From d3301874083874f8a0ac88aa1bb7da6b62df34d2 Mon Sep 17 00:00:00 2001 From: Mike Anderson Date: Thu, 16 Jun 2005 11:12:38 -0700 Subject: [SCSI] host state model update: replace old host bitmap state Migrate the current SCSI host state model to a model like SCSI device is using. Signed-off-by: Mike Anderson Rejections fixed up and Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 0fc8b48f052b..e9c451ba71fc 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -75,7 +75,7 @@ int scsi_eh_scmd_add(struct scsi_cmnd *scmd, int eh_flag) scmd->eh_eflags |= eh_flag; list_add_tail(&scmd->eh_entry, &shost->eh_cmd_q); - set_bit(SHOST_RECOVERY, &shost->shost_state); + scsi_host_set_state(shost, SHOST_RECOVERY); shost->host_failed++; scsi_eh_wakeup(shost); spin_unlock_irqrestore(shost->host_lock, flags); @@ -197,7 +197,8 @@ int scsi_block_when_processing_errors(struct scsi_device *sdev) { int online; - wait_event(sdev->host->host_wait, (!test_bit(SHOST_RECOVERY, &sdev->host->shost_state))); + wait_event(sdev->host->host_wait, (sdev->host->shost_state != + SHOST_RECOVERY)); online = scsi_device_online(sdev); @@ -1458,7 +1459,7 @@ static void scsi_restart_operations(struct Scsi_Host *shost) SCSI_LOG_ERROR_RECOVERY(3, printk("%s: waking up host to restart\n", __FUNCTION__)); - clear_bit(SHOST_RECOVERY, &shost->shost_state); + scsi_host_set_state(shost, SHOST_RUNNING); wake_up(&shost->host_wait); -- cgit v1.2.1 From 33aa687db90dd8541bd5e9a762eebf880eaee767 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sun, 28 Aug 2005 11:31:14 -0500 Subject: [SCSI] convert SPI transport class to scsi_execute This one's slightly more difficult. The transport class uses REQ_FAILFAST, so another interface (scsi_execute) had to be invented to take the extra flag. Also, the sense functions are shifted around to allow spi_execute to place data directly into a struct scsi_sense_hdr. With this change, there's probably a lot of unnecessary sense buffer allocation going on which we can fix later. Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index e9c451ba71fc..2686d5672e5e 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -1847,12 +1847,16 @@ EXPORT_SYMBOL(scsi_reset_provider); int scsi_normalize_sense(const u8 *sense_buffer, int sb_len, struct scsi_sense_hdr *sshdr) { - if (!sense_buffer || !sb_len || (sense_buffer[0] & 0x70) != 0x70) + if (!sense_buffer || !sb_len) return 0; memset(sshdr, 0, sizeof(struct scsi_sense_hdr)); sshdr->response_code = (sense_buffer[0] & 0x7f); + + if (!scsi_sense_valid(sshdr)) + return 0; + if (sshdr->response_code >= 0x72) { /* * descriptor format -- cgit v1.2.1 From e47373ec1c9aab9ee134f4e2b8249957e9f4c7ef Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 30 Mar 2005 15:05:45 -0500 Subject: [SCSI] return success after retries in scsi_eh_tur The problem lies in the way the error handler uses TEST UNIT READY to tell whether error recovery has succeeded. The scsi_eh_tur function gives up after one round of retrying; after that it decides that more error recovery is needed. However TUR is liable to report sense data indicating a retry is needed when in fact error recovery has succeeded. A typical example might be SK=2, ASC=4, ASCQ=1 (Logical unit in process of becoming ready). The mere fact that we were able to get a sensible reply to the TUR should indicate that the device is working well enough to stop error recovery. I ran across a case back in January where this happened. A CD-ROM drive timed out the INQUIRY command, and a device reset fixed the blockage. But then the drive kept responding with 2/4/1 -- because it was spinning up I suppose -- until the error handler gave up and placed it offline. If the initial INQUIRY had received the 2/4/1 instead, everything would have worked okay. It doesn't seem reasonable for things to fail just because the error handler had started running. Signed-off-by: Alan Stern Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index e9c451ba71fc..688bce740786 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -776,9 +776,11 @@ retry_tur: __FUNCTION__, scmd, rtn)); if (rtn == SUCCESS) return 0; - else if (rtn == NEEDS_RETRY) + else if (rtn == NEEDS_RETRY) { if (retry_cnt--) goto retry_tur; + return 0; + } return 1; } -- cgit v1.2.1 From c5478def7a3a2dba9ceda452c2aa3539514d30a9 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 6 Sep 2005 14:04:26 +0200 Subject: [SCSI] switch EH thread startup to the kthread API Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 29 ++--------------------------- 1 file changed, 2 insertions(+), 27 deletions(-) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 688bce740786..ebe74ccb518a 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include @@ -1585,16 +1586,8 @@ int scsi_error_handler(void *data) int rtn; DECLARE_MUTEX_LOCKED(sem); - /* - * Flush resources - */ - - daemonize("scsi_eh_%d", shost->host_no); - current->flags |= PF_NOFREEZE; - shost->eh_wait = &sem; - shost->ehandler = current; /* * Wake up the thread that created us. @@ -1602,8 +1595,6 @@ int scsi_error_handler(void *data) SCSI_LOG_ERROR_RECOVERY(3, printk("Wake up parent of" " scsi_eh_%d\n",shost->host_no)); - complete(shost->eh_notify); - while (1) { /* * If we get a signal, it means we are supposed to go @@ -1624,7 +1615,7 @@ int scsi_error_handler(void *data) * semaphores isn't unreasonable. */ down_interruptible(&sem); - if (shost->eh_kill) + if (kthread_should_stop()) break; SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler" @@ -1663,22 +1654,6 @@ int scsi_error_handler(void *data) * Make sure that nobody tries to wake us up again. */ shost->eh_wait = NULL; - - /* - * Knock this down too. From this point on, the host is flying - * without a pilot. If this is because the module is being unloaded, - * that's fine. If the user sent a signal to this thing, we are - * potentially in real danger. - */ - shost->eh_active = 0; - shost->ehandler = NULL; - - /* - * If anyone is waiting for us to exit (i.e. someone trying to unload - * a driver), then wake up that process to let them know we are on - * the way out the door. - */ - complete_and_exit(shost->eh_notify, 0); return 0; } -- cgit v1.2.1 From fe1b2d544d71300f8e2d151c3c77a130d13a58be Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 6 Sep 2005 14:15:37 +0200 Subject: [SCSI] unexport scsi_add_timer/scsi_delete_timer Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'drivers/scsi/scsi_error.c') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index ebe74ccb518a..ae28bcb79242 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -116,7 +116,6 @@ void scsi_add_timer(struct scsi_cmnd *scmd, int timeout, add_timer(&scmd->eh_timeout); } -EXPORT_SYMBOL(scsi_add_timer); /** * scsi_delete_timer - Delete/cancel timer for a given function. @@ -144,7 +143,6 @@ int scsi_delete_timer(struct scsi_cmnd *scmd) return rtn; } -EXPORT_SYMBOL(scsi_delete_timer); /** * scsi_times_out - Timeout function for normal scsi commands. -- cgit v1.2.1