From 6f52ac29712f3eec192599249b12612360948646 Mon Sep 17 00:00:00 2001
From: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:33 +0100
Subject: [S390] cio: make sense id procedure work with partial hardware
 response

In some cases the current sense id procedure trips over incomplete
hardware responses. In these cases, checking against the preset value
of 0xFFFF is not enough. More critically, the VM DIAG call will always be
considered to have provided data after such an incident, even if it was not
successful at all.

The solution is to always initialize the control unit data before doing a
sense id call. Check the condition code before considering the control unit
data. And initialize again, before evaluating the VM data.

Signed-off-by: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/device_id.c | 107 ++++++++++++++++++++++++++-----------------
 1 file changed, 64 insertions(+), 43 deletions(-)

diff --git a/drivers/s390/cio/device_id.c b/drivers/s390/cio/device_id.c
index 918b8b89cf9..dc4d87f77f6 100644
--- a/drivers/s390/cio/device_id.c
+++ b/drivers/s390/cio/device_id.c
@@ -26,17 +26,18 @@
 #include "ioasm.h"
 #include "io_sch.h"
 
-/*
- * Input :
- *   devno - device number
- *   ps	   - pointer to sense ID data area
- * Output : none
+/**
+ * vm_vdev_to_cu_type - Convert vm virtual device into control unit type
+ *			for certain devices.
+ * @class: virtual device class
+ * @type: virtual device type
+ *
+ * Returns control unit type if a match was made or %0xffff otherwise.
  */
-static void
-VM_virtual_device_info (__u16 devno, struct senseid *ps)
+static int vm_vdev_to_cu_type(int class, int type)
 {
 	static struct {
-		int vrdcvcla, vrdcvtyp, cu_type;
+		int class, type, cu_type;
 	} vm_devices[] = {
 		{ 0x08, 0x01, 0x3480 },
 		{ 0x08, 0x02, 0x3430 },
@@ -68,8 +69,26 @@ VM_virtual_device_info (__u16 devno, struct senseid *ps)
 		{ 0x40, 0xc0, 0x5080 },
 		{ 0x80, 0x00, 0x3215 },
 	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(vm_devices); i++)
+		if (class == vm_devices[i].class && type == vm_devices[i].type)
+			return vm_devices[i].cu_type;
+
+	return 0xffff;
+}
+
+/**
+ * diag_get_dev_info - retrieve device information via DIAG X'210'
+ * @devno: device number
+ * @ps: pointer to sense ID data area
+ *
+ * Returns zero on success, non-zero otherwise.
+ */
+static int diag_get_dev_info(u16 devno, struct senseid *ps)
+{
 	struct diag210 diag_data;
-	int ccode, i;
+	int ccode;
 
 	CIO_TRACE_EVENT (4, "VMvdinf");
 
@@ -79,21 +98,21 @@ VM_virtual_device_info (__u16 devno, struct senseid *ps)
 	};
 
 	ccode = diag210 (&diag_data);
-	ps->reserved = 0xff;
+	if ((ccode == 0) || (ccode == 2)) {
+		ps->reserved = 0xff;
 
-	/* Special case for bloody osa devices. */
-	if (diag_data.vrdcvcla == 0x02 &&
-	    diag_data.vrdcvtyp == 0x20) {
-		ps->cu_type = 0x3088;
-		ps->cu_model = 0x60;
-		return;
-	}
-	for (i = 0; i < ARRAY_SIZE(vm_devices); i++)
-		if (diag_data.vrdcvcla == vm_devices[i].vrdcvcla &&
-		    diag_data.vrdcvtyp == vm_devices[i].vrdcvtyp) {
-			ps->cu_type = vm_devices[i].cu_type;
-			return;
+		/* Special case for osa devices. */
+		if (diag_data.vrdcvcla == 0x02 && diag_data.vrdcvtyp == 0x20) {
+			ps->cu_type = 0x3088;
+			ps->cu_model = 0x60;
+			return 0;
 		}
+		ps->cu_type = vm_vdev_to_cu_type(diag_data.vrdcvcla,
+						diag_data.vrdcvtyp);
+		if (ps->cu_type != 0xffff)
+			return 0;
+	}
+
 	CIO_MSG_EVENT(0, "DIAG X'210' for device %04X returned (cc = %d):"
 		      "vdev class : %02X, vdev type : %04X \n ...  "
 		      "rdev class : %02X, rdev type : %04X, "
@@ -102,6 +121,8 @@ VM_virtual_device_info (__u16 devno, struct senseid *ps)
 		      diag_data.vrdcvcla, diag_data.vrdcvtyp,
 		      diag_data.vrdcrccl, diag_data.vrdccrty,
 		      diag_data.vrdccrmd);
+
+	return -ENODEV;
 }
 
 /*
@@ -130,6 +151,7 @@ __ccw_device_sense_id_start(struct ccw_device *cdev)
 	/* Try on every path. */
 	ret = -ENODEV;
 	while (cdev->private->imask != 0) {
+		cdev->private->senseid.cu_type = 0xFFFF;
 		if ((sch->opm & cdev->private->imask) != 0 &&
 		    cdev->private->iretry > 0) {
 			cdev->private->iretry--;
@@ -153,7 +175,6 @@ ccw_device_sense_id_start(struct ccw_device *cdev)
 	int ret;
 
 	memset (&cdev->private->senseid, 0, sizeof (struct senseid));
-	cdev->private->senseid.cu_type = 0xFFFF;
 	cdev->private->imask = 0x80;
 	cdev->private->iretry = 5;
 	ret = __ccw_device_sense_id_start(cdev);
@@ -173,13 +194,7 @@ ccw_device_check_sense_id(struct ccw_device *cdev)
 
 	sch = to_subchannel(cdev->dev.parent);
 	irb = &cdev->private->irb;
-	/* Did we get a proper answer ? */
-	if (cdev->private->senseid.cu_type != 0xFFFF && 
-	    cdev->private->senseid.reserved == 0xFF) {
-		if (irb->scsw.count < sizeof (struct senseid) - 8)
-			cdev->private->flags.esid = 1;
-		return 0; /* Success */
-	}
+
 	/* Check the error cases. */
 	if (irb->scsw.fctl & (SCSW_FCTL_HALT_FUNC | SCSW_FCTL_CLEAR_FUNC)) {
 		/* Retry Sense ID if requested. */
@@ -231,6 +246,15 @@ ccw_device_check_sense_id(struct ccw_device *cdev)
 				      sch->schid.ssid, sch->schid.sch_no);
 		return -EACCES;
 	}
+
+	/* Did we get a proper answer ? */
+	if (irb->scsw.cc == 0 && cdev->private->senseid.cu_type != 0xFFFF &&
+	    cdev->private->senseid.reserved == 0xFF) {
+		if (irb->scsw.count < sizeof(struct senseid) - 8)
+			cdev->private->flags.esid = 1;
+		return 0; /* Success */
+	}
+
 	/* Hmm, whatever happened, try again. */
 	CIO_MSG_EVENT(2, "SenseID : start_IO() for device %04x on "
 		      "subchannel 0.%x.%04x returns status %02X%02X\n",
@@ -283,20 +307,17 @@ ccw_device_sense_id_irq(struct ccw_device *cdev, enum dev_event dev_event)
 			break;
 		/* fall through. */
 	default:		/* Sense ID failed. Try asking VM. */
-		if (MACHINE_IS_VM) {
-			VM_virtual_device_info (cdev->private->dev_id.devno,
+		if (MACHINE_IS_VM)
+			ret = diag_get_dev_info(cdev->private->dev_id.devno,
 						&cdev->private->senseid);
-			if (cdev->private->senseid.cu_type != 0xFFFF) {
-				/* Got the device information from VM. */
-				ccw_device_sense_id_done(cdev, 0);
-				return;
-			}
-		}
-		/*
-		 * If we can't couldn't identify the device type we
-		 *  consider the device "not operational".
-		 */
-		ccw_device_sense_id_done(cdev, -ENODEV);
+		else
+			/*
+			 * If we can't couldn't identify the device type we
+			 *  consider the device "not operational".
+			 */
+			ret = -ENODEV;
+
+		ccw_device_sense_id_done(cdev, ret);
 		break;
 	}
 }
-- 
cgit v1.2.3


From b9c9a21a7c8faeff1d13a23d2c57a5d4b512cfa0 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:34 +0100
Subject: [S390] cio: Clean up chsc response code handling.

This provides unified return codes for common response codes and
also makes the debug feature messages more similar and informational.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/chsc.c | 147 ++++++++++++++++++------------------------------
 1 file changed, 55 insertions(+), 92 deletions(-)

diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c
index e7ba16a74ef..007aaeb4f53 100644
--- a/drivers/s390/cio/chsc.c
+++ b/drivers/s390/cio/chsc.c
@@ -26,6 +26,25 @@
 
 static void *sei_page;
 
+static int chsc_error_from_response(int response)
+{
+	switch (response) {
+	case 0x0001:
+		return 0;
+	case 0x0002:
+	case 0x0003:
+	case 0x0006:
+	case 0x0007:
+	case 0x0008:
+	case 0x000a:
+		return -EINVAL;
+	case 0x0004:
+		return -EOPNOTSUPP;
+	default:
+		return -EIO;
+	}
+}
+
 struct chsc_ssd_area {
 	struct chsc_header request;
 	u16 :10;
@@ -75,11 +94,11 @@ int chsc_get_ssd_info(struct subchannel_id schid, struct chsc_ssd_info *ssd)
 		ret = (ccode == 3) ? -ENODEV : -EBUSY;
 		goto out_free;
 	}
-	if (ssd_area->response.code != 0x0001) {
+	ret = chsc_error_from_response(ssd_area->response.code);
+	if (ret != 0) {
 		CIO_MSG_EVENT(2, "chsc: ssd failed for 0.%x.%04x (rc=%04x)\n",
 			      schid.ssid, schid.sch_no,
 			      ssd_area->response.code);
-		ret = -EIO;
 		goto out_free;
 	}
 	if (!ssd_area->sch_valid) {
@@ -717,36 +736,15 @@ __chsc_do_secm(struct channel_subsystem *css, int enable, void *page)
 		return (ccode == 3) ? -ENODEV : -EBUSY;
 
 	switch (secm_area->response.code) {
-	case 0x0001: /* Success. */
-		ret = 0;
-		break;
-	case 0x0003: /* Invalid block. */
-	case 0x0007: /* Invalid format. */
-	case 0x0008: /* Other invalid block. */
-		CIO_CRW_EVENT(2, "Error in chsc request block!\n");
-		ret = -EINVAL;
-		break;
-	case 0x0004: /* Command not provided in model. */
-		CIO_CRW_EVENT(2, "Model does not provide secm\n");
-		ret = -EOPNOTSUPP;
-		break;
-	case 0x0102: /* cub adresses incorrect */
-		CIO_CRW_EVENT(2, "Invalid addresses in chsc request block\n");
-		ret = -EINVAL;
-		break;
-	case 0x0103: /* key error */
-		CIO_CRW_EVENT(2, "Access key error in secm\n");
+	case 0x0102:
+	case 0x0103:
 		ret = -EINVAL;
-		break;
-	case 0x0105: /* error while starting */
-		CIO_CRW_EVENT(2, "Error while starting channel measurement\n");
-		ret = -EIO;
-		break;
 	default:
-		CIO_CRW_EVENT(2, "Unknown CHSC response %d\n",
-			      secm_area->response.code);
-		ret = -EIO;
+		ret = chsc_error_from_response(secm_area->response.code);
 	}
+	if (ret != 0)
+		CIO_CRW_EVENT(2, "chsc: secm failed (rc=%04x)\n",
+			      secm_area->response.code);
 	return ret;
 }
 
@@ -827,27 +825,14 @@ int chsc_determine_channel_path_description(struct chp_id chpid,
 		goto out;
 	}
 
-	switch (scpd_area->response.code) {
-	case 0x0001: /* Success. */
+	ret = chsc_error_from_response(scpd_area->response.code);
+	if (ret == 0)
+		/* Success. */
 		memcpy(desc, &scpd_area->desc,
 		       sizeof(struct channel_path_desc));
-		ret = 0;
-		break;
-	case 0x0003: /* Invalid block. */
-	case 0x0007: /* Invalid format. */
-	case 0x0008: /* Other invalid block. */
-		CIO_CRW_EVENT(2, "Error in chsc request block!\n");
-		ret = -EINVAL;
-		break;
-	case 0x0004: /* Command not provided in model. */
-		CIO_CRW_EVENT(2, "Model does not provide scpd\n");
-		ret = -EOPNOTSUPP;
-		break;
-	default:
-		CIO_CRW_EVENT(2, "Unknown CHSC response %d\n",
+	else
+		CIO_CRW_EVENT(2, "chsc: scpd failed (rc=%04x)\n",
 			      scpd_area->response.code);
-		ret = -EIO;
-	}
 out:
 	free_page((unsigned long)scpd_area);
 	return ret;
@@ -923,8 +908,9 @@ int chsc_get_channel_measurement_chars(struct channel_path *chp)
 		goto out;
 	}
 
-	switch (scmc_area->response.code) {
-	case 0x0001: /* Success. */
+	ret = chsc_error_from_response(scmc_area->response.code);
+	if (ret == 0) {
+		/* Success. */
 		if (!scmc_area->not_valid) {
 			chp->cmg = scmc_area->cmg;
 			chp->shared = scmc_area->shared;
@@ -935,22 +921,9 @@ int chsc_get_channel_measurement_chars(struct channel_path *chp)
 			chp->cmg = -1;
 			chp->shared = -1;
 		}
-		ret = 0;
-		break;
-	case 0x0003: /* Invalid block. */
-	case 0x0007: /* Invalid format. */
-	case 0x0008: /* Invalid bit combination. */
-		CIO_CRW_EVENT(2, "Error in chsc request block!\n");
-		ret = -EINVAL;
-		break;
-	case 0x0004: /* Command not provided. */
-		CIO_CRW_EVENT(2, "Model does not provide scmc\n");
-		ret = -EOPNOTSUPP;
-		break;
-	default:
-		CIO_CRW_EVENT(2, "Unknown CHSC response %d\n",
+	} else {
+		CIO_CRW_EVENT(2, "chsc: scmc failed (rc=%04x)\n",
 			      scmc_area->response.code);
-		ret = -EIO;
 	}
 out:
 	free_page((unsigned long)scmc_area);
@@ -1002,21 +975,17 @@ chsc_enable_facility(int operation_code)
 		ret = (ret == 3) ? -ENODEV : -EBUSY;
 		goto out;
 	}
+
 	switch (sda_area->response.code) {
-	case 0x0001: /* everything ok */
-		ret = 0;
-		break;
-	case 0x0003: /* invalid request block */
-	case 0x0007:
-		ret = -EINVAL;
-		break;
-	case 0x0004: /* command not provided */
-	case 0x0101: /* facility not provided */
+	case 0x0101:
 		ret = -EOPNOTSUPP;
 		break;
-	default: /* something went wrong */
-		ret = -EIO;
+	default:
+		ret = chsc_error_from_response(sda_area->response.code);
 	}
+	if (ret != 0)
+		CIO_CRW_EVENT(2, "chsc: sda (oc=%x) failed (rc=%04x)\n",
+			      operation_code, sda_area->response.code);
  out:
 	free_page((unsigned long)sda_area);
 	return ret;
@@ -1041,33 +1010,27 @@ chsc_determine_css_characteristics(void)
 	} __attribute__ ((packed)) *scsc_area;
 
 	scsc_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
-	if (!scsc_area) {
-		CIO_MSG_EVENT(0, "Was not able to determine available "
-			      "CHSCs due to no memory.\n");
+	if (!scsc_area)
 		return -ENOMEM;
-	}
 
 	scsc_area->request.length = 0x0010;
 	scsc_area->request.code = 0x0010;
 
 	result = chsc(scsc_area);
 	if (result) {
-		CIO_MSG_EVENT(0, "Was not able to determine available CHSCs, "
-			      "cc=%i.\n", result);
-		result = -EIO;
+		result = (result == 3) ? -ENODEV : -EBUSY;
 		goto exit;
 	}
 
-	if (scsc_area->response.code != 1) {
-		CIO_MSG_EVENT(0, "Was not able to determine "
-			      "available CHSCs.\n");
-		result = -EIO;
-		goto exit;
-	}
-	memcpy(&css_general_characteristics, scsc_area->general_char,
-	       sizeof(css_general_characteristics));
-	memcpy(&css_chsc_characteristics, scsc_area->chsc_char,
-	       sizeof(css_chsc_characteristics));
+	result = chsc_error_from_response(scsc_area->response.code);
+	if (result == 0) {
+		memcpy(&css_general_characteristics, scsc_area->general_char,
+		       sizeof(css_general_characteristics));
+		memcpy(&css_chsc_characteristics, scsc_area->chsc_char,
+		       sizeof(css_chsc_characteristics));
+	} else
+		CIO_CRW_EVENT(2, "chsc: scsc failed (rc=%04x)\n",
+			      scsc_area->response.code);
 exit:
 	free_page ((unsigned long) scsc_area);
 	return result;
-- 
cgit v1.2.3


From 2fffc9355e6240466d1af764b0dcdede52085f7c Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:35 +0100
Subject: [S390] cio: Update documentation.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 Documentation/DocBook/s390-drivers.tmpl | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/Documentation/DocBook/s390-drivers.tmpl b/Documentation/DocBook/s390-drivers.tmpl
index 3d2f31b99dd..4acc73240a6 100644
--- a/Documentation/DocBook/s390-drivers.tmpl
+++ b/Documentation/DocBook/s390-drivers.tmpl
@@ -59,7 +59,7 @@
    <title>Introduction</title>
   <para>
     This document describes the interfaces available for device drivers that
-    drive s390 based channel attached devices. This includes interfaces for
+    drive s390 based channel attached I/O devices. This includes interfaces for
     interaction with the hardware and interfaces for interacting with the
     common driver core. Those interfaces are provided by the s390 common I/O
     layer.
@@ -86,9 +86,10 @@
 	The ccw bus typically contains the majority of devices available to
 	a s390 system. Named after the channel command word (ccw), the basic
 	command structure used to address its devices, the ccw bus contains
-	so-called channel attached devices. They are addressed via subchannels,
-	visible on the css bus. A device driver, however, will never interact
-	with the subchannel directly, but only via the device on the ccw bus,
+	so-called channel attached devices. They are addressed via I/O
+	subchannels, visible on the css bus. A device driver for
+	channel-attached devices, however, will never interact	with the
+	subchannel directly, but only via the I/O device on the ccw bus,
 	the ccw device.
   </para>
     <sect1 id="channelIO">
@@ -116,7 +117,6 @@
 !Iinclude/asm-s390/ccwdev.h
 !Edrivers/s390/cio/device.c
 !Edrivers/s390/cio/device_ops.c
-!Edrivers/s390/cio/airq.c
     </sect1>
     <sect1 id="cmf">
      <title>The channel-measurement facility</title>
@@ -147,4 +147,15 @@
    </sect1>
   </chapter>
 
+  <chapter id="genericinterfaces">
+   <title>Generic interfaces</title>
+  <para>
+	Some interfaces are available to other drivers that do not necessarily
+	have anything to do with the busses described above, but still are
+	indirectly using basic infrastructure in the common I/O layer.
+	One example is the support for adapter interrupts.
+  </para>
+!Edrivers/s390/cio/airq.c
+  </chapter>
+
 </book>
-- 
cgit v1.2.3


From 01bc8ad165490458a8feb744c8f401c1a7098e3a Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:36 +0100
Subject: [S390] cio: Add shutdown callback for ccwgroup.

This intendeds to make proper shutdown of qeth devices easier.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/ccwgroup.c | 12 ++++++++++++
 include/asm-s390/ccwgroup.h |  2 ++
 2 files changed, 14 insertions(+)

diff --git a/drivers/s390/cio/ccwgroup.c b/drivers/s390/cio/ccwgroup.c
index 3964056a9a4..03914fa8117 100644
--- a/drivers/s390/cio/ccwgroup.c
+++ b/drivers/s390/cio/ccwgroup.c
@@ -391,12 +391,24 @@ ccwgroup_remove (struct device *dev)
 	return 0;
 }
 
+static void ccwgroup_shutdown(struct device *dev)
+{
+	struct ccwgroup_device *gdev;
+	struct ccwgroup_driver *gdrv;
+
+	gdev = to_ccwgroupdev(dev);
+	gdrv = to_ccwgroupdrv(dev->driver);
+	if (gdrv && gdrv->shutdown)
+		gdrv->shutdown(gdev);
+}
+
 static struct bus_type ccwgroup_bus_type = {
 	.name   = "ccwgroup",
 	.match  = ccwgroup_bus_match,
 	.uevent = ccwgroup_uevent,
 	.probe  = ccwgroup_probe,
 	.remove = ccwgroup_remove,
+	.shutdown = ccwgroup_shutdown,
 };
 
 /**
diff --git a/include/asm-s390/ccwgroup.h b/include/asm-s390/ccwgroup.h
index 7109c7cab87..289053ef5e6 100644
--- a/include/asm-s390/ccwgroup.h
+++ b/include/asm-s390/ccwgroup.h
@@ -37,6 +37,7 @@ struct ccwgroup_device {
  * @remove: function called on remove
  * @set_online: function called when device is set online
  * @set_offline: function called when device is set offline
+ * @shutdown: function called when device is shut down
  * @driver: embedded driver structure
  */
 struct ccwgroup_driver {
@@ -49,6 +50,7 @@ struct ccwgroup_driver {
 	void (*remove) (struct ccwgroup_device *);
 	int (*set_online) (struct ccwgroup_device *);
 	int (*set_offline) (struct ccwgroup_device *);
+	void (*shutdown)(struct ccwgroup_device *);
 
 	struct device_driver driver;
 };
-- 
cgit v1.2.3


From 2485579bf5d3ea30d39b251defa1620ad77168bd Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:37 +0100
Subject: [S390] DEBUG_PAGEALLOC support for s390.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig.debug       |  8 ++++++++
 arch/s390/kernel/traps.c      |  5 ++++-
 arch/s390/mm/init.c           | 27 +++++++++++++++++++++++++++
 include/asm-s390/cacheflush.h |  4 ++++
 4 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/arch/s390/Kconfig.debug b/arch/s390/Kconfig.debug
index 2283933a9a9..4599fa06bd8 100644
--- a/arch/s390/Kconfig.debug
+++ b/arch/s390/Kconfig.debug
@@ -6,4 +6,12 @@ config TRACE_IRQFLAGS_SUPPORT
 
 source "lib/Kconfig.debug"
 
+config DEBUG_PAGEALLOC
+	bool "Debug page memory allocations"
+	depends on DEBUG_KERNEL
+	help
+	  Unmap pages from the kernel linear mapping after free_pages().
+	  This results in a slowdown, but helps to find certain types of
+	  memory corruptions.
+
 endmenu
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 52b8342c6bf..1a2fdb6991d 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -271,7 +271,10 @@ void die(const char * str, struct pt_regs * regs, long err)
 	printk("PREEMPT ");
 #endif
 #ifdef CONFIG_SMP
-	printk("SMP");
+	printk("SMP ");
+#endif
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	printk("DEBUG_PAGEALLOC");
 #endif
 	printk("\n");
 	notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV);
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index b234bb4a6da..983ec6ec0e7 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -167,6 +167,33 @@ void __init mem_init(void)
 	       PFN_ALIGN((unsigned long)&_eshared) - 1);
 }
 
+#ifdef CONFIG_DEBUG_PAGEALLOC
+void kernel_map_pages(struct page *page, int numpages, int enable)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long address;
+	int i;
+
+	for (i = 0; i < numpages; i++) {
+		address = page_to_phys(page + i);
+		pgd = pgd_offset_k(address);
+		pud = pud_offset(pgd, address);
+		pmd = pmd_offset(pud, address);
+		pte = pte_offset_kernel(pmd, address);
+		if (!enable) {
+			ptep_invalidate(address, pte);
+			continue;
+		}
+		*pte = mk_pte_phys(address, __pgprot(_PAGE_TYPE_RW));
+		/* Flush cpu write queue. */
+		mb();
+	}
+}
+#endif
+
 void free_initmem(void)
 {
         unsigned long addr;
diff --git a/include/asm-s390/cacheflush.h b/include/asm-s390/cacheflush.h
index f7cade8083f..49d5af916d0 100644
--- a/include/asm-s390/cacheflush.h
+++ b/include/asm-s390/cacheflush.h
@@ -24,4 +24,8 @@
 #define copy_from_user_page(vma, page, vaddr, dst, src, len) \
 	memcpy(dst, src, len)
 
+#ifdef CONFIG_DEBUG_PAGEALLOC
+void kernel_map_pages(struct page *page, int numpages, int enable);
+#endif
+
 #endif /* _S390_CACHEFLUSH_H */
-- 
cgit v1.2.3


From a817a61f85c75181bde6c3d83ae11a24b089dd6a Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:38 +0100
Subject: [S390] Fix linker script.

Fixes this warning:
vmlinux: warning: allocated section `.text' not in segment

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/vmlinux.lds.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index 7d43c3cd3ef..b4607155e8d 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -35,7 +35,7 @@ SECTIONS
 		KPROBES_TEXT
 		*(.fixup)
 		*(.gnu.warning)
-	} = 0x0700
+	} :text = 0x0700
 
 	_etext = .;		/* End of text section */
 
-- 
cgit v1.2.3


From 37c5f719e71882b759fa8acbdd11d5ca3a7965bb Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:39 +0100
Subject: [S390] Fix smp_call_function_mask semantics.

Make sure func isn't called on the local cpu just like on all other
architectures that implement this function.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index aa37fa15451..f5046fd4930 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -225,12 +225,11 @@ EXPORT_SYMBOL(smp_call_function_single);
  * You must not call this function with disabled interrupts or from a
  * hardware interrupt handler or from a bottom half handler.
  */
-int
-smp_call_function_mask(cpumask_t mask,
-			void (*func)(void *), void *info,
-			int wait)
+int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
+			   int wait)
 {
 	preempt_disable();
+	cpu_clear(smp_processor_id(), mask);
 	__smp_call_function_map(func, info, 0, wait, mask);
 	preempt_enable();
 	return 0;
-- 
cgit v1.2.3


From 2bc89b5ece48dc888734e8760ba5ad8566431912 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:40 +0100
Subject: [S390] Fix couple of section mismatches.

Fix couple of section mismatches. And since we touch the code
anyway change the IPL code to use C99 initializers.

Cc: Michael Holzheu <holzheu@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/entry.S   |  7 ++-----
 arch/s390/kernel/entry64.S |  7 ++-----
 arch/s390/kernel/ipl.c     | 27 ++++++++++++++++++---------
 arch/s390/kernel/setup.c   |  2 +-
 arch/s390/kernel/smp.c     |  6 +++---
 arch/s390/mm/vmem.c        |  2 +-
 6 files changed, 27 insertions(+), 24 deletions(-)

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 1a6dac8df6f..6766e37fe8e 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -11,6 +11,7 @@
 
 #include <linux/sys.h>
 #include <linux/linkage.h>
+#include <linux/init.h>
 #include <asm/cache.h>
 #include <asm/lowcore.h>
 #include <asm/errno.h>
@@ -830,9 +831,7 @@ mcck_return:
  * Restart interruption handler, kick starter for additional CPUs
  */
 #ifdef CONFIG_SMP
-#ifndef CONFIG_HOTPLUG_CPU
-	.section .init.text,"ax"
-#endif
+	__CPUINIT
 	.globl restart_int_handler
 restart_int_handler:
 	l	%r15,__LC_SAVE_AREA+60	# load ksp
@@ -845,9 +844,7 @@ restart_int_handler:
 	br	%r14			# branch to start_secondary
 restart_addr:
 	.long	start_secondary
-#ifndef CONFIG_HOTPLUG_CPU
 	.previous
-#endif
 #else
 /*
  * If we do not run with SMP enabled, let the new CPU crash ...
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index a3e47b893f0..efde6e178f6 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -11,6 +11,7 @@
 
 #include <linux/sys.h>
 #include <linux/linkage.h>
+#include <linux/init.h>
 #include <asm/cache.h>
 #include <asm/lowcore.h>
 #include <asm/errno.h>
@@ -801,9 +802,7 @@ mcck_return:
  * Restart interruption handler, kick starter for additional CPUs
  */
 #ifdef CONFIG_SMP
-#ifndef CONFIG_HOTPLUG_CPU
-	.section .init.text,"ax"
-#endif
+	__CPUINIT
 	.globl restart_int_handler
 restart_int_handler:
 	lg	%r15,__LC_SAVE_AREA+120 # load ksp
@@ -814,9 +813,7 @@ restart_int_handler:
 	lmg	%r6,%r15,__SF_GPRS(%r15) # load registers from clone
 	stosm	__SF_EMPTY(%r15),0x04	# now we can turn dat on
 	jg	start_secondary
-#ifndef CONFIG_HOTPLUG_CPU
 	.previous
-#endif
 #else
 /*
  * If we do not run with SMP enabled, let the new CPU crash ...
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index db28cca81fe..60acdc266db 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -439,7 +439,7 @@ static void ipl_run(struct shutdown_trigger *trigger)
 		reipl_ccw_dev(&ipl_info.data.ccw.dev_id);
 }
 
-static int ipl_init(void)
+static int __init ipl_init(void)
 {
 	int rc;
 
@@ -471,8 +471,11 @@ out:
 	return 0;
 }
 
-static struct shutdown_action ipl_action = {SHUTDOWN_ACTION_IPL_STR, ipl_run,
-					    ipl_init};
+static struct shutdown_action __refdata ipl_action = {
+	.name	= SHUTDOWN_ACTION_IPL_STR,
+	.fn	= ipl_run,
+	.init	= ipl_init,
+};
 
 /*
  * reipl shutdown action: Reboot Linux on shutdown.
@@ -792,7 +795,7 @@ static int __init reipl_fcp_init(void)
 	return 0;
 }
 
-static int reipl_init(void)
+static int __init reipl_init(void)
 {
 	int rc;
 
@@ -819,8 +822,11 @@ static int reipl_init(void)
 	return 0;
 }
 
-static struct shutdown_action reipl_action = {SHUTDOWN_ACTION_REIPL_STR,
-					      reipl_run, reipl_init};
+static struct shutdown_action __refdata reipl_action = {
+	.name	= SHUTDOWN_ACTION_REIPL_STR,
+	.fn	= reipl_run,
+	.init	= reipl_init,
+};
 
 /*
  * dump shutdown action: Dump Linux on shutdown.
@@ -998,7 +1004,7 @@ static int __init dump_fcp_init(void)
 	return 0;
 }
 
-static int dump_init(void)
+static int __init dump_init(void)
 {
 	int rc;
 
@@ -1020,8 +1026,11 @@ static int dump_init(void)
 	return 0;
 }
 
-static struct shutdown_action dump_action = {SHUTDOWN_ACTION_DUMP_STR,
-					     dump_run, dump_init};
+static struct shutdown_action __refdata dump_action = {
+	.name	= SHUTDOWN_ACTION_DUMP_STR,
+	.fn	= dump_run,
+	.init	= dump_init,
+};
 
 /*
  * vmcmd shutdown action: Trigger vm command on shutdown.
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 766c783bd7a..2d266f45e73 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -77,7 +77,7 @@ unsigned long machine_flags = 0;
 unsigned long elf_hwcap = 0;
 char elf_platform[ELF_PLATFORM_SIZE];
 
-struct mem_chunk __initdata memory_chunk[MEMORY_CHUNKS];
+struct mem_chunk __meminitdata memory_chunk[MEMORY_CHUNKS];
 volatile int __cpu_logical_map[NR_CPUS]; /* logical cpu to cpu address */
 static unsigned long __initdata memory_end;
 
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index f5046fd4930..85060659fb1 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -1007,7 +1007,7 @@ static struct notifier_block __cpuinitdata smp_cpu_nb = {
 	.notifier_call = smp_cpu_notify,
 };
 
-static int smp_add_present_cpu(int cpu)
+static int __devinit smp_add_present_cpu(int cpu)
 {
 	struct cpu *c = &per_cpu(cpu_devices, cpu);
 	struct sys_device *s = &c->sysdev;
@@ -1035,8 +1035,8 @@ out:
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static ssize_t rescan_store(struct sys_device *dev, const char *buf,
-			    size_t count)
+static ssize_t __ref rescan_store(struct sys_device *dev,
+				  const char *buf, size_t count)
 {
 	cpumask_t newcpus;
 	int cpu;
diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index 79d13a166a3..07e04677724 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -62,7 +62,7 @@ void __meminit memmap_init(unsigned long size, int nid, unsigned long zone,
 	}
 }
 
-static void __init_refok *vmem_alloc_pages(unsigned int order)
+static void __ref *vmem_alloc_pages(unsigned int order)
 {
 	if (slab_is_available())
 		return (void *)__get_free_pages(GFP_KERNEL, order);
-- 
cgit v1.2.3


From 8c0933eeb701eb8f526d88b1915af7bb35748e7b Mon Sep 17 00:00:00 2001
From: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:41 +0100
Subject: [S390] console: allow vt220 console to be the only console

Fix console detection logic to support configurations in which the
vt220 console is the only available Linux console.

Signed-off-by: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 2d266f45e73..72d20f70f8a 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -145,7 +145,7 @@ __setup("condev=", condev_setup);
 
 static int __init conmode_setup(char *str)
 {
-#if defined(CONFIG_SCLP_CONSOLE)
+#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
 	if (strncmp(str, "hwc", 4) == 0 || strncmp(str, "sclp", 5) == 0)
                 SET_CONSOLE_SCLP;
 #endif
@@ -183,7 +183,7 @@ static void __init conmode_default(void)
 		 */
 		cpcmd("TERM CONMODE 3215", NULL, 0, NULL);
 		if (ptr == NULL) {
-#if defined(CONFIG_SCLP_CONSOLE)
+#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
 			SET_CONSOLE_SCLP;
 #endif
 			return;
@@ -193,7 +193,7 @@ static void __init conmode_default(void)
 			SET_CONSOLE_3270;
 #elif defined(CONFIG_TN3215_CONSOLE)
 			SET_CONSOLE_3215;
-#elif defined(CONFIG_SCLP_CONSOLE)
+#elif defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
 			SET_CONSOLE_SCLP;
 #endif
 		} else if (strncmp(ptr + 8, "3215", 4) == 0) {
@@ -201,7 +201,7 @@ static void __init conmode_default(void)
 			SET_CONSOLE_3215;
 #elif defined(CONFIG_TN3270_CONSOLE)
 			SET_CONSOLE_3270;
-#elif defined(CONFIG_SCLP_CONSOLE)
+#elif defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
 			SET_CONSOLE_SCLP;
 #endif
 		}
@@ -212,7 +212,7 @@ static void __init conmode_default(void)
 		SET_CONSOLE_3270;
 #endif
 	} else {
-#if defined(CONFIG_SCLP_CONSOLE)
+#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
 		SET_CONSOLE_SCLP;
 #endif
 	}
-- 
cgit v1.2.3


From b6b40c532a36f91df9c21caf9baba3b635e99d4e Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:42 +0100
Subject: [S390] Define GENERIC_LOCKBREAK.

Fix compile error:

  CC      arch/s390/kernel/asm-offsets.s
In file included from
arch/s390/kernel/asm-offsets.c:7:
include/linux/sched.h: In function 'spin_needbreak':
include/linux/sched.h:1931: error: implicit declaration of function '__raw_spin_is_contended'
make[2]: *** [arch/s390/kernel/asm-offsets.s] Error 1

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 82cbffd0365..f0e7ccf1b20 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -47,6 +47,11 @@ config NO_IOMEM
 config NO_DMA
 	def_bool y
 
+config GENERIC_LOCKBREAK
+	bool
+	default y
+	depends on SMP && PREEMPT
+
 mainmenu "Linux Kernel Configuration"
 
 config S390
-- 
cgit v1.2.3


From 0abbf05cdd69d74f92628bf444cd210ba046f6eb Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:43 +0100
Subject: [S390] Cleanup & optimize bitops.

The bitops header is now a bit shorter and easier to understand since
it uses less inline assembly. It requires some tricks to persuade the
compiler to generate decent code. The ffz/ffs functions now use the
_zb_findmap/_sb_findmap table as well.
With this cleanup the new bitops for ext4 can be implemented with a
few lines, instead of another large inline assembly.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/bitops.h | 515 ++++++++++++++++++++--------------------------
 1 file changed, 219 insertions(+), 296 deletions(-)

diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h
index dba6fecad0b..95d93955a9b 100644
--- a/include/asm-s390/bitops.h
+++ b/include/asm-s390/bitops.h
@@ -440,242 +440,256 @@ __constant_test_bit(unsigned long nr, const volatile unsigned long *addr) {
  __test_bit((nr),(addr)) )
 
 /*
- * ffz = Find First Zero in word. Undefined if no zero exists,
- * so code should check against ~0UL first..
+ * Optimized find bit helper functions.
  */
-static inline unsigned long ffz(unsigned long word)
+
+/**
+ * __ffz_word_loop - find byte offset of first long != -1UL
+ * @addr: pointer to array of unsigned long
+ * @size: size of the array in bits
+ */
+static inline unsigned long __ffz_word_loop(const unsigned long *addr,
+					    unsigned long size)
+{
+	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
+	unsigned long bytes = 0;
+
+	asm volatile(
+#ifndef __s390x__
+		"	ahi	%1,31\n"
+		"	srl	%1,5\n"
+		"0:	c	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,4(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#else
+		"	aghi	%1,63\n"
+		"	srlg	%1,%1,6\n"
+		"0:	cg	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,8(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#endif
+		: "+a" (bytes), "+d" (size)
+		: "d" (-1UL), "a" (addr), "m" (*(addrtype *) addr)
+		: "cc" );
+	return bytes;
+}
+
+/**
+ * __ffs_word_loop - find byte offset of first long != 0UL
+ * @addr: pointer to array of unsigned long
+ * @size: size of the array in bits
+ */
+static inline unsigned long __ffs_word_loop(const unsigned long *addr,
+					    unsigned long size)
 {
-        unsigned long bit = 0;
+	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
+	unsigned long bytes = 0;
 
+	asm volatile(
+#ifndef __s390x__
+		"	ahi	%1,31\n"
+		"	srl	%1,5\n"
+		"0:	c	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,4(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#else
+		"	aghi	%1,63\n"
+		"	srlg	%1,%1,6\n"
+		"0:	cg	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,8(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#endif
+		: "+a" (bytes), "+a" (size)
+		: "d" (0UL), "a" (addr), "m" (*(addrtype *) addr)
+		: "cc" );
+	return bytes;
+}
+
+/**
+ * __ffz_word - add number of the first unset bit
+ * @nr: base value the bit number is added to
+ * @word: the word that is searched for unset bits
+ */
+static inline unsigned long __ffz_word(unsigned long nr, unsigned long word)
+{
 #ifdef __s390x__
 	if (likely((word & 0xffffffff) == 0xffffffff)) {
 		word >>= 32;
-		bit += 32;
+		nr += 32;
 	}
 #endif
 	if (likely((word & 0xffff) == 0xffff)) {
 		word >>= 16;
-		bit += 16;
+		nr += 16;
 	}
 	if (likely((word & 0xff) == 0xff)) {
 		word >>= 8;
-		bit += 8;
+		nr += 8;
 	}
-	return bit + _zb_findmap[word & 0xff];
+	return nr + _zb_findmap[(unsigned char) word];
 }
 
-/*
- * __ffs = find first bit in word. Undefined if no bit exists,
- * so code should check against 0UL first..
+/**
+ * __ffs_word - add number of the first set bit
+ * @nr: base value the bit number is added to
+ * @word: the word that is searched for set bits
  */
-static inline unsigned long __ffs (unsigned long word)
+static inline unsigned long __ffs_word(unsigned long nr, unsigned long word)
 {
-	unsigned long bit = 0;
-
 #ifdef __s390x__
 	if (likely((word & 0xffffffff) == 0)) {
 		word >>= 32;
-		bit += 32;
+		nr += 32;
 	}
 #endif
 	if (likely((word & 0xffff) == 0)) {
 		word >>= 16;
-		bit += 16;
+		nr += 16;
 	}
 	if (likely((word & 0xff) == 0)) {
 		word >>= 8;
-		bit += 8;
+		nr += 8;
 	}
-	return bit + _sb_findmap[word & 0xff];
+	return nr + _sb_findmap[(unsigned char) word];
 }
 
-/*
- * Find-bit routines..
- */
 
-#ifndef __s390x__
+/**
+ * __load_ulong_be - load big endian unsigned long
+ * @p: pointer to array of unsigned long
+ * @offset: byte offset of source value in the array
+ */
+static inline unsigned long __load_ulong_be(const unsigned long *p,
+					    unsigned long offset)
+{
+	p = (unsigned long *)((unsigned long) p + offset);
+	return *p;
+}
 
-static inline int
-find_first_zero_bit(const unsigned long * addr, unsigned long size)
+/**
+ * __load_ulong_le - load little endian unsigned long
+ * @p: pointer to array of unsigned long
+ * @offset: byte offset of source value in the array
+ */
+static inline unsigned long __load_ulong_le(const unsigned long *p,
+					    unsigned long offset)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-	unsigned long cmp, count;
-        unsigned int res;
+	unsigned long word;
 
-        if (!size)
-                return 0;
+	p = (unsigned long *)((unsigned long) p + offset);
+#ifndef __s390x__
 	asm volatile(
-		"	lhi	%1,-1\n"
-		"	lr	%2,%3\n"
-		"	slr	%0,%0\n"
-		"	ahi	%2,31\n"
-		"	srl	%2,5\n"
-		"0:	c	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	la	%0,4(%0)\n"
-		"	brct	%2,0b\n"
-		"	lr	%0,%3\n"
-		"	j	4f\n"
-		"1:	l	%2,0(%0,%4)\n"
-		"	sll	%0,3\n"
-		"	lhi	%1,0xff\n"
-		"	tml	%2,0xffff\n"
-		"	jno	2f\n"
-		"	ahi	%0,16\n"
-		"	srl	%2,16\n"
-		"2:	tml	%2,0x00ff\n"
-		"	jno	3f\n"
-		"	ahi	%0,8\n"
-		"	srl	%2,8\n"
-		"3:	nr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	alr	%0,%2\n"
-		"4:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
+		"	ic	%0,0(%1)\n"
+		"	icm	%0,2,1(%1)\n"
+		"	icm	%0,4,2(%1)\n"
+		"	icm	%0,8,3(%1)"
+		: "=&d" (word) : "a" (p), "m" (*p) : "cc");
+#else
+	asm volatile(
+		"	lrvg	%0,%1"
+		: "=d" (word) : "m" (*p) );
+#endif
+	return word;
 }
 
-static inline int
-find_first_bit(const unsigned long * addr, unsigned long size)
+/*
+ * The various find bit functions.
+ */
+
+/*
+ * ffz - find first zero in word.
+ * @word: The word to search
+ *
+ * Undefined if no zero exists, so code should check against ~0UL first.
+ */
+static inline unsigned long ffz(unsigned long word)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-	unsigned long cmp, count;
-        unsigned int res;
+	return __ffz_word(0, word);
+}
 
-        if (!size)
-                return 0;
-	asm volatile(
-		"	slr	%1,%1\n"
-		"	lr	%2,%3\n"
-		"	slr	%0,%0\n"
-		"	ahi	%2,31\n"
-		"	srl	%2,5\n"
-		"0:	c	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	la	%0,4(%0)\n"
-		"	brct	%2,0b\n"
-		"	lr	%0,%3\n"
-		"	j	4f\n"
-		"1:	l	%2,0(%0,%4)\n"
-		"	sll	%0,3\n"
-		"	lhi	%1,0xff\n"
-		"	tml	%2,0xffff\n"
-		"	jnz	2f\n"
-		"	ahi	%0,16\n"
-		"	srl	%2,16\n"
-		"2:	tml	%2,0x00ff\n"
-		"	jnz	3f\n"
-		"	ahi	%0,8\n"
-		"	srl	%2,8\n"
-		"3:	nr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	alr	%0,%2\n"
-		"4:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_sb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
+/**
+ * __ffs - find first bit in word.
+ * @word: The word to search
+ *
+ * Undefined if no bit exists, so code should check against 0 first.
+ */
+static inline unsigned long __ffs (unsigned long word)
+{
+	return __ffs_word(0, word);
 }
 
-#else /* __s390x__ */
+/**
+ * ffs - find first bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ */
+static inline int ffs(int x)
+{
+	if (!x)
+		return 0;
+	return __ffs_word(1, x);
+}
 
-static inline unsigned long
-find_first_zero_bit(const unsigned long * addr, unsigned long size)
+/**
+ * find_first_zero_bit - find the first zero bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit-number of the first zero bit, not the number of the byte
+ * containing a bit.
+ */
+static inline unsigned long find_first_zero_bit(const unsigned long *addr,
+						unsigned long size)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-        unsigned long res, cmp, count;
+	unsigned long bytes, bits;
 
         if (!size)
                 return 0;
-	asm volatile(
-		"	lghi	%1,-1\n"
-		"	lgr	%2,%3\n"
-		"	slgr	%0,%0\n"
-		"	aghi	%2,63\n"
-		"	srlg	%2,%2,6\n"
-		"0:	cg	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	la	%0,8(%0)\n"
-		"	brct	%2,0b\n"
-		"	lgr	%0,%3\n"
-		"	j	5f\n"
-		"1:	lg	%2,0(%0,%4)\n"
-		"	sllg	%0,%0,3\n"
-		"	clr	%2,%1\n"
-		"	jne	2f\n"
-		"	aghi	%0,32\n"
-		"	srlg	%2,%2,32\n"
-		"2:	lghi	%1,0xff\n"
-		"	tmll	%2,0xffff\n"
-		"	jno	3f\n"
-		"	aghi	%0,16\n"
-		"	srl	%2,16\n"
-		"3:	tmll	%2,0x00ff\n"
-		"	jno	4f\n"
-		"	aghi	%0,8\n"
-		"	srl	%2,8\n"
-		"4:	ngr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	algr	%0,%2\n"
-		"5:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
-}
-
-static inline unsigned long
-find_first_bit(const unsigned long * addr, unsigned long size)
+	bytes = __ffz_word_loop(addr, size);
+	bits = __ffz_word(bytes*8, __load_ulong_be(addr, bytes));
+	return (bits < size) ? bits : size;
+}
+
+/**
+ * find_first_bit - find the first set bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit-number of the first set bit, not the number of the byte
+ * containing a bit.
+ */
+static inline unsigned long find_first_bit(const unsigned long * addr,
+					   unsigned long size)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-        unsigned long res, cmp, count;
+	unsigned long bytes, bits;
 
         if (!size)
                 return 0;
-	asm volatile(
-		"	slgr	%1,%1\n"
-		"	lgr	%2,%3\n"
-		"	slgr	%0,%0\n"
-		"	aghi	%2,63\n"
-		"	srlg	%2,%2,6\n"
-		"0:	cg	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	aghi	%0,8\n"
-		"	brct	%2,0b\n"
-		"	lgr	%0,%3\n"
-		"	j	5f\n"
-		"1:	lg	%2,0(%0,%4)\n"
-		"	sllg	%0,%0,3\n"
-		"	clr	%2,%1\n"
-		"	jne	2f\n"
-		"	aghi	%0,32\n"
-		"	srlg	%2,%2,32\n"
-		"2:	lghi	%1,0xff\n"
-		"	tmll	%2,0xffff\n"
-		"	jnz	3f\n"
-		"	aghi	%0,16\n"
-		"	srl	%2,16\n"
-		"3:	tmll	%2,0x00ff\n"
-		"	jnz	4f\n"
-		"	aghi	%0,8\n"
-		"	srl	%2,8\n"
-		"4:	ngr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	algr	%0,%2\n"
-		"5:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_sb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
+	bytes = __ffs_word_loop(addr, size);
+	bits = __ffs_word(bytes*8, __load_ulong_be(addr, bytes));
+	return (bits < size) ? bits : size;
 }
 
-#endif /* __s390x__ */
-
-static inline int
-find_next_zero_bit (const unsigned long * addr, unsigned long size,
-		    unsigned long offset)
+/**
+ * find_next_zero_bit - find the first zero bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The maximum size to search
+ */
+static inline int find_next_zero_bit (const unsigned long * addr,
+				      unsigned long size,
+				      unsigned long offset)
 {
         const unsigned long *p;
 	unsigned long bit, set;
@@ -688,10 +702,10 @@ find_next_zero_bit (const unsigned long * addr, unsigned long size,
 	p = addr + offset / __BITOPS_WORDSIZE;
 	if (bit) {
 		/*
-		 * s390 version of ffz returns __BITOPS_WORDSIZE
+		 * __ffz_word returns __BITOPS_WORDSIZE
 		 * if no zero bit is present in the word.
 		 */
-		set = ffz(*p >> bit) + bit;
+		set = __ffz_word(0, *p >> bit) + bit;
 		if (set >= size)
 			return size + offset;
 		if (set < __BITOPS_WORDSIZE)
@@ -703,9 +717,15 @@ find_next_zero_bit (const unsigned long * addr, unsigned long size,
 	return offset + find_first_zero_bit(p, size);
 }
 
-static inline int
-find_next_bit (const unsigned long * addr, unsigned long size,
-	       unsigned long offset)
+/**
+ * find_next_bit - find the first set bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The maximum size to search
+ */
+static inline int find_next_bit (const unsigned long * addr,
+				 unsigned long size,
+				 unsigned long offset)
 {
         const unsigned long *p;
 	unsigned long bit, set;
@@ -718,10 +738,10 @@ find_next_bit (const unsigned long * addr, unsigned long size,
 	p = addr + offset / __BITOPS_WORDSIZE;
 	if (bit) {
 		/*
-		 * s390 version of __ffs returns __BITOPS_WORDSIZE
+		 * __ffs_word returns __BITOPS_WORDSIZE
 		 * if no one bit is present in the word.
 		 */
-		set = __ffs(*p & (~0UL << bit));
+		set = __ffs_word(0, *p & (~0UL << bit));
 		if (set >= size)
 			return size + offset;
 		if (set < __BITOPS_WORDSIZE)
@@ -744,8 +764,6 @@ static inline int sched_find_first_bit(unsigned long *b)
 	return find_first_bit(b, 140);
 }
 
-#include <asm-generic/bitops/ffs.h>
-
 #include <asm-generic/bitops/fls.h>
 #include <asm-generic/bitops/fls64.h>
 
@@ -775,105 +793,22 @@ static inline int sched_find_first_bit(unsigned long *b)
 #define ext2_find_next_bit(addr, size, off) \
 	generic_find_next_le_bit((unsigned long *)(addr), (size), (off))
 
-#ifndef __s390x__
-
-static inline int 
-ext2_find_first_zero_bit(void *vaddr, unsigned int size)
+static inline int ext2_find_first_zero_bit(void *vaddr, unsigned int size)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-	unsigned long cmp, count;
-        unsigned int res;
+	unsigned long bytes, bits;
 
         if (!size)
                 return 0;
-	asm volatile(
-		"	lhi	%1,-1\n"
-		"	lr	%2,%3\n"
-		"	ahi	%2,31\n"
-		"	srl	%2,5\n"
-		"	slr	%0,%0\n"
-		"0:	cl	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	ahi	%0,4\n"
-		"	brct	%2,0b\n"
-		"	lr	%0,%3\n"
-		"	j	4f\n"
-		"1:	l	%2,0(%0,%4)\n"
-		"	sll	%0,3\n"
-		"	ahi	%0,24\n"
-		"	lhi	%1,0xff\n"
-		"	tmh	%2,0xffff\n"
-		"	jo	2f\n"
-		"	ahi	%0,-16\n"
-		"	srl	%2,16\n"
-		"2:	tml	%2,0xff00\n"
-		"	jo	3f\n"
-		"	ahi	%0,-8\n"
-		"	srl	%2,8\n"
-		"3:	nr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	alr	%0,%2\n"
-		"4:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (vaddr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) vaddr) : "cc");
-        return (res < size) ? res : size;
+	bytes = __ffz_word_loop(vaddr, size);
+	bits = __ffz_word(bytes*8, __load_ulong_le(vaddr, bytes));
+	return (bits < size) ? bits : size;
 }
 
-#else /* __s390x__ */
-
-static inline unsigned long
-ext2_find_first_zero_bit(void *vaddr, unsigned long size)
-{
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-        unsigned long res, cmp, count;
-
-        if (!size)
-                return 0;
-	asm volatile(
-		"	lghi	%1,-1\n"
-		"	lgr	%2,%3\n"
-		"	aghi	%2,63\n"
-		"	srlg	%2,%2,6\n"
-		"	slgr	%0,%0\n"
-		"0:	clg	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	aghi	%0,8\n"
-		"	brct	%2,0b\n"
-		"	lgr	%0,%3\n"
-		"	j	5f\n"
-		"1:	cl	%1,0(%0,%4)\n"
-		"	jne	2f\n"
-		"	aghi	%0,4\n"
-		"2:	l	%2,0(%0,%4)\n"
-		"	sllg	%0,%0,3\n"
-		"	aghi	%0,24\n"
-		"	lghi	%1,0xff\n"
-		"	tmlh	%2,0xffff\n"
-		"	jo	3f\n"
-		"	aghi	%0,-16\n"
-		"	srl	%2,16\n"
-		"3:	tmll	%2,0xff00\n"
-		"	jo	4f\n"
-		"	aghi	%0,-8\n"
-		"	srl	%2,8\n"
-		"4:	ngr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	algr	%0,%2\n"
-		"5:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (vaddr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) vaddr) : "cc");
-        return (res < size) ? res : size;
-}
-
-#endif /* __s390x__ */
-
-static inline int
-ext2_find_next_zero_bit(void *vaddr, unsigned long size, unsigned long offset)
+static inline int ext2_find_next_zero_bit(void *vaddr, unsigned long size,
+					  unsigned long offset)
 {
         unsigned long *addr = vaddr, *p;
-	unsigned long word, bit, set;
+	unsigned long bit, set;
 
         if (offset >= size)
                 return size;
@@ -882,23 +817,11 @@ ext2_find_next_zero_bit(void *vaddr, unsigned long size, unsigned long offset)
 	size -= offset;
 	p = addr + offset / __BITOPS_WORDSIZE;
         if (bit) {
-#ifndef __s390x__
-		asm volatile(
-			"	ic	%0,0(%1)\n"
-			"	icm	%0,2,1(%1)\n"
-			"	icm	%0,4,2(%1)\n"
-			"	icm	%0,8,3(%1)"
-			: "=&a" (word) : "a" (p), "m" (*p) : "cc");
-#else
-		asm volatile(
-			"	lrvg	%0,%1"
-			: "=a" (word) : "m" (*p) );
-#endif
 		/*
 		 * s390 version of ffz returns __BITOPS_WORDSIZE
 		 * if no zero bit is present in the word.
 		 */
-		set = ffz(word >> bit) + bit;
+		set = ffz(__load_ulong_le(p, 0) >> bit) + bit;
 		if (set >= size)
 			return size + offset;
 		if (set < __BITOPS_WORDSIZE)
-- 
cgit v1.2.3


From 67fe9251bba510572feb6c3357636148bbd17e30 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:44 +0100
Subject: [S390] Implement ext2_find_next_bit.

Fixes this compile error:

fs/ext4/mballoc.c:
	In function 'ext4_mb_generate_buddy':
fs/ext4/mballoc.c:954:
	error: implicit declaration of function 'generic_find_next_le_bit'

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 include/asm-s390/bitops.h | 43 +++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 41 insertions(+), 2 deletions(-)

diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h
index 95d93955a9b..882db054110 100644
--- a/include/asm-s390/bitops.h
+++ b/include/asm-s390/bitops.h
@@ -790,8 +790,6 @@ static inline int sched_find_first_bit(unsigned long *b)
 	test_and_clear_bit((nr)^(__BITOPS_WORDSIZE - 8), (unsigned long *)addr)
 #define ext2_test_bit(nr, addr)      \
 	test_bit((nr)^(__BITOPS_WORDSIZE - 8), (unsigned long *)addr)
-#define ext2_find_next_bit(addr, size, off) \
-	generic_find_next_le_bit((unsigned long *)(addr), (size), (off))
 
 static inline int ext2_find_first_zero_bit(void *vaddr, unsigned int size)
 {
@@ -833,6 +831,47 @@ static inline int ext2_find_next_zero_bit(void *vaddr, unsigned long size,
 	return offset + ext2_find_first_zero_bit(p, size);
 }
 
+static inline unsigned long ext2_find_first_bit(void *vaddr,
+						unsigned long size)
+{
+	unsigned long bytes, bits;
+
+	if (!size)
+		return 0;
+	bytes = __ffs_word_loop(vaddr, size);
+	bits = __ffs_word(bytes*8, __load_ulong_le(vaddr, bytes));
+	return (bits < size) ? bits : size;
+}
+
+static inline int ext2_find_next_bit(void *vaddr, unsigned long size,
+				     unsigned long offset)
+{
+	unsigned long *addr = vaddr, *p;
+	unsigned long bit, set;
+
+	if (offset >= size)
+		return size;
+	bit = offset & (__BITOPS_WORDSIZE - 1);
+	offset -= bit;
+	size -= offset;
+	p = addr + offset / __BITOPS_WORDSIZE;
+	if (bit) {
+		/*
+		 * s390 version of ffz returns __BITOPS_WORDSIZE
+		 * if no zero bit is present in the word.
+		 */
+		set = ffs(__load_ulong_le(p, 0) >> bit) + bit;
+		if (set >= size)
+			return size + offset;
+		if (set < __BITOPS_WORDSIZE)
+			return set + offset;
+		offset += __BITOPS_WORDSIZE;
+		size -= __BITOPS_WORDSIZE;
+		p++;
+	}
+	return offset + ext2_find_first_bit(p, size);
+}
+
 #include <asm-generic/bitops/minix.h>
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From a3afe70b83fdbbd4d757d2911900d168bc798a31 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:45 +0100
Subject: [S390] latencytop s390 support.

Cc: Holger Wolf <wolf@linux.vnet.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig             |  3 +++
 arch/s390/kernel/stacktrace.c | 31 +++++++++++++++++++++++--------
 2 files changed, 26 insertions(+), 8 deletions(-)

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index f0e7ccf1b20..92a4f7b4323 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -16,6 +16,9 @@ config LOCKDEP_SUPPORT
 config STACKTRACE_SUPPORT
 	def_bool y
 
+config HAVE_LATENCYTOP_SUPPORT
+	def_bool y
+
 config RWSEM_GENERIC_SPINLOCK
 	bool
 
diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index da692472996..85e46a5d0e0 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -14,7 +14,8 @@
 static unsigned long save_context_stack(struct stack_trace *trace,
 					unsigned long sp,
 					unsigned long low,
-					unsigned long high)
+					unsigned long high,
+					int savesched)
 {
 	struct stack_frame *sf;
 	struct pt_regs *regs;
@@ -47,10 +48,12 @@ static unsigned long save_context_stack(struct stack_trace *trace,
 			return sp;
 		regs = (struct pt_regs *)sp;
 		addr = regs->psw.addr & PSW_ADDR_INSN;
-		if (!trace->skip)
-			trace->entries[trace->nr_entries++] = addr;
-		else
-			trace->skip--;
+		if (savesched || !in_sched_functions(addr)) {
+			if (!trace->skip)
+				trace->entries[trace->nr_entries++] = addr;
+			else
+				trace->skip--;
+		}
 		if (trace->nr_entries >= trace->max_entries)
 			return sp;
 		low = sp;
@@ -66,15 +69,27 @@ void save_stack_trace(struct stack_trace *trace)
 	orig_sp = sp & PSW_ADDR_INSN;
 	new_sp = save_context_stack(trace, orig_sp,
 				    S390_lowcore.panic_stack - PAGE_SIZE,
-				    S390_lowcore.panic_stack);
+				    S390_lowcore.panic_stack, 1);
 	if (new_sp != orig_sp)
 		return;
 	new_sp = save_context_stack(trace, new_sp,
 				    S390_lowcore.async_stack - ASYNC_SIZE,
-				    S390_lowcore.async_stack);
+				    S390_lowcore.async_stack, 1);
 	if (new_sp != orig_sp)
 		return;
 	save_context_stack(trace, new_sp,
 			   S390_lowcore.thread_info,
-			   S390_lowcore.thread_info + THREAD_SIZE);
+			   S390_lowcore.thread_info + THREAD_SIZE, 1);
+}
+
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
+{
+	unsigned long sp, low, high;
+
+	sp = tsk->thread.ksp & PSW_ADDR_INSN;
+	low = (unsigned long) task_stack_page(tsk);
+	high = (unsigned long) task_pt_regs(tsk);
+	save_context_stack(trace, sp, low, high, 0);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
-- 
cgit v1.2.3


From 6c5f57c7884a7e0806ae9af86de243321cab4953 Mon Sep 17 00:00:00 2001
From: Stefan Haberland <stefan.haberland@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:46 +0100
Subject: [S390] dasd: add ifcc handling

Adding interface control check (ifcc) handling in error recovery.
First retry up to 255 times and if all retries fail try an alternate
path if possible.

Signed-off-by: Stefan Haberland <stefan.haberland@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/block/dasd.c          | 17 ++++-------
 drivers/s390/block/dasd_3990_erp.c | 62 ++++++++++++++++++++++++++++----------
 2 files changed, 52 insertions(+), 27 deletions(-)

diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index d640427c74c..ab4f64c4982 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -1057,12 +1057,11 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
 		if (device->features & DASD_FEATURE_ERPLOG) {
 			dasd_log_sense(cqr, irb);
 		}
-		/* If we have no sense data, or we just don't want complex ERP
-		 * for this request, but if we have retries left, then just
-		 * reset this request and retry it in the fastpath
+		/*
+		 * If we don't want complex ERP for this request, then just
+		 * reset this and retry it in the fastpath
 		 */
-		if (!(cqr->irb.esw.esw0.erw.cons &&
-		      test_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags)) &&
+		if (!test_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags) &&
 		    cqr->retries > 0) {
 			DEV_MESSAGE(KERN_DEBUG, device,
 				    "default ERP in fastpath (%i retries left)",
@@ -1742,12 +1741,8 @@ restart:
 
 		/*  Process requests that may be recovered */
 		if (cqr->status == DASD_CQR_NEED_ERP) {
-			if (cqr->irb.esw.esw0.erw.cons &&
-			    test_bit(DASD_CQR_FLAGS_USE_ERP,
-				     &cqr->flags)) {
-				erp_fn = base->discipline->erp_action(cqr);
-				erp_fn(cqr);
-			}
+			erp_fn = base->discipline->erp_action(cqr);
+			erp_fn(cqr);
 			goto restart;
 		}
 
diff --git a/drivers/s390/block/dasd_3990_erp.c b/drivers/s390/block/dasd_3990_erp.c
index c361ab69ec0..f69714a0e9e 100644
--- a/drivers/s390/block/dasd_3990_erp.c
+++ b/drivers/s390/block/dasd_3990_erp.c
@@ -164,7 +164,7 @@ dasd_3990_erp_alternate_path(struct dasd_ccw_req * erp)
 
 		/* reset status to submit the request again... */
 		erp->status = DASD_CQR_FILLED;
-		erp->retries = 1;
+		erp->retries = 10;
 	} else {
 		DEV_MESSAGE(KERN_ERR, device,
 			    "No alternate channel path left (lpum=%x / "
@@ -301,8 +301,7 @@ dasd_3990_erp_action_4(struct dasd_ccw_req * erp, char *sense)
 		erp->function = dasd_3990_erp_action_4;
 
 	} else {
-
-		if (sense[25] == 0x1D) {	/* state change pending */
+		if (sense && (sense[25] == 0x1D)) { /* state change pending */
 
 			DEV_MESSAGE(KERN_INFO, device,
 				    "waiting for state change pending "
@@ -311,7 +310,7 @@ dasd_3990_erp_action_4(struct dasd_ccw_req * erp, char *sense)
 
 			dasd_3990_erp_block_queue(erp, 30*HZ);
 
-                } else if (sense[25] == 0x1E) {	/* busy */
+		} else if (sense && (sense[25] == 0x1E)) {	/* busy */
 			DEV_MESSAGE(KERN_INFO, device,
 				    "busy - redriving request later, "
 				    "%d retries left",
@@ -2119,6 +2118,34 @@ dasd_3990_erp_inspect_32(struct dasd_ccw_req * erp, char *sense)
  *****************************************************************************
  */
 
+/*
+ * DASD_3990_ERP_CONTROL_CHECK
+ *
+ * DESCRIPTION
+ *   Does a generic inspection if a control check occured and sets up
+ *   the related error recovery procedure
+ *
+ * PARAMETER
+ *   erp		pointer to the currently created default ERP
+ *
+ * RETURN VALUES
+ *   erp_filled		pointer to the erp
+ */
+
+static struct dasd_ccw_req *
+dasd_3990_erp_control_check(struct dasd_ccw_req *erp)
+{
+	struct dasd_device *device = erp->startdev;
+
+	if (erp->refers->irb.scsw.cstat & (SCHN_STAT_INTF_CTRL_CHK
+					   | SCHN_STAT_CHN_CTRL_CHK)) {
+		DEV_MESSAGE(KERN_DEBUG, device, "%s",
+			    "channel or interface control check");
+		erp = dasd_3990_erp_action_4(erp, NULL);
+	}
+	return erp;
+}
+
 /*
  * DASD_3990_ERP_INSPECT
  *
@@ -2145,8 +2172,11 @@ dasd_3990_erp_inspect(struct dasd_ccw_req * erp)
 	if (erp_new)
 		return erp_new;
 
+	/* check if no concurrent sens is available */
+	if (!erp->refers->irb.esw.esw0.erw.cons)
+		erp_new = dasd_3990_erp_control_check(erp);
 	/* distinguish between 24 and 32 byte sense data */
-	if (sense[27] & DASD_SENSE_BIT_0) {
+	else if (sense[27] & DASD_SENSE_BIT_0) {
 
 		/* inspect the 24 byte sense data */
 		erp_new = dasd_3990_erp_inspect_24(erp, sense);
@@ -2285,6 +2315,17 @@ dasd_3990_erp_error_match(struct dasd_ccw_req *cqr1, struct dasd_ccw_req *cqr2)
 		//	return 0;	/* CCW doesn't match */
 	}
 
+	if (cqr1->irb.esw.esw0.erw.cons != cqr2->irb.esw.esw0.erw.cons)
+		return 0;
+
+	if ((cqr1->irb.esw.esw0.erw.cons == 0) &&
+	    (cqr2->irb.esw.esw0.erw.cons == 0))	{
+		if ((cqr1->irb.scsw.cstat & (SCHN_STAT_INTF_CTRL_CHK |
+					     SCHN_STAT_CHN_CTRL_CHK)) ==
+		    (cqr2->irb.scsw.cstat & (SCHN_STAT_INTF_CTRL_CHK |
+					     SCHN_STAT_CHN_CTRL_CHK)))
+			return 1; /* match with ifcc*/
+	}
 	/* check sense data; byte 0-2,25,27 */
 	if (!((memcmp (cqr1->irb.ecw, cqr2->irb.ecw, 3) == 0) &&
 	      (cqr1->irb.ecw[27] == cqr2->irb.ecw[27]) &&
@@ -2560,17 +2601,6 @@ dasd_3990_erp_action(struct dasd_ccw_req * cqr)
 
 		return cqr;
 	}
-	/* check if sense data are available */
-	if (!cqr->irb.ecw) {
-		DEV_MESSAGE(KERN_DEBUG, device,
-			    "ERP called witout sense data avail ..."
-			    "request %p - NO ERP possible", cqr);
-
-		cqr->status = DASD_CQR_FAILED;
-
-		return cqr;
-
-	}
 
 	/* check if error happened before */
 	erp = dasd_3990_erp_in_erp(cqr);
-- 
cgit v1.2.3


From fe6b8e76d920b93fd445382aff7ff24082af8874 Mon Sep 17 00:00:00 2001
From: Stefan Weinhuber <wein@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:47 +0100
Subject: [S390] dasd: fix panic caused by alias device offline

When an alias device is set offline while it is in use this may
result in a panic in the cleanup part of the dasd_block_tasklet.
The problem here is that there may exist some ccw requests that were
originally created for the alias device and transferred to the base
device when the alias was set offline. When these request are
cleaned up later, the discipline pointer in the alias device may not
be valid anymore. To fix this use the base device discipline to find
the cleanup function.

Signed-off-by: Stefan Weinhuber <wein@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/block/dasd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index ab4f64c4982..d984e0fae63 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -1706,7 +1706,7 @@ static void __dasd_cleanup_cqr(struct dasd_ccw_req *cqr)
 
 	req = (struct request *) cqr->callback_data;
 	dasd_profile_end(cqr->block, cqr, req);
-	status = cqr->memdev->discipline->free_cp(cqr, req);
+	status = cqr->block->base->discipline->free_cp(cqr, req);
 	if (status <= 0)
 		error = status ? status : -EIO;
 	dasd_end_request(req, error);
-- 
cgit v1.2.3


From e35e1fadb4585e3143fab34dd4f5070698b3305b Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:48 +0100
Subject: [S390] sclp_tty/sclp_vt220: Fix scheduling while atomic

Under load the following bug message appeared while using sysrq-t:

BUG: scheduling while atomic: bash/3662/0x00000004
0000000000105b74 000000003ba17740 0000000000000002 0000000000000000
       000000003ba177e0 000000003ba17758 000000003ba17758 0000000000105bfe
       0000000000817ba8 000000003f2a5350 0000000000000000 0000000000000000
       000000003ba17740 000000000000000c 000000003ba17740 000000003ba177b0
       0000000000568630 0000000000105bfe 000000003ba17740 000000003ba17790
Call Trace:
([<0000000000105b74>] show_trace+0x13c/0x158)
 [<0000000000105c58>] show_stack+0xc8/0xfc
 [<0000000000105cbc>] dump_stack+0x30/0x40
 [<000000000012a0c8>] __schedule_bug+0x84/0x94
 [<000000000056234e>] schedule+0x5ea/0x970
 [<0000000000477cd2>] __sclp_vt220_write+0x1f6/0x3ec
 [<0000000000477f00>] sclp_vt220_con_write+0x38/0x48
 [<0000000000130b4a>] __call_console_drivers+0xbe/0xd8
 [<0000000000130bf0>] _call_console_drivers+0x8c/0xd0
 [<0000000000130eea>] release_console_sem+0x1a6/0x2fc
 [<0000000000131786>] vprintk+0x262/0x480
 [<00000000001319fa>] printk+0x56/0x68
 [<0000000000125aaa>] print_cfs_rq+0x45e/0x4a4
 [<000000000012614e>] sched_debug_show+0x65e/0xee8
 [<000000000012a8fc>] show_state_filter+0x1cc/0x1f0
 [<000000000044d39c>] sysrq_handle_showstate+0x2c/0x3c
 [<000000000044d1fe>] __handle_sysrq+0xae/0x18c
 [<00000000002001f2>] write_sysrq_trigger+0x8a/0x90
 [<00000000001f7862>] proc_reg_write+0x9a/0xc4
 [<00000000001a83d4>] vfs_write+0xb8/0x174
 [<00000000001a8b88>] sys_write+0x58/0x8c
 [<0000000000112e7c>] sysc_noemu+0x10/0x16
 [<0000020000116f68>] 0x20000116f68

The problem seems to be, that with a full console buffer, release_console_sem
disables interrupts with spin_lock_irqsave and then calls the console function
without enabling interrupts. __sclp_vt220_write checks for in_interrupt, to
decide if it can schedule. It should check for in_atomic instead.

The same is true for sclp_tty.c.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/sclp_tty.c   | 2 +-
 drivers/s390/char/sclp_vt220.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/s390/char/sclp_tty.c b/drivers/s390/char/sclp_tty.c
index e3b3d390b4a..2e616e33891 100644
--- a/drivers/s390/char/sclp_tty.c
+++ b/drivers/s390/char/sclp_tty.c
@@ -332,7 +332,7 @@ sclp_tty_write_string(const unsigned char *str, int count)
 		if (sclp_ttybuf == NULL) {
 			while (list_empty(&sclp_tty_pages)) {
 				spin_unlock_irqrestore(&sclp_tty_lock, flags);
-				if (in_interrupt())
+				if (in_atomic())
 					sclp_sync_wait();
 				else
 					wait_event(sclp_tty_waitq,
diff --git a/drivers/s390/char/sclp_vt220.c b/drivers/s390/char/sclp_vt220.c
index 40cd21bc5cc..68071622d4b 100644
--- a/drivers/s390/char/sclp_vt220.c
+++ b/drivers/s390/char/sclp_vt220.c
@@ -400,7 +400,7 @@ __sclp_vt220_write(const unsigned char *buf, int count, int do_schedule,
 			while (list_empty(&sclp_vt220_empty)) {
 				spin_unlock_irqrestore(&sclp_vt220_lock,
 						       flags);
-				if (in_interrupt())
+				if (in_atomic())
 					sclp_sync_wait();
 				else
 					wait_event(sclp_vt220_waitq,
-- 
cgit v1.2.3


From 0189103c69f47712a0c542a8bc28ff46ebe53a8a Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:49 +0100
Subject: [S390] Remove BUILD_BUG_ON() in vmem code.

Remove BUILD_BUG_ON() in vmem code since it causes build failures if
the size of struct page increases. Instead calculate at compile time
the address of the highest physical address that can be added to the
1:1 mapping.
This supposed to fix a build failure with the page owner tracking leak
detector patches as reported by akpm.

page-owner-tracking-leak-detector-broken-on-s390.patch can be removed
from -mm again when this is merged.

Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c   |  2 +-
 arch/s390/mm/vmem.c        |  3 +--
 include/asm-s390/pgtable.h | 12 +++++++++---
 3 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 72d20f70f8a..29ae165d174 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -528,7 +528,7 @@ static void __init setup_memory_end(void)
 	memory_size = 0;
 	memory_end &= PAGE_MASK;
 
-	max_mem = memory_end ? min(VMALLOC_START, memory_end) : VMALLOC_START;
+	max_mem = memory_end ? min(VMEM_MAX_PHYS, memory_end) : VMEM_MAX_PHYS;
 	memory_end = min(max_mem, memory_end);
 
 	/*
diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index 07e04677724..7c1287ccf78 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -250,7 +250,7 @@ static int insert_memory_segment(struct memory_segment *seg)
 {
 	struct memory_segment *tmp;
 
-	if (seg->start + seg->size >= VMALLOC_START ||
+	if (seg->start + seg->size >= VMEM_MAX_PHYS ||
 	    seg->start + seg->size < seg->start)
 		return -ERANGE;
 
@@ -360,7 +360,6 @@ void __init vmem_map_init(void)
 {
 	int i;
 
-	BUILD_BUG_ON((unsigned long)VMEM_MAP + VMEM_MAP_SIZE > VMEM_MAP_MAX);
 	NODE_DATA(0)->node_mem_map = VMEM_MAP;
 	for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++)
 		vmem_add_mem(memory_chunk[i].addr, memory_chunk[i].size);
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index 79b9eab1a0c..3f520754e71 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -115,15 +115,21 @@ extern char empty_zero_page[PAGE_SIZE];
 #ifndef __s390x__
 #define VMALLOC_START	0x78000000UL
 #define VMALLOC_END	0x7e000000UL
-#define VMEM_MAP_MAX	0x80000000UL
+#define VMEM_MAP_END	0x80000000UL
 #else /* __s390x__ */
 #define VMALLOC_START	0x3e000000000UL
 #define VMALLOC_END	0x3e040000000UL
-#define VMEM_MAP_MAX	0x40000000000UL
+#define VMEM_MAP_END	0x40000000000UL
 #endif /* __s390x__ */
 
+/*
+ * VMEM_MAX_PHYS is the highest physical address that can be added to the 1:1
+ * mapping. This needs to be calculated at compile time since the size of the
+ * VMEM_MAP is static but the size of struct page can change.
+ */
+#define VMEM_MAX_PHYS	min(VMALLOC_START, ((VMEM_MAP_END - VMALLOC_END) / \
+			  sizeof(struct page) * PAGE_SIZE) & ~((16 << 20) - 1))
 #define VMEM_MAP	((struct page *) VMALLOC_END)
-#define VMEM_MAP_SIZE	((VMALLOC_START / PAGE_SIZE) * sizeof(struct page))
 
 /*
  * A 31 bit pagetable entry of S390 has following format:
-- 
cgit v1.2.3


From c5411dba58c28736d25cffef65da1e01ed7d1423 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 5 Feb 2008 16:50:50 +0100
Subject: [S390] dcss: Initialize workqueue before using it.

In case a dcss segment cannot be loaded blk_cleanup_queue
will be called before blk_queue_make_request, leaving the
struct work unplug_work of the request queue uninitialized
before it is used.
That leads also to the lockdep message below.
To avoid that call blk_queue_make_request right after the
request_queue has been allocated.
This makes sure that the struct work is always initialized
before it is used.

INFO: trying to register non-static key.
the code is fine but needs lockdep annotation.
turning off the locking correctness validator.
CPU: 2 Not tainted 2.6.24 #6
Process swapper (pid: 1, task: 000000000f854038, ksp: 000000000f85f980)
040000000f85f860 000000000f85f880 0000000000000002 0000000000000000
       000000000f85f920 000000000f85f898 000000000f85f898 000000000001622e
       0000000000000000 000000000f85f980 0000000000000000 0000000000000000
       000000000f85f880 000000000000000c 000000000f85f880 000000000f85f8f0
       0000000000342908 000000000001622e 000000000f85f880 000000000f85f8d0
Call Trace:
([<000000000001619e>] show_trace+0xda/0x104)
 [<0000000000016288>] show_stack+0xc0/0xf8
 [<00000000000163d0>] dump_stack+0xb0/0xc0
 [<000000000006e4ea>] __lock_acquire+0x47e/0x1160
 [<000000000006f27c>] lock_acquire+0xb0/0xd8
 [<000000000005a522>] __cancel_work_timer+0x9e/0x240
 [<000000000005a72e>] cancel_work_sync+0x2a/0x3c
 [<0000000000165c46>] kblockd_flush_work+0x26/0x34
 [<0000000000169034>] blk_sync_queue+0x38/0x48
 [<0000000000169080>] blk_release_queue+0x3c/0xa8
 [<000000000017bce8>] kobject_cleanup+0x58/0xac
 [<000000000017bd66>] kobject_release+0x2a/0x38
 [<000000000017d28e>] kref_put+0x6e/0x94
 [<000000000017bc80>] kobject_put+0x38/0x48
 [<00000000001653be>] blk_put_queue+0x2a/0x38
 [<0000000000168fee>] blk_cleanup_queue+0x82/0x90
 [<0000000000213e7e>] dcssblk_add_store+0x34e/0x700
 [<00000000005243b8>] dcssblk_init+0x1a0/0x308
 [<000000000050a3c2>] kernel_init+0x1b2/0x3a4
 [<000000000001ac82>] kernel_thread_starter+0x6/0xc
 [<000000000001ac7c>] kernel_thread_starter+0x0/0xc

INFO: lockdep is turned off.

Cc: Gerald Schaefer <geraldsc@de.ibm.com>
Cc: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/block/dcssblk.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c
index 7779bfce1c3..3faf0538b32 100644
--- a/drivers/s390/block/dcssblk.c
+++ b/drivers/s390/block/dcssblk.c
@@ -415,6 +415,8 @@ dcssblk_add_store(struct device *dev, struct device_attribute *attr, const char
 	dev_info->gd->queue = dev_info->dcssblk_queue;
 	dev_info->gd->private_data = dev_info;
 	dev_info->gd->driverfs_dev = &dev_info->dev;
+	blk_queue_make_request(dev_info->dcssblk_queue, dcssblk_make_request);
+	blk_queue_hardsect_size(dev_info->dcssblk_queue, 4096);
 	/*
 	 * load the segment
 	 */
@@ -472,9 +474,6 @@ dcssblk_add_store(struct device *dev, struct device_attribute *attr, const char
 	if (rc)
 		goto unregister_dev;
 
-	blk_queue_make_request(dev_info->dcssblk_queue, dcssblk_make_request);
-	blk_queue_hardsect_size(dev_info->dcssblk_queue, 4096);
-
 	add_disk(dev_info->gd);
 
 	switch (dev_info->segment_type) {
-- 
cgit v1.2.3