All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Stephen M. Cameron" <scameron@beardog.cce.hp.com>
To: axboe@kernel.dk
Cc: mikem@beardog.cce.hp.com, akpm@linux-foundation.org,
	thenzl@redhat.com, linux-kernel@vger.kernel.org,
	smcameron@yahoo.com
Subject: [PATCH 11/16] cciss: use new doorbell-bit-5 reset method
Date: Tue, 03 May 2011 14:53:46 -0500	[thread overview]
Message-ID: <20110503195346.5154.2503.stgit@beardog.cce.hp.com> (raw)
In-Reply-To: <20110503194919.5154.78352.stgit@beardog.cce.hp.com>

From: Stephen M. Cameron <scameron@beardog.cce.hp.com>

The bit-2-doorbell reset method seemed to cause (survivable) NMIs
on some systems and (unsurvivable) IOCK NMIs on some G7 servers.
Firmware guys implemented a new doorbell method to alleviate these
problems triggered by bit 5 of the doorbell register.  We want to
use it if it's available.

Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
---
 drivers/block/cciss.c     |   25 ++++++++++++++-----------
 drivers/block/cciss_cmd.h |    2 ++
 2 files changed, 16 insertions(+), 11 deletions(-)

diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index b9f8658..df58c59 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -4384,7 +4384,7 @@ static __devinit int cciss_message(struct pci_dev *pdev, unsigned char opcode, u
 #define cciss_noop(p) cciss_message(p, 3, 0)
 
 static int cciss_controller_hard_reset(struct pci_dev *pdev,
-	void * __iomem vaddr, bool use_doorbell)
+	void * __iomem vaddr, u32 use_doorbell)
 {
 	u16 pmcsr;
 	int pos;
@@ -4395,7 +4395,7 @@ static int cciss_controller_hard_reset(struct pci_dev *pdev,
 		 * other way using the doorbell register.
 		 */
 		dev_info(&pdev->dev, "using doorbell to reset controller\n");
-		writel(DOORBELL_CTLR_RESET, vaddr + SA5_DOORBELL);
+		writel(use_doorbell, vaddr + SA5_DOORBELL);
 		msleep(1000);
 	} else { /* Try to do it the PCI power state way */
 
@@ -4499,7 +4499,7 @@ static __devinit int cciss_kdump_hard_reset_controller(struct pci_dev *pdev)
 	u32 misc_fw_support;
 	int rc;
 	CfgTable_struct __iomem *cfgtable;
-	bool use_doorbell;
+	u32 use_doorbell;
 	u32 board_id;
 	u16 command_register;
 
@@ -4560,15 +4560,18 @@ static __devinit int cciss_kdump_hard_reset_controller(struct pci_dev *pdev)
 	if (rc)
 		goto unmap_vaddr;
 
-	/* If reset via doorbell register is supported, use that. */
-	misc_fw_support = readl(&cfgtable->misc_fw_support);
-	use_doorbell = misc_fw_support & MISC_FW_DOORBELL_RESET;
-
-	/* The doorbell reset seems to cause lockups on some Smart
-	 * Arrays (e.g. P410, P410i, maybe others).  Until this is
-	 * fixed or at least isolated, avoid the doorbell reset.
+	/* If reset via doorbell register is supported, use that.
+	 * There are two such methods.  Favor the newest method.
 	 */
-	use_doorbell = 0;
+	misc_fw_support = readl(&cfgtable->misc_fw_support);
+	use_doorbell = misc_fw_support & MISC_FW_DOORBELL_RESET2;
+	if (use_doorbell) {
+		use_doorbell = DOORBELL_CTLR_RESET2;
+	} else {
+		use_doorbell = misc_fw_support & MISC_FW_DOORBELL_RESET;
+		if (use_doorbell)
+			use_doorbell = DOORBELL_CTLR_RESET;
+	}
 
 	rc = cciss_controller_hard_reset(pdev, vaddr, use_doorbell);
 	if (rc)
diff --git a/drivers/block/cciss_cmd.h b/drivers/block/cciss_cmd.h
index 3b20c31..d9be6b4 100644
--- a/drivers/block/cciss_cmd.h
+++ b/drivers/block/cciss_cmd.h
@@ -53,6 +53,7 @@
 #define CFGTBL_ChangeReq        0x00000001l
 #define CFGTBL_AccCmds          0x00000001l
 #define DOORBELL_CTLR_RESET     0x00000004l
+#define DOORBELL_CTLR_RESET2    0x00000020l
 
 #define CFGTBL_Trans_Simple     0x00000002l
 #define CFGTBL_Trans_Performant 0x00000004l
@@ -243,6 +244,7 @@ typedef struct _CfgTable_struct {
   u8		   reserved[0x78 - 0x58];
   u32		   misc_fw_support; /* offset 0x78 */
 #define MISC_FW_DOORBELL_RESET (0x02)
+#define MISC_FW_DOORBELL_RESET2 (0x10)
 	u8	   driver_version[32];
 } CfgTable_struct;
 


  parent reply	other threads:[~2011-05-03 20:01 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-03 19:52 [PATCH 00/16] cciss: May 3, 2011 updates Stephen M. Cameron
2011-05-03 19:52 ` [PATCH 01/16] cciss: add readl after writel in interrupt mask setting code Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 02/16] cciss: do a better job of detecting controller reset failure Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 03/16] cciss: factor out command pool allocation functions Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 04/16] cciss: factor out scatterlist " Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 05/16] cciss: factor out irq request code Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 06/16] cciss: fix reply pool and block fetch table memory leaks Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 07/16] cciss: get rid of message related magic numbers Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 08/16] cciss: increase time to wait for board reset to start Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 09/16] cciss: clarify messages around reset behavior Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 10/16] cciss: increase timeouts for post-reset no-ops Stephen M. Cameron
2011-05-03 19:53 ` Stephen M. Cameron [this message]
2011-05-03 19:53 ` [PATCH 12/16] cciss: do soft reset if hard reset is broken Stephen M. Cameron
2011-05-03 19:53 ` [PATCH 13/16] cciss: remove superfluous sleeps around reset code Stephen M. Cameron
2011-05-03 19:54 ` [PATCH 14/16] cciss: do not attempt PCI power management reset method if we know it won't work Stephen M. Cameron
2011-05-03 19:54 ` [PATCH 15/16] cciss: do not use bit 2 doorbell reset Stephen M. Cameron
2011-05-03 19:54 ` [PATCH 16/16] cciss: add cciss_tape_cmds module paramter Stephen M. Cameron
2011-05-06 14:29 ` [PATCH 00/16] cciss: May 3, 2011 updates Jens Axboe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110503195346.5154.2503.stgit@beardog.cce.hp.com \
    --to=scameron@beardog.cce.hp.com \
    --cc=akpm@linux-foundation.org \
    --cc=axboe@kernel.dk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mikem@beardog.cce.hp.com \
    --cc=smcameron@yahoo.com \
    --cc=thenzl@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.