public inbox for linux-scsi@vger.kernel.org
 help / color / mirror / Atom feed
From: Christof Schmitt <christof.schmitt@de.ibm.com>
To: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: linux-scsi@vger.kernel.org, linux-s390@vger.kernel.org,
	schwidefsky@de.ibm.com, heiko.carstens@de.ibm.com,
	Swen Schillig <swen@vnet.ibm.com>,
	Christof Schmitt <christof.schmitt@de.ibm.com>
Subject: [patch 12/18] [PATCH] zfcp: Remove PCI flag
Date: Mon, 02 Mar 2009 13:09:05 +0100	[thread overview]
Message-ID: <20090302121003.642226000@de.ibm.com> (raw)
In-Reply-To: 20090302120853.279447000@de.ibm.com

[-- Attachment #1: q.diff --]
[-- Type: text/plain, Size: 3100 bytes --]

From: Christof Schmitt <christof.schmitt@de.ibm.com>

The usage of the PCI flag to trigger interrupts is optional. Even
without setting the flag, qdio still receives interrupts to continue
working on the queue.  Remove the PCI flag from zfcp, it is not
necessary.

Acked-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
--- a/drivers/s390/scsi/zfcp_qdio.c	2009-03-02 09:47:50.000000000 +0100
+++ b/drivers/s390/scsi/zfcp_qdio.c	2009-03-02 09:47:50.000000000 +0100
@@ -11,9 +11,6 @@
 
 #include "zfcp_ext.h"
 
-/* FIXME(tune): free space should be one max. SBAL chain plus what? */
-#define ZFCP_QDIO_PCI_INTERVAL	(QDIO_MAX_BUFFERS_PER_Q \
-				- (FSF_MAX_SBALS_PER_REQ + 4))
 #define QBUFF_PER_PAGE		(PAGE_SIZE / sizeof(struct qdio_buffer))
 
 static int zfcp_qdio_buffers_enqueue(struct qdio_buffer **sbal)
@@ -364,23 +361,12 @@ int zfcp_qdio_send(struct zfcp_fsf_req *
 	struct zfcp_qdio_queue *req_q = &adapter->req_q;
 	int first = fsf_req->sbal_first;
 	int count = fsf_req->sbal_number;
-	int retval, pci, pci_batch;
-	struct qdio_buffer_element *sbale;
-
-	/* acknowledgements for transferred buffers */
-	pci_batch = adapter->req_q_pci_batch + count;
-	if (unlikely(pci_batch >= ZFCP_QDIO_PCI_INTERVAL)) {
-		pci_batch %= ZFCP_QDIO_PCI_INTERVAL;
-		pci = first + count - (pci_batch + 1);
-		pci %= QDIO_MAX_BUFFERS_PER_Q;
-		sbale = zfcp_qdio_sbale(req_q, pci, 0);
-		sbale->flags |= SBAL_FLAGS0_PCI;
-	}
+	int retval;
+	unsigned int qdio_flags = QDIO_FLAG_SYNC_OUTPUT;
 
 	zfcp_qdio_account(adapter);
 
-	retval = do_QDIO(adapter->ccw_device, QDIO_FLAG_SYNC_OUTPUT, 0, first,
-			 count);
+	retval = do_QDIO(adapter->ccw_device, qdio_flags, 0, first, count);
 	if (unlikely(retval)) {
 		zfcp_qdio_zero_sbals(req_q->sbal, first, count);
 		return retval;
@@ -390,7 +376,6 @@ int zfcp_qdio_send(struct zfcp_fsf_req *
 	atomic_sub(count, &req_q->count);
 	req_q->first += count;
 	req_q->first %= QDIO_MAX_BUFFERS_PER_Q;
-	adapter->req_q_pci_batch = pci_batch;
 	return 0;
 }
 
@@ -461,7 +446,6 @@ void zfcp_qdio_close(struct zfcp_adapter
 	}
 	req_q->first = 0;
 	atomic_set(&req_q->count, 0);
-	adapter->req_q_pci_batch = 0;
 	adapter->resp_q.first = 0;
 	atomic_set(&adapter->resp_q.count, 0);
 }
@@ -499,7 +483,6 @@ int zfcp_qdio_open(struct zfcp_adapter *
 	/* set index of first avalable SBALS / number of available SBALS */
 	adapter->req_q.first = 0;
 	atomic_set(&adapter->req_q.count, QDIO_MAX_BUFFERS_PER_Q);
-	adapter->req_q_pci_batch = 0;
 
 	return 0;
 
--- a/drivers/s390/scsi/zfcp_def.h	2009-03-02 09:47:50.000000000 +0100
+++ b/drivers/s390/scsi/zfcp_def.h	2009-03-02 09:55:14.000000000 +0100
@@ -443,8 +443,6 @@ struct zfcp_adapter {
 	spinlock_t		req_list_lock;	   /* request list lock */
 	struct zfcp_qdio_queue	req_q;		   /* request queue */
 	spinlock_t		req_q_lock;	   /* for operations on queue */
-	int			req_q_pci_batch;   /* SBALs since PCI indication
-						      was last set */
 	ktime_t			req_q_time; /* time of last fill level change */
 	u64			req_q_util; /* for accounting */
 	spinlock_t		qdio_stat_lock;


  parent reply	other threads:[~2009-03-02 12:10 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-03-02 12:08 [patch 00/18] zfcp updates for 2.6.30 Christof Schmitt
2009-03-02 12:08 ` [patch 01/18] [PATCH] zfcp: Remove some port flags Christof Schmitt
2009-03-02 12:08 ` [patch 02/18] [PATCH] zfcp: Remove UNIT_REGISTERED status flag Christof Schmitt
2009-03-02 12:08 ` [patch 03/18] [Patch] zfcp: add measurement data for average qdio queue utilisation Christof Schmitt
2009-03-02 12:08 ` [patch 04/18] [PATCH] zfcp: Simplify latency lock handling Christof Schmitt
2009-03-02 12:08 ` [patch 05/18] [PATCH] zfcp: Only increment req_id for successfully issued requests Christof Schmitt
2009-03-02 12:08 ` [patch 06/18] [PATCH] zfcp: Wait for free SBALs when possible Christof Schmitt
2009-03-02 12:09 ` [patch 07/18] [PATCH] zfcp: Improve reliability of SCSI eh handlers in zfcp Christof Schmitt
2009-03-02 12:09 ` [patch 08/18] [PATCH] zfcp: Send ELS ADISC from workqueue Christof Schmitt
2009-03-02 12:09 ` [patch 09/18] [PATCH] zfcp: remove undefined subtype for status read response Christof Schmitt
2009-03-02 12:09 ` [patch 10/18] [PATCH] zfcp: prevent adapter close on initial adapter open Christof Schmitt
2009-03-02 12:09 ` [patch 11/18] [PATCH] zfcp: replace current ERP logging with a more convenient version Christof Schmitt
2009-03-02 12:09 ` Christof Schmitt [this message]
2009-03-02 12:09 ` [patch 13/18] [PATCH] zfcp: Report fc_host_port_type as NPIV Christof Schmitt
2009-03-02 12:09 ` [patch 14/18] [PATCH] zfcp: incorrect reaction on incoming RSCN Christof Schmitt
2009-03-02 12:09 ` [patch 15/18] [PATCH] zfcp: Block FC transport rports early on errors Christof Schmitt
2009-03-02 12:09 ` [patch 16/18] [PATCH] zfcp: erp failed status bit will not be set Christof Schmitt
2009-03-02 12:09 ` [patch 17/18] [PATCH] zfcp: fix queue, scheduled work processing Christof Schmitt
2009-03-02 12:09 ` [patch 18/18] [PATCH] zfcp: Ensure all work is cancelled on adapter dequeue Christof Schmitt

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=20090302121003.642226000@de.ibm.com \
    --to=christof.schmitt@de.ibm.com \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=heiko.carstens@de.ibm.com \
    --cc=linux-s390@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=schwidefsky@de.ibm.com \
    --cc=swen@vnet.ibm.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox