivers/ide/pdc4030.c

@@ -648,6 +648,12 @@ ide_startstop_t promise_rw_disk (ide_drive_t *drive, struct request *rq, unsigne
 
 	memset(&taskfile, 0, sizeof(struct hd_drive_task_hdr));
 
+	/* The four drives on the two logical (one physical) interfaces
+	   are distinguished by writing the drive number (0-3) to the
+	   Feature register.
+	   FIXME: Is promise_selectproc now redundant??
+	 */
+	taskfile.feature    = (drive->channel->unit << 1) + drive->select.b.unit;
 	taskfile.sector_count	= rq->nr_sectors;
 	taskfile.sector_number	= block;
 	taskfile.low_cylinder	= (block>>=8);

include/linux/ide.h

@@ -273,19 +273,22 @@ typedef union {
 	} b;
 } special_t;
 
-#define IDE_MAX_TAG	32		/* spec says 32 max */
+#define IDE_MAX_TAG		(32)		/* spec says 32 max */
+#define IDE_INACTIVE_TAG	(-1)
 
 struct ata_request;
 typedef struct ide_tag_info_s {
 	unsigned long tag_mask;			/* next tag bit mask */
 	struct ata_request *ar[IDE_MAX_TAG];	/* in-progress requests */
 	int active_tag;				/* current active tag */
+
 	int queued;				/* current depth */
 
 	/*
 	 * stats ->
 	 */
 	int max_depth;				/* max depth ever */
+
 	int max_last_depth;			/* max since last check */
 
 	/*
@@ -295,14 +298,19 @@ typedef struct ide_tag_info_s {
 	 */
 	int immed_rel;
 	int immed_comp;
+	unsigned long oldest_command;
 } ide_tag_info_t;
 
 #define IDE_GET_AR(drive, tag)	((drive)->tcq->ar[(tag)])
 #define IDE_CUR_TAG(drive)	(IDE_GET_AR((drive), (drive)->tcq->active_tag))
-#define IDE_SET_CUR_TAG(drive, tag)	((drive)->tcq->active_tag = (tag))
+#define IDE_SET_CUR_TAG(drive, tag)	\
+	do {						\
+		((drive)->tcq->active_tag = (tag));	\
+		if ((tag) == IDE_INACTIVE_TAG)		\
+			HWGROUP((drive))->rq = NULL;	\
+	} while (0);
 
-#define IDE_CUR_AR(drive)	\
-	((drive)->using_tcq ? IDE_CUR_TAG((drive)) : HWGROUP((drive))->rq->special)
+#define IDE_CUR_AR(drive)	(HWGROUP((drive))->rq->special)
 
 struct ide_settings_s;
 
@@ -359,6 +367,7 @@ typedef struct ide_drive_s {
 	unsigned autotune	: 2;	/* 1=autotune, 2=noautotune, 0=default */
 	unsigned remap_0_to_1	: 2;	/* 0=remap if ezdrive, 1=remap, 2=noremap */
 	unsigned ata_flash	: 1;	/* 1=present, 0=default */
+	unsigned service_pending: 1;
 	unsigned	addressing;	/* : 2; 0=28-bit, 1=48-bit, 2=64-bit */
 	byte		scsi;		/* 0=default, 1=skip current ide-subdriver for ide-scsi emulation */
 	select_t	select;		/* basic drive/head select reg value */
@@ -546,7 +555,7 @@ extern void ide_unregister(struct ata_channel *hwif);
 typedef enum {
 	ide_stopped,	/* no drive operation was started */
 	ide_started,	/* a drive operation was started, and a handler was set */
-	ide_released	/* started and released bus */
+	ide_released,	/* started, handler set, bus released */
 } ide_startstop_t;
 
 /*
@@ -980,6 +989,11 @@ extern void revalidate_drives(void);
 #define ATA_AR_SETUP	2
 #define ATA_AR_RETURN	4
 
+/*
+ * if turn-around time is longer than this, halve queue depth
+ */
+#define ATA_AR_MAX_TURNAROUND	(3 * HZ)
+
 #define list_ata_entry(entry) list_entry((entry), struct ata_request, ar_queue)
 
 static inline void ata_ar_init(ide_drive_t *drive, struct ata_request *ar)
@@ -1026,7 +1040,7 @@ static inline void ata_ar_put(ide_drive_t *drive, struct ata_request *ar)
 	ar->ar_rq = NULL;
 }
 
-extern inline int ide_get_tag(ide_drive_t *drive)
+static inline int ide_get_tag(ide_drive_t *drive)
 {
 	int tag = ffz(drive->tcq->tag_mask);
 
@@ -1043,12 +1057,28 @@ extern inline int ide_get_tag(ide_drive_t *drive)
 }
 
 #ifdef CONFIG_BLK_DEV_IDE_TCQ
-# define ide_pending_commands(drive)	((drive)->using_tcq && (drive)->tcq->queued)
+static inline int ide_pending_commands(ide_drive_t *drive)
+{
+	if (!drive->tcq)
+		return 0;
+
+	return drive->tcq->queued;
+}
+
+static inline int ide_can_queue(ide_drive_t *drive)
+{
+	if (!drive->tcq)
+		return 1;
+
+	return drive->tcq->queued < drive->queue_depth;
+}
 #else
-# define ide_pending_commands(drive)	0
+#define ide_pending_commands(drive)	(0)
+#define ide_can_queue(drive)		(1)
 #endif
 
 int ide_build_commandlist(ide_drive_t *);
+int ide_init_commandlist(ide_drive_t *);
 void ide_teardown_commandlist(ide_drive_t *);
 int ide_tcq_dmaproc(ide_dma_action_t, ide_drive_t *);
 ide_startstop_t ide_start_tag(ide_dma_action_t, ide_drive_t *, struct ata_request *);