public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
To: linux-kernel@vger.kernel.org, linux-s390@vger.kernel.org
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>,
	Sebastian Ott <sebott@linux.vnet.ibm.com>,
	Martin Schwidefsky <schwidefsky@de.ibm.com>
Subject: [patch 05/32] [PATCH] cio: wait for channel report
Date: Wed, 24 Feb 2010 09:44:35 +0100	[thread overview]
Message-ID: <20100224084447.955310300@de.ibm.com> (raw)
In-Reply-To: 20100224084430.193562869@de.ibm.com

[-- Attachment #1: 104-cio-wait-channel-report.diff --]
[-- Type: text/plain, Size: 3408 bytes --]

From: Sebastian Ott <sebott@linux.vnet.ibm.com>

To fetch a pending channel report word (crw) we use a kernel
thread which triggers stcrw and sleeps on a semaphore. The s390
machine check handler uses crw_handle_channel_report to handle
one crw if needed.

This patch replaces the semaphore with a waitqueue (to block the
kernel thread) and an atomic_t (to count the number of pending
requests).

By this we achieve the ability to force this thread to check for
a pending crw (independent on when it is triggered by the machine
check handler) and wait for this action to finish.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---

 arch/s390/include/asm/crw.h |    1 +
 drivers/s390/cio/crw.c      |   29 +++++++++++++++--------------
 2 files changed, 16 insertions(+), 14 deletions(-)

Index: quilt-2.6/arch/s390/include/asm/crw.h
===================================================================
--- quilt-2.6.orig/arch/s390/include/asm/crw.h	2010-02-24 09:28:13.000000000 +0100
+++ quilt-2.6/arch/s390/include/asm/crw.h	2010-02-24 09:44:23.000000000 +0100
@@ -32,6 +32,7 @@
 extern int crw_register_handler(int rsc, crw_handler_t handler);
 extern void crw_unregister_handler(int rsc);
 extern void crw_handle_channel_report(void);
+void crw_wait_for_channel_report(void);
 
 #define NR_RSCS 16
 
Index: quilt-2.6/drivers/s390/cio/crw.c
===================================================================
--- quilt-2.6.orig/drivers/s390/cio/crw.c	2010-02-24 09:28:13.000000000 +0100
+++ quilt-2.6/drivers/s390/cio/crw.c	2010-02-24 09:44:23.000000000 +0100
@@ -8,15 +8,16 @@
  *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
  */
 
-#include <linux/semaphore.h>
 #include <linux/mutex.h>
 #include <linux/kthread.h>
 #include <linux/init.h>
+#include <linux/wait.h>
 #include <asm/crw.h>
 
-static struct semaphore crw_semaphore;
 static DEFINE_MUTEX(crw_handler_mutex);
 static crw_handler_t crw_handlers[NR_RSCS];
+static atomic_t crw_nr_req = ATOMIC_INIT(0);
+static DECLARE_WAIT_QUEUE_HEAD(crw_handler_wait_q);
 
 /**
  * crw_register_handler() - register a channel report word handler
@@ -59,12 +60,14 @@
 static int crw_collect_info(void *unused)
 {
 	struct crw crw[2];
-	int ccode;
+	int ccode, signal;
 	unsigned int chain;
-	int ignore;
 
 repeat:
-	ignore = down_interruptible(&crw_semaphore);
+	signal = wait_event_interruptible(crw_handler_wait_q,
+					  atomic_read(&crw_nr_req) > 0);
+	if (unlikely(signal))
+		atomic_inc(&crw_nr_req);
 	chain = 0;
 	while (1) {
 		crw_handler_t handler;
@@ -122,25 +125,23 @@
 		/* chain is always 0 or 1 here. */
 		chain = crw[chain].chn ? chain + 1 : 0;
 	}
+	if (atomic_dec_and_test(&crw_nr_req))
+		wake_up(&crw_handler_wait_q);
 	goto repeat;
 	return 0;
 }
 
 void crw_handle_channel_report(void)
 {
-	up(&crw_semaphore);
+	atomic_inc(&crw_nr_req);
+	wake_up(&crw_handler_wait_q);
 }
 
-/*
- * Separate initcall needed for semaphore initialization since
- * crw_handle_channel_report might be called before crw_machine_check_init.
- */
-static int __init crw_init_semaphore(void)
+void crw_wait_for_channel_report(void)
 {
-	init_MUTEX_LOCKED(&crw_semaphore);
-	return 0;
+	crw_handle_channel_report();
+	wait_event(crw_handler_wait_q, atomic_read(&crw_nr_req) == 0);
 }
-pure_initcall(crw_init_semaphore);
 
 /*
  * Machine checks for the channel subsystem must be enabled


  parent reply	other threads:[~2010-02-24  8:51 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-24  8:44 [patch 00/32] s390 patch queue for 2.6.34 Martin Schwidefsky
2010-02-24  8:44 ` [patch 01/32] [PATCH] uaccess: implement strict user copy checks Martin Schwidefsky
2010-02-24  8:44 ` [patch 02/32] [PATCH] adapt text to cu3088-removal Martin Schwidefsky
2010-02-24  8:44 ` [patch 03/32] [PATCH] cio: consolidate workqueues Martin Schwidefsky
2010-02-24  8:44 ` [patch 04/32] [PATCH] cio: introduce cio_settle Martin Schwidefsky
2010-02-24  8:44 ` Martin Schwidefsky [this message]
2010-02-24  8:44 ` [patch 06/32] [PATCH] cio: make wait_events interruptible Martin Schwidefsky
2010-02-24  8:44 ` [patch 07/32] [PATCH] use inline assembly contraints available with gcc 3.3.3 Martin Schwidefsky
2010-02-24  8:44 ` [patch 08/32] [PATCH] zfcpdump: remove cross arch dump support Martin Schwidefsky
2010-02-24  8:44 ` [patch 09/32] [PATCH] smp: rename and add lowcore defines Martin Schwidefsky
2010-02-24  8:44 ` [patch 10/32] [PATCH] smp: always reboot on cpu 0 Martin Schwidefsky
2010-02-24  8:44 ` [patch 11/32] [PATCH] qdio: account processed SBAL during queue scan Martin Schwidefsky
2010-02-24  8:44 ` [patch 12/32] [PATCH] qdio: optimize cache line usage of struct qdio_irq Martin Schwidefsky
2010-02-24  8:44 ` [patch 13/32] [PATCH] add MACHINE_IS_LPAR flag Martin Schwidefsky
2010-02-24  8:44 ` [patch 14/32] [PATCH] sysinfo: fix SYSIB 3,2,2 structure Martin Schwidefsky
2010-02-24  8:44 ` [patch 15/32] [PATCH] spinlock: check virtual cpu running status Martin Schwidefsky
2010-02-24  8:44 ` [patch 16/32] [PATCH] smp: rework sigp code Martin Schwidefsky
2010-02-24  8:44 ` [patch 17/32] [PATCH] Replace ENOTSUPP usage with EOPNOTSUPP Martin Schwidefsky
2010-02-24  8:44 ` [patch 18/32] [PATCH] free_initmem: reduce code duplication Martin Schwidefsky
2010-02-24  8:44 ` [patch 19/32] [PATCH] ccw_device_notify: improve return codes Martin Schwidefsky
2010-02-24  8:44 ` [patch 20/32] [PATCH] Cleanup struct _lowcore usage and defines Martin Schwidefsky
2010-02-24  8:44 ` [patch 21/32] [PATCH] bug: use relative pointers in bug table entries Martin Schwidefsky
2010-02-24  8:44 ` [patch 22/32] [PATCH] use kprobes_built_in() in mm/fault code Martin Schwidefsky
2010-02-24  8:44 ` [patch 23/32] [PATCH] dasd: fix online/offline race Martin Schwidefsky
2010-02-24  8:44 ` [patch 24/32] [PATCH] add z9-ec/z10 instruction to kernel disassembler Martin Schwidefsky
2010-02-24  8:44 ` [patch 25/32] [PATCH] seq_file: convert drivers/s390/ Martin Schwidefsky
2010-02-24  8:44 ` [patch 26/32] [PATCH] codepage conversion of kernel parameter line Martin Schwidefsky
2010-02-24  8:44 ` [patch 27/32] [PATCH] Define new s390 ELF note sections in elf.h Martin Schwidefsky
2010-02-24  8:44 ` [patch 28/32] [PATCH] add support for compressed kernels Martin Schwidefsky
2010-02-24  8:44 ` [patch 29/32] [PATCH] correct vdso version string Martin Schwidefsky
2010-02-24  8:45 ` [patch 30/32] [PATCH] dasd: fix refcounting Martin Schwidefsky
2010-02-24  8:45 ` [patch 31/32] [PATCH] dasd: correct offline processing Martin Schwidefsky
2010-02-24  8:45 ` [patch 32/32] [PATCH] cio: trigger subchannel event at resume time Martin Schwidefsky

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=20100224084447.955310300@de.ibm.com \
    --to=schwidefsky@de.ibm.com \
    --cc=heiko.carstens@de.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=sebott@linux.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