Skip to contentSkip to conversation

libata: fix command queue leak when xlat_func fails

PATCH Applied John W. Linville 2 participants 6 messages Feb 18, 2005

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/5 libata: fix command queue leak when xlat_func fails

ata_scsi_translate allocates from the libata command queue by calling ata_scsi_qc_new. If xlat_func returns non-zero, control jumps to err_out which fails to free the allocated command. Fix is to add a new API to free unused commands.

---

 drivers/scsi/libata-core.c |   18 ++++++++++++++++++
 drivers/scsi/libata-scsi.c |    1 +
 drivers/scsi/libata.h      |    1 +
 3 files changed, 20 insertions(+)
drivers/scsi/libata-scsi.c renamed from drivers/scsi/libata-scsi.c.bugfix +1 -0
@@ -967,6 +967,7 @@ static void ata_scsi_translate(struct at
 	return;
 
 err_out:
+	ata_qc_free(qc);
 	ata_bad_cdb(cmd, done);
 	DPRINTK("EXIT - badcmd\n");
 }
drivers/scsi/libata-core.c renamed from drivers/scsi/libata-core.c.bugfix +18 -0
@@ -2687,6 +2687,24 @@ static void __ata_qc_complete(struct ata
 }
 
 /**
+ *	ata_qc_free - free unused ata_queued_cmd
+ *	@qc: Command to complete
+ *
+ *	Designed to free unused ata_queued_cmd object
+ *	in case something prevents using it.
+ *
+ *	LOCKING:
+ *
+ */
+void ata_qc_free(struct ata_queued_cmd *qc)
+{
+	assert(qc != NULL);	/* ata_qc_from_tag _might_ return NULL */
+	assert(qc->waiting == NULL);	/* nothing should be waiting */
+
+	__ata_qc_complete(qc);
+}
+
+/**
  *	ata_qc_complete - Complete an active ATA command
  *	@qc: Command to complete
  *	@drv_stat: ATA status register contents
drivers/scsi/libata.h renamed from drivers/scsi/libata.h.bugfix +1 -0
@@ -37,6 +37,7 @@ struct ata_scsi_args {
 /* libata-core.c */
 extern struct ata_queued_cmd *ata_qc_new_init(struct ata_port *ap,
 				      struct ata_device *dev);
+extern void ata_qc_free(struct ata_queued_cmd *qc);
 extern int ata_qc_issue(struct ata_queued_cmd *qc);
 extern int ata_check_atapi_dma(struct ata_queued_cmd *qc);
 extern int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg);
Signature · 2 lines
John W. Linville
[email protected]
#1 lore raw

PATCH 2/5 libata: sync SMART ioctls with ATA pass thru spec (T10/04-262r7)

Minor modifications to the SMART-related ioctls in libata, in compliance with the latest ATA pass thru spec (T10/04-262r7).

---

 drivers/scsi/libata-scsi.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
drivers/scsi/libata-scsi.c renamed from drivers/scsi/libata-scsi.c.orig +4 -2
@@ -109,14 +109,16 @@ int ata_cmd_ioctl(struct scsi_device *sc
 			return -ENOMEM;
 
 		scsi_cmd[1]  = (4 << 1); /* PIO Data-in */
+		scsi_cmd[2]  = 0x0e;     /* no off.line or cc, read from dev,
+		                            block count in sector count field */
 		sreq->sr_data_direction = DMA_FROM_DEVICE;
 	} else {
 		scsi_cmd[1]  = (3 << 1); /* Non-data */
+		/* scsi_cmd[2] is already 0 -- no off.line, cc, or data xfer */
 		sreq->sr_data_direction = DMA_NONE;
 	}
 
 	scsi_cmd[0] = ATA_16;
-	scsi_cmd[2] = 0x1f;     /* no off.line or cc, yes all registers */
 
 	scsi_cmd[4] = args[2];
 	if (args[0] == WIN_SMART) { /* hack -- ide driver does this too... */
@@ -179,7 +181,7 @@ int ata_task_ioctl(struct scsi_device *s
 	memset(scsi_cmd, 0, sizeof(scsi_cmd));
 	scsi_cmd[0]  = ATA_16;
 	scsi_cmd[1]  = (3 << 1); /* Non-data */
-	scsi_cmd[2]  = 0x1f;     /* no off.line or cc, yes all registers */
+	/* scsi_cmd[2] is already 0 -- no off.line, cc, or data xfer */
 	scsi_cmd[4]  = args[1];
 	scsi_cmd[6]  = args[2];
 	scsi_cmd[8]  = args[3];
Signature · 2 lines
John W. Linville
[email protected]
#2 lore raw

PATCH 3/5 libata: filter SET_FEATURES - XFER MODE from ATA pass thru

Filter-out attempts to issue a SET_FEATURES - XFER MODE command via the ATA pass thru mechanism.

---

 drivers/scsi/libata-scsi.c |   11 +++++++++++
 1 files changed, 11 insertions(+)
drivers/scsi/libata-scsi.c renamed from drivers/scsi/libata-scsi.c.filter +11 -0
@@ -1764,6 +1764,17 @@ ata_scsi_pass_thru(struct ata_queued_cmd
 	}
 
 	/*
+	 * Filter SET_FEATURES - XFER MODE command -- otherwise,
+	 * SET_FEATURES - XFER MODE must be preceded/succeeded
+	 * by an update to hardware-specific registers for each
+	 * controller (i.e. the reason for ->set_piomode(),
+	 * ->set_dmamode(), and ->post_set_mode() hooks).
+	 */
+	if ((tf->command == ATA_CMD_SET_FEATURES)
+	 && (tf->feature == SETFEATURES_XFER))
+		return 1;
+
+	/*
 	 * Set flags so that all registers will be written,
 	 * and pass on write indication (used for PIO/DMA
 	 * setup.)
Signature · 2 lines
John W. Linville
[email protected]
#3 lore raw

PATCH 4/5 libata: minor style changes in ata_scsi_pass_thru

Remove some curlies around single-line if statements.

---

 drivers/scsi/libata-scsi.c |    9 +++------
 1 files changed, 3 insertions(+), 6 deletions(-)
drivers/scsi/libata-scsi.c renamed from drivers/scsi/libata-scsi.c.style +3 -6
@@ -1713,9 +1713,8 @@ ata_scsi_pass_thru(struct ata_queued_cmd
 	struct ata_taskfile *tf = &(qc->tf);
 	struct scsi_cmnd *cmd = qc->scsicmd;
 
-	if ((tf->protocol = ata_scsi_map_proto(scsicmd[1])) == ATA_PROT_UNKNOWN) {
+	if ((tf->protocol = ata_scsi_map_proto(scsicmd[1])) == ATA_PROT_UNKNOWN)
 		return 1;
-	}
 
 	/*
 	 * 12 and 16 byte CDBs use different offsets to
@@ -1734,9 +1733,8 @@ ata_scsi_pass_thru(struct ata_queued_cmd
 			tf->hob_lbam = scsicmd[9];
 			tf->hob_lbah = scsicmd[11];
 			tf->flags |= ATA_TFLAG_LBA48 ;
-		} else {
+		} else
 			tf->flags &= ~ATA_TFLAG_LBA48 ;
-		}
 
 		/*
 		 * Always copy low byte, device and command registers.
@@ -1781,9 +1779,8 @@ ata_scsi_pass_thru(struct ata_queued_cmd
 	 */
 	tf->flags |= (ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE) ;
 
-	if (cmd->sc_data_direction == SCSI_DATA_WRITE) {
+	if (cmd->sc_data_direction == SCSI_DATA_WRITE)
 		tf->flags |= ATA_TFLAG_WRITE;
-	}
 
 	/*
 	 * Set transfer length.
Signature · 2 lines
John W. Linville
[email protected]
#4 lore raw

PATCH 5/5 libata: update ATA pass thru opcodes

Update ATA pass thru opcodes to match what is specified in the current ATA pass thru spec (T10/04-262r7).

---
This part was controversial before, because these opcodes seemed to overlap with some existing opcodes. So I contacted the author of the spec, Curtis Stevens: (quoted without permision)

"The command values in 04-262r7 are correct. There was an erroneous value published in the minutes that subsequently got published in SPC. This has been corrected.

Command sets are supposed to be qualified by the Inquiry Peripheral Device Type. I am a little concerned that some portions of linux depend on opcodes being the same for all PDTs. Using reusing opcodes will continue to happen more frequently as new applications emerge and new devices are created. The SCSI opcoded are already overloaded to the point that Service Actions have become necessary to extend the protocol."

Honestly, I don't know if that clears things up or not. But, Curtis seems pretty adamant that the values in the patch below (which match rev 7 of the spec) are correct. IMHO, at worst these values are no more wrong than the ones already there.

So, apply this now, or apply it later... :-)

 include/scsi/scsi.h |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)
include/scsi/scsi.h renamed from include/scsi/scsi.h.orig +3 -3
@@ -113,9 +113,9 @@ extern const char *const scsi_device_typ
 /* values for service action in */
 #define	SAI_READ_CAPACITY_16  0x10
 
-/* Temporary values for T10/04-262 until official values are allocated */
-#define	ATA_16		      0x85	/* 16-byte pass-thru [0x85 == unused]*/
-#define	ATA_12		      0xb3	/* 12-byte pass-thru [0xb3 == obsolete set limits command] */
+/* Values for T10/04-262r7 */
+#define	ATA_16		      0x85	/* 16-byte pass-thru */
+#define	ATA_12		      0xa1	/* 12-byte pass-thru */
 
 /*
  *  SCSI Architecture Model (SAM) Status codes. Taken from SAM-3 draft
Signature · 2 lines
John W. Linville
[email protected]