blob: 4e7b55a14b1a46d40920fde48460fa2400c46f84 [file] [log] [blame]
// SPDX-License-Identifier: GPL-2.0
/*
* Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
* Horst Hummel <Horst.Hummel@de.ibm.com>
* Carsten Otte <Cotte@de.ibm.com>
* Martin Schwidefsky <schwidefsky@de.ibm.com>
* Bugreports.to..: <Linux390@de.ibm.com>
* Copyright IBM Corp. 1999, 2009
* EMC Symmetrix ioctl Copyright EMC Corporation, 2008
* Author.........: Nigel Hislop <hislop_nigel@emc.com>
*/
#define KMSG_COMPONENT "dasd-eckd"
#include <linux/stddef.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/hdreg.h> /* HDIO_GETGEO */
#include <linux/bio.h>
#include <linux/module.h>
#include <linux/compat.h>
#include <linux/init.h>
#include <linux/seq_file.h>
#include <asm/css_chars.h>
#include <asm/debug.h>
#include <asm/idals.h>
#include <asm/ebcdic.h>
#include <asm/io.h>
#include <linux/uaccess.h>
#include <asm/cio.h>
#include <asm/ccwdev.h>
#include <asm/itcw.h>
#include <asm/schid.h>
#include <asm/chpid.h>
#include "dasd_int.h"
#include "dasd_eckd.h"
#ifdef PRINTK_HEADER
#undef PRINTK_HEADER
#endif /* PRINTK_HEADER */
#define PRINTK_HEADER "dasd(eckd):"
#define ECKD_C0(i) (i->home_bytes)
#define ECKD_F(i) (i->formula)
#define ECKD_F1(i) (ECKD_F(i)==0x01?(i->factors.f_0x01.f1):\
(i->factors.f_0x02.f1))
#define ECKD_F2(i) (ECKD_F(i)==0x01?(i->factors.f_0x01.f2):\
(i->factors.f_0x02.f2))
#define ECKD_F3(i) (ECKD_F(i)==0x01?(i->factors.f_0x01.f3):\
(i->factors.f_0x02.f3))
#define ECKD_F4(i) (ECKD_F(i)==0x02?(i->factors.f_0x02.f4):0)
#define ECKD_F5(i) (ECKD_F(i)==0x02?(i->factors.f_0x02.f5):0)
#define ECKD_F6(i) (i->factor6)
#define ECKD_F7(i) (i->factor7)
#define ECKD_F8(i) (i->factor8)
/*
* raw track access always map to 64k in memory
* so it maps to 16 blocks of 4k per track
*/
#define DASD_RAW_BLOCK_PER_TRACK 16
#define DASD_RAW_BLOCKSIZE 4096
/* 64k are 128 x 512 byte sectors */
#define DASD_RAW_SECTORS_PER_TRACK 128
MODULE_LICENSE("GPL");
static struct dasd_discipline dasd_eckd_discipline;
/* The ccw bus type uses this table to find devices that it sends to
* dasd_eckd_probe */
static struct ccw_device_id dasd_eckd_ids[] = {
{ CCW_DEVICE_DEVTYPE (0x3990, 0, 0x3390, 0), .driver_info = 0x1},
{ CCW_DEVICE_DEVTYPE (0x2105, 0, 0x3390, 0), .driver_info = 0x2},
{ CCW_DEVICE_DEVTYPE (0x3880, 0, 0x3380, 0), .driver_info = 0x3},
{ CCW_DEVICE_DEVTYPE (0x3990, 0, 0x3380, 0), .driver_info = 0x4},
{ CCW_DEVICE_DEVTYPE (0x2105, 0, 0x3380, 0), .driver_info = 0x5},
{ CCW_DEVICE_DEVTYPE (0x9343, 0, 0x9345, 0), .driver_info = 0x6},
{ CCW_DEVICE_DEVTYPE (0x2107, 0, 0x3390, 0), .driver_info = 0x7},
{ CCW_DEVICE_DEVTYPE (0x2107, 0, 0x3380, 0), .driver_info = 0x8},
{ CCW_DEVICE_DEVTYPE (0x1750, 0, 0x3390, 0), .driver_info = 0x9},
{ CCW_DEVICE_DEVTYPE (0x1750, 0, 0x3380, 0), .driver_info = 0xa},
{ /* end of list */ },
};
MODULE_DEVICE_TABLE(ccw, dasd_eckd_ids);
static struct ccw_driver dasd_eckd_driver; /* see below */
static void *rawpadpage;
#define INIT_CQR_OK 0
#define INIT_CQR_UNFORMATTED 1
#define INIT_CQR_ERROR 2
/* emergency request for reserve/release */
static struct {
struct dasd_ccw_req cqr;
struct ccw1 ccw;
char data[32];
} *dasd_reserve_req;
static DEFINE_MUTEX(dasd_reserve_mutex);
/* definitions for the path verification worker */
struct path_verification_work_data {
struct work_struct worker;
struct dasd_device *device;
struct dasd_ccw_req cqr;
struct ccw1 ccw;
__u8 rcd_buffer[DASD_ECKD_RCD_DATA_SIZE];
int isglobal;
__u8 tbvpm;
};
static struct path_verification_work_data *path_verification_worker;
static DEFINE_MUTEX(dasd_path_verification_mutex);
struct check_attention_work_data {
struct work_struct worker;
struct dasd_device *device;
__u8 lpum;
};
static int prepare_itcw(struct itcw *, unsigned int, unsigned int, int,
struct dasd_device *, struct dasd_device *,
unsigned int, int, unsigned int, unsigned int,
unsigned int, unsigned int);
/* initial attempt at a probe function. this can be simplified once
* the other detection code is gone */
static int
dasd_eckd_probe (struct ccw_device *cdev)
{
int ret;
/* set ECKD specific ccw-device options */
ret = ccw_device_set_options(cdev, CCWDEV_ALLOW_FORCE |
CCWDEV_DO_PATHGROUP | CCWDEV_DO_MULTIPATH);
if (ret) {
DBF_EVENT_DEVID(DBF_WARNING, cdev, "%s",
"dasd_eckd_probe: could not set "
"ccw-device options");
return ret;
}
ret = dasd_generic_probe(cdev, &dasd_eckd_discipline);
return ret;
}
static int
dasd_eckd_set_online(struct ccw_device *cdev)
{
return dasd_generic_set_online(cdev, &dasd_eckd_discipline);
}
static const int sizes_trk0[] = { 28, 148, 84 };
#define LABEL_SIZE 140
/* head and record addresses of count_area read in analysis ccw */
static const int count_area_head[] = { 0, 0, 0, 0, 2 };
static const int count_area_rec[] = { 1, 2, 3, 4, 1 };
static inline unsigned int
round_up_multiple(unsigned int no, unsigned int mult)
{
int rem = no % mult;
return (rem ? no - rem + mult : no);
}
static inline unsigned int
ceil_quot(unsigned int d1, unsigned int d2)
{
return (d1 + (d2 - 1)) / d2;
}
static unsigned int
recs_per_track(struct dasd_eckd_characteristics * rdc,
unsigned int kl, unsigned int dl)
{
int dn, kn;
switch (rdc->dev_type) {
case 0x3380:
if (kl)
return 1499 / (15 + 7 + ceil_quot(kl + 12, 32) +
ceil_quot(dl + 12, 32));
else
return 1499 / (15 + ceil_quot(dl + 12, 32));
case 0x3390:
dn = ceil_quot(dl + 6, 232) + 1;
if (kl) {
kn = ceil_quot(kl + 6, 232) + 1;
return 1729 / (10 + 9 + ceil_quot(kl + 6 * kn, 34) +
9 + ceil_quot(dl + 6 * dn, 34));
} else
return 1729 / (10 + 9 + ceil_quot(dl + 6 * dn, 34));
case 0x9345:
dn = ceil_quot(dl + 6, 232) + 1;
if (kl) {
kn = ceil_quot(kl + 6, 232) + 1;
return 1420 / (18 + 7 + ceil_quot(kl + 6 * kn, 34) +
ceil_quot(dl + 6 * dn, 34));
} else
return 1420 / (18 + 7 + ceil_quot(dl + 6 * dn, 34));
}
return 0;
}
static void set_ch_t(struct ch_t *geo, __u32 cyl, __u8 head)
{
geo->cyl = (__u16) cyl;
geo->head = cyl >> 16;
geo->head <<= 4;
geo->head |= head;
}
static int set_timestamp(struct ccw1 *ccw, struct DE_eckd_data *data,
struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
int rc;
rc = get_phys_clock(&data->ep_sys_time);
/*
* Ignore return code if XRC is not supported or
* sync clock is switched off
*/
if ((rc && !private->rdc_data.facilities.XRC_supported) ||
rc == -EOPNOTSUPP || rc == -EACCES)
return 0;
/* switch on System Time Stamp - needed for XRC Support */
data->ga_extended |= 0x08; /* switch on 'Time Stamp Valid' */
data->ga_extended |= 0x02; /* switch on 'Extended Parameter' */
if (ccw) {
ccw->count = sizeof(struct DE_eckd_data);
ccw->flags |= CCW_FLAG_SLI;
}
return rc;
}
static int
define_extent(struct ccw1 *ccw, struct DE_eckd_data *data, unsigned int trk,
unsigned int totrk, int cmd, struct dasd_device *device,
int blksize)
{
struct dasd_eckd_private *private = device->private;
u16 heads, beghead, endhead;
u32 begcyl, endcyl;
int rc = 0;
if (ccw) {
ccw->cmd_code = DASD_ECKD_CCW_DEFINE_EXTENT;
ccw->flags = 0;
ccw->count = 16;
ccw->cda = (__u32)__pa(data);
}
memset(data, 0, sizeof(struct DE_eckd_data));
switch (cmd) {
case DASD_ECKD_CCW_READ_HOME_ADDRESS:
case DASD_ECKD_CCW_READ_RECORD_ZERO:
case DASD_ECKD_CCW_READ:
case DASD_ECKD_CCW_READ_MT:
case DASD_ECKD_CCW_READ_CKD:
case DASD_ECKD_CCW_READ_CKD_MT:
case DASD_ECKD_CCW_READ_KD:
case DASD_ECKD_CCW_READ_KD_MT:
data->mask.perm = 0x1;
data->attributes.operation = private->attrib.operation;
break;
case DASD_ECKD_CCW_READ_COUNT:
data->mask.perm = 0x1;
data->attributes.operation = DASD_BYPASS_CACHE;
break;
case DASD_ECKD_CCW_READ_TRACK:
case DASD_ECKD_CCW_READ_TRACK_DATA:
data->mask.perm = 0x1;
data->attributes.operation = private->attrib.operation;
data->blk_size = 0;
break;
case DASD_ECKD_CCW_WRITE:
case DASD_ECKD_CCW_WRITE_MT:
case DASD_ECKD_CCW_WRITE_KD:
case DASD_ECKD_CCW_WRITE_KD_MT:
data->mask.perm = 0x02;
data->attributes.operation = private->attrib.operation;
rc = set_timestamp(ccw, data, device);
break;
case DASD_ECKD_CCW_WRITE_CKD:
case DASD_ECKD_CCW_WRITE_CKD_MT:
data->attributes.operation = DASD_BYPASS_CACHE;
rc = set_timestamp(ccw, data, device);
break;
case DASD_ECKD_CCW_ERASE:
case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
data->mask.perm = 0x3;
data->mask.auth = 0x1;
data->attributes.operation = DASD_BYPASS_CACHE;
rc = set_timestamp(ccw, data, device);
break;
case DASD_ECKD_CCW_WRITE_FULL_TRACK:
data->mask.perm = 0x03;
data->attributes.operation = private->attrib.operation;
data->blk_size = 0;
break;
case DASD_ECKD_CCW_WRITE_TRACK_DATA:
data->mask.perm = 0x02;
data->attributes.operation = private->attrib.operation;
data->blk_size = blksize;
rc = set_timestamp(ccw, data, device);
break;
default:
dev_err(&device->cdev->dev,
"0x%x is not a known command\n", cmd);
break;
}
data->attributes.mode = 0x3; /* ECKD */
if ((private->rdc_data.cu_type == 0x2105 ||
private->rdc_data.cu_type == 0x2107 ||
private->rdc_data.cu_type == 0x1750)
&& !(private->uses_cdl && trk < 2))
data->ga_extended |= 0x40; /* Regular Data Format Mode */
heads = private->rdc_data.trk_per_cyl;
begcyl = trk / heads;
beghead = trk % heads;
endcyl = totrk / heads;
endhead = totrk % heads;
/* check for sequential prestage - enhance cylinder range */
if (data->attributes.operation == DASD_SEQ_PRESTAGE ||
data->attributes.operation == DASD_SEQ_ACCESS) {
if (endcyl + private->attrib.nr_cyl < private->real_cyl)
endcyl += private->attrib.nr_cyl;
else
endcyl = (private->real_cyl - 1);
}
set_ch_t(&data->beg_ext, begcyl, beghead);
set_ch_t(&data->end_ext, endcyl, endhead);
return rc;
}
static void locate_record_ext(struct ccw1 *ccw, struct LRE_eckd_data *data,
unsigned int trk, unsigned int rec_on_trk,
int count, int cmd, struct dasd_device *device,
unsigned int reclen, unsigned int tlf)
{
struct dasd_eckd_private *private = device->private;
int sector;
int dn, d;
if (ccw) {
ccw->cmd_code = DASD_ECKD_CCW_LOCATE_RECORD_EXT;
ccw->flags = 0;
if (cmd == DASD_ECKD_CCW_WRITE_FULL_TRACK)
ccw->count = 22;
else
ccw->count = 20;
ccw->cda = (__u32)__pa(data);
}
memset(data, 0, sizeof(*data));
sector = 0;
if (rec_on_trk) {
switch (private->rdc_data.dev_type) {
case 0x3390:
dn = ceil_quot(reclen + 6, 232);
d = 9 + ceil_quot(reclen + 6 * (dn + 1), 34);
sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
break;
case 0x3380:
d = 7 + ceil_quot(reclen + 12, 32);
sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
break;
}
}
data->sector = sector;
/* note: meaning of count depends on the operation
* for record based I/O it's the number of records, but for
* track based I/O it's the number of tracks
*/
data->count = count;
switch (cmd) {
case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
data->operation.orientation = 0x3;
data->operation.operation = 0x03;
break;
case DASD_ECKD_CCW_READ_HOME_ADDRESS:
data->operation.orientation = 0x3;
data->operation.operation = 0x16;
break;
case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
data->operation.orientation = 0x1;
data->operation.operation = 0x03;
data->count++;
break;
case DASD_ECKD_CCW_READ_RECORD_ZERO:
data->operation.orientation = 0x3;
data->operation.operation = 0x16;
data->count++;
break;
case DASD_ECKD_CCW_WRITE:
case DASD_ECKD_CCW_WRITE_MT:
case DASD_ECKD_CCW_WRITE_KD:
case DASD_ECKD_CCW_WRITE_KD_MT:
data->auxiliary.length_valid = 0x1;
data->length = reclen;
data->operation.operation = 0x01;
break;
case DASD_ECKD_CCW_WRITE_CKD:
case DASD_ECKD_CCW_WRITE_CKD_MT:
data->auxiliary.length_valid = 0x1;
data->length = reclen;
data->operation.operation = 0x03;
break;
case DASD_ECKD_CCW_WRITE_FULL_TRACK:
data->operation.orientation = 0x0;
data->operation.operation = 0x3F;
data->extended_operation = 0x11;
data->length = 0;
data->extended_parameter_length = 0x02;
if (data->count > 8) {
data->extended_parameter[0] = 0xFF;
data->extended_parameter[1] = 0xFF;
data->extended_parameter[1] <<= (16 - count);
} else {
data->extended_parameter[0] = 0xFF;
data->extended_parameter[0] <<= (8 - count);
data->extended_parameter[1] = 0x00;
}
data->sector = 0xFF;
break;
case DASD_ECKD_CCW_WRITE_TRACK_DATA:
data->auxiliary.length_valid = 0x1;
data->length = reclen; /* not tlf, as one might think */
data->operation.operation = 0x3F;
data->extended_operation = 0x23;
break;
case DASD_ECKD_CCW_READ:
case DASD_ECKD_CCW_READ_MT:
case DASD_ECKD_CCW_READ_KD:
case DASD_ECKD_CCW_READ_KD_MT:
data->auxiliary.length_valid = 0x1;
data->length = reclen;
data->operation.operation = 0x06;
break;
case DASD_ECKD_CCW_READ_CKD:
case DASD_ECKD_CCW_READ_CKD_MT:
data->auxiliary.length_valid = 0x1;
data->length = reclen;
data->operation.operation = 0x16;
break;
case DASD_ECKD_CCW_READ_COUNT:
data->operation.operation = 0x06;
break;
case DASD_ECKD_CCW_READ_TRACK:
data->operation.orientation = 0x1;
data->operation.operation = 0x0C;
data->extended_parameter_length = 0;
data->sector = 0xFF;
break;
case DASD_ECKD_CCW_READ_TRACK_DATA:
data->auxiliary.length_valid = 0x1;
data->length = tlf;
data->operation.operation = 0x0C;
break;
case DASD_ECKD_CCW_ERASE:
data->length = reclen;
data->auxiliary.length_valid = 0x1;
data->operation.operation = 0x0b;
break;
default:
DBF_DEV_EVENT(DBF_ERR, device,
"fill LRE unknown opcode 0x%x", cmd);
BUG();
}
set_ch_t(&data->seek_addr,
trk / private->rdc_data.trk_per_cyl,
trk % private->rdc_data.trk_per_cyl);
data->search_arg.cyl = data->seek_addr.cyl;
data->search_arg.head = data->seek_addr.head;
data->search_arg.record = rec_on_trk;
}
static int prefix_LRE(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
unsigned int trk, unsigned int totrk, int cmd,
struct dasd_device *basedev, struct dasd_device *startdev,
unsigned int format, unsigned int rec_on_trk, int count,
unsigned int blksize, unsigned int tlf)
{
struct dasd_eckd_private *basepriv, *startpriv;
struct LRE_eckd_data *lredata;
struct DE_eckd_data *dedata;
int rc = 0;
basepriv = basedev->private;
startpriv = startdev->private;
dedata = &pfxdata->define_extent;
lredata = &pfxdata->locate_record;
ccw->cmd_code = DASD_ECKD_CCW_PFX;
ccw->flags = 0;
if (cmd == DASD_ECKD_CCW_WRITE_FULL_TRACK) {
ccw->count = sizeof(*pfxdata) + 2;
ccw->cda = (__u32) __pa(pfxdata);
memset(pfxdata, 0, sizeof(*pfxdata) + 2);
} else {
ccw->count = sizeof(*pfxdata);
ccw->cda = (__u32) __pa(pfxdata);
memset(pfxdata, 0, sizeof(*pfxdata));
}
/* prefix data */
if (format > 1) {
DBF_DEV_EVENT(DBF_ERR, basedev,
"PFX LRE unknown format 0x%x", format);
BUG();
return -EINVAL;
}
pfxdata->format = format;
pfxdata->base_address = basepriv->ned->unit_addr;
pfxdata->base_lss = basepriv->ned->ID;
pfxdata->validity.define_extent = 1;
/* private uid is kept up to date, conf_data may be outdated */
if (startpriv->uid.type == UA_BASE_PAV_ALIAS)
pfxdata->validity.verify_base = 1;
if (startpriv->uid.type == UA_HYPER_PAV_ALIAS) {
pfxdata->validity.verify_base = 1;
pfxdata->validity.hyper_pav = 1;
}
rc = define_extent(NULL, dedata, trk, totrk, cmd, basedev, blksize);
/*
* For some commands the System Time Stamp is set in the define extent
* data when XRC is supported. The validity of the time stamp must be
* reflected in the prefix data as well.
*/
if (dedata->ga_extended & 0x08 && dedata->ga_extended & 0x02)
pfxdata->validity.time_stamp = 1; /* 'Time Stamp Valid' */
if (format == 1) {
locate_record_ext(NULL, lredata, trk, rec_on_trk, count, cmd,
basedev, blksize, tlf);
}
return rc;
}
static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
unsigned int trk, unsigned int totrk, int cmd,
struct dasd_device *basedev, struct dasd_device *startdev)
{
return prefix_LRE(ccw, pfxdata, trk, totrk, cmd, basedev, startdev,
0, 0, 0, 0, 0);
}
static void
locate_record(struct ccw1 *ccw, struct LO_eckd_data *data, unsigned int trk,
unsigned int rec_on_trk, int no_rec, int cmd,
struct dasd_device * device, int reclen)
{
struct dasd_eckd_private *private = device->private;
int sector;
int dn, d;
DBF_DEV_EVENT(DBF_INFO, device,
"Locate: trk %d, rec %d, no_rec %d, cmd %d, reclen %d",
trk, rec_on_trk, no_rec, cmd, reclen);
ccw->cmd_code = DASD_ECKD_CCW_LOCATE_RECORD;
ccw->flags = 0;
ccw->count = 16;
ccw->cda = (__u32) __pa(data);
memset(data, 0, sizeof(struct LO_eckd_data));
sector = 0;
if (rec_on_trk) {
switch (private->rdc_data.dev_type) {
case 0x3390:
dn = ceil_quot(reclen + 6, 232);
d = 9 + ceil_quot(reclen + 6 * (dn + 1), 34);
sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
break;
case 0x3380:
d = 7 + ceil_quot(reclen + 12, 32);
sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
break;
}
}
data->sector = sector;
data->count = no_rec;
switch (cmd) {
case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
data->operation.orientation = 0x3;
data->operation.operation = 0x03;
break;
case DASD_ECKD_CCW_READ_HOME_ADDRESS:
data->operation.orientation = 0x3;
data->operation.operation = 0x16;
break;
case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
data->operation.orientation = 0x1;
data->operation.operation = 0x03;
data->count++;
break;
case DASD_ECKD_CCW_READ_RECORD_ZERO:
data->operation.orientation = 0x3;
data->operation.operation = 0x16;
data->count++;
break;
case DASD_ECKD_CCW_WRITE:
case DASD_ECKD_CCW_WRITE_MT:
case DASD_ECKD_CCW_WRITE_KD:
case DASD_ECKD_CCW_WRITE_KD_MT:
data->auxiliary.last_bytes_used = 0x1;
data->length = reclen;
data->operation.operation = 0x01;
break;
case DASD_ECKD_CCW_WRITE_CKD:
case DASD_ECKD_CCW_WRITE_CKD_MT:
data->auxiliary.last_bytes_used = 0x1;
data->length = reclen;
data->operation.operation = 0x03;
break;
case DASD_ECKD_CCW_READ:
case DASD_ECKD_CCW_READ_MT:
case DASD_ECKD_CCW_READ_KD:
case DASD_ECKD_CCW_READ_KD_MT:
data->auxiliary.last_bytes_used = 0x1;
data->length = reclen;
data->operation.operation = 0x06;
break;
case DASD_ECKD_CCW_READ_CKD:
case DASD_ECKD_CCW_READ_CKD_MT:
data->auxiliary.last_bytes_used = 0x1;
data->length = reclen;
data->operation.operation = 0x16;
break;
case DASD_ECKD_CCW_READ_COUNT:
data->operation.operation = 0x06;
break;
case DASD_ECKD_CCW_ERASE:
data->length = reclen;
data->auxiliary.last_bytes_used = 0x1;
data->operation.operation = 0x0b;
break;
default:
DBF_DEV_EVENT(DBF_ERR, device, "unknown locate record "
"opcode 0x%x", cmd);
}
set_ch_t(&data->seek_addr,
trk / private->rdc_data.trk_per_cyl,
trk % private->rdc_data.trk_per_cyl);
data->search_arg.cyl = data->seek_addr.cyl;
data->search_arg.head = data->seek_addr.head;
data->search_arg.record = rec_on_trk;
}
/*
* Returns 1 if the block is one of the special blocks that needs
* to get read/written with the KD variant of the command.
* That is DASD_ECKD_READ_KD_MT instead of DASD_ECKD_READ_MT and
* DASD_ECKD_WRITE_KD_MT instead of DASD_ECKD_WRITE_MT.
* Luckily the KD variants differ only by one bit (0x08) from the
* normal variant. So don't wonder about code like:
* if (dasd_eckd_cdl_special(blk_per_trk, recid))
* ccw->cmd_code |= 0x8;
*/
static inline int
dasd_eckd_cdl_special(int blk_per_trk, int recid)
{
if (recid < 3)
return 1;
if (recid < blk_per_trk)
return 0;
if (recid < 2 * blk_per_trk)
return 1;
return 0;
}
/*
* Returns the record size for the special blocks of the cdl format.
* Only returns something useful if dasd_eckd_cdl_special is true
* for the recid.
*/
static inline int
dasd_eckd_cdl_reclen(int recid)
{
if (recid < 3)
return sizes_trk0[recid];
return LABEL_SIZE;
}
/* create unique id from private structure. */
static void create_uid(struct dasd_eckd_private *private)
{
int count;
struct dasd_uid *uid;
uid = &private->uid;
memset(uid, 0, sizeof(struct dasd_uid));
memcpy(uid->vendor, private->ned->HDA_manufacturer,
sizeof(uid->vendor) - 1);
EBCASC(uid->vendor, sizeof(uid->vendor) - 1);
memcpy(uid->serial, private->ned->HDA_location,
sizeof(uid->serial) - 1);
EBCASC(uid->serial, sizeof(uid->serial) - 1);
uid->ssid = private->gneq->subsystemID;
uid->real_unit_addr = private->ned->unit_addr;
if (private->sneq) {
uid->type = private->sneq->sua_flags;
if (uid->type == UA_BASE_PAV_ALIAS)
uid->base_unit_addr = private->sneq->base_unit_addr;
} else {
uid->type = UA_BASE_DEVICE;
}
if (private->vdsneq) {
for (count = 0; count < 16; count++) {
sprintf(uid->vduit+2*count, "%02x",
private->vdsneq->uit[count]);
}
}
}
/*
* Generate device unique id that specifies the physical device.
*/
static int dasd_eckd_generate_uid(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
unsigned long flags;
if (!private)
return -ENODEV;
if (!private->ned || !private->gneq)
return -ENODEV;
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
create_uid(private);
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
return 0;
}
static int dasd_eckd_get_uid(struct dasd_device *device, struct dasd_uid *uid)
{
struct dasd_eckd_private *private = device->private;
unsigned long flags;
if (private) {
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
*uid = private->uid;
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
return 0;
}
return -EINVAL;
}
/*
* compare device UID with data of a given dasd_eckd_private structure
* return 0 for match
*/
static int dasd_eckd_compare_path_uid(struct dasd_device *device,
struct dasd_eckd_private *private)
{
struct dasd_uid device_uid;
create_uid(private);
dasd_eckd_get_uid(device, &device_uid);
return memcmp(&device_uid, &private->uid, sizeof(struct dasd_uid));
}
static void dasd_eckd_fill_rcd_cqr(struct dasd_device *device,
struct dasd_ccw_req *cqr,
__u8 *rcd_buffer,
__u8 lpm)
{
struct ccw1 *ccw;
/*
* buffer has to start with EBCDIC "V1.0" to show
* support for virtual device SNEQ
*/
rcd_buffer[0] = 0xE5;
rcd_buffer[1] = 0xF1;
rcd_buffer[2] = 0x4B;
rcd_buffer[3] = 0xF0;
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_RCD;
ccw->flags = 0;
ccw->cda = (__u32)(addr_t)rcd_buffer;
ccw->count = DASD_ECKD_RCD_DATA_SIZE;
cqr->magic = DASD_ECKD_MAGIC;
cqr->startdev = device;
cqr->memdev = device;
cqr->block = NULL;
cqr->expires = 10*HZ;
cqr->lpm = lpm;
cqr->retries = 256;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
set_bit(DASD_CQR_VERIFY_PATH, &cqr->flags);
}
/*
* Wakeup helper for read_conf
* if the cqr is not done and needs some error recovery
* the buffer has to be re-initialized with the EBCDIC "V1.0"
* to show support for virtual device SNEQ
*/
static void read_conf_cb(struct dasd_ccw_req *cqr, void *data)
{
struct ccw1 *ccw;
__u8 *rcd_buffer;
if (cqr->status != DASD_CQR_DONE) {
ccw = cqr->cpaddr;
rcd_buffer = (__u8 *)((addr_t) ccw->cda);
memset(rcd_buffer, 0, sizeof(*rcd_buffer));
rcd_buffer[0] = 0xE5;
rcd_buffer[1] = 0xF1;
rcd_buffer[2] = 0x4B;
rcd_buffer[3] = 0xF0;
}
dasd_wakeup_cb(cqr, data);
}
static int dasd_eckd_read_conf_immediately(struct dasd_device *device,
struct dasd_ccw_req *cqr,
__u8 *rcd_buffer,
__u8 lpm)
{
struct ciw *ciw;
int rc;
/*
* sanity check: scan for RCD command in extended SenseID data
* some devices do not support RCD
*/
ciw = ccw_device_get_ciw(device->cdev, CIW_TYPE_RCD);
if (!ciw || ciw->cmd != DASD_ECKD_CCW_RCD)
return -EOPNOTSUPP;
dasd_eckd_fill_rcd_cqr(device, cqr, rcd_buffer, lpm);
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
set_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags);
cqr->retries = 5;
cqr->callback = read_conf_cb;
rc = dasd_sleep_on_immediatly(cqr);
return rc;
}
static int dasd_eckd_read_conf_lpm(struct dasd_device *device,
void **rcd_buffer,
int *rcd_buffer_size, __u8 lpm)
{
struct ciw *ciw;
char *rcd_buf = NULL;
int ret;
struct dasd_ccw_req *cqr;
/*
* sanity check: scan for RCD command in extended SenseID data
* some devices do not support RCD
*/
ciw = ccw_device_get_ciw(device->cdev, CIW_TYPE_RCD);
if (!ciw || ciw->cmd != DASD_ECKD_CCW_RCD) {
ret = -EOPNOTSUPP;
goto out_error;
}
rcd_buf = kzalloc(DASD_ECKD_RCD_DATA_SIZE, GFP_KERNEL | GFP_DMA);
if (!rcd_buf) {
ret = -ENOMEM;
goto out_error;
}
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* RCD */,
0, /* use rcd_buf as data ara */
device, NULL);
if (IS_ERR(cqr)) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Could not allocate RCD request");
ret = -ENOMEM;
goto out_error;
}
dasd_eckd_fill_rcd_cqr(device, cqr, rcd_buf, lpm);
cqr->callback = read_conf_cb;
ret = dasd_sleep_on(cqr);
/*
* on success we update the user input parms
*/
dasd_sfree_request(cqr, cqr->memdev);
if (ret)
goto out_error;
*rcd_buffer_size = DASD_ECKD_RCD_DATA_SIZE;
*rcd_buffer = rcd_buf;
return 0;
out_error:
kfree(rcd_buf);
*rcd_buffer = NULL;
*rcd_buffer_size = 0;
return ret;
}
static int dasd_eckd_identify_conf_parts(struct dasd_eckd_private *private)
{
struct dasd_sneq *sneq;
int i, count;
private->ned = NULL;
private->sneq = NULL;
private->vdsneq = NULL;
private->gneq = NULL;
count = private->conf_len / sizeof(struct dasd_sneq);
sneq = (struct dasd_sneq *)private->conf_data;
for (i = 0; i < count; ++i) {
if (sneq->flags.identifier == 1 && sneq->format == 1)
private->sneq = sneq;
else if (sneq->flags.identifier == 1 && sneq->format == 4)
private->vdsneq = (struct vd_sneq *)sneq;
else if (sneq->flags.identifier == 2)
private->gneq = (struct dasd_gneq *)sneq;
else if (sneq->flags.identifier == 3 && sneq->res1 == 1)
private->ned = (struct dasd_ned *)sneq;
sneq++;
}
if (!private->ned || !private->gneq) {
private->ned = NULL;
private->sneq = NULL;
private->vdsneq = NULL;
private->gneq = NULL;
return -EINVAL;
}
return 0;
};
static unsigned char dasd_eckd_path_access(void *conf_data, int conf_len)
{
struct dasd_gneq *gneq;
int i, count, found;
count = conf_len / sizeof(*gneq);
gneq = (struct dasd_gneq *)conf_data;
found = 0;
for (i = 0; i < count; ++i) {
if (gneq->flags.identifier == 2) {
found = 1;
break;
}
gneq++;
}
if (found)
return ((char *)gneq)[18] & 0x07;
else
return 0;
}
static void dasd_eckd_clear_conf_data(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
int i;
private->conf_data = NULL;
private->conf_len = 0;
for (i = 0; i < 8; i++) {
kfree(device->path[i].conf_data);
device->path[i].conf_data = NULL;
device->path[i].cssid = 0;
device->path[i].ssid = 0;
device->path[i].chpid = 0;
}
}
static int dasd_eckd_read_conf(struct dasd_device *device)
{
void *conf_data;
int conf_len, conf_data_saved;
int rc, path_err, pos;
__u8 lpm, opm;
struct dasd_eckd_private *private, path_private;
struct dasd_uid *uid;
char print_path_uid[60], print_device_uid[60];
struct channel_path_desc_fmt0 *chp_desc;
struct subchannel_id sch_id;
private = device->private;
opm = ccw_device_get_path_mask(device->cdev);
ccw_device_get_schid(device->cdev, &sch_id);
conf_data_saved = 0;
path_err = 0;
/* get configuration data per operational path */
for (lpm = 0x80; lpm; lpm>>= 1) {
if (!(lpm & opm))
continue;
rc = dasd_eckd_read_conf_lpm(device, &conf_data,
&conf_len, lpm);
if (rc && rc != -EOPNOTSUPP) { /* -EOPNOTSUPP is ok */
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Read configuration data returned "
"error %d", rc);
return rc;
}
if (conf_data == NULL) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"No configuration data "
"retrieved");
/* no further analysis possible */
dasd_path_add_opm(device, opm);
continue; /* no error */
}
/* save first valid configuration data */
if (!conf_data_saved) {
/* initially clear previously stored conf_data */
dasd_eckd_clear_conf_data(device);
private->conf_data = conf_data;
private->conf_len = conf_len;
if (dasd_eckd_identify_conf_parts(private)) {
private->conf_data = NULL;
private->conf_len = 0;
kfree(conf_data);
continue;
}
pos = pathmask_to_pos(lpm);
/* store per path conf_data */
device->path[pos].conf_data = conf_data;
device->path[pos].cssid = sch_id.cssid;
device->path[pos].ssid = sch_id.ssid;
chp_desc = ccw_device_get_chp_desc(device->cdev, pos);
if (chp_desc)
device->path[pos].chpid = chp_desc->chpid;
kfree(chp_desc);
/*
* build device UID that other path data
* can be compared to it
*/
dasd_eckd_generate_uid(device);
conf_data_saved++;
} else {
path_private.conf_data = conf_data;
path_private.conf_len = DASD_ECKD_RCD_DATA_SIZE;
if (dasd_eckd_identify_conf_parts(
&path_private)) {
path_private.conf_data = NULL;
path_private.conf_len = 0;
kfree(conf_data);
continue;
}
if (dasd_eckd_compare_path_uid(
device, &path_private)) {
uid = &path_private.uid;
if (strlen(uid->vduit) > 0)
snprintf(print_path_uid,
sizeof(print_path_uid),
"%s.%s.%04x.%02x.%s",
uid->vendor, uid->serial,
uid->ssid, uid->real_unit_addr,
uid->vduit);
else
snprintf(print_path_uid,
sizeof(print_path_uid),
"%s.%s.%04x.%02x",
uid->vendor, uid->serial,
uid->ssid,
uid->real_unit_addr);
uid = &private->uid;
if (strlen(uid->vduit) > 0)
snprintf(print_device_uid,
sizeof(print_device_uid),
"%s.%s.%04x.%02x.%s",
uid->vendor, uid->serial,
uid->ssid, uid->real_unit_addr,
uid->vduit);
else
snprintf(print_device_uid,
sizeof(print_device_uid),
"%s.%s.%04x.%02x",
uid->vendor, uid->serial,
uid->ssid,
uid->real_unit_addr);
dev_err(&device->cdev->dev,
"Not all channel paths lead to "
"the same device, path %02X leads to "
"device %s instead of %s\n", lpm,
print_path_uid, print_device_uid);
path_err = -EINVAL;
dasd_path_add_cablepm(device, lpm);
continue;
}
pos = pathmask_to_pos(lpm);
/* store per path conf_data */
device->path[pos].conf_data = conf_data;
device->path[pos].cssid = sch_id.cssid;
device->path[pos].ssid = sch_id.ssid;
chp_desc = ccw_device_get_chp_desc(device->cdev, pos);
if (chp_desc)
device->path[pos].chpid = chp_desc->chpid;
kfree(chp_desc);
path_private.conf_data = NULL;
path_private.conf_len = 0;
}
switch (dasd_eckd_path_access(conf_data, conf_len)) {
case 0x02:
dasd_path_add_nppm(device, lpm);
break;
case 0x03:
dasd_path_add_ppm(device, lpm);
break;
}
if (!dasd_path_get_opm(device)) {
dasd_path_set_opm(device, lpm);
dasd_generic_path_operational(device);
} else {
dasd_path_add_opm(device, lpm);
}
}
return path_err;
}
static u32 get_fcx_max_data(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
int fcx_in_css, fcx_in_gneq, fcx_in_features;
int tpm, mdc;
if (dasd_nofcx)
return 0;
/* is transport mode supported? */
fcx_in_css = css_general_characteristics.fcx;
fcx_in_gneq = private->gneq->reserved2[7] & 0x04;
fcx_in_features = private->features.feature[40] & 0x80;
tpm = fcx_in_css && fcx_in_gneq && fcx_in_features;
if (!tpm)
return 0;
mdc = ccw_device_get_mdc(device->cdev, 0);
if (mdc < 0) {
dev_warn(&device->cdev->dev, "Detecting the maximum supported data size for zHPF requests failed\n");
return 0;
} else {
return (u32)mdc * FCX_MAX_DATA_FACTOR;
}
}
static int verify_fcx_max_data(struct dasd_device *device, __u8 lpm)
{
struct dasd_eckd_private *private = device->private;
int mdc;
u32 fcx_max_data;
if (private->fcx_max_data) {
mdc = ccw_device_get_mdc(device->cdev, lpm);
if ((mdc < 0)) {
dev_warn(&device->cdev->dev,
"Detecting the maximum data size for zHPF "
"requests failed (rc=%d) for a new path %x\n",
mdc, lpm);
return mdc;
}
fcx_max_data = (u32)mdc * FCX_MAX_DATA_FACTOR;
if (fcx_max_data < private->fcx_max_data) {
dev_warn(&device->cdev->dev,
"The maximum data size for zHPF requests %u "
"on a new path %x is below the active maximum "
"%u\n", fcx_max_data, lpm,
private->fcx_max_data);
return -EACCES;
}
}
return 0;
}
static int rebuild_device_uid(struct dasd_device *device,
struct path_verification_work_data *data)
{
struct dasd_eckd_private *private = device->private;
__u8 lpm, opm = dasd_path_get_opm(device);
int rc = -ENODEV;
for (lpm = 0x80; lpm; lpm >>= 1) {
if (!(lpm & opm))
continue;
memset(&data->rcd_buffer, 0, sizeof(data->rcd_buffer));
memset(&data->cqr, 0, sizeof(data->cqr));
data->cqr.cpaddr = &data->ccw;
rc = dasd_eckd_read_conf_immediately(device, &data->cqr,
data->rcd_buffer,
lpm);
if (rc) {
if (rc == -EOPNOTSUPP) /* -EOPNOTSUPP is ok */
continue;
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Read configuration data "
"returned error %d", rc);
break;
}
memcpy(private->conf_data, data->rcd_buffer,
DASD_ECKD_RCD_DATA_SIZE);
if (dasd_eckd_identify_conf_parts(private)) {
rc = -ENODEV;
} else /* first valid path is enough */
break;
}
if (!rc)
rc = dasd_eckd_generate_uid(device);
return rc;
}
static void do_path_verification_work(struct work_struct *work)
{
struct path_verification_work_data *data;
struct dasd_device *device;
struct dasd_eckd_private path_private;
struct dasd_uid *uid;
__u8 path_rcd_buf[DASD_ECKD_RCD_DATA_SIZE];
__u8 lpm, opm, npm, ppm, epm, hpfpm, cablepm;
unsigned long flags;
char print_uid[60];
int rc;
data = container_of(work, struct path_verification_work_data, worker);
device = data->device;
/* delay path verification until device was resumed */
if (test_bit(DASD_FLAG_SUSPENDED, &device->flags)) {
schedule_work(work);
return;
}
/* check if path verification already running and delay if so */
if (test_and_set_bit(DASD_FLAG_PATH_VERIFY, &device->flags)) {
schedule_work(work);
return;
}
opm = 0;
npm = 0;
ppm = 0;
epm = 0;
hpfpm = 0;
cablepm = 0;
for (lpm = 0x80; lpm; lpm >>= 1) {
if (!(lpm & data->tbvpm))
continue;
memset(&data->rcd_buffer, 0, sizeof(data->rcd_buffer));
memset(&data->cqr, 0, sizeof(data->cqr));
data->cqr.cpaddr = &data->ccw;
rc = dasd_eckd_read_conf_immediately(device, &data->cqr,
data->rcd_buffer,
lpm);
if (!rc) {
switch (dasd_eckd_path_access(data->rcd_buffer,
DASD_ECKD_RCD_DATA_SIZE)
) {
case 0x02:
npm |= lpm;
break;
case 0x03:
ppm |= lpm;
break;
}
opm |= lpm;
} else if (rc == -EOPNOTSUPP) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"path verification: No configuration "
"data retrieved");
opm |= lpm;
} else if (rc == -EAGAIN) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"path verification: device is stopped,"
" try again later");
epm |= lpm;
} else {
dev_warn(&device->cdev->dev,
"Reading device feature codes failed "
"(rc=%d) for new path %x\n", rc, lpm);
continue;
}
if (verify_fcx_max_data(device, lpm)) {
opm &= ~lpm;
npm &= ~lpm;
ppm &= ~lpm;
hpfpm |= lpm;
continue;
}
/*
* save conf_data for comparison after
* rebuild_device_uid may have changed
* the original data
*/
memcpy(&path_rcd_buf, data->rcd_buffer,
DASD_ECKD_RCD_DATA_SIZE);
path_private.conf_data = (void *) &path_rcd_buf;
path_private.conf_len = DASD_ECKD_RCD_DATA_SIZE;
if (dasd_eckd_identify_conf_parts(&path_private)) {
path_private.conf_data = NULL;
path_private.conf_len = 0;
continue;
}
/*
* compare path UID with device UID only if at least
* one valid path is left
* in other case the device UID may have changed and
* the first working path UID will be used as device UID
*/
if (dasd_path_get_opm(device) &&
dasd_eckd_compare_path_uid(device, &path_private)) {
/*
* the comparison was not successful
* rebuild the device UID with at least one
* known path in case a z/VM hyperswap command
* has changed the device
*
* after this compare again
*
* if either the rebuild or the recompare fails
* the path can not be used
*/
if (rebuild_device_uid(device, data) ||
dasd_eckd_compare_path_uid(
device, &path_private)) {
uid = &path_private.uid;
if (strlen(uid->vduit) > 0)
snprintf(print_uid, sizeof(print_uid),
"%s.%s.%04x.%02x.%s",
uid->vendor, uid->serial,
uid->ssid, uid->real_unit_addr,
uid->vduit);
else
snprintf(print_uid, sizeof(print_uid),
"%s.%s.%04x.%02x",
uid->vendor, uid->serial,
uid->ssid,
uid->real_unit_addr);
dev_err(&device->cdev->dev,
"The newly added channel path %02X "
"will not be used because it leads "
"to a different device %s\n",
lpm, print_uid);
opm &= ~lpm;
npm &= ~lpm;
ppm &= ~lpm;
cablepm |= lpm;
continue;
}
}
/*
* There is a small chance that a path is lost again between
* above path verification and the following modification of
* the device opm mask. We could avoid that race here by using
* yet another path mask, but we rather deal with this unlikely
* situation in dasd_start_IO.
*/
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
if (!dasd_path_get_opm(device) && opm) {
dasd_path_set_opm(device, opm);
dasd_generic_path_operational(device);
} else {
dasd_path_add_opm(device, opm);
}
dasd_path_add_nppm(device, npm);
dasd_path_add_ppm(device, ppm);
dasd_path_add_tbvpm(device, epm);
dasd_path_add_cablepm(device, cablepm);
dasd_path_add_nohpfpm(device, hpfpm);
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
}
clear_bit(DASD_FLAG_PATH_VERIFY, &device->flags);
dasd_put_device(device);
if (data->isglobal)
mutex_unlock(&dasd_path_verification_mutex);
else
kfree(data);
}
static int dasd_eckd_verify_path(struct dasd_device *device, __u8 lpm)
{
struct path_verification_work_data *data;
data = kmalloc(sizeof(*data), GFP_ATOMIC | GFP_DMA);
if (!data) {
if (mutex_trylock(&dasd_path_verification_mutex)) {
data = path_verification_worker;
data->isglobal = 1;
} else
return -ENOMEM;
} else {
memset(data, 0, sizeof(*data));
data->isglobal = 0;
}
INIT_WORK(&data->worker, do_path_verification_work);
dasd_get_device(device);
data->device = device;
data->tbvpm = lpm;
schedule_work(&data->worker);
return 0;
}
static void dasd_eckd_reset_path(struct dasd_device *device, __u8 pm)
{
struct dasd_eckd_private *private = device->private;
unsigned long flags;
if (!private->fcx_max_data)
private->fcx_max_data = get_fcx_max_data(device);
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
dasd_path_set_tbvpm(device, pm ? : dasd_path_get_notoperpm(device));
dasd_schedule_device_bh(device);
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
}
static int dasd_eckd_read_features(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
struct dasd_psf_prssd_data *prssdp;
struct dasd_rssd_features *features;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
int rc;
memset(&private->features, 0, sizeof(struct dasd_rssd_features));
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ + 1 /* RSSD */,
(sizeof(struct dasd_psf_prssd_data) +
sizeof(struct dasd_rssd_features)),
device, NULL);
if (IS_ERR(cqr)) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s", "Could not "
"allocate initialization request");
return PTR_ERR(cqr);
}
cqr->startdev = device;
cqr->memdev = device;
cqr->block = NULL;
cqr->retries = 256;
cqr->expires = 10 * HZ;
/* Prepare for Read Subsystem Data */
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
prssdp->order = PSF_ORDER_PRSSD;
prssdp->suborder = 0x41; /* Read Feature Codes */
/* all other bytes of prssdp must be zero */
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->count = sizeof(struct dasd_psf_prssd_data);
ccw->flags |= CCW_FLAG_CC;
ccw->cda = (__u32)(addr_t) prssdp;
/* Read Subsystem Data - feature codes */
features = (struct dasd_rssd_features *) (prssdp + 1);
memset(features, 0, sizeof(struct dasd_rssd_features));
ccw++;
ccw->cmd_code = DASD_ECKD_CCW_RSSD;
ccw->count = sizeof(struct dasd_rssd_features);
ccw->cda = (__u32)(addr_t) features;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
rc = dasd_sleep_on(cqr);
if (rc == 0) {
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
features = (struct dasd_rssd_features *) (prssdp + 1);
memcpy(&private->features, features,
sizeof(struct dasd_rssd_features));
} else
dev_warn(&device->cdev->dev, "Reading device feature codes"
" failed with rc=%d\n", rc);
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* Build CP for Perform Subsystem Function - SSC.
*/
static struct dasd_ccw_req *dasd_eckd_build_psf_ssc(struct dasd_device *device,
int enable_pav)
{
struct dasd_ccw_req *cqr;
struct dasd_psf_ssc_data *psf_ssc_data;
struct ccw1 *ccw;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ ,
sizeof(struct dasd_psf_ssc_data),
device, NULL);
if (IS_ERR(cqr)) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Could not allocate PSF-SSC request");
return cqr;
}
psf_ssc_data = (struct dasd_psf_ssc_data *)cqr->data;
psf_ssc_data->order = PSF_ORDER_SSC;
psf_ssc_data->suborder = 0xc0;
if (enable_pav) {
psf_ssc_data->suborder |= 0x08;
psf_ssc_data->reserved[0] = 0x88;
}
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->cda = (__u32)(addr_t)psf_ssc_data;
ccw->count = 66;
cqr->startdev = device;
cqr->memdev = device;
cqr->block = NULL;
cqr->retries = 256;
cqr->expires = 10*HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
return cqr;
}
/*
* Perform Subsystem Function.
* It is necessary to trigger CIO for channel revalidation since this
* call might change behaviour of DASD devices.
*/
static int
dasd_eckd_psf_ssc(struct dasd_device *device, int enable_pav,
unsigned long flags)
{
struct dasd_ccw_req *cqr;
int rc;
cqr = dasd_eckd_build_psf_ssc(device, enable_pav);
if (IS_ERR(cqr))
return PTR_ERR(cqr);
/*
* set flags e.g. turn on failfast, to prevent blocking
* the calling function should handle failed requests
*/
cqr->flags |= flags;
rc = dasd_sleep_on(cqr);
if (!rc)
/* trigger CIO to reprobe devices */
css_schedule_reprobe();
else if (cqr->intrc == -EAGAIN)
rc = -EAGAIN;
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* Valide storage server of current device.
*/
static int dasd_eckd_validate_server(struct dasd_device *device,
unsigned long flags)
{
struct dasd_eckd_private *private = device->private;
int enable_pav, rc;
if (private->uid.type == UA_BASE_PAV_ALIAS ||
private->uid.type == UA_HYPER_PAV_ALIAS)
return 0;
if (dasd_nopav || MACHINE_IS_VM)
enable_pav = 0;
else
enable_pav = 1;
rc = dasd_eckd_psf_ssc(device, enable_pav, flags);
/* may be requested feature is not available on server,
* therefore just report error and go ahead */
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "PSF-SSC for SSID %04x "
"returned rc=%d", private->uid.ssid, rc);
return rc;
}
/*
* worker to do a validate server in case of a lost pathgroup
*/
static void dasd_eckd_do_validate_server(struct work_struct *work)
{
struct dasd_device *device = container_of(work, struct dasd_device,
kick_validate);
unsigned long flags = 0;
set_bit(DASD_CQR_FLAGS_FAILFAST, &flags);
if (dasd_eckd_validate_server(device, flags)
== -EAGAIN) {
/* schedule worker again if failed */
schedule_work(&device->kick_validate);
return;
}
dasd_put_device(device);
}
static void dasd_eckd_kick_validate_server(struct dasd_device *device)
{
dasd_get_device(device);
/* exit if device not online or in offline processing */
if (test_bit(DASD_FLAG_OFFLINE, &device->flags) ||
device->state < DASD_STATE_ONLINE) {
dasd_put_device(device);
return;
}
/* queue call to do_validate_server to the kernel event daemon. */
if (!schedule_work(&device->kick_validate))
dasd_put_device(device);
}
/*
* Check device characteristics.
* If the device is accessible using ECKD discipline, the device is enabled.
*/
static int
dasd_eckd_check_characteristics(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
struct dasd_block *block;
struct dasd_uid temp_uid;
int rc, i;
int readonly;
unsigned long value;
/* setup work queue for validate server*/
INIT_WORK(&device->kick_validate, dasd_eckd_do_validate_server);
/* setup work queue for summary unit check */
INIT_WORK(&device->suc_work, dasd_alias_handle_summary_unit_check);
if (!ccw_device_is_pathgroup(device->cdev)) {
dev_warn(&device->cdev->dev,
"A channel path group could not be established\n");
return -EIO;
}
if (!ccw_device_is_multipath(device->cdev)) {
dev_info(&device->cdev->dev,
"The DASD is not operating in multipath mode\n");
}
if (!private) {
private = kzalloc(sizeof(*private), GFP_KERNEL | GFP_DMA);
if (!private) {
dev_warn(&device->cdev->dev,
"Allocating memory for private DASD data "
"failed\n");
return -ENOMEM;
}
device->private = private;
} else {
memset(private, 0, sizeof(*private));
}
/* Invalidate status of initial analysis. */
private->init_cqr_status = -1;
/* Set default cache operations. */
private->attrib.operation = DASD_NORMAL_CACHE;
private->attrib.nr_cyl = 0;
/* Read Configuration Data */
rc = dasd_eckd_read_conf(device);
if (rc)
goto out_err1;
/* set some default values */
device->default_expires = DASD_EXPIRES;
device->default_retries = DASD_RETRIES;
device->path_thrhld = DASD_ECKD_PATH_THRHLD;
device->path_interval = DASD_ECKD_PATH_INTERVAL;
if (private->gneq) {
value = 1;
for (i = 0; i < private->gneq->timeout.value; i++)
value = 10 * value;
value = value * private->gneq->timeout.number;
/* do not accept useless values */
if (value != 0 && value <= DASD_EXPIRES_MAX)
device->default_expires = value;
}
dasd_eckd_get_uid(device, &temp_uid);
if (temp_uid.type == UA_BASE_DEVICE) {
block = dasd_alloc_block();
if (IS_ERR(block)) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"could not allocate dasd "
"block structure");
rc = PTR_ERR(block);
goto out_err1;
}
device->block = block;
block->base = device;
}
/* register lcu with alias handling, enable PAV */
rc = dasd_alias_make_device_known_to_lcu(device);
if (rc)
goto out_err2;
dasd_eckd_validate_server(device, 0);
/* device may report different configuration data after LCU setup */
rc = dasd_eckd_read_conf(device);
if (rc)
goto out_err3;
/* Read Feature Codes */
dasd_eckd_read_features(device);
/* Read Device Characteristics */
rc = dasd_generic_read_dev_chars(device, DASD_ECKD_MAGIC,
&private->rdc_data, 64);
if (rc) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Read device characteristic failed, rc=%d", rc);
goto out_err3;
}
if ((device->features & DASD_FEATURE_USERAW) &&
!(private->rdc_data.facilities.RT_in_LR)) {
dev_err(&device->cdev->dev, "The storage server does not "
"support raw-track access\n");
rc = -EINVAL;
goto out_err3;
}
/* find the valid cylinder size */
if (private->rdc_data.no_cyl == LV_COMPAT_CYL &&
private->rdc_data.long_no_cyl)
private->real_cyl = private->rdc_data.long_no_cyl;
else
private->real_cyl = private->rdc_data.no_cyl;
private->fcx_max_data = get_fcx_max_data(device);
readonly = dasd_device_is_ro(device);
if (readonly)
set_bit(DASD_FLAG_DEVICE_RO, &device->flags);
dev_info(&device->cdev->dev, "New DASD %04X/%02X (CU %04X/%02X) "
"with %d cylinders, %d heads, %d sectors%s\n",
private->rdc_data.dev_type,
private->rdc_data.dev_model,
private->rdc_data.cu_type,
private->rdc_data.cu_model.model,
private->real_cyl,
private->rdc_data.trk_per_cyl,
private->rdc_data.sec_per_trk,
readonly ? ", read-only device" : "");
return 0;
out_err3:
dasd_alias_disconnect_device_from_lcu(device);
out_err2:
dasd_free_block(device->block);
device->block = NULL;
out_err1:
kfree(private->conf_data);
kfree(device->private);
device->private = NULL;
return rc;
}
static void dasd_eckd_uncheck_device(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
int i;
if (!private)
return;
dasd_alias_disconnect_device_from_lcu(device);
private->ned = NULL;
private->sneq = NULL;
private->vdsneq = NULL;
private->gneq = NULL;
private->conf_len = 0;
for (i = 0; i < 8; i++) {
kfree(device->path[i].conf_data);
if ((__u8 *)device->path[i].conf_data ==
private->conf_data) {
private->conf_data = NULL;
private->conf_len = 0;
}
device->path[i].conf_data = NULL;
device->path[i].cssid = 0;
device->path[i].ssid = 0;
device->path[i].chpid = 0;
}
kfree(private->conf_data);
private->conf_data = NULL;
}
static struct dasd_ccw_req *
dasd_eckd_analysis_ccw(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
struct eckd_count *count_data;
struct LO_eckd_data *LO_data;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
int cplength, datasize;
int i;
cplength = 8;
datasize = sizeof(struct DE_eckd_data) + 2*sizeof(struct LO_eckd_data);
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize, device,
NULL);
if (IS_ERR(cqr))
return cqr;
ccw = cqr->cpaddr;
/* Define extent for the first 3 tracks. */
define_extent(ccw++, cqr->data, 0, 2,
DASD_ECKD_CCW_READ_COUNT, device, 0);
LO_data = cqr->data + sizeof(struct DE_eckd_data);
/* Locate record for the first 4 records on track 0. */
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, LO_data++, 0, 0, 4,
DASD_ECKD_CCW_READ_COUNT, device, 0);
count_data = private->count_area;
for (i = 0; i < 4; i++) {
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = DASD_ECKD_CCW_READ_COUNT;
ccw->flags = 0;
ccw->count = 8;
ccw->cda = (__u32)(addr_t) count_data;
ccw++;
count_data++;
}
/* Locate record for the first record on track 2. */
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, LO_data++, 2, 0, 1,
DASD_ECKD_CCW_READ_COUNT, device, 0);
/* Read count ccw. */
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = DASD_ECKD_CCW_READ_COUNT;
ccw->flags = 0;
ccw->count = 8;
ccw->cda = (__u32)(addr_t) count_data;
cqr->block = NULL;
cqr->startdev = device;
cqr->memdev = device;
cqr->retries = 255;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
return cqr;
}
/* differentiate between 'no record found' and any other error */
static int dasd_eckd_analysis_evaluation(struct dasd_ccw_req *init_cqr)
{
char *sense;
if (init_cqr->status == DASD_CQR_DONE)
return INIT_CQR_OK;
else if (init_cqr->status == DASD_CQR_NEED_ERP ||
init_cqr->status == DASD_CQR_FAILED) {
sense = dasd_get_sense(&init_cqr->irb);
if (sense && (sense[1] & SNS1_NO_REC_FOUND))
return INIT_CQR_UNFORMATTED;
else
return INIT_CQR_ERROR;
} else
return INIT_CQR_ERROR;
}
/*
* This is the callback function for the init_analysis cqr. It saves
* the status of the initial analysis ccw before it frees it and kicks
* the device to continue the startup sequence. This will call
* dasd_eckd_do_analysis again (if the devices has not been marked
* for deletion in the meantime).
*/
static void dasd_eckd_analysis_callback(struct dasd_ccw_req *init_cqr,
void *data)
{
struct dasd_device *device = init_cqr->startdev;
struct dasd_eckd_private *private = device->private;
private->init_cqr_status = dasd_eckd_analysis_evaluation(init_cqr);
dasd_sfree_request(init_cqr, device);
dasd_kick_device(device);
}
static int dasd_eckd_start_analysis(struct dasd_block *block)
{
struct dasd_ccw_req *init_cqr;
init_cqr = dasd_eckd_analysis_ccw(block->base);
if (IS_ERR(init_cqr))
return PTR_ERR(init_cqr);
init_cqr->callback = dasd_eckd_analysis_callback;
init_cqr->callback_data = NULL;
init_cqr->expires = 5*HZ;
/* first try without ERP, so we can later handle unformatted
* devices as special case
*/
clear_bit(DASD_CQR_FLAGS_USE_ERP, &init_cqr->flags);
init_cqr->retries = 0;
dasd_add_request_head(init_cqr);
return -EAGAIN;
}
static int dasd_eckd_end_analysis(struct dasd_block *block)
{
struct dasd_device *device = block->base;
struct dasd_eckd_private *private = device->private;
struct eckd_count *count_area;
unsigned int sb, blk_per_trk;
int status, i;
struct dasd_ccw_req *init_cqr;
status = private->init_cqr_status;
private->init_cqr_status = -1;
if (status == INIT_CQR_ERROR) {
/* try again, this time with full ERP */
init_cqr = dasd_eckd_analysis_ccw(device);
dasd_sleep_on(init_cqr);
status = dasd_eckd_analysis_evaluation(init_cqr);
dasd_sfree_request(init_cqr, device);
}
if (device->features & DASD_FEATURE_USERAW) {
block->bp_block = DASD_RAW_BLOCKSIZE;
blk_per_trk = DASD_RAW_BLOCK_PER_TRACK;
block->s2b_shift = 3;
goto raw;
}
if (status == INIT_CQR_UNFORMATTED) {
dev_warn(&device->cdev->dev, "The DASD is not formatted\n");
return -EMEDIUMTYPE;
} else if (status == INIT_CQR_ERROR) {
dev_err(&device->cdev->dev,
"Detecting the DASD disk layout failed because "
"of an I/O error\n");
return -EIO;
}
private->uses_cdl = 1;
/* Check Track 0 for Compatible Disk Layout */
count_area = NULL;
for (i = 0; i < 3; i++) {
if (private->count_area[i].kl != 4 ||
private->count_area[i].dl != dasd_eckd_cdl_reclen(i) - 4 ||
private->count_area[i].cyl != 0 ||
private->count_area[i].head != count_area_head[i] ||
private->count_area[i].record != count_area_rec[i]) {
private->uses_cdl = 0;
break;
}
}
if (i == 3)
count_area = &private->count_area[4];
if (private->uses_cdl == 0) {
for (i = 0; i < 5; i++) {
if ((private->count_area[i].kl != 0) ||
(private->count_area[i].dl !=
private->count_area[0].dl) ||
private->count_area[i].cyl != 0 ||
private->count_area[i].head != count_area_head[i] ||
private->count_area[i].record != count_area_rec[i])
break;
}
if (i == 5)
count_area = &private->count_area[0];
} else {
if (private->count_area[3].record == 1)
dev_warn(&device->cdev->dev,
"Track 0 has no records following the VTOC\n");
}
if (count_area != NULL && count_area->kl == 0) {
/* we found notthing violating our disk layout */
if (dasd_check_blocksize(count_area->dl) == 0)
block->bp_block = count_area->dl;
}
if (block->bp_block == 0) {
dev_warn(&device->cdev->dev,
"The disk layout of the DASD is not supported\n");
return -EMEDIUMTYPE;
}
block->s2b_shift = 0; /* bits to shift 512 to get a block */
for (sb = 512; sb < block->bp_block; sb = sb << 1)
block->s2b_shift++;
blk_per_trk = recs_per_track(&private->rdc_data, 0, block->bp_block);
raw:
block->blocks = (private->real_cyl *
private->rdc_data.trk_per_cyl *
blk_per_trk);
dev_info(&device->cdev->dev,
"DASD with %d KB/block, %d KB total size, %d KB/track, "
"%s\n", (block->bp_block >> 10),
((private->real_cyl *
private->rdc_data.trk_per_cyl *
blk_per_trk * (block->bp_block >> 9)) >> 1),
((blk_per_trk * block->bp_block) >> 10),
private->uses_cdl ?
"compatible disk layout" : "linux disk layout");
return 0;
}
static int dasd_eckd_do_analysis(struct dasd_block *block)
{
struct dasd_eckd_private *private = block->base->private;
if (private->init_cqr_status < 0)
return dasd_eckd_start_analysis(block);
else
return dasd_eckd_end_analysis(block);
}
static int dasd_eckd_basic_to_ready(struct dasd_device *device)
{
return dasd_alias_add_device(device);
};
static int dasd_eckd_online_to_ready(struct dasd_device *device)
{
if (cancel_work_sync(&device->reload_device))
dasd_put_device(device);
if (cancel_work_sync(&device->kick_validate))
dasd_put_device(device);
return 0;
};
static int dasd_eckd_basic_to_known(struct dasd_device *device)
{
return dasd_alias_remove_device(device);
};
static int
dasd_eckd_fill_geometry(struct dasd_block *block, struct hd_geometry *geo)
{
struct dasd_eckd_private *private = block->base->private;
if (dasd_check_blocksize(block->bp_block) == 0) {
geo->sectors = recs_per_track(&private->rdc_data,
0, block->bp_block);
}
geo->cylinders = private->rdc_data.no_cyl;
geo->heads = private->rdc_data.trk_per_cyl;
return 0;
}
/*
* Build the TCW request for the format check
*/
static struct dasd_ccw_req *
dasd_eckd_build_check_tcw(struct dasd_device *base, struct format_data_t *fdata,
int enable_pav, struct eckd_count *fmt_buffer,
int rpt)
{
struct dasd_eckd_private *start_priv;
struct dasd_device *startdev = NULL;
struct tidaw *last_tidaw = NULL;
struct dasd_ccw_req *cqr;
struct itcw *itcw;
int itcw_size;
int count;
int rc;
int i;
if (enable_pav)
startdev = dasd_alias_get_start_dev(base);
if (!startdev)
startdev = base;
start_priv = startdev->private;
count = rpt * (fdata->stop_unit - fdata->start_unit + 1);
/*
* we're adding 'count' amount of tidaw to the itcw.
* calculate the corresponding itcw_size
*/
itcw_size = itcw_calc_size(0, count, 0);
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 0, itcw_size, startdev,
NULL);
if (IS_ERR(cqr))
return cqr;
start_priv->count++;
itcw = itcw_init(cqr->data, itcw_size, ITCW_OP_READ, 0, count, 0);
if (IS_ERR(itcw)) {
rc = -EINVAL;
goto out_err;
}
cqr->cpaddr = itcw_get_tcw(itcw);
rc = prepare_itcw(itcw, fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_READ_COUNT_MT, base, startdev, 0, count,
sizeof(struct eckd_count),
count * sizeof(struct eckd_count), 0, rpt);
if (rc)
goto out_err;
for (i = 0; i < count; i++) {
last_tidaw = itcw_add_tidaw(itcw, 0, fmt_buffer++,
sizeof(struct eckd_count));
if (IS_ERR(last_tidaw)) {
rc = -EINVAL;
goto out_err;
}
}
last_tidaw->flags |= TIDAW_FLAGS_LAST;
itcw_finalize(itcw);
cqr->cpmode = 1;
cqr->startdev = startdev;
cqr->memdev = startdev;
cqr->basedev = base;
cqr->retries = startdev->default_retries;
cqr->expires = startdev->default_expires * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
/* Set flags to suppress output for expected errors */
set_bit(DASD_CQR_SUPPRESS_FP, &cqr->flags);
set_bit(DASD_CQR_SUPPRESS_IL, &cqr->flags);
return cqr;
out_err:
dasd_sfree_request(cqr, startdev);
return ERR_PTR(rc);
}
/*
* Build the CCW request for the format check
*/
static struct dasd_ccw_req *
dasd_eckd_build_check(struct dasd_device *base, struct format_data_t *fdata,
int enable_pav, struct eckd_count *fmt_buffer, int rpt)
{
struct dasd_eckd_private *start_priv;
struct dasd_eckd_private *base_priv;
struct dasd_device *startdev = NULL;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
void *data;
int cplength, datasize;
int use_prefix;
int count;
int i;
if (enable_pav)
startdev = dasd_alias_get_start_dev(base);
if (!startdev)
startdev = base;
start_priv = startdev->private;
base_priv = base->private;
count = rpt * (fdata->stop_unit - fdata->start_unit + 1);
use_prefix = base_priv->features.feature[8] & 0x01;
if (use_prefix) {
cplength = 1;
datasize = sizeof(struct PFX_eckd_data);
} else {
cplength = 2;
datasize = sizeof(struct DE_eckd_data) +
sizeof(struct LO_eckd_data);
}
cplength += count;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize,
startdev, NULL);
if (IS_ERR(cqr))
return cqr;
start_priv->count++;
data = cqr->data;
ccw = cqr->cpaddr;
if (use_prefix) {
prefix_LRE(ccw++, data, fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_READ_COUNT, base, startdev, 1, 0,
count, 0, 0);
} else {
define_extent(ccw++, data, fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_READ_COUNT, startdev, 0);
data += sizeof(struct DE_eckd_data);
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, data, fdata->start_unit, 0, count,
DASD_ECKD_CCW_READ_COUNT, base, 0);
}
for (i = 0; i < count; i++) {
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = DASD_ECKD_CCW_READ_COUNT;
ccw->flags = CCW_FLAG_SLI;
ccw->count = 8;
ccw->cda = (__u32)(addr_t) fmt_buffer;
ccw++;
fmt_buffer++;
}
cqr->startdev = startdev;
cqr->memdev = startdev;
cqr->basedev = base;
cqr->retries = DASD_RETRIES;
cqr->expires = startdev->default_expires * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
/* Set flags to suppress output for expected errors */
set_bit(DASD_CQR_SUPPRESS_NRF, &cqr->flags);
return cqr;
}
static struct dasd_ccw_req *
dasd_eckd_build_format(struct dasd_device *base,
struct format_data_t *fdata,
int enable_pav)
{
struct dasd_eckd_private *base_priv;
struct dasd_eckd_private *start_priv;
struct dasd_device *startdev = NULL;
struct dasd_ccw_req *fcp;
struct eckd_count *ect;
struct ch_t address;
struct ccw1 *ccw;
void *data;
int rpt;
int cplength, datasize;
int i, j;
int intensity = 0;
int r0_perm;
int nr_tracks;
int use_prefix;
if (enable_pav)
startdev = dasd_alias_get_start_dev(base);
if (!startdev)
startdev = base;
start_priv = startdev->private;
base_priv = base->private;
rpt = recs_per_track(&base_priv->rdc_data, 0, fdata->blksize);
nr_tracks = fdata->stop_unit - fdata->start_unit + 1;
/*
* fdata->intensity is a bit string that tells us what to do:
* Bit 0: write record zero
* Bit 1: write home address, currently not supported
* Bit 2: invalidate tracks
* Bit 3: use OS/390 compatible disk layout (cdl)
* Bit 4: do not allow storage subsystem to modify record zero
* Only some bit combinations do make sense.
*/
if (fdata->intensity & 0x10) {
r0_perm = 0;
intensity = fdata->intensity & ~0x10;
} else {
r0_perm = 1;
intensity = fdata->intensity;
}
use_prefix = base_priv->features.feature[8] & 0x01;
switch (intensity) {
case 0x00: /* Normal format */
case 0x08: /* Normal format, use cdl. */
cplength = 2 + (rpt*nr_tracks);
if (use_prefix)
datasize = sizeof(struct PFX_eckd_data) +
sizeof(struct LO_eckd_data) +
rpt * nr_tracks * sizeof(struct eckd_count);
else
datasize = sizeof(struct DE_eckd_data) +
sizeof(struct LO_eckd_data) +
rpt * nr_tracks * sizeof(struct eckd_count);
break;
case 0x01: /* Write record zero and format track. */
case 0x09: /* Write record zero and format track, use cdl. */
cplength = 2 + rpt * nr_tracks;
if (use_prefix)
datasize = sizeof(struct PFX_eckd_data) +
sizeof(struct LO_eckd_data) +
sizeof(struct eckd_count) +
rpt * nr_tracks * sizeof(struct eckd_count);
else
datasize = sizeof(struct DE_eckd_data) +
sizeof(struct LO_eckd_data) +
sizeof(struct eckd_count) +
rpt * nr_tracks * sizeof(struct eckd_count);
break;
case 0x04: /* Invalidate track. */
case 0x0c: /* Invalidate track, use cdl. */
cplength = 3;
if (use_prefix)
datasize = sizeof(struct PFX_eckd_data) +
sizeof(struct LO_eckd_data) +
sizeof(struct eckd_count);
else
datasize = sizeof(struct DE_eckd_data) +
sizeof(struct LO_eckd_data) +
sizeof(struct eckd_count);
break;
default:
dev_warn(&startdev->cdev->dev,
"An I/O control call used incorrect flags 0x%x\n",
fdata->intensity);
return ERR_PTR(-EINVAL);
}
/* Allocate the format ccw request. */
fcp = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength,
datasize, startdev, NULL);
if (IS_ERR(fcp))
return fcp;
start_priv->count++;
data = fcp->data;
ccw = fcp->cpaddr;
switch (intensity & ~0x08) {
case 0x00: /* Normal format. */
if (use_prefix) {
prefix(ccw++, (struct PFX_eckd_data *) data,
fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_WRITE_CKD, base, startdev);
/* grant subsystem permission to format R0 */
if (r0_perm)
((struct PFX_eckd_data *)data)
->define_extent.ga_extended |= 0x04;
data += sizeof(struct PFX_eckd_data);
} else {
define_extent(ccw++, (struct DE_eckd_data *) data,
fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_WRITE_CKD, startdev, 0);
/* grant subsystem permission to format R0 */
if (r0_perm)
((struct DE_eckd_data *) data)
->ga_extended |= 0x04;
data += sizeof(struct DE_eckd_data);
}
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, (struct LO_eckd_data *) data,
fdata->start_unit, 0, rpt*nr_tracks,
DASD_ECKD_CCW_WRITE_CKD, base,
fdata->blksize);
data += sizeof(struct LO_eckd_data);
break;
case 0x01: /* Write record zero + format track. */
if (use_prefix) {
prefix(ccw++, (struct PFX_eckd_data *) data,
fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_WRITE_RECORD_ZERO,
base, startdev);
data += sizeof(struct PFX_eckd_data);
} else {
define_extent(ccw++, (struct DE_eckd_data *) data,
fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_WRITE_RECORD_ZERO, startdev, 0);
data += sizeof(struct DE_eckd_data);
}
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, (struct LO_eckd_data *) data,
fdata->start_unit, 0, rpt * nr_tracks + 1,
DASD_ECKD_CCW_WRITE_RECORD_ZERO, base,
base->block->bp_block);
data += sizeof(struct LO_eckd_data);
break;
case 0x04: /* Invalidate track. */
if (use_prefix) {
prefix(ccw++, (struct PFX_eckd_data *) data,
fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_WRITE_CKD, base, startdev);
data += sizeof(struct PFX_eckd_data);
} else {
define_extent(ccw++, (struct DE_eckd_data *) data,
fdata->start_unit, fdata->stop_unit,
DASD_ECKD_CCW_WRITE_CKD, startdev, 0);
data += sizeof(struct DE_eckd_data);
}
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, (struct LO_eckd_data *) data,
fdata->start_unit, 0, 1,
DASD_ECKD_CCW_WRITE_CKD, base, 8);
data += sizeof(struct LO_eckd_data);
break;
}
for (j = 0; j < nr_tracks; j++) {
/* calculate cylinder and head for the current track */
set_ch_t(&address,
(fdata->start_unit + j) /
base_priv->rdc_data.trk_per_cyl,
(fdata->start_unit + j) %
base_priv->rdc_data.trk_per_cyl);
if (intensity & 0x01) { /* write record zero */
ect = (struct eckd_count *) data;
data += sizeof(struct eckd_count);
ect->cyl = address.cyl;
ect->head = address.head;
ect->record = 0;
ect->kl = 0;
ect->dl = 8;
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = DASD_ECKD_CCW_WRITE_RECORD_ZERO;
ccw->flags = CCW_FLAG_SLI;
ccw->count = 8;
ccw->cda = (__u32)(addr_t) ect;
ccw++;
}
if ((intensity & ~0x08) & 0x04) { /* erase track */
ect = (struct eckd_count *) data;
data += sizeof(struct eckd_count);
ect->cyl = address.cyl;
ect->head = address.head;
ect->record = 1;
ect->kl = 0;
ect->dl = 0;
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = DASD_ECKD_CCW_WRITE_CKD;
ccw->flags = CCW_FLAG_SLI;
ccw->count = 8;
ccw->cda = (__u32)(addr_t) ect;
} else { /* write remaining records */
for (i = 0; i < rpt; i++) {
ect = (struct eckd_count *) data;
data += sizeof(struct eckd_count);
ect->cyl = address.cyl;
ect->head = address.head;
ect->record = i + 1;
ect->kl = 0;
ect->dl = fdata->blksize;
/*
* Check for special tracks 0-1
* when formatting CDL
*/
if ((intensity & 0x08) &&
address.cyl == 0 && address.head == 0) {
if (i < 3) {
ect->kl = 4;
ect->dl = sizes_trk0[i] - 4;
}
}
if ((intensity & 0x08) &&
address.cyl == 0 && address.head == 1) {
ect->kl = 44;
ect->dl = LABEL_SIZE - 44;
}
ccw[-1].flags |= CCW_FLAG_CC;
if (i != 0 || j == 0)
ccw->cmd_code =
DASD_ECKD_CCW_WRITE_CKD;
else
ccw->cmd_code =
DASD_ECKD_CCW_WRITE_CKD_MT;
ccw->flags = CCW_FLAG_SLI;
ccw->count = 8;
ccw->cda = (__u32)(addr_t) ect;
ccw++;
}
}
}
fcp->startdev = startdev;
fcp->memdev = startdev;
fcp->basedev = base;
fcp->retries = 256;
fcp->expires = startdev->default_expires * HZ;
fcp->buildclk = get_tod_clock();
fcp->status = DASD_CQR_FILLED;
return fcp;
}
/*
* Wrapper function to build a CCW request depending on input data
*/
static struct dasd_ccw_req *
dasd_eckd_format_build_ccw_req(struct dasd_device *base,
struct format_data_t *fdata, int enable_pav,
int tpm, struct eckd_count *fmt_buffer, int rpt)
{
struct dasd_ccw_req *ccw_req;
if (!fmt_buffer) {
ccw_req = dasd_eckd_build_format(base, fdata, enable_pav);
} else {
if (tpm)
ccw_req = dasd_eckd_build_check_tcw(base, fdata,
enable_pav,
fmt_buffer, rpt);
else
ccw_req = dasd_eckd_build_check(base, fdata, enable_pav,
fmt_buffer, rpt);
}
return ccw_req;
}
/*
* Sanity checks on format_data
*/
static int dasd_eckd_format_sanity_checks(struct dasd_device *base,
struct format_data_t *fdata)
{
struct dasd_eckd_private *private = base->private;
if (fdata->start_unit >=
(private->real_cyl * private->rdc_data.trk_per_cyl)) {
dev_warn(&base->cdev->dev,
"Start track number %u used in formatting is too big\n",
fdata->start_unit);
return -EINVAL;
}
if (fdata->stop_unit >=
(private->real_cyl * private->rdc_data.trk_per_cyl)) {
dev_warn(&base->cdev->dev,
"Stop track number %u used in formatting is too big\n",
fdata->stop_unit);
return -EINVAL;
}
if (fdata->start_unit > fdata->stop_unit) {
dev_warn(&base->cdev->dev,
"Start track %u used in formatting exceeds end track\n",
fdata->start_unit);
return -EINVAL;
}
if (dasd_check_blocksize(fdata->blksize) != 0) {
dev_warn(&base->cdev->dev,
"The DASD cannot be formatted with block size %u\n",
fdata->blksize);
return -EINVAL;
}
return 0;
}
/*
* This function will process format_data originally coming from an IOCTL
*/
static int dasd_eckd_format_process_data(struct dasd_device *base,
struct format_data_t *fdata,
int enable_pav, int tpm,
struct eckd_count *fmt_buffer, int rpt,
struct irb *irb)
{
struct dasd_eckd_private *private = base->private;
struct dasd_ccw_req *cqr, *n;
struct list_head format_queue;
struct dasd_device *device;
char *sense = NULL;
int old_start, old_stop, format_step;
int step, retry;
int rc;
rc = dasd_eckd_format_sanity_checks(base, fdata);
if (rc)
return rc;
INIT_LIST_HEAD(&format_queue);
old_start = fdata->start_unit;
old_stop = fdata->stop_unit;
if (!tpm && fmt_buffer != NULL) {
/* Command Mode / Format Check */
format_step = 1;
} else if (tpm && fmt_buffer != NULL) {
/* Transport Mode / Format Check */
format_step = DASD_CQR_MAX_CCW / rpt;
} else {
/* Normal Formatting */
format_step = DASD_CQR_MAX_CCW /
recs_per_track(&private->rdc_data, 0, fdata->blksize);
}
do {
retry = 0;
while (fdata->start_unit <= old_stop) {
step = fdata->stop_unit - fdata->start_unit + 1;
if (step > format_step) {
fdata->stop_unit =
fdata->start_unit + format_step - 1;
}
cqr = dasd_eckd_format_build_ccw_req(base, fdata,
enable_pav, tpm,
fmt_buffer, rpt);
if (IS_ERR(cqr)) {
rc = PTR_ERR(cqr);
if (rc == -ENOMEM) {
if (list_empty(&format_queue))
goto out;
/*
* not enough memory available, start
* requests retry after first requests
* were finished
*/
retry = 1;
break;
}
goto out_err;
}
list_add_tail(&cqr->blocklist, &format_queue);
if (fmt_buffer) {
step = fdata->stop_unit - fdata->start_unit + 1;
fmt_buffer += rpt * step;
}
fdata->start_unit = fdata->stop_unit + 1;
fdata->stop_unit = old_stop;
}
rc = dasd_sleep_on_queue(&format_queue);
out_err:
list_for_each_entry_safe(cqr, n, &format_queue, blocklist) {
device = cqr->startdev;
private = device->private;
if (cqr->status == DASD_CQR_FAILED) {
/*
* Only get sense data if called by format
* check
*/
if (fmt_buffer && irb) {
sense = dasd_get_sense(&cqr->irb);
memcpy(irb, &cqr->irb, sizeof(*irb));
}
rc = -EIO;
}
list_del_init(&cqr->blocklist);
dasd_sfree_request(cqr, device);
private->count--;
}
if (rc && rc != -EIO)
goto out;
if (rc == -EIO) {
/*
* In case fewer than the expected records are on the
* track, we will most likely get a 'No Record Found'
* error (in command mode) or a 'File Protected' error
* (in transport mode). Those particular cases shouldn't
* pass the -EIO to the IOCTL, therefore reset the rc
* and continue.
*/
if (sense &&
(sense[1] & SNS1_NO_REC_FOUND ||
sense[1] & SNS1_FILE_PROTECTED))
retry = 1;
else
goto out;
}
} while (retry);
out:
fdata->start_unit = old_start;
fdata->stop_unit = old_stop;
return rc;
}
static int dasd_eckd_format_device(struct dasd_device *base,
struct format_data_t *fdata, int enable_pav)
{
return dasd_eckd_format_process_data(base, fdata, enable_pav, 0, NULL,
0, NULL);
}
/*
* Helper function to count consecutive records of a single track.
*/
static int dasd_eckd_count_records(struct eckd_count *fmt_buffer, int start,
int max)
{
int head;
int i;
head = fmt_buffer[start].head;
/*
* There are 3 conditions where we stop counting:
* - if data reoccurs (same head and record may reoccur), which may
* happen due to the way DASD_ECKD_CCW_READ_COUNT works
* - when the head changes, because we're iterating over several tracks
* then (DASD_ECKD_CCW_READ_COUNT_MT)
* - when we've reached the end of sensible data in the buffer (the
* record will be 0 then)
*/
for (i = start; i < max; i++) {
if (i > start) {
if ((fmt_buffer[i].head == head &&
fmt_buffer[i].record == 1) ||
fmt_buffer[i].head != head ||
fmt_buffer[i].record == 0)
break;
}
}
return i - start;
}
/*
* Evaluate a given range of tracks. Data like number of records, blocksize,
* record ids, and key length are compared with expected data.
*
* If a mismatch occurs, the corresponding error bit is set, as well as
* additional information, depending on the error.
*/
static void dasd_eckd_format_evaluate_tracks(struct eckd_count *fmt_buffer,
struct format_check_t *cdata,
int rpt_max, int rpt_exp,
int trk_per_cyl, int tpm)
{
struct ch_t geo;
int max_entries;
int count = 0;
int trkcount;
int blksize;
int pos = 0;
int i, j;
int kl;
trkcount = cdata->expect.stop_unit - cdata->expect.start_unit + 1;
max_entries = trkcount * rpt_max;
for (i = cdata->expect.start_unit; i <= cdata->expect.stop_unit; i++) {
/* Calculate the correct next starting position in the buffer */
if (tpm) {
while (fmt_buffer[pos].record == 0 &&
fmt_buffer[pos].dl == 0) {
if (pos++ > max_entries)
break;
}
} else {
if (i != cdata->expect.start_unit)
pos += rpt_max - count;
}
/* Calculate the expected geo values for the current track */
set_ch_t(&geo, i / trk_per_cyl, i % trk_per_cyl);
/* Count and check number of records */
count = dasd_eckd_count_records(fmt_buffer, pos, pos + rpt_max);
if (count < rpt_exp) {
cdata->result = DASD_FMT_ERR_TOO_FEW_RECORDS;
break;
}
if (count > rpt_exp) {
cdata->result = DASD_FMT_ERR_TOO_MANY_RECORDS;
break;
}
for (j = 0; j < count; j++, pos++) {
blksize = cdata->expect.blksize;
kl = 0;
/*
* Set special values when checking CDL formatted
* devices.
*/
if ((cdata->expect.intensity & 0x08) &&
geo.cyl == 0 && geo.head == 0) {
if (j < 3) {
blksize = sizes_trk0[j] - 4;
kl = 4;
}
}
if ((cdata->expect.intensity & 0x08) &&
geo.cyl == 0 && geo.head == 1) {
blksize = LABEL_SIZE - 44;
kl = 44;
}
/* Check blocksize */
if (fmt_buffer[pos].dl != blksize) {
cdata->result = DASD_FMT_ERR_BLKSIZE;
goto out;
}
/* Check if key length is 0 */
if (fmt_buffer[pos].kl != kl) {
cdata->result = DASD_FMT_ERR_KEY_LENGTH;
goto out;
}
/* Check if record_id is correct */
if (fmt_buffer[pos].cyl != geo.cyl ||
fmt_buffer[pos].head != geo.head ||
fmt_buffer[pos].record != (j + 1)) {
cdata->result = DASD_FMT_ERR_RECORD_ID;
goto out;
}
}
}
out:
/*
* In case of no errors, we need to decrease by one
* to get the correct positions.
*/
if (!cdata->result) {
i--;
pos--;
}
cdata->unit = i;
cdata->num_records = count;
cdata->rec = fmt_buffer[pos].record;
cdata->blksize = fmt_buffer[pos].dl;
cdata->key_length = fmt_buffer[pos].kl;
}
/*
* Check the format of a range of tracks of a DASD.
*/
static int dasd_eckd_check_device_format(struct dasd_device *base,
struct format_check_t *cdata,
int enable_pav)
{
struct dasd_eckd_private *private = base->private;
struct eckd_count *fmt_buffer;
struct irb irb;
int rpt_max, rpt_exp;
int fmt_buffer_size;
int trk_per_cyl;
int trkcount;
int tpm = 0;
int rc;
trk_per_cyl = private->rdc_data.trk_per_cyl;
/* Get maximum and expected amount of records per track */
rpt_max = recs_per_track(&private->rdc_data, 0, 512) + 1;
rpt_exp = recs_per_track(&private->rdc_data, 0, cdata->expect.blksize);
trkcount = cdata->expect.stop_unit - cdata->expect.start_unit + 1;
fmt_buffer_size = trkcount * rpt_max * sizeof(struct eckd_count);
fmt_buffer = kzalloc(fmt_buffer_size, GFP_KERNEL | GFP_DMA);
if (!fmt_buffer)
return -ENOMEM;
/*
* A certain FICON feature subset is needed to operate in transport
* mode. Additionally, the support for transport mode is implicitly
* checked by comparing the buffer size with fcx_max_data. As long as
* the buffer size is smaller we can operate in transport mode and
* process multiple tracks. If not, only one track at once is being
* processed using command mode.
*/
if ((private->features.feature[40] & 0x04) &&
fmt_buffer_size <= private->fcx_max_data)
tpm = 1;
rc = dasd_eckd_format_process_data(base, &cdata->expect, enable_pav,
tpm, fmt_buffer, rpt_max, &irb);
if (rc && rc != -EIO)
goto out;
if (rc == -EIO) {
/*
* If our first attempt with transport mode enabled comes back
* with an incorrect length error, we're going to retry the
* check with command mode.
*/
if (tpm && scsw_cstat(&irb.scsw) == 0x40) {
tpm = 0;
rc = dasd_eckd_format_process_data(base, &cdata->expect,
enable_pav, tpm,
fmt_buffer, rpt_max,
&irb);
if (rc)
goto out;
} else {
goto out;
}
}
dasd_eckd_format_evaluate_tracks(fmt_buffer, cdata, rpt_max, rpt_exp,
trk_per_cyl, tpm);
out:
kfree(fmt_buffer);
return rc;
}
static void dasd_eckd_handle_terminated_request(struct dasd_ccw_req *cqr)
{
if (cqr->retries < 0) {
cqr->status = DASD_CQR_FAILED;
return;
}
cqr->status = DASD_CQR_FILLED;
if (cqr->block && (cqr->startdev != cqr->block->base)) {
dasd_eckd_reset_ccw_to_base_io(cqr);
cqr->startdev = cqr->block->base;
cqr->lpm = dasd_path_get_opm(cqr->block->base);
}
};
static dasd_erp_fn_t
dasd_eckd_erp_action(struct dasd_ccw_req * cqr)
{
struct dasd_device *device = (struct dasd_device *) cqr->startdev;
struct ccw_device *cdev = device->cdev;
switch (cdev->id.cu_type) {
case 0x3990:
case 0x2105:
case 0x2107:
case 0x1750:
return dasd_3990_erp_action;
case 0x9343:
case 0x3880:
default:
return dasd_default_erp_action;
}
}
static dasd_erp_fn_t
dasd_eckd_erp_postaction(struct dasd_ccw_req * cqr)
{
return dasd_default_erp_postaction;
}
static void dasd_eckd_check_for_device_change(struct dasd_device *device,
struct dasd_ccw_req *cqr,
struct irb *irb)
{
char mask;
char *sense = NULL;
struct dasd_eckd_private *private = device->private;
/* first of all check for state change pending interrupt */
mask = DEV_STAT_ATTENTION | DEV_STAT_DEV_END | DEV_STAT_UNIT_EXCEP;
if ((scsw_dstat(&irb->scsw) & mask) == mask) {
/*
* for alias only, not in offline processing
* and only if not suspended
*/
if (!device->block && private->lcu &&
device->state == DASD_STATE_ONLINE &&
!test_bit(DASD_FLAG_OFFLINE, &device->flags) &&
!test_bit(DASD_FLAG_SUSPENDED, &device->flags)) {
/* schedule worker to reload device */
dasd_reload_device(device);
}
dasd_generic_handle_state_change(device);
return;
}
sense = dasd_get_sense(irb);
if (!sense)
return;
/* summary unit check */
if ((sense[27] & DASD_SENSE_BIT_0) && (sense[7] == 0x0D) &&
(scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK)) {
if (test_and_set_bit(DASD_FLAG_SUC, &device->flags)) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"eckd suc: device already notified");
return;
}
sense = dasd_get_sense(irb);
if (!sense) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"eckd suc: no reason code available");
clear_bit(DASD_FLAG_SUC, &device->flags);
return;
}
private->suc_reason = sense[8];
DBF_DEV_EVENT(DBF_NOTICE, device, "%s %x",
"eckd handle summary unit check: reason",
private->suc_reason);
dasd_get_device(device);
if (!schedule_work(&device->suc_work))
dasd_put_device(device);
return;
}
/* service information message SIM */
if (!cqr && !(sense[27] & DASD_SENSE_BIT_0) &&
((sense[6] & DASD_SIM_SENSE) == DASD_SIM_SENSE)) {
dasd_3990_erp_handle_sim(device, sense);
return;
}
/* loss of device reservation is handled via base devices only
* as alias devices may be used with several bases
*/
if (device->block && (sense[27] & DASD_SENSE_BIT_0) &&
(sense[7] == 0x3F) &&
(scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK) &&
test_bit(DASD_FLAG_IS_RESERVED, &device->flags)) {
if (device->features & DASD_FEATURE_FAILONSLCK)
set_bit(DASD_FLAG_LOCK_STOLEN, &device->flags);
clear_bit(DASD_FLAG_IS_RESERVED, &device->flags);
dev_err(&device->cdev->dev,
"The device reservation was lost\n");
}
}
static struct dasd_ccw_req *dasd_eckd_build_cp_cmd_single(
struct dasd_device *startdev,
struct dasd_block *block,
struct request *req,
sector_t first_rec,
sector_t last_rec,
sector_t first_trk,
sector_t last_trk,
unsigned int first_offs,
unsigned int last_offs,
unsigned int blk_per_trk,
unsigned int blksize)
{
struct dasd_eckd_private *private;
unsigned long *idaws;
struct LO_eckd_data *LO_data;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
struct req_iterator iter;
struct bio_vec bv;
char *dst;
unsigned int off;
int count, cidaw, cplength, datasize;
sector_t recid;
unsigned char cmd, rcmd;
int use_prefix;
struct dasd_device *basedev;
basedev = block->base;
private = basedev->private;
if (rq_data_dir(req) == READ)
cmd = DASD_ECKD_CCW_READ_MT;
else if (rq_data_dir(req) == WRITE)
cmd = DASD_ECKD_CCW_WRITE_MT;
else
return ERR_PTR(-EINVAL);
/* Check struct bio and count the number of blocks for the request. */
count = 0;
cidaw = 0;
rq_for_each_segment(bv, req, iter) {
if (bv.bv_len & (blksize - 1))
/* Eckd can only do full blocks. */
return ERR_PTR(-EINVAL);
count += bv.bv_len >> (block->s2b_shift + 9);
if (idal_is_needed (page_address(bv.bv_page), bv.bv_len))
cidaw += bv.bv_len >> (block->s2b_shift + 9);
}
/* Paranoia. */
if (count != last_rec - first_rec + 1)
return ERR_PTR(-EINVAL);
/* use the prefix command if available */
use_prefix = private->features.feature[8] & 0x01;
if (use_prefix) {
/* 1x prefix + number of blocks */
cplength = 2 + count;
/* 1x prefix + cidaws*sizeof(long) */
datasize = sizeof(struct PFX_eckd_data) +
sizeof(struct LO_eckd_data) +
cidaw * sizeof(unsigned long);
} else {
/* 1x define extent + 1x locate record + number of blocks */
cplength = 2 + count;
/* 1x define extent + 1x locate record + cidaws*sizeof(long) */
datasize = sizeof(struct DE_eckd_data) +
sizeof(struct LO_eckd_data) +
cidaw * sizeof(unsigned long);
}
/* Find out the number of additional locate record ccws for cdl. */
if (private->uses_cdl && first_rec < 2*blk_per_trk) {
if (last_rec >= 2*blk_per_trk)
count = 2*blk_per_trk - first_rec;
cplength += count;
datasize += count*sizeof(struct LO_eckd_data);
}
/* Allocate the ccw request. */
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize,
startdev, blk_mq_rq_to_pdu(req));
if (IS_ERR(cqr))
return cqr;
ccw = cqr->cpaddr;
/* First ccw is define extent or prefix. */
if (use_prefix) {
if (prefix(ccw++, cqr->data, first_trk,
last_trk, cmd, basedev, startdev) == -EAGAIN) {
/* Clock not in sync and XRC is enabled.
* Try again later.
*/
dasd_sfree_request(cqr, startdev);
return ERR_PTR(-EAGAIN);
}
idaws = (unsigned long *) (cqr->data +
sizeof(struct PFX_eckd_data));
} else {
if (define_extent(ccw++, cqr->data, first_trk,
last_trk, cmd, basedev, 0) == -EAGAIN) {
/* Clock not in sync and XRC is enabled.
* Try again later.
*/
dasd_sfree_request(cqr, startdev);
return ERR_PTR(-EAGAIN);
}
idaws = (unsigned long *) (cqr->data +
sizeof(struct DE_eckd_data));
}
/* Build locate_record+read/write/ccws. */
LO_data = (struct LO_eckd_data *) (idaws + cidaw);
recid = first_rec;
if (private->uses_cdl == 0 || recid > 2*blk_per_trk) {
/* Only standard blocks so there is just one locate record. */
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, LO_data++, first_trk, first_offs + 1,
last_rec - recid + 1, cmd, basedev, blksize);
}
rq_for_each_segment(bv, req, iter) {
dst = page_address(bv.bv_page) + bv.bv_offset;
if (dasd_page_cache) {
char *copy = kmem_cache_alloc(dasd_page_cache,
GFP_DMA | __GFP_NOWARN);
if (copy && rq_data_dir(req) == WRITE)
memcpy(copy + bv.bv_offset, dst, bv.bv_len);
if (copy)
dst = copy + bv.bv_offset;
}
for (off = 0; off < bv.bv_len; off += blksize) {
sector_t trkid = recid;
unsigned int recoffs = sector_div(trkid, blk_per_trk);
rcmd = cmd;
count = blksize;
/* Locate record for cdl special block ? */
if (private->uses_cdl && recid < 2*blk_per_trk) {
if (dasd_eckd_cdl_special(blk_per_trk, recid)){
rcmd |= 0x8;
count = dasd_eckd_cdl_reclen(recid);
if (count < blksize &&
rq_data_dir(req) == READ)
memset(dst + count, 0xe5,
blksize - count);
}
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, LO_data++,
trkid, recoffs + 1,
1, rcmd, basedev, count);
}
/* Locate record for standard blocks ? */
if (private->uses_cdl && recid == 2*blk_per_trk) {
ccw[-1].flags |= CCW_FLAG_CC;
locate_record(ccw++, LO_data++,
trkid, recoffs + 1,
last_rec - recid + 1,
cmd, basedev, count);
}
/* Read/write ccw. */
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = rcmd;
ccw->count = count;
if (idal_is_needed(dst, blksize)) {
ccw->cda = (__u32)(addr_t) idaws;
ccw->flags = CCW_FLAG_IDA;
idaws = idal_create_words(idaws, dst, blksize);
} else {
ccw->cda = (__u32)(addr_t) dst;
ccw->flags = 0;
}
ccw++;
dst += blksize;
recid++;
}
}
if (blk_noretry_request(req) ||
block->base->features & DASD_FEATURE_FAILFAST)
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->startdev = startdev;
cqr->memdev = startdev;
cqr->block = block;
cqr->expires = startdev->default_expires * HZ; /* default 5 minutes */
cqr->lpm = dasd_path_get_ppm(startdev);
cqr->retries = startdev->default_retries;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
return cqr;
}
static struct dasd_ccw_req *dasd_eckd_build_cp_cmd_track(
struct dasd_device *startdev,
struct dasd_block *block,
struct request *req,
sector_t first_rec,
sector_t last_rec,
sector_t first_trk,
sector_t last_trk,
unsigned int first_offs,
unsigned int last_offs,
unsigned int blk_per_trk,
unsigned int blksize)
{
unsigned long *idaws;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
struct req_iterator iter;
struct bio_vec bv;
char *dst, *idaw_dst;
unsigned int cidaw, cplength, datasize;
unsigned int tlf;
sector_t recid;
unsigned char cmd;
struct dasd_device *basedev;
unsigned int trkcount, count, count_to_trk_end;
unsigned int idaw_len, seg_len, part_len, len_to_track_end;
unsigned char new_track, end_idaw;
sector_t trkid;
unsigned int recoffs;
basedev = block->base;
if (rq_data_dir(req) == READ)
cmd = DASD_ECKD_CCW_READ_TRACK_DATA;
else if (rq_data_dir(req) == WRITE)
cmd = DASD_ECKD_CCW_WRITE_TRACK_DATA;
else
return ERR_PTR(-EINVAL);
/* Track based I/O needs IDAWs for each page, and not just for
* 64 bit addresses. We need additional idals for pages
* that get filled from two tracks, so we use the number
* of records as upper limit.
*/
cidaw = last_rec - first_rec + 1;
trkcount = last_trk - first_trk + 1;
/* 1x prefix + one read/write ccw per track */
cplength = 1 + trkcount;
datasize = sizeof(struct PFX_eckd_data) + cidaw * sizeof(unsigned long);
/* Allocate the ccw request. */
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength, datasize,
startdev, blk_mq_rq_to_pdu(req));
if (IS_ERR(cqr))
return cqr;
ccw = cqr->cpaddr;
/* transfer length factor: how many bytes to read from the last track */
if (first_trk == last_trk)
tlf = last_offs - first_offs + 1;
else
tlf = last_offs + 1;
tlf *= blksize;
if (prefix_LRE(ccw++, cqr->data, first_trk,
last_trk, cmd, basedev, startdev,
1 /* format */, first_offs + 1,
trkcount, blksize,
tlf) == -EAGAIN) {
/* Clock not in sync and XRC is enabled.
* Try again later.
*/
dasd_sfree_request(cqr, startdev);
return ERR_PTR(-EAGAIN);
}
/*
* The translation of request into ccw programs must meet the
* following conditions:
* - all idaws but the first and the last must address full pages
* (or 2K blocks on 31-bit)
* - the scope of a ccw and it's idal ends with the track boundaries
*/
idaws = (unsigned long *) (cqr->data + sizeof(struct PFX_eckd_data));
recid = first_rec;
new_track = 1;
end_idaw = 0;
len_to_track_end = 0;
idaw_dst = NULL;
idaw_len = 0;
rq_for_each_segment(bv, req, iter) {
dst = page_address(bv.bv_page) + bv.bv_offset;
seg_len = bv.bv_len;
while (seg_len) {
if (new_track) {
trkid = recid;
recoffs = sector_div(trkid, blk_per_trk);
count_to_trk_end = blk_per_trk - recoffs;
count = min((last_rec - recid + 1),
(sector_t)count_to_trk_end);
len_to_track_end = count * blksize;
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = cmd;
ccw->count = len_to_track_end;
ccw->cda = (__u32)(addr_t)idaws;
ccw->flags = CCW_FLAG_IDA;
ccw++;
recid += count;
new_track = 0;
/* first idaw for a ccw may start anywhere */
if (!idaw_dst)
idaw_dst = dst;
}
/* If we start a new idaw, we must make sure that it
* starts on an IDA_BLOCK_SIZE boundary.
* If we continue an idaw, we must make sure that the
* current segment begins where the so far accumulated
* idaw ends
*/
if (!idaw_dst) {
if (__pa(dst) & (IDA_BLOCK_SIZE-1)) {
dasd_sfree_request(cqr, startdev);
return ERR_PTR(-ERANGE);
} else
idaw_dst = dst;
}
if ((idaw_dst + idaw_len) != dst) {
dasd_sfree_request(cqr, startdev);
return ERR_PTR(-ERANGE);
}
part_len = min(seg_len, len_to_track_end);
seg_len -= part_len;
dst += part_len;
idaw_len += part_len;
len_to_track_end -= part_len;
/* collected memory area ends on an IDA_BLOCK border,
* -> create an idaw
* idal_create_words will handle cases where idaw_len
* is larger then IDA_BLOCK_SIZE
*/
if (!(__pa(idaw_dst + idaw_len) & (IDA_BLOCK_SIZE-1)))
end_idaw = 1;
/* We also need to end the idaw at track end */
if (!len_to_track_end) {
new_track = 1;
end_idaw = 1;
}
if (end_idaw) {
idaws = idal_create_words(idaws, idaw_dst,
idaw_len);
idaw_dst = NULL;
idaw_len = 0;
end_idaw = 0;
}
}
}
if (blk_noretry_request(req) ||
block->base->features & DASD_FEATURE_FAILFAST)
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->startdev = startdev;
cqr->memdev = startdev;
cqr->block = block;
cqr->expires = startdev->default_expires * HZ; /* default 5 minutes */
cqr->lpm = dasd_path_get_ppm(startdev);
cqr->retries = startdev->default_retries;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
return cqr;
}
static int prepare_itcw(struct itcw *itcw,
unsigned int trk, unsigned int totrk, int cmd,
struct dasd_device *basedev,
struct dasd_device *startdev,
unsigned int rec_on_trk, int count,
unsigned int blksize,
unsigned int total_data_size,
unsigned int tlf,
unsigned int blk_per_trk)
{
struct PFX_eckd_data pfxdata;
struct dasd_eckd_private *basepriv, *startpriv;
struct DE_eckd_data *dedata;
struct LRE_eckd_data *lredata;
struct dcw *dcw;
u32 begcyl, endcyl;
u16 heads, beghead, endhead;
u8 pfx_cmd;
int rc = 0;
int sector = 0;
int dn, d;
/* setup prefix data */
basepriv = basedev->private;
startpriv = startdev->private;
dedata = &pfxdata.define_extent;
lredata = &pfxdata.locate_record;
memset(&pfxdata, 0, sizeof(pfxdata));
pfxdata.format = 1; /* PFX with LRE */
pfxdata.base_address = basepriv->ned->unit_addr;
pfxdata.base_lss = basepriv->ned->ID;
pfxdata.validity.define_extent = 1;
/* private uid is kept up to date, conf_data may be outdated */
if (startpriv->uid.type == UA_BASE_PAV_ALIAS)
pfxdata.validity.verify_base = 1;
if (startpriv->uid.type == UA_HYPER_PAV_ALIAS) {
pfxdata.validity.verify_base = 1;
pfxdata.validity.hyper_pav = 1;
}
switch (cmd) {
case DASD_ECKD_CCW_READ_TRACK_DATA:
dedata->mask.perm = 0x1;
dedata->attributes.operation = basepriv->attrib.operation;
dedata->blk_size = blksize;
dedata->ga_extended |= 0x42;
lredata->operation.orientation = 0x0;
lredata->operation.operation = 0x0C;
lredata->auxiliary.check_bytes = 0x01;
pfx_cmd = DASD_ECKD_CCW_PFX_READ;
break;
case DASD_ECKD_CCW_WRITE_TRACK_DATA:
dedata->mask.perm = 0x02;
dedata->attributes.operation = basepriv->attrib.operation;
dedata->blk_size = blksize;
rc = set_timestamp(NULL, dedata, basedev);
dedata->ga_extended |= 0x42;
lredata->operation.orientation = 0x0;
lredata->operation.operation = 0x3F;
lredata->extended_operation = 0x23;
lredata->auxiliary.check_bytes = 0x2;
/*
* If XRC is supported the System Time Stamp is set. The
* validity of the time stamp must be reflected in the prefix
* data as well.
*/
if (dedata->ga_extended & 0x08 && dedata->ga_extended & 0x02)
pfxdata.validity.time_stamp = 1; /* 'Time Stamp Valid' */
pfx_cmd = DASD_ECKD_CCW_PFX;
break;
case DASD_ECKD_CCW_READ_COUNT_MT:
dedata->mask.perm = 0x1;
dedata->attributes.operation = DASD_BYPASS_CACHE;
dedata->ga_extended |= 0x42;
dedata->blk_size = blksize;
lredata->operation.orientation = 0x2;
lredata->operation.operation = 0x16;
lredata->auxiliary.check_bytes = 0x01;
pfx_cmd = DASD_ECKD_CCW_PFX_READ;
break;
default:
DBF_DEV_EVENT(DBF_ERR, basedev,
"prepare itcw, unknown opcode 0x%x", cmd);
BUG();
break;
}
if (rc)
return rc;
dedata->attributes.mode = 0x3; /* ECKD */
heads = basepriv->rdc_data.trk_per_cyl;
begcyl = trk / heads;
beghead = trk % heads;
endcyl = totrk / heads;
endhead = totrk % heads;
/* check for sequential prestage - enhance cylinder range */
if (dedata->attributes.operation == DASD_SEQ_PRESTAGE ||
dedata->attributes.operation == DASD_SEQ_ACCESS) {
if (endcyl + basepriv->attrib.nr_cyl < basepriv->real_cyl)
endcyl += basepriv->attrib.nr_cyl;
else
endcyl = (basepriv->real_cyl - 1);
}
set_ch_t(&dedata->beg_ext, begcyl, beghead);
set_ch_t(&dedata->end_ext, endcyl, endhead);
dedata->ep_format = 0x20; /* records per track is valid */
dedata->ep_rec_per_track = blk_per_trk;
if (rec_on_trk) {
switch (basepriv->rdc_data.dev_type) {
case 0x3390:
dn = ceil_quot(blksize + 6, 232);
d = 9 + ceil_quot(blksize + 6 * (dn + 1), 34);
sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
break;
case 0x3380:
d = 7 + ceil_quot(blksize + 12, 32);
sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
break;
}
}
if (cmd == DASD_ECKD_CCW_READ_COUNT_MT) {
lredata->auxiliary.length_valid = 0;
lredata->auxiliary.length_scope = 0;
lredata->sector = 0xff;
} else {
lredata->auxiliary.length_valid = 1;
lredata->auxiliary.length_scope = 1;
lredata->sector = sector;
}
lredata->auxiliary.imbedded_ccw_valid = 1;
lredata->length = tlf;
lredata->imbedded_ccw = cmd;
lredata->count = count;
set_ch_t(&lredata->seek_addr, begcyl, beghead);
lredata->search_arg.cyl = lredata->seek_addr.cyl;
lredata->search_arg.head = lredata->seek_addr.head;
lredata->search_arg.record = rec_on_trk;
dcw = itcw_add_dcw(itcw, pfx_cmd, 0,
&pfxdata, sizeof(pfxdata), total_data_size);
return PTR_ERR_OR_ZERO(dcw);
}
static struct dasd_ccw_req *dasd_eckd_build_cp_tpm_track(
struct dasd_device *startdev,
struct dasd_block *block,
struct request *req,
sector_t first_rec,
sector_t last_rec,
sector_t first_trk,
sector_t last_trk,
unsigned int first_offs,
unsigned int last_offs,
unsigned int blk_per_trk,
unsigned int blksize)
{
struct dasd_ccw_req *cqr;
struct req_iterator iter;
struct bio_vec bv;
char *dst;
unsigned int trkcount, ctidaw;
unsigned char cmd;
struct dasd_device *basedev;
unsigned int tlf;
struct itcw *itcw;
struct tidaw *last_tidaw = NULL;
int itcw_op;
size_t itcw_size;
u8 tidaw_flags;
unsigned int seg_len, part_len, len_to_track_end;
unsigned char new_track;
sector_t recid, trkid;
unsigned int offs;
unsigned int count, count_to_trk_end;
int ret;
basedev = block->base;
if (rq_data_dir(req) == READ) {
cmd = DASD_ECKD_CCW_READ_TRACK_DATA;
itcw_op = ITCW_OP_READ;
} else if (rq_data_dir(req) == WRITE) {
cmd = DASD_ECKD_CCW_WRITE_TRACK_DATA;
itcw_op = ITCW_OP_WRITE;
} else
return ERR_PTR(-EINVAL);
/* trackbased I/O needs address all memory via TIDAWs,
* not just for 64 bit addresses. This allows us to map
* each segment directly to one tidaw.
* In the case of write requests, additional tidaws may
* be needed when a segment crosses a track boundary.
*/
trkcount = last_trk - first_trk + 1;
ctidaw = 0;
rq_for_each_segment(bv, req, iter) {
++ctidaw;
}
if (rq_data_dir(req) == WRITE)
ctidaw += (last_trk - first_trk);
/* Allocate the ccw request. */
itcw_size = itcw_calc_size(0, ctidaw, 0);
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 0, itcw_size, startdev,
blk_mq_rq_to_pdu(req));
if (IS_ERR(cqr))
return cqr;
/* transfer length factor: how many bytes to read from the last track */
if (first_trk == last_trk)
tlf = last_offs - first_offs + 1;
else
tlf = last_offs + 1;
tlf *= blksize;
itcw = itcw_init(cqr->data, itcw_size, itcw_op, 0, ctidaw, 0);
if (IS_ERR(itcw)) {
ret = -EINVAL;
goto out_error;
}
cqr->cpaddr = itcw_get_tcw(itcw);
if (prepare_itcw(itcw, first_trk, last_trk,
cmd, basedev, startdev,
first_offs + 1,
trkcount, blksize,
(last_rec - first_rec + 1) * blksize,
tlf, blk_per_trk) == -EAGAIN) {
/* Clock not in sync and XRC is enabled.
* Try again later.
*/
ret = -EAGAIN;
goto out_error;
}
len_to_track_end = 0;
/*
* A tidaw can address 4k of memory, but must not cross page boundaries
* We can let the block layer handle this by setting
* blk_queue_segment_boundary to page boundaries and
* blk_max_segment_size to page size when setting up the request queue.
* For write requests, a TIDAW must not cross track boundaries, because
* we have to set the CBC flag on the last tidaw for each track.
*/
if (rq_data_dir(req) == WRITE) {
new_track = 1;
recid = first_rec;
rq_for_each_segment(bv, req, iter) {
dst = page_address(bv.bv_page) + bv.bv_offset;
seg_len = bv.bv_len;
while (seg_len) {
if (new_track) {
trkid = recid;
offs = sector_div(trkid, blk_per_trk);
count_to_trk_end = blk_per_trk - offs;
count = min((last_rec - recid + 1),
(sector_t)count_to_trk_end);
len_to_track_end = count * blksize;
recid += count;
new_track = 0;
}
part_len = min(seg_len, len_to_track_end);
seg_len -= part_len;
len_to_track_end -= part_len;
/* We need to end the tidaw at track end */
if (!len_to_track_end) {
new_track = 1;
tidaw_flags = TIDAW_FLAGS_INSERT_CBC;
} else
tidaw_flags = 0;
last_tidaw = itcw_add_tidaw(itcw, tidaw_flags,
dst, part_len);
if (IS_ERR(last_tidaw)) {
ret = -EINVAL;
goto out_error;
}
dst += part_len;
}
}
} else {
rq_for_each_segment(bv, req, iter) {
dst = page_address(bv.bv_page) + bv.bv_offset;
last_tidaw = itcw_add_tidaw(itcw, 0x00,
dst, bv.bv_len);
if (IS_ERR(last_tidaw)) {
ret = -EINVAL;
goto out_error;
}
}
}
last_tidaw->flags |= TIDAW_FLAGS_LAST;
last_tidaw->flags &= ~TIDAW_FLAGS_INSERT_CBC;
itcw_finalize(itcw);
if (blk_noretry_request(req) ||
block->base->features & DASD_FEATURE_FAILFAST)
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->cpmode = 1;
cqr->startdev = startdev;
cqr->memdev = startdev;
cqr->block = block;
cqr->expires = startdev->default_expires * HZ; /* default 5 minutes */
cqr->lpm = dasd_path_get_ppm(startdev);
cqr->retries = startdev->default_retries;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
return cqr;
out_error:
dasd_sfree_request(cqr, startdev);
return ERR_PTR(ret);
}
static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
struct dasd_block *block,
struct request *req)
{
int cmdrtd, cmdwtd;
int use_prefix;
int fcx_multitrack;
struct dasd_eckd_private *private;
struct dasd_device *basedev;
sector_t first_rec, last_rec;
sector_t first_trk, last_trk;
unsigned int first_offs, last_offs;
unsigned int blk_per_trk, blksize;
int cdlspecial;
unsigned int data_size;
struct dasd_ccw_req *cqr;
basedev = block->base;
private = basedev->private;
/* Calculate number of blocks/records per track. */
blksize = block->bp_block;
blk_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
if (blk_per_trk == 0)
return ERR_PTR(-EINVAL);
/* Calculate record id of first and last block. */
first_rec = first_trk = blk_rq_pos(req) >> block->s2b_shift;
first_offs = sector_div(first_trk, blk_per_trk);
last_rec = last_trk =
(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
last_offs = sector_div(last_trk, blk_per_trk);
cdlspecial = (private->uses_cdl && first_rec < 2*blk_per_trk);
fcx_multitrack = private->features.feature[40] & 0x20;
data_size = blk_rq_bytes(req);
if (data_size % blksize)
return ERR_PTR(-EINVAL);
/* tpm write request add CBC data on each track boundary */
if (rq_data_dir(req) == WRITE)
data_size += (last_trk - first_trk) * 4;
/* is read track data and write track data in command mode supported? */
cmdrtd = private->features.feature[9] & 0x20;
cmdwtd = private->features.feature[12] & 0x40;
use_prefix = private->features.feature[8] & 0x01;
cqr = NULL;
if (cdlspecial || dasd_page_cache) {
/* do nothing, just fall through to the cmd mode single case */
} else if ((data_size <= private->fcx_max_data)
&& (fcx_multitrack || (first_trk == last_trk))) {
cqr = dasd_eckd_build_cp_tpm_track(startdev, block, req,
first_rec, last_rec,
first_trk, last_trk,
first_offs, last_offs,
blk_per_trk, blksize);
if (IS_ERR(cqr) && (PTR_ERR(cqr) != -EAGAIN) &&
(PTR_ERR(cqr) != -ENOMEM))
cqr = NULL;
} else if (use_prefix &&
(((rq_data_dir(req) == READ) && cmdrtd) ||
((rq_data_dir(req) == WRITE) && cmdwtd))) {
cqr = dasd_eckd_build_cp_cmd_track(startdev, block, req,
first_rec, last_rec,
first_trk, last_trk,
first_offs, last_offs,
blk_per_trk, blksize);
if (IS_ERR(cqr) && (PTR_ERR(cqr) != -EAGAIN) &&
(PTR_ERR(cqr) != -ENOMEM))
cqr = NULL;
}
if (!cqr)
cqr = dasd_eckd_build_cp_cmd_single(startdev, block, req,
first_rec, last_rec,
first_trk, last_trk,
first_offs, last_offs,
blk_per_trk, blksize);
return cqr;
}
static struct dasd_ccw_req *dasd_eckd_build_cp_raw(struct dasd_device *startdev,
struct dasd_block *block,
struct request *req)
{
sector_t start_padding_sectors, end_sector_offset, end_padding_sectors;
unsigned int seg_len, len_to_track_end;
unsigned int cidaw, cplength, datasize;
sector_t first_trk, last_trk, sectors;
struct dasd_eckd_private *base_priv;
struct dasd_device *basedev;
struct req_iterator iter;
struct dasd_ccw_req *cqr;
unsigned int first_offs;
unsigned int trkcount;
unsigned long *idaws;
unsigned int size;
unsigned char cmd;
struct bio_vec bv;
struct ccw1 *ccw;
int use_prefix;
void *data;
char *dst;
/*
* raw track access needs to be mutiple of 64k and on 64k boundary
* For read requests we can fix an incorrect alignment by padding
* the request with dummy pages.
*/
start_padding_sectors = blk_rq_pos(req) % DASD_RAW_SECTORS_PER_TRACK;
end_sector_offset = (blk_rq_pos(req) + blk_rq_sectors(req)) %
DASD_RAW_SECTORS_PER_TRACK;
end_padding_sectors = (DASD_RAW_SECTORS_PER_TRACK - end_sector_offset) %
DASD_RAW_SECTORS_PER_TRACK;
basedev = block->base;
if ((start_padding_sectors || end_padding_sectors) &&
(rq_data_dir(req) == WRITE)) {
DBF_DEV_EVENT(DBF_ERR, basedev,
"raw write not track aligned (%lu,%lu) req %p",
start_padding_sectors, end_padding_sectors, req);
return ERR_PTR(-EINVAL);
}
first_trk = blk_rq_pos(req) / DASD_RAW_SECTORS_PER_TRACK;
last_trk = (blk_rq_pos(req) + blk_rq_sectors(req) - 1) /
DASD_RAW_SECTORS_PER_TRACK;
trkcount = last_trk - first_trk + 1;
first_offs = 0;
if (rq_data_dir(req) == READ)
cmd = DASD_ECKD_CCW_READ_TRACK;
else if (rq_data_dir(req) == WRITE)
cmd = DASD_ECKD_CCW_WRITE_FULL_TRACK;
else
return ERR_PTR(-EINVAL);
/*
* Raw track based I/O needs IDAWs for each page,
* and not just for 64 bit addresses.
*/
cidaw = trkcount * DASD_RAW_BLOCK_PER_TRACK;
/*
* struct PFX_eckd_data and struct LRE_eckd_data can have up to 2 bytes
* of extended parameter. This is needed for write full track.
*/
base_priv = basedev->private;
use_prefix = base_priv->features.feature[8] & 0x01;
if (use_prefix) {
cplength = 1 + trkcount;
size = sizeof(struct PFX_eckd_data) + 2;
} else {
cplength = 2 + trkcount;
size = sizeof(struct DE_eckd_data) +
sizeof(struct LRE_eckd_data) + 2;
}
size = ALIGN(size, 8);
datasize = size + cidaw * sizeof(unsigned long);
/* Allocate the ccw request. */
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, cplength,
datasize, startdev, blk_mq_rq_to_pdu(req));
if (IS_ERR(cqr))
return cqr;
ccw = cqr->cpaddr;
data = cqr->data;
if (use_prefix) {
prefix_LRE(ccw++, data, first_trk, last_trk, cmd, basedev,
startdev, 1, first_offs + 1, trkcount, 0, 0);
} else {
define_extent(ccw++, data, first_trk, last_trk, cmd, basedev, 0);
ccw[-1].flags |= CCW_FLAG_CC;
data += sizeof(struct DE_eckd_data);
locate_record_ext(ccw++, data, first_trk, first_offs + 1,
trkcount, cmd, basedev, 0, 0);
}
idaws = (unsigned long *)(cqr->data + size);
len_to_track_end = 0;
if (start_padding_sectors) {
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = cmd;
/* maximum 3390 track size */
ccw->count = 57326;
/* 64k map to one track */
len_to_track_end = 65536 - start_padding_sectors * 512;
ccw->cda = (__u32)(addr_t)idaws;
ccw->flags |= CCW_FLAG_IDA;
ccw->flags |= CCW_FLAG_SLI;
ccw++;
for (sectors = 0; sectors < start_padding_sectors; sectors += 8)
idaws = idal_create_words(idaws, rawpadpage, PAGE_SIZE);
}
rq_for_each_segment(bv, req, iter) {
dst = page_address(bv.bv_page) + bv.bv_offset;
seg_len = bv.bv_len;
if (cmd == DASD_ECKD_CCW_READ_TRACK)
memset(dst, 0, seg_len);
if (!len_to_track_end) {
ccw[-1].flags |= CCW_FLAG_CC;
ccw->cmd_code = cmd;
/* maximum 3390 track size */
ccw->count = 57326;
/* 64k map to one track */
len_to_track_end = 65536;
ccw->cda = (__u32)(addr_t)idaws;
ccw->flags |= CCW_FLAG_IDA;
ccw->flags |= CCW_FLAG_SLI;
ccw++;
}
len_to_track_end -= seg_len;
idaws = idal_create_words(idaws, dst, seg_len);
}
for (sectors = 0; sectors < end_padding_sectors; sectors += 8)
idaws = idal_create_words(idaws, rawpadpage, PAGE_SIZE);
if (blk_noretry_request(req) ||
block->base->features & DASD_FEATURE_FAILFAST)
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->startdev = startdev;
cqr->memdev = startdev;
cqr->block = block;
cqr->expires = startdev->default_expires * HZ;
cqr->lpm = dasd_path_get_ppm(startdev);
cqr->retries = startdev->default_retries;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
return cqr;
}
static int
dasd_eckd_free_cp(struct dasd_ccw_req *cqr, struct request *req)
{
struct dasd_eckd_private *private;
struct ccw1 *ccw;
struct req_iterator iter;
struct bio_vec bv;
char *dst, *cda;
unsigned int blksize, blk_per_trk, off;
sector_t recid;
int status;
if (!dasd_page_cache)
goto out;
private = cqr->block->base->private;
blksize = cqr->block->bp_block;
blk_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
recid = blk_rq_pos(req) >> cqr->block->s2b_shift;
ccw = cqr->cpaddr;
/* Skip over define extent & locate record. */
ccw++;
if (private->uses_cdl == 0 || recid > 2*blk_per_trk)
ccw++;
rq_for_each_segment(bv, req, iter) {
dst = page_address(bv.bv_page) + bv.bv_offset;
for (off = 0; off < bv.bv_len; off += blksize) {
/* Skip locate record. */
if (private->uses_cdl && recid <= 2*blk_per_trk)
ccw++;
if (dst) {
if (ccw->flags & CCW_FLAG_IDA)
cda = *((char **)((addr_t) ccw->cda));
else
cda = (char *)((addr_t) ccw->cda);
if (dst != cda) {
if (rq_data_dir(req) == READ)
memcpy(dst, cda, bv.bv_len);
kmem_cache_free(dasd_page_cache,
(void *)((addr_t)cda & PAGE_MASK));
}
dst = NULL;
}
ccw++;
recid++;
}
}
out:
status = cqr->status == DASD_CQR_DONE;
dasd_sfree_request(cqr, cqr->memdev);
return status;
}
/*
* Modify ccw/tcw in cqr so it can be started on a base device.
*
* Note that this is not enough to restart the cqr!
* Either reset cqr->startdev as well (summary unit check handling)
* or restart via separate cqr (as in ERP handling).
*/
void dasd_eckd_reset_ccw_to_base_io(struct dasd_ccw_req *cqr)
{
struct ccw1 *ccw;
struct PFX_eckd_data *pfxdata;
struct tcw *tcw;
struct tccb *tccb;
struct dcw *dcw;
if (cqr->cpmode == 1) {
tcw = cqr->cpaddr;
tccb = tcw_get_tccb(tcw);
dcw = (struct dcw *)&tccb->tca[0];
pfxdata = (struct PFX_eckd_data *)&dcw->cd[0];
pfxdata->validity.verify_base = 0;
pfxdata->validity.hyper_pav = 0;
} else {
ccw = cqr->cpaddr;
pfxdata = cqr->data;
if (ccw->cmd_code == DASD_ECKD_CCW_PFX) {
pfxdata->validity.verify_base = 0;
pfxdata->validity.hyper_pav = 0;
}
}
}
#define DASD_ECKD_CHANQ_MAX_SIZE 4
static struct dasd_ccw_req *dasd_eckd_build_alias_cp(struct dasd_device *base,
struct dasd_block *block,
struct request *req)
{
struct dasd_eckd_private *private;
struct dasd_device *startdev;
unsigned long flags;
struct dasd_ccw_req *cqr;
startdev = dasd_alias_get_start_dev(base);
if (!startdev)
startdev = base;
private = startdev->private;
if (private->count >= DASD_ECKD_CHANQ_MAX_SIZE)
return ERR_PTR(-EBUSY);
spin_lock_irqsave(get_ccwdev_lock(startdev->cdev), flags);
private->count++;
if ((base->features & DASD_FEATURE_USERAW))
cqr = dasd_eckd_build_cp_raw(startdev, block, req);
else
cqr = dasd_eckd_build_cp(startdev, block, req);
if (IS_ERR(cqr))
private->count--;
spin_unlock_irqrestore(get_ccwdev_lock(startdev->cdev), flags);
return cqr;
}
static int dasd_eckd_free_alias_cp(struct dasd_ccw_req *cqr,
struct request *req)
{
struct dasd_eckd_private *private;
unsigned long flags;
spin_lock_irqsave(get_ccwdev_lock(cqr->memdev->cdev), flags);
private = cqr->memdev->private;
private->count--;
spin_unlock_irqrestore(get_ccwdev_lock(cqr->memdev->cdev), flags);
return dasd_eckd_free_cp(cqr, req);
}
static int
dasd_eckd_fill_info(struct dasd_device * device,
struct dasd_information2_t * info)
{
struct dasd_eckd_private *private = device->private;
info->label_block = 2;
info->FBA_layout = private->uses_cdl ? 0 : 1;
info->format = private->uses_cdl ? DASD_FORMAT_CDL : DASD_FORMAT_LDL;
info->characteristics_size = sizeof(private->rdc_data);
memcpy(info->characteristics, &private->rdc_data,
sizeof(private->rdc_data));
info->confdata_size = min((unsigned long)private->conf_len,
sizeof(info->configuration_data));
memcpy(info->configuration_data, private->conf_data,
info->confdata_size);
return 0;
}
/*
* SECTION: ioctl functions for eckd devices.
*/
/*
* Release device ioctl.
* Buils a channel programm to releases a prior reserved
* (see dasd_eckd_reserve) device.
*/
static int
dasd_eckd_release(struct dasd_device *device)
{
struct dasd_ccw_req *cqr;
int rc;
struct ccw1 *ccw;
int useglobal;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
useglobal = 0;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 32, device, NULL);
if (IS_ERR(cqr)) {
mutex_lock(&dasd_reserve_mutex);
useglobal = 1;
cqr = &dasd_reserve_req->cqr;
memset(cqr, 0, sizeof(*cqr));
memset(&dasd_reserve_req->ccw, 0,
sizeof(dasd_reserve_req->ccw));
cqr->cpaddr = &dasd_reserve_req->ccw;
cqr->data = &dasd_reserve_req->data;
cqr->magic = DASD_ECKD_MAGIC;
}
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_RELEASE;
ccw->flags |= CCW_FLAG_SLI;
ccw->count = 32;
ccw->cda = (__u32)(addr_t) cqr->data;
cqr->startdev = device;
cqr->memdev = device;
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->retries = 2; /* set retry counter to enable basic ERP */
cqr->expires = 2 * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
rc = dasd_sleep_on_immediatly(cqr);
if (!rc)
clear_bit(DASD_FLAG_IS_RESERVED, &device->flags);
if (useglobal)
mutex_unlock(&dasd_reserve_mutex);
else
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* Reserve device ioctl.
* Options are set to 'synchronous wait for interrupt' and
* 'timeout the request'. This leads to a terminate IO if
* the interrupt is outstanding for a certain time.
*/
static int
dasd_eckd_reserve(struct dasd_device *device)
{
struct dasd_ccw_req *cqr;
int rc;
struct ccw1 *ccw;
int useglobal;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
useglobal = 0;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 32, device, NULL);
if (IS_ERR(cqr)) {
mutex_lock(&dasd_reserve_mutex);
useglobal = 1;
cqr = &dasd_reserve_req->cqr;
memset(cqr, 0, sizeof(*cqr));
memset(&dasd_reserve_req->ccw, 0,
sizeof(dasd_reserve_req->ccw));
cqr->cpaddr = &dasd_reserve_req->ccw;
cqr->data = &dasd_reserve_req->data;
cqr->magic = DASD_ECKD_MAGIC;
}
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_RESERVE;
ccw->flags |= CCW_FLAG_SLI;
ccw->count = 32;
ccw->cda = (__u32)(addr_t) cqr->data;
cqr->startdev = device;
cqr->memdev = device;
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->retries = 2; /* set retry counter to enable basic ERP */
cqr->expires = 2 * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
rc = dasd_sleep_on_immediatly(cqr);
if (!rc)
set_bit(DASD_FLAG_IS_RESERVED, &device->flags);
if (useglobal)
mutex_unlock(&dasd_reserve_mutex);
else
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* Steal lock ioctl - unconditional reserve device.
* Buils a channel programm to break a device's reservation.
* (unconditional reserve)
*/
static int
dasd_eckd_steal_lock(struct dasd_device *device)
{
struct dasd_ccw_req *cqr;
int rc;
struct ccw1 *ccw;
int useglobal;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
useglobal = 0;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1, 32, device, NULL);
if (IS_ERR(cqr)) {
mutex_lock(&dasd_reserve_mutex);
useglobal = 1;
cqr = &dasd_reserve_req->cqr;
memset(cqr, 0, sizeof(*cqr));
memset(&dasd_reserve_req->ccw, 0,
sizeof(dasd_reserve_req->ccw));
cqr->cpaddr = &dasd_reserve_req->ccw;
cqr->data = &dasd_reserve_req->data;
cqr->magic = DASD_ECKD_MAGIC;
}
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_SLCK;
ccw->flags |= CCW_FLAG_SLI;
ccw->count = 32;
ccw->cda = (__u32)(addr_t) cqr->data;
cqr->startdev = device;
cqr->memdev = device;
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
cqr->retries = 2; /* set retry counter to enable basic ERP */
cqr->expires = 2 * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
rc = dasd_sleep_on_immediatly(cqr);
if (!rc)
set_bit(DASD_FLAG_IS_RESERVED, &device->flags);
if (useglobal)
mutex_unlock(&dasd_reserve_mutex);
else
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* SNID - Sense Path Group ID
* This ioctl may be used in situations where I/O is stalled due to
* a reserve, so if the normal dasd_smalloc_request fails, we use the
* preallocated dasd_reserve_req.
*/
static int dasd_eckd_snid(struct dasd_device *device,
void __user *argp)
{
struct dasd_ccw_req *cqr;
int rc;
struct ccw1 *ccw;
int useglobal;
struct dasd_snid_ioctl_data usrparm;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (copy_from_user(&usrparm, argp, sizeof(usrparm)))
return -EFAULT;
useglobal = 0;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1,
sizeof(struct dasd_snid_data), device,
NULL);
if (IS_ERR(cqr)) {
mutex_lock(&dasd_reserve_mutex);
useglobal = 1;
cqr = &dasd_reserve_req->cqr;
memset(cqr, 0, sizeof(*cqr));
memset(&dasd_reserve_req->ccw, 0,
sizeof(dasd_reserve_req->ccw));
cqr->cpaddr = &dasd_reserve_req->ccw;
cqr->data = &dasd_reserve_req->data;
cqr->magic = DASD_ECKD_MAGIC;
}
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_SNID;
ccw->flags |= CCW_FLAG_SLI;
ccw->count = 12;
ccw->cda = (__u32)(addr_t) cqr->data;
cqr->startdev = device;
cqr->memdev = device;
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
set_bit(DASD_CQR_ALLOW_SLOCK, &cqr->flags);
cqr->retries = 5;
cqr->expires = 10 * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
cqr->lpm = usrparm.path_mask;
rc = dasd_sleep_on_immediatly(cqr);
/* verify that I/O processing didn't modify the path mask */
if (!rc && usrparm.path_mask && (cqr->lpm != usrparm.path_mask))
rc = -EIO;
if (!rc) {
usrparm.data = *((struct dasd_snid_data *)cqr->data);
if (copy_to_user(argp, &usrparm, sizeof(usrparm)))
rc = -EFAULT;
}
if (useglobal)
mutex_unlock(&dasd_reserve_mutex);
else
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* Read performance statistics
*/
static int
dasd_eckd_performance(struct dasd_device *device, void __user *argp)
{
struct dasd_psf_prssd_data *prssdp;
struct dasd_rssd_perf_stats_t *stats;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
int rc;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ + 1 /* RSSD */,
(sizeof(struct dasd_psf_prssd_data) +
sizeof(struct dasd_rssd_perf_stats_t)),
device, NULL);
if (IS_ERR(cqr)) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Could not allocate initialization request");
return PTR_ERR(cqr);
}
cqr->startdev = device;
cqr->memdev = device;
cqr->retries = 0;
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
cqr->expires = 10 * HZ;
/* Prepare for Read Subsystem Data */
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
prssdp->order = PSF_ORDER_PRSSD;
prssdp->suborder = 0x01; /* Performance Statistics */
prssdp->varies[1] = 0x01; /* Perf Statistics for the Subsystem */
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->count = sizeof(struct dasd_psf_prssd_data);
ccw->flags |= CCW_FLAG_CC;
ccw->cda = (__u32)(addr_t) prssdp;
/* Read Subsystem Data - Performance Statistics */
stats = (struct dasd_rssd_perf_stats_t *) (prssdp + 1);
memset(stats, 0, sizeof(struct dasd_rssd_perf_stats_t));
ccw++;
ccw->cmd_code = DASD_ECKD_CCW_RSSD;
ccw->count = sizeof(struct dasd_rssd_perf_stats_t);
ccw->cda = (__u32)(addr_t) stats;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
rc = dasd_sleep_on(cqr);
if (rc == 0) {
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
stats = (struct dasd_rssd_perf_stats_t *) (prssdp + 1);
if (copy_to_user(argp, stats,
sizeof(struct dasd_rssd_perf_stats_t)))
rc = -EFAULT;
}
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* Get attributes (cache operations)
* Returnes the cache attributes used in Define Extend (DE).
*/
static int
dasd_eckd_get_attrib(struct dasd_device *device, void __user *argp)
{
struct dasd_eckd_private *private = device->private;
struct attrib_data_t attrib = private->attrib;
int rc;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (!argp)
return -EINVAL;
rc = 0;
if (copy_to_user(argp, (long *) &attrib,
sizeof(struct attrib_data_t)))
rc = -EFAULT;
return rc;
}
/*
* Set attributes (cache operations)
* Stores the attributes for cache operation to be used in Define Extend (DE).
*/
static int
dasd_eckd_set_attrib(struct dasd_device *device, void __user *argp)
{
struct dasd_eckd_private *private = device->private;
struct attrib_data_t attrib;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (!argp)
return -EINVAL;
if (copy_from_user(&attrib, argp, sizeof(struct attrib_data_t)))
return -EFAULT;
private->attrib = attrib;
dev_info(&device->cdev->dev,
"The DASD cache mode was set to %x (%i cylinder prestage)\n",
private->attrib.operation, private->attrib.nr_cyl);
return 0;
}
/*
* Issue syscall I/O to EMC Symmetrix array.
* CCWs are PSF and RSSD
*/
static int dasd_symm_io(struct dasd_device *device, void __user *argp)
{
struct dasd_symmio_parms usrparm;
char *psf_data, *rssd_result;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
char psf0, psf1;
int rc;
if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RAWIO))
return -EACCES;
psf0 = psf1 = 0;
/* Copy parms from caller */
rc = -EFAULT;
if (copy_from_user(&usrparm, argp, sizeof(usrparm)))
goto out;
if (is_compat_task()) {
/* Make sure pointers are sane even on 31 bit. */
rc = -EINVAL;
if ((usrparm.psf_data >> 32) != 0)
goto out;
if ((usrparm.rssd_result >> 32) != 0)
goto out;
usrparm.psf_data &= 0x7fffffffULL;
usrparm.rssd_result &= 0x7fffffffULL;
}
/* alloc I/O data area */
psf_data = kzalloc(usrparm.psf_data_len, GFP_KERNEL | GFP_DMA);
rssd_result = kzalloc(usrparm.rssd_result_len, GFP_KERNEL | GFP_DMA);
if (!psf_data || !rssd_result) {
rc = -ENOMEM;
goto out_free;
}
/* get syscall header from user space */
rc = -EFAULT;
if (copy_from_user(psf_data,
(void __user *)(unsigned long) usrparm.psf_data,
usrparm.psf_data_len))
goto out_free;
psf0 = psf_data[0];
psf1 = psf_data[1];
/* setup CCWs for PSF + RSSD */
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 2, 0, device, NULL);
if (IS_ERR(cqr)) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Could not allocate initialization request");
rc = PTR_ERR(cqr);
goto out_free;
}
cqr->startdev = device;
cqr->memdev = device;
cqr->retries = 3;
cqr->expires = 10 * HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
/* Build the ccws */
ccw = cqr->cpaddr;
/* PSF ccw */
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->count = usrparm.psf_data_len;
ccw->flags |= CCW_FLAG_CC;
ccw->cda = (__u32)(addr_t) psf_data;
ccw++;
/* RSSD ccw */
ccw->cmd_code = DASD_ECKD_CCW_RSSD;
ccw->count = usrparm.rssd_result_len;
ccw->flags = CCW_FLAG_SLI ;
ccw->cda = (__u32)(addr_t) rssd_result;
rc = dasd_sleep_on(cqr);
if (rc)
goto out_sfree;
rc = -EFAULT;
if (copy_to_user((void __user *)(unsigned long) usrparm.rssd_result,
rssd_result, usrparm.rssd_result_len))
goto out_sfree;
rc = 0;
out_sfree:
dasd_sfree_request(cqr, cqr->memdev);
out_free:
kfree(rssd_result);
kfree(psf_data);
out:
DBF_DEV_EVENT(DBF_WARNING, device,
"Symmetrix ioctl (0x%02x 0x%02x): rc=%d",
(int) psf0, (int) psf1, rc);
return rc;
}
static int
dasd_eckd_ioctl(struct dasd_block *block, unsigned int cmd, void __user *argp)
{
struct dasd_device *device = block->base;
switch (cmd) {
case BIODASDGATTR:
return dasd_eckd_get_attrib(device, argp);
case BIODASDSATTR:
return dasd_eckd_set_attrib(device, argp);
case BIODASDPSRD:
return dasd_eckd_performance(device, argp);
case BIODASDRLSE:
return dasd_eckd_release(device);
case BIODASDRSRV:
return dasd_eckd_reserve(device);
case BIODASDSLCK:
return dasd_eckd_steal_lock(device);
case BIODASDSNID:
return dasd_eckd_snid(device, argp);
case BIODASDSYMMIO:
return dasd_symm_io(device, argp);
default:
return -ENOTTY;
}
}
/*
* Dump the range of CCWs into 'page' buffer
* and return number of printed chars.
*/
static int
dasd_eckd_dump_ccw_range(struct ccw1 *from, struct ccw1 *to, char *page)
{
int len, count;
char *datap;
len = 0;
while (from <= to) {
len += sprintf(page + len, PRINTK_HEADER
" CCW %p: %08X %08X DAT:",
from, ((int *) from)[0], ((int *) from)[1]);
/* get pointer to data (consider IDALs) */
if (from->flags & CCW_FLAG_IDA)
datap = (char *) *((addr_t *) (addr_t) from->cda);
else
datap = (char *) ((addr_t) from->cda);
/* dump data (max 32 bytes) */
for (count = 0; count < from->count && count < 32; count++) {
if (count % 8 == 0) len += sprintf(page + len, " ");
if (count % 4 == 0) len += sprintf(page + len, " ");
len += sprintf(page + len, "%02x", datap[count]);
}
len += sprintf(page + len, "\n");
from++;
}
return len;
}
static void
dasd_eckd_dump_sense_dbf(struct dasd_device *device, struct irb *irb,
char *reason)
{
u64 *sense;
u64 *stat;
sense = (u64 *) dasd_get_sense(irb);
stat = (u64 *) &irb->scsw;
if (sense) {
DBF_DEV_EVENT(DBF_EMERG, device, "%s: %016llx %08x : "
"%016llx %016llx %016llx %016llx",
reason, *stat, *((u32 *) (stat + 1)),
sense[0], sense[1], sense[2], sense[3]);
} else {
DBF_DEV_EVENT(DBF_EMERG, device, "%s: %016llx %08x : %s",
reason, *stat, *((u32 *) (stat + 1)),
"NO VALID SENSE");
}
}
/*
* Print sense data and related channel program.
* Parts are printed because printk buffer is only 1024 bytes.
*/
static void dasd_eckd_dump_sense_ccw(struct dasd_device *device,
struct dasd_ccw_req *req, struct irb *irb)
{
char *page;
struct ccw1 *first, *last, *fail, *from, *to;
int len, sl, sct;
page = (char *) get_zeroed_page(GFP_ATOMIC);
if (page == NULL) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"No memory to dump sense data\n");
return;
}
/* dump the sense data */
len = sprintf(page, PRINTK_HEADER
" I/O status report for device %s:\n",
dev_name(&device->cdev->dev));
len += sprintf(page + len, PRINTK_HEADER
" in req: %p CC:%02X FC:%02X AC:%02X SC:%02X DS:%02X "
"CS:%02X RC:%d\n",
req, scsw_cc(&irb->scsw), scsw_fctl(&irb->scsw),
scsw_actl(&irb->scsw), scsw_stctl(&irb->scsw),
scsw_dstat(&irb->scsw), scsw_cstat(&irb->scsw),
req ? req->intrc : 0);
len += sprintf(page + len, PRINTK_HEADER
" device %s: Failing CCW: %p\n",
dev_name(&device->cdev->dev),
(void *) (addr_t) irb->scsw.cmd.cpa);
if (irb->esw.esw0.erw.cons) {
for (sl = 0; sl < 4; sl++) {
len += sprintf(page + len, PRINTK_HEADER
" Sense(hex) %2d-%2d:",
(8 * sl), ((8 * sl) + 7));
for (sct = 0; sct < 8; sct++) {
len += sprintf(page + len, " %02x",
irb->ecw[8 * sl + sct]);
}
len += sprintf(page + len, "\n");
}
if (irb->ecw[27] & DASD_SENSE_BIT_0) {
/* 24 Byte Sense Data */
sprintf(page + len, PRINTK_HEADER
" 24 Byte: %x MSG %x, "
"%s MSGb to SYSOP\n",
irb->ecw[7] >> 4, irb->ecw[7] & 0x0f,
irb->ecw[1] & 0x10 ? "" : "no");
} else {
/* 32 Byte Sense Data */
sprintf(page + len, PRINTK_HEADER
" 32 Byte: Format: %x "
"Exception class %x\n",
irb->ecw[6] & 0x0f, irb->ecw[22] >> 4);
}
} else {
sprintf(page + len, PRINTK_HEADER
" SORRY - NO VALID SENSE AVAILABLE\n");
}
printk(KERN_ERR "%s", page);
if (req) {
/* req == NULL for unsolicited interrupts */
/* dump the Channel Program (max 140 Bytes per line) */
/* Count CCW and print first CCWs (maximum 1024 % 140 = 7) */
first = req->cpaddr;
for (last = first; last->flags & (CCW_FLAG_CC | CCW_FLAG_DC); last++);
to = min(first + 6, last);
len = sprintf(page, PRINTK_HEADER
" Related CP in req: %p\n", req);
dasd_eckd_dump_ccw_range(first, to, page + len);
printk(KERN_ERR "%s", page);
/* print failing CCW area (maximum 4) */
/* scsw->cda is either valid or zero */
len = 0;
from = ++to;
fail = (struct ccw1 *)(addr_t)
irb->scsw.cmd.cpa; /* failing CCW */
if (from < fail - 2) {
from = fail - 2; /* there is a gap - print header */
len += sprintf(page, PRINTK_HEADER "......\n");
}
to = min(fail + 1, last);
len += dasd_eckd_dump_ccw_range(from, to, page + len);
/* print last CCWs (maximum 2) */
from = max(from, ++to);
if (from < last - 1) {
from = last - 1; /* there is a gap - print header */
len += sprintf(page + len, PRINTK_HEADER "......\n");
}
len += dasd_eckd_dump_ccw_range(from, last, page + len);
if (len > 0)
printk(KERN_ERR "%s", page);
}
free_page((unsigned long) page);
}
/*
* Print sense data from a tcw.
*/
static void dasd_eckd_dump_sense_tcw(struct dasd_device *device,
struct dasd_ccw_req *req, struct irb *irb)
{
char *page;
int len, sl, sct, residual;
struct tsb *tsb;
u8 *sense, *rcq;
page = (char *) get_zeroed_page(GFP_ATOMIC);
if (page == NULL) {
DBF_DEV_EVENT(DBF_WARNING, device, " %s",
"No memory to dump sense data");
return;
}
/* dump the sense data */
len = sprintf(page, PRINTK_HEADER
" I/O status report for device %s:\n",
dev_name(&device->cdev->dev));
len += sprintf(page + len, PRINTK_HEADER
" in req: %p CC:%02X FC:%02X AC:%02X SC:%02X DS:%02X "
"CS:%02X fcxs:%02X schxs:%02X RC:%d\n",
req, scsw_cc(&irb->scsw), scsw_fctl(&irb->scsw),
scsw_actl(&irb->scsw), scsw_stctl(&irb->scsw),
scsw_dstat(&irb->scsw), scsw_cstat(&irb->scsw),
irb->scsw.tm.fcxs,
(irb->scsw.tm.ifob << 7) | irb->scsw.tm.sesq,
req ? req->intrc : 0);
len += sprintf(page + len, PRINTK_HEADER
" device %s: Failing TCW: %p\n",
dev_name(&device->cdev->dev),
(void *) (addr_t) irb->scsw.tm.tcw);
tsb = NULL;
sense = NULL;
if (irb->scsw.tm.tcw && (irb->scsw.tm.fcxs & 0x01))
tsb = tcw_get_tsb(
(struct tcw *)(unsigned long)irb->scsw.tm.tcw);
if (tsb) {
len += sprintf(page + len, PRINTK_HEADER
" tsb->length %d\n", tsb->length);
len += sprintf(page + len, PRINTK_HEADER
" tsb->flags %x\n", tsb->flags);
len += sprintf(page + len, PRINTK_HEADER
" tsb->dcw_offset %d\n", tsb->dcw_offset);
len += sprintf(page + len, PRINTK_HEADER
" tsb->count %d\n", tsb->count);
residual = tsb->count - 28;
len += sprintf(page + len, PRINTK_HEADER
" residual %d\n", residual);
switch (tsb->flags & 0x07) {
case 1: /* tsa_iostat */
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.iostat.dev_time %d\n",
tsb->tsa.iostat.dev_time);
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.iostat.def_time %d\n",
tsb->tsa.iostat.def_time);
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.iostat.queue_time %d\n",
tsb->tsa.iostat.queue_time);
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.iostat.dev_busy_time %d\n",
tsb->tsa.iostat.dev_busy_time);
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.iostat.dev_act_time %d\n",
tsb->tsa.iostat.dev_act_time);
sense = tsb->tsa.iostat.sense;
break;
case 2: /* ts_ddpc */
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.ddpc.rc %d\n", tsb->tsa.ddpc.rc);
for (sl = 0; sl < 2; sl++) {
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.ddpc.rcq %2d-%2d: ",
(8 * sl), ((8 * sl) + 7));
rcq = tsb->tsa.ddpc.rcq;
for (sct = 0; sct < 8; sct++) {
len += sprintf(page + len, " %02x",
rcq[8 * sl + sct]);
}
len += sprintf(page + len, "\n");
}
sense = tsb->tsa.ddpc.sense;
break;
case 3: /* tsa_intrg */
len += sprintf(page + len, PRINTK_HEADER
" tsb->tsa.intrg.: not supported yet\n");
break;
}
if (sense) {
for (sl = 0; sl < 4; sl++) {
len += sprintf(page + len, PRINTK_HEADER
" Sense(hex) %2d-%2d:",
(8 * sl), ((8 * sl) + 7));
for (sct = 0; sct < 8; sct++) {
len += sprintf(page + len, " %02x",
sense[8 * sl + sct]);
}
len += sprintf(page + len, "\n");
}
if (sense[27] & DASD_SENSE_BIT_0) {
/* 24 Byte Sense Data */
sprintf(page + len, PRINTK_HEADER
" 24 Byte: %x MSG %x, "
"%s MSGb to SYSOP\n",
sense[7] >> 4, sense[7] & 0x0f,
sense[1] & 0x10 ? "" : "no");
} else {
/* 32 Byte Sense Data */
sprintf(page + len, PRINTK_HEADER
" 32 Byte: Format: %x "
"Exception class %x\n",
sense[6] & 0x0f, sense[22] >> 4);
}
} else {
sprintf(page + len, PRINTK_HEADER
" SORRY - NO VALID SENSE AVAILABLE\n");
}
} else {
sprintf(page + len, PRINTK_HEADER
" SORRY - NO TSB DATA AVAILABLE\n");
}
printk(KERN_ERR "%s", page);
free_page((unsigned long) page);
}
static void dasd_eckd_dump_sense(struct dasd_device *device,
struct dasd_ccw_req *req, struct irb *irb)
{
u8 *sense = dasd_get_sense(irb);
if (scsw_is_tm(&irb->scsw)) {
/*
* In some cases the 'File Protected' or 'Incorrect Length'
* error might be expected and log messages shouldn't be written
* then. Check if the according suppress bit is set.
*/
if (sense && (sense[1] & SNS1_FILE_PROTECTED) &&
test_bit(DASD_CQR_SUPPRESS_FP, &req->flags))
return;
if (scsw_cstat(&irb->scsw) == 0x40 &&
test_bit(DASD_CQR_SUPPRESS_IL, &req->flags))
return;
dasd_eckd_dump_sense_tcw(device, req, irb);
} else {
/*
* In some cases the 'Command Reject' or 'No Record Found'
* error might be expected and log messages shouldn't be
* written then. Check if the according suppress bit is set.
*/
if (sense && sense[0] & SNS0_CMD_REJECT &&
test_bit(DASD_CQR_SUPPRESS_CR, &req->flags))
return;
if (sense && sense[1] & SNS1_NO_REC_FOUND &&
test_bit(DASD_CQR_SUPPRESS_NRF, &req->flags))
return;
dasd_eckd_dump_sense_ccw(device, req, irb);
}
}
static int dasd_eckd_pm_freeze(struct dasd_device *device)
{
/*
* the device should be disconnected from our LCU structure
* on restore we will reconnect it and reread LCU specific
* information like PAV support that might have changed
*/
dasd_alias_remove_device(device);
dasd_alias_disconnect_device_from_lcu(device);
return 0;
}
static int dasd_eckd_restore_device(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
struct dasd_eckd_characteristics temp_rdc_data;
int rc;
struct dasd_uid temp_uid;
unsigned long flags;
unsigned long cqr_flags = 0;
/* Read Configuration Data */
rc = dasd_eckd_read_conf(device);
if (rc) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Read configuration data failed, rc=%d", rc);
goto out_err;
}
dasd_eckd_get_uid(device, &temp_uid);
/* Generate device unique id */
rc = dasd_eckd_generate_uid(device);
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
if (memcmp(&private->uid, &temp_uid, sizeof(struct dasd_uid)) != 0)
dev_err(&device->cdev->dev, "The UID of the DASD has "
"changed\n");
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
if (rc)
goto out_err;
/* register lcu with alias handling, enable PAV if this is a new lcu */
rc = dasd_alias_make_device_known_to_lcu(device);
if (rc)
goto out_err;
set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr_flags);
dasd_eckd_validate_server(device, cqr_flags);
/* RE-Read Configuration Data */
rc = dasd_eckd_read_conf(device);
if (rc) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Read configuration data failed, rc=%d", rc);
goto out_err2;
}
/* Read Feature Codes */
dasd_eckd_read_features(device);
/* Read Device Characteristics */
rc = dasd_generic_read_dev_chars(device, DASD_ECKD_MAGIC,
&temp_rdc_data, 64);
if (rc) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Read device characteristic failed, rc=%d", rc);
goto out_err2;
}
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
memcpy(&private->rdc_data, &temp_rdc_data, sizeof(temp_rdc_data));
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
/* add device to alias management */
dasd_alias_add_device(device);
return 0;
out_err2:
dasd_alias_disconnect_device_from_lcu(device);
out_err:
return -1;
}
static int dasd_eckd_reload_device(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
int rc, old_base;
char print_uid[60];
struct dasd_uid uid;
unsigned long flags;
/*
* remove device from alias handling to prevent new requests
* from being scheduled on the wrong alias device
*/
dasd_alias_remove_device(device);
spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
old_base = private->uid.base_unit_addr;
spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
/* Read Configuration Data */
rc = dasd_eckd_read_conf(device);
if (rc)
goto out_err;
rc = dasd_eckd_generate_uid(device);
if (rc)
goto out_err;
/*
* update unit address configuration and
* add device to alias management
*/
dasd_alias_update_add_device(device);
dasd_eckd_get_uid(device, &uid);
if (old_base != uid.base_unit_addr) {
if (strlen(uid.vduit) > 0)
snprintf(print_uid, sizeof(print_uid),
"%s.%s.%04x.%02x.%s", uid.vendor, uid.serial,
uid.ssid, uid.base_unit_addr, uid.vduit);
else
snprintf(print_uid, sizeof(print_uid),
"%s.%s.%04x.%02x", uid.vendor, uid.serial,
uid.ssid, uid.base_unit_addr);
dev_info(&device->cdev->dev,
"An Alias device was reassigned to a new base device "
"with UID: %s\n", print_uid);
}
return 0;
out_err:
return -1;
}
static int dasd_eckd_read_message_buffer(struct dasd_device *device,
struct dasd_rssd_messages *messages,
__u8 lpum)
{
struct dasd_rssd_messages *message_buf;
struct dasd_psf_prssd_data *prssdp;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
int rc;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ + 1 /* RSSD */,
(sizeof(struct dasd_psf_prssd_data) +
sizeof(struct dasd_rssd_messages)),
device, NULL);
if (IS_ERR(cqr)) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"Could not allocate read message buffer request");
return PTR_ERR(cqr);
}
cqr->lpm = lpum;
retry:
cqr->startdev = device;
cqr->memdev = device;
cqr->block = NULL;
cqr->expires = 10 * HZ;
set_bit(DASD_CQR_VERIFY_PATH, &cqr->flags);
/* dasd_sleep_on_immediatly does not do complex error
* recovery so clear erp flag and set retry counter to
* do basic erp */
clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
cqr->retries = 256;
/* Prepare for Read Subsystem Data */
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
prssdp->order = PSF_ORDER_PRSSD;
prssdp->suborder = 0x03; /* Message Buffer */
/* all other bytes of prssdp must be zero */
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->count = sizeof(struct dasd_psf_prssd_data);
ccw->flags |= CCW_FLAG_CC;
ccw->flags |= CCW_FLAG_SLI;
ccw->cda = (__u32)(addr_t) prssdp;
/* Read Subsystem Data - message buffer */
message_buf = (struct dasd_rssd_messages *) (prssdp + 1);
memset(message_buf, 0, sizeof(struct dasd_rssd_messages));
ccw++;
ccw->cmd_code = DASD_ECKD_CCW_RSSD;
ccw->count = sizeof(struct dasd_rssd_messages);
ccw->flags |= CCW_FLAG_SLI;
ccw->cda = (__u32)(addr_t) message_buf;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
rc = dasd_sleep_on_immediatly(cqr);
if (rc == 0) {
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
message_buf = (struct dasd_rssd_messages *)
(prssdp + 1);
memcpy(messages, message_buf,
sizeof(struct dasd_rssd_messages));
} else if (cqr->lpm) {
/*
* on z/VM we might not be able to do I/O on the requested path
* but instead we get the required information on any path
* so retry with open path mask
*/
cqr->lpm = 0;
goto retry;
} else
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Reading messages failed with rc=%d\n"
, rc);
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
static int dasd_eckd_query_host_access(struct dasd_device *device,
struct dasd_psf_query_host_access *data)
{
struct dasd_eckd_private *private = device->private;
struct dasd_psf_query_host_access *host_access;
struct dasd_psf_prssd_data *prssdp;
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
int rc;
/* not available for HYPER PAV alias devices */
if (!device->block && private->lcu->pav == HYPER_PAV)
return -EOPNOTSUPP;
/* may not be supported by the storage server */
if (!(private->features.feature[14] & 0x80))
return -EOPNOTSUPP;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ + 1 /* RSSD */,
sizeof(struct dasd_psf_prssd_data) + 1,
device, NULL);
if (IS_ERR(cqr)) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"Could not allocate read message buffer request");
return PTR_ERR(cqr);
}
host_access = kzalloc(sizeof(*host_access), GFP_KERNEL | GFP_DMA);
if (!host_access) {
dasd_sfree_request(cqr, device);
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"Could not allocate host_access buffer");
return -ENOMEM;
}
cqr->startdev = device;
cqr->memdev = device;
cqr->block = NULL;
cqr->retries = 256;
cqr->expires = 10 * HZ;
/* Prepare for Read Subsystem Data */
prssdp = (struct dasd_psf_prssd_data *) cqr->data;
memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
prssdp->order = PSF_ORDER_PRSSD;
prssdp->suborder = PSF_SUBORDER_QHA; /* query host access */
/* LSS and Volume that will be queried */
prssdp->lss = private->ned->ID;
prssdp->volume = private->ned->unit_addr;
/* all other bytes of prssdp must be zero */
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->count = sizeof(struct dasd_psf_prssd_data);
ccw->flags |= CCW_FLAG_CC;
ccw->flags |= CCW_FLAG_SLI;
ccw->cda = (__u32)(addr_t) prssdp;
/* Read Subsystem Data - query host access */
ccw++;
ccw->cmd_code = DASD_ECKD_CCW_RSSD;
ccw->count = sizeof(struct dasd_psf_query_host_access);
ccw->flags |= CCW_FLAG_SLI;
ccw->cda = (__u32)(addr_t) host_access;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
/* the command might not be supported, suppress error message */
__set_bit(DASD_CQR_SUPPRESS_CR, &cqr->flags);
rc = dasd_sleep_on_interruptible(cqr);
if (rc == 0) {
*data = *host_access;
} else {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev,
"Reading host access data failed with rc=%d\n",
rc);
rc = -EOPNOTSUPP;
}
dasd_sfree_request(cqr, cqr->memdev);
kfree(host_access);
return rc;
}
/*
* return number of grouped devices
*/
static int dasd_eckd_host_access_count(struct dasd_device *device)
{
struct dasd_psf_query_host_access *access;
struct dasd_ckd_path_group_entry *entry;
struct dasd_ckd_host_information *info;
int count = 0;
int rc, i;
access = kzalloc(sizeof(*access), GFP_NOIO);
if (!access) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"Could not allocate access buffer");
return -ENOMEM;
}
rc = dasd_eckd_query_host_access(device, access);
if (rc) {
kfree(access);
return rc;
}
info = (struct dasd_ckd_host_information *)
access->host_access_information;
for (i = 0; i < info->entry_count; i++) {
entry = (struct dasd_ckd_path_group_entry *)
(info->entry + i * info->entry_size);
if (entry->status_flags & DASD_ECKD_PG_GROUPED)
count++;
}
kfree(access);
return count;
}
/*
* write host access information to a sequential file
*/
static int dasd_hosts_print(struct dasd_device *device, struct seq_file *m)
{
struct dasd_psf_query_host_access *access;
struct dasd_ckd_path_group_entry *entry;
struct dasd_ckd_host_information *info;
char sysplex[9] = "";
int rc, i;
access = kzalloc(sizeof(*access), GFP_NOIO);
if (!access) {
DBF_EVENT_DEVID(DBF_WARNING, device->cdev, "%s",
"Could not allocate access buffer");
return -ENOMEM;
}
rc = dasd_eckd_query_host_access(device, access);
if (rc) {
kfree(access);
return rc;
}
info = (struct dasd_ckd_host_information *)
access->host_access_information;
for (i = 0; i < info->entry_count; i++) {
entry = (struct dasd_ckd_path_group_entry *)
(info->entry + i * info->entry_size);
/* PGID */
seq_printf(m, "pgid %*phN\n", 11, entry->pgid);
/* FLAGS */
seq_printf(m, "status_flags %02x\n", entry->status_flags);
/* SYSPLEX NAME */
memcpy(&sysplex, &entry->sysplex_name, sizeof(sysplex) - 1);
EBCASC(sysplex, sizeof(sysplex));
seq_printf(m, "sysplex_name %8s\n", sysplex);
/* SUPPORTED CYLINDER */
seq_printf(m, "supported_cylinder %d\n", entry->cylinder);
/* TIMESTAMP */
seq_printf(m, "timestamp %lu\n", (unsigned long)
entry->timestamp);
}
kfree(access);
return 0;
}
/*
* Perform Subsystem Function - CUIR response
*/
static int
dasd_eckd_psf_cuir_response(struct dasd_device *device, int response,
__u32 message_id, __u8 lpum)
{
struct dasd_psf_cuir_response *psf_cuir;
int pos = pathmask_to_pos(lpum);
struct dasd_ccw_req *cqr;
struct ccw1 *ccw;
int rc;
cqr = dasd_smalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */ ,
sizeof(struct dasd_psf_cuir_response),
device, NULL);
if (IS_ERR(cqr)) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Could not allocate PSF-CUIR request");
return PTR_ERR(cqr);
}
psf_cuir = (struct dasd_psf_cuir_response *)cqr->data;
psf_cuir->order = PSF_ORDER_CUIR_RESPONSE;
psf_cuir->cc = response;
psf_cuir->chpid = device->path[pos].chpid;
psf_cuir->message_id = message_id;
psf_cuir->cssid = device->path[pos].cssid;
psf_cuir->ssid = device->path[pos].ssid;
ccw = cqr->cpaddr;
ccw->cmd_code = DASD_ECKD_CCW_PSF;
ccw->cda = (__u32)(addr_t)psf_cuir;
ccw->flags = CCW_FLAG_SLI;
ccw->count = sizeof(struct dasd_psf_cuir_response);
cqr->startdev = device;
cqr->memdev = device;
cqr->block = NULL;
cqr->retries = 256;
cqr->expires = 10*HZ;
cqr->buildclk = get_tod_clock();
cqr->status = DASD_CQR_FILLED;
set_bit(DASD_CQR_VERIFY_PATH, &cqr->flags);
rc = dasd_sleep_on(cqr);
dasd_sfree_request(cqr, cqr->memdev);
return rc;
}
/*
* return configuration data that is referenced by record selector
* if a record selector is specified or per default return the
* conf_data pointer for the path specified by lpum
*/
static struct dasd_conf_data *dasd_eckd_get_ref_conf(struct dasd_device *device,
__u8 lpum,
struct dasd_cuir_message *cuir)
{
struct dasd_conf_data *conf_data;
int path, pos;
if (cuir->record_selector == 0)
goto out;
for (path = 0x80, pos = 0; path; path >>= 1, pos++) {
conf_data = device->path[pos].conf_data;
if (conf_data->gneq.record_selector ==
cuir->record_selector)
return conf_data;
}
out:
return device->path[pathmask_to_pos(lpum)].conf_data;
}
/*
* This function determines the scope of a reconfiguration request by
* analysing the path and device selection data provided in the CUIR request.
* Returns a path mask containing CUIR affected paths for the give device.
*
* If the CUIR request does not contain the required information return the
* path mask of the path the attention message for the CUIR request was reveived
* on.
*/
static int dasd_eckd_cuir_scope(struct dasd_device *device, __u8 lpum,
struct dasd_cuir_message *cuir)
{
struct dasd_conf_data *ref_conf_data;
unsigned long bitmask = 0, mask = 0;
struct dasd_conf_data *conf_data;
unsigned int pos, path;
char *ref_gneq, *gneq;
char *ref_ned, *ned;
int tbcpm = 0;
/* if CUIR request does not specify the scope use the path
the attention message was presented on */
if (!cuir->ned_map ||
!(cuir->neq_map[0] | cuir->neq_map[1] | cuir->neq_map[2]))
return lpum;
/* get reference conf data */
ref_conf_data = dasd_eckd_get_ref_conf(device, lpum, cuir);
/* reference ned is determined by ned_map field */
pos = 8 - ffs(cuir->ned_map);
ref_ned = (char *)&ref_conf_data->neds[pos];
ref_gneq = (char *)&ref_conf_data->gneq;
/* transfer 24 bit neq_map to mask */
mask = cuir->neq_map[2];
mask |= cuir->neq_map[1] << 8;
mask |= cuir->neq_map[0] << 16;
for (path = 0; path < 8; path++) {
/* initialise data per path */
bitmask = mask;
conf_data = device->path[path].conf_data;
pos = 8 - ffs(cuir->ned_map);
ned = (char *) &conf_data->neds[pos];
/* compare reference ned and per path ned */
if (memcmp(ref_ned, ned, sizeof(*ned)) != 0)
continue;
gneq = (char *)&conf_data->gneq;
/* compare reference gneq and per_path gneq under
24 bit mask where mask bit 0 equals byte 7 of
the gneq and mask bit 24 equals byte 31 */
while (bitmask) {
pos = ffs(bitmask) - 1;
if (memcmp(&ref_gneq[31 - pos], &gneq[31 - pos], 1)
!= 0)
break;
clear_bit(pos, &bitmask);
}
if (bitmask)
continue;
/* device and path match the reference values
add path to CUIR scope */
tbcpm |= 0x80 >> path;
}
return tbcpm;
}
static void dasd_eckd_cuir_notify_user(struct dasd_device *device,
unsigned long paths, int action)
{
int pos;
while (paths) {
/* get position of bit in mask */
pos = 8 - ffs(paths);
/* get channel path descriptor from this position */
if (action == CUIR_QUIESCE)
pr_warn("Service on the storage server caused path %x.%02x to go offline",
device->path[pos].cssid,
device->path[pos].chpid);
else if (action == CUIR_RESUME)
pr_info("Path %x.%02x is back online after service on the storage server",
device->path[pos].cssid,
device->path[pos].chpid);
clear_bit(7 - pos, &paths);
}
}
static int dasd_eckd_cuir_remove_path(struct dasd_device *device, __u8 lpum,
struct dasd_cuir_message *cuir)
{
unsigned long tbcpm;
tbcpm = dasd_eckd_cuir_scope(device, lpum, cuir);
/* nothing to do if path is not in use */
if (!(dasd_path_get_opm(device) & tbcpm))
return 0;
if (!(dasd_path_get_opm(device) & ~tbcpm)) {
/* no path would be left if the CUIR action is taken
return error */
return -EINVAL;
}
/* remove device from operational path mask */
dasd_path_remove_opm(device, tbcpm);
dasd_path_add_cuirpm(device, tbcpm);
return tbcpm;
}
/*
* walk through all devices and build a path mask to quiesce them
* return an error if the last path to a device would be removed
*
* if only part of the devices are quiesced and an error
* occurs no onlining necessary, the storage server will
* notify the already set offline devices again
*/
static int dasd_eckd_cuir_quiesce(struct dasd_device *device, __u8 lpum,
struct dasd_cuir_message *cuir)
{
struct dasd_eckd_private *private = device->private;
struct alias_pav_group *pavgroup, *tempgroup;
struct dasd_device *dev, *n;
unsigned long paths = 0;
unsigned long flags;
int tbcpm;
/* active devices */
list_for_each_entry_safe(dev, n, &private->lcu->active_devices,
alias_list) {
spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
spin_unlock_irqrestore(get_ccwdev_lock(dev->cdev), flags);
if (tbcpm < 0)
goto out_err;
paths |= tbcpm;
}
/* inactive devices */
list_for_each_entry_safe(dev, n, &private->lcu->inactive_devices,
alias_list) {
spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
spin_unlock_irqrestore(get_ccwdev_lock(dev->cdev), flags);
if (tbcpm < 0)
goto out_err;
paths |= tbcpm;
}
/* devices in PAV groups */
list_for_each_entry_safe(pavgroup, tempgroup,
&private->lcu->grouplist, group) {
list_for_each_entry_safe(dev, n, &pavgroup->baselist,
alias_list) {
spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
spin_unlock_irqrestore(
get_ccwdev_lock(dev->cdev), flags);
if (tbcpm < 0)
goto out_err;
paths |= tbcpm;
}
list_for_each_entry_safe(dev, n, &pavgroup->aliaslist,
alias_list) {
spin_lock_irqsave(get_ccwdev_lock(dev->cdev), flags);
tbcpm = dasd_eckd_cuir_remove_path(dev, lpum, cuir);
spin_unlock_irqrestore(
get_ccwdev_lock(dev->cdev), flags);
if (tbcpm < 0)
goto out_err;
paths |= tbcpm;
}
}
/* notify user about all paths affected by CUIR action */
dasd_eckd_cuir_notify_user(device, paths, CUIR_QUIESCE);
return 0;
out_err:
return tbcpm;
}
static int dasd_eckd_cuir_resume(struct dasd_device *device, __u8 lpum,
struct dasd_cuir_message *cuir)
{
struct dasd_eckd_private *private = device->private;
struct alias_pav_group *pavgroup, *tempgroup;
struct dasd_device *dev, *n;
unsigned long paths = 0;
int tbcpm;
/*
* the path may have been added through a generic path event before
* only trigger path verification if the path is not already in use
*/
list_for_each_entry_safe(dev, n,
&private->lcu->active_devices,
alias_list) {
tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
paths |= tbcpm;
if (!(dasd_path_get_opm(dev) & tbcpm)) {
dasd_path_add_tbvpm(dev, tbcpm);
dasd_schedule_device_bh(dev);
}
}
list_for_each_entry_safe(dev, n,
&private->lcu->inactive_devices,
alias_list) {
tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
paths |= tbcpm;
if (!(dasd_path_get_opm(dev) & tbcpm)) {
dasd_path_add_tbvpm(dev, tbcpm);
dasd_schedule_device_bh(dev);
}
}
/* devices in PAV groups */
list_for_each_entry_safe(pavgroup, tempgroup,
&private->lcu->grouplist,
group) {
list_for_each_entry_safe(dev, n,
&pavgroup->baselist,
alias_list) {
tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
paths |= tbcpm;
if (!(dasd_path_get_opm(dev) & tbcpm)) {
dasd_path_add_tbvpm(dev, tbcpm);
dasd_schedule_device_bh(dev);
}
}
list_for_each_entry_safe(dev, n,
&pavgroup->aliaslist,
alias_list) {
tbcpm = dasd_eckd_cuir_scope(dev, lpum, cuir);
paths |= tbcpm;
if (!(dasd_path_get_opm(dev) & tbcpm)) {
dasd_path_add_tbvpm(dev, tbcpm);
dasd_schedule_device_bh(dev);
}
}
}
/* notify user about all paths affected by CUIR action */
dasd_eckd_cuir_notify_user(device, paths, CUIR_RESUME);
return 0;
}
static void dasd_eckd_handle_cuir(struct dasd_device *device, void *messages,
__u8 lpum)
{
struct dasd_cuir_message *cuir = messages;
int response;
DBF_DEV_EVENT(DBF_WARNING, device,
"CUIR request: %016llx %016llx %016llx %08x",
((u64 *)cuir)[0], ((u64 *)cuir)[1], ((u64 *)cuir)[2],
((u32 *)cuir)[3]);
if (cuir->code == CUIR_QUIESCE) {
/* quiesce */
if (dasd_eckd_cuir_quiesce(device, lpum, cuir))
response = PSF_CUIR_LAST_PATH;
else
response = PSF_CUIR_COMPLETED;
} else if (cuir->code == CUIR_RESUME) {
/* resume */
dasd_eckd_cuir_resume(device, lpum, cuir);
response = PSF_CUIR_COMPLETED;
} else
response = PSF_CUIR_NOT_SUPPORTED;
dasd_eckd_psf_cuir_response(device, response,
cuir->message_id, lpum);
DBF_DEV_EVENT(DBF_WARNING, device,
"CUIR response: %d on message ID %08x", response,
cuir->message_id);
/* to make sure there is no attention left schedule work again */
device->discipline->check_attention(device, lpum);
}
static void dasd_eckd_check_attention_work(struct work_struct *work)
{
struct check_attention_work_data *data;
struct dasd_rssd_messages *messages;
struct dasd_device *device;
int rc;
data = container_of(work, struct check_attention_work_data, worker);
device = data->device;
messages = kzalloc(sizeof(*messages), GFP_KERNEL);
if (!messages) {
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Could not allocate attention message buffer");
goto out;
}
rc = dasd_eckd_read_message_buffer(device, messages, data->lpum);
if (rc)
goto out;
if (messages->length == ATTENTION_LENGTH_CUIR &&
messages->format == ATTENTION_FORMAT_CUIR)
dasd_eckd_handle_cuir(device, messages, data->lpum);
out:
dasd_put_device(device);
kfree(messages);
kfree(data);
}
static int dasd_eckd_check_attention(struct dasd_device *device, __u8 lpum)
{
struct check_attention_work_data *data;
data = kzalloc(sizeof(*data), GFP_ATOMIC);
if (!data)
return -ENOMEM;
INIT_WORK(&data->worker, dasd_eckd_check_attention_work);
dasd_get_device(device);
data->device = device;
data->lpum = lpum;
schedule_work(&data->worker);
return 0;
}
static int dasd_eckd_disable_hpf_path(struct dasd_device *device, __u8 lpum)
{
if (~lpum & dasd_path_get_opm(device)) {
dasd_path_add_nohpfpm(device, lpum);
dasd_path_remove_opm(device, lpum);
dev_err(&device->cdev->dev,
"Channel path %02X lost HPF functionality and is disabled\n",
lpum);
return 1;
}
return 0;
}
static void dasd_eckd_disable_hpf_device(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
dev_err(&device->cdev->dev,
"High Performance FICON disabled\n");
private->fcx_max_data = 0;
}
static int dasd_eckd_hpf_enabled(struct dasd_device *device)
{
struct dasd_eckd_private *private = device->private;
return private->fcx_max_data ? 1 : 0;
}
static void dasd_eckd_handle_hpf_error(struct dasd_device *device,
struct irb *irb)
{
struct dasd_eckd_private *private = device->private;
if (!private->fcx_max_data) {
/* sanity check for no HPF, the error makes no sense */
DBF_DEV_EVENT(DBF_WARNING, device, "%s",
"Trying to disable HPF for a non HPF device");
return;
}
if (irb->scsw.tm.sesq == SCSW_SESQ_DEV_NOFCX) {
dasd_eckd_disable_hpf_device(device);
} else if (irb->scsw.tm.sesq == SCSW_SESQ_PATH_NOFCX) {
if (dasd_eckd_disable_hpf_path(device, irb->esw.esw1.lpum))
return;
dasd_eckd_disable_hpf_device(device);
dasd_path_set_tbvpm(device,
dasd_path_get_hpfpm(device));
}
/*
* prevent that any new I/O ist started on the device and schedule a
* requeue of existing requests
*/
dasd_device_set_stop_bits(device, DASD_STOPPED_NOT_ACC);
dasd_schedule_requeue(device);
}
static struct ccw_driver dasd_eckd_driver = {
.driver = {
.name = "dasd-eckd",
.owner = THIS_MODULE,
},
.ids = dasd_eckd_ids,
.probe = dasd_eckd_probe,
.remove = dasd_generic_remove,
.set_offline = dasd_generic_set_offline,
.set_online = dasd_eckd_set_online,
.notify = dasd_generic_notify,
.path_event = dasd_generic_path_event,
.shutdown = dasd_generic_shutdown,
.freeze = dasd_generic_pm_freeze,
.thaw = dasd_generic_restore_device,
.restore = dasd_generic_restore_device,
.uc_handler = dasd_generic_uc_handler,
.int_class = IRQIO_DAS,
};
/*
* max_blocks is dependent on the amount of storage that is available
* in the static io buffer for each device. Currently each device has
* 8192 bytes (=2 pages). For 64 bit one dasd_mchunkt_t structure has
* 24 bytes, the struct dasd_ccw_req has 136 bytes and each block can use
* up to 16 bytes (8 for the ccw and 8 for the idal pointer). In
* addition we have one define extent ccw + 16 bytes of data and one
* locate record ccw + 16 bytes of data. That makes:
* (8192 - 24 - 136 - 8 - 16 - 8 - 16) / 16 = 499 blocks at maximum.
* We want to fit two into the available memory so that we can immediately
* start the next request if one finishes off. That makes 249.5 blocks
* for one request. Give a little safety and the result is 240.
*/
static struct dasd_discipline dasd_eckd_discipline = {
.owner = THIS_MODULE,
.name = "ECKD",
.ebcname = "ECKD",
.max_blocks = 190,
.check_device = dasd_eckd_check_characteristics,
.uncheck_device = dasd_eckd_uncheck_device,
.do_analysis = dasd_eckd_do_analysis,
.verify_path = dasd_eckd_verify_path,
.basic_to_ready = dasd_eckd_basic_to_ready,
.online_to_ready = dasd_eckd_online_to_ready,
.basic_to_known = dasd_eckd_basic_to_known,
.fill_geometry = dasd_eckd_fill_geometry,
.start_IO = dasd_start_IO,
.term_IO = dasd_term_IO,
.handle_terminated_request = dasd_eckd_handle_terminated_request,
.format_device = dasd_eckd_format_device,
.check_device_format = dasd_eckd_check_device_format,
.erp_action = dasd_eckd_erp_action,
.erp_postaction = dasd_eckd_erp_postaction,
.check_for_device_change = dasd_eckd_check_for_device_change,
.build_cp = dasd_eckd_build_alias_cp,
.free_cp = dasd_eckd_free_alias_cp,
.dump_sense = dasd_eckd_dump_sense,
.dump_sense_dbf = dasd_eckd_dump_sense_dbf,
.fill_info = dasd_eckd_fill_info,
.ioctl = dasd_eckd_ioctl,
.freeze = dasd_eckd_pm_freeze,
.restore = dasd_eckd_restore_device,
.reload = dasd_eckd_reload_device,
.get_uid = dasd_eckd_get_uid,
.kick_validate = dasd_eckd_kick_validate_server,
.check_attention = dasd_eckd_check_attention,
.host_access_count = dasd_eckd_host_access_count,
.hosts_print = dasd_hosts_print,
.handle_hpf_error = dasd_eckd_handle_hpf_error,
.disable_hpf = dasd_eckd_disable_hpf_device,
.hpf_enabled = dasd_eckd_hpf_enabled,
.reset_path = dasd_eckd_reset_path,
};
static int __init
dasd_eckd_init(void)
{
int ret;
ASCEBC(dasd_eckd_discipline.ebcname, 4);
dasd_reserve_req = kmalloc(sizeof(*dasd_reserve_req),
GFP_KERNEL | GFP_DMA);
if (!dasd_reserve_req)
return -ENOMEM;
path_verification_worker = kmalloc(sizeof(*path_verification_worker),
GFP_KERNEL | GFP_DMA);
if (!path_verification_worker) {
kfree(dasd_reserve_req);
return -ENOMEM;
}
rawpadpage = (void *)__get_free_page(GFP_KERNEL);
if (!rawpadpage) {
kfree(path_verification_worker);
kfree(dasd_reserve_req);
return -ENOMEM;
}
ret = ccw_driver_register(&dasd_eckd_driver);
if (!ret)
wait_for_device_probe();
else {
kfree(path_verification_worker);
kfree(dasd_reserve_req);
free_page((unsigned long)rawpadpage);
}
return ret;
}
static void __exit
dasd_eckd_cleanup(void)
{
ccw_driver_unregister(&dasd_eckd_driver);
kfree(path_verification_worker);
kfree(dasd_reserve_req);
free_page((unsigned long)rawpadpage);
}
module_init(dasd_eckd_init);
module_exit(dasd_eckd_cleanup);