linux_dsm_epyc7002/drivers/scsi/ppa.c
Ming Lei 57ef4e5109 scsi: ppa: use sg helper to iterate over scatterlist
Unlike the legacy I/O path, scsi-mq preallocates a large array to hold
the scatterlist for each request. This static allocation can consume
substantial amounts of memory on modern controllers which support a
large number of concurrently outstanding requests.

To facilitate a switch to a smaller static allocation combined with a
dynamic allocation for requests that need it, we need to make sure all
SCSI drivers handle chained scatterlists correctly.

Convert remaining drivers that directly dereference the scatterlist
array to using the iterator functions.

[mkp: clarified commit message]

Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-20 15:21:33 -04:00

1169 lines
26 KiB
C

/* ppa.c -- low level driver for the IOMEGA PPA3
* parallel port SCSI host adapter.
*
* (The PPA3 is the embedded controller in the ZIP drive.)
*
* (c) 1995,1996 Grant R. Guenther, grant@torque.net,
* under the terms of the GNU General Public License.
*
*/
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/blkdev.h>
#include <linux/parport.h>
#include <linux/workqueue.h>
#include <linux/delay.h>
#include <linux/jiffies.h>
#include <asm/io.h>
#include <scsi/scsi.h>
#include <scsi/scsi_cmnd.h>
#include <scsi/scsi_device.h>
#include <scsi/scsi_host.h>
static void ppa_reset_pulse(unsigned int base);
typedef struct {
struct pardevice *dev; /* Parport device entry */
int base; /* Actual port address */
int mode; /* Transfer mode */
struct scsi_cmnd *cur_cmd; /* Current queued command */
struct delayed_work ppa_tq; /* Polling interrupt stuff */
unsigned long jstart; /* Jiffies at start */
unsigned long recon_tmo; /* How many usecs to wait for reconnection (6th bit) */
unsigned int failed:1; /* Failure flag */
unsigned wanted:1; /* Parport sharing busy flag */
unsigned int dev_no; /* Device number */
wait_queue_head_t *waiting;
struct Scsi_Host *host;
struct list_head list;
} ppa_struct;
#include "ppa.h"
static inline ppa_struct *ppa_dev(struct Scsi_Host *host)
{
return *(ppa_struct **)&host->hostdata;
}
static DEFINE_SPINLOCK(arbitration_lock);
static void got_it(ppa_struct *dev)
{
dev->base = dev->dev->port->base;
if (dev->cur_cmd)
dev->cur_cmd->SCp.phase = 1;
else
wake_up(dev->waiting);
}
static void ppa_wakeup(void *ref)
{
ppa_struct *dev = (ppa_struct *) ref;
unsigned long flags;
spin_lock_irqsave(&arbitration_lock, flags);
if (dev->wanted) {
parport_claim(dev->dev);
got_it(dev);
dev->wanted = 0;
}
spin_unlock_irqrestore(&arbitration_lock, flags);
return;
}
static int ppa_pb_claim(ppa_struct *dev)
{
unsigned long flags;
int res = 1;
spin_lock_irqsave(&arbitration_lock, flags);
if (parport_claim(dev->dev) == 0) {
got_it(dev);
res = 0;
}
dev->wanted = res;
spin_unlock_irqrestore(&arbitration_lock, flags);
return res;
}
static void ppa_pb_dismiss(ppa_struct *dev)
{
unsigned long flags;
int wanted;
spin_lock_irqsave(&arbitration_lock, flags);
wanted = dev->wanted;
dev->wanted = 0;
spin_unlock_irqrestore(&arbitration_lock, flags);
if (!wanted)
parport_release(dev->dev);
}
static inline void ppa_pb_release(ppa_struct *dev)
{
parport_release(dev->dev);
}
/*
* Start of Chipset kludges
*/
/* This is to give the ppa driver a way to modify the timings (and other
* parameters) by writing to the /proc/scsi/ppa/0 file.
* Very simple method really... (To simple, no error checking :( )
* Reason: Kernel hackers HATE having to unload and reload modules for
* testing...
* Also gives a method to use a script to obtain optimum timings (TODO)
*/
static inline int ppa_write_info(struct Scsi_Host *host, char *buffer, int length)
{
ppa_struct *dev = ppa_dev(host);
unsigned long x;
if ((length > 5) && (strncmp(buffer, "mode=", 5) == 0)) {
x = simple_strtoul(buffer + 5, NULL, 0);
dev->mode = x;
return length;
}
if ((length > 10) && (strncmp(buffer, "recon_tmo=", 10) == 0)) {
x = simple_strtoul(buffer + 10, NULL, 0);
dev->recon_tmo = x;
printk(KERN_INFO "ppa: recon_tmo set to %ld\n", x);
return length;
}
printk(KERN_WARNING "ppa /proc: invalid variable\n");
return -EINVAL;
}
static int ppa_show_info(struct seq_file *m, struct Scsi_Host *host)
{
ppa_struct *dev = ppa_dev(host);
seq_printf(m, "Version : %s\n", PPA_VERSION);
seq_printf(m, "Parport : %s\n", dev->dev->port->name);
seq_printf(m, "Mode : %s\n", PPA_MODE_STRING[dev->mode]);
#if PPA_DEBUG > 0
seq_printf(m, "recon_tmo : %lu\n", dev->recon_tmo);
#endif
return 0;
}
static int device_check(ppa_struct *dev);
#if PPA_DEBUG > 0
#define ppa_fail(x,y) printk("ppa: ppa_fail(%i) from %s at line %d\n",\
y, __func__, __LINE__); ppa_fail_func(x,y);
static inline void ppa_fail_func(ppa_struct *dev, int error_code)
#else
static inline void ppa_fail(ppa_struct *dev, int error_code)
#endif
{
/* If we fail a device then we trash status / message bytes */
if (dev->cur_cmd) {
dev->cur_cmd->result = error_code << 16;
dev->failed = 1;
}
}
/*
* Wait for the high bit to be set.
*
* In principle, this could be tied to an interrupt, but the adapter
* doesn't appear to be designed to support interrupts. We spin on
* the 0x80 ready bit.
*/
static unsigned char ppa_wait(ppa_struct *dev)
{
int k;
unsigned short ppb = dev->base;
unsigned char r;
k = PPA_SPIN_TMO;
/* Wait for bit 6 and 7 - PJC */
for (r = r_str(ppb); ((r & 0xc0) != 0xc0) && (k); k--) {
udelay(1);
r = r_str(ppb);
}
/*
* return some status information.
* Semantics: 0xc0 = ZIP wants more data
* 0xd0 = ZIP wants to send more data
* 0xe0 = ZIP is expecting SCSI command data
* 0xf0 = end of transfer, ZIP is sending status
*/
if (k)
return (r & 0xf0);
/* Counter expired - Time out occurred */
ppa_fail(dev, DID_TIME_OUT);
printk(KERN_WARNING "ppa timeout in ppa_wait\n");
return 0; /* command timed out */
}
/*
* Clear EPP Timeout Bit
*/
static inline void epp_reset(unsigned short ppb)
{
int i;
i = r_str(ppb);
w_str(ppb, i);
w_str(ppb, i & 0xfe);
}
/*
* Wait for empty ECP fifo (if we are in ECP fifo mode only)
*/
static inline void ecp_sync(ppa_struct *dev)
{
int i, ppb_hi = dev->dev->port->base_hi;
if (ppb_hi == 0)
return;
if ((r_ecr(ppb_hi) & 0xe0) == 0x60) { /* mode 011 == ECP fifo mode */
for (i = 0; i < 100; i++) {
if (r_ecr(ppb_hi) & 0x01)
return;
udelay(5);
}
printk(KERN_WARNING "ppa: ECP sync failed as data still present in FIFO.\n");
}
}
static int ppa_byte_out(unsigned short base, const char *buffer, int len)
{
int i;
for (i = len; i; i--) {
w_dtr(base, *buffer++);
w_ctr(base, 0xe);
w_ctr(base, 0xc);
}
return 1; /* All went well - we hope! */
}
static int ppa_byte_in(unsigned short base, char *buffer, int len)
{
int i;
for (i = len; i; i--) {
*buffer++ = r_dtr(base);
w_ctr(base, 0x27);
w_ctr(base, 0x25);
}
return 1; /* All went well - we hope! */
}
static int ppa_nibble_in(unsigned short base, char *buffer, int len)
{
for (; len; len--) {
unsigned char h;
w_ctr(base, 0x4);
h = r_str(base) & 0xf0;
w_ctr(base, 0x6);
*buffer++ = h | ((r_str(base) & 0xf0) >> 4);
}
return 1; /* All went well - we hope! */
}
static int ppa_out(ppa_struct *dev, char *buffer, int len)
{
int r;
unsigned short ppb = dev->base;
r = ppa_wait(dev);
if ((r & 0x50) != 0x40) {
ppa_fail(dev, DID_ERROR);
return 0;
}
switch (dev->mode) {
case PPA_NIBBLE:
case PPA_PS2:
/* 8 bit output, with a loop */
r = ppa_byte_out(ppb, buffer, len);
break;
case PPA_EPP_32:
case PPA_EPP_16:
case PPA_EPP_8:
epp_reset(ppb);
w_ctr(ppb, 0x4);
#ifdef CONFIG_SCSI_IZIP_EPP16
if (!(((long) buffer | len) & 0x01))
outsw(ppb + 4, buffer, len >> 1);
#else
if (!(((long) buffer | len) & 0x03))
outsl(ppb + 4, buffer, len >> 2);
#endif
else
outsb(ppb + 4, buffer, len);
w_ctr(ppb, 0xc);
r = !(r_str(ppb) & 0x01);
w_ctr(ppb, 0xc);
ecp_sync(dev);
break;
default:
printk(KERN_ERR "PPA: bug in ppa_out()\n");
r = 0;
}
return r;
}
static int ppa_in(ppa_struct *dev, char *buffer, int len)
{
int r;
unsigned short ppb = dev->base;
r = ppa_wait(dev);
if ((r & 0x50) != 0x50) {
ppa_fail(dev, DID_ERROR);
return 0;
}
switch (dev->mode) {
case PPA_NIBBLE:
/* 4 bit input, with a loop */
r = ppa_nibble_in(ppb, buffer, len);
w_ctr(ppb, 0xc);
break;
case PPA_PS2:
/* 8 bit input, with a loop */
w_ctr(ppb, 0x25);
r = ppa_byte_in(ppb, buffer, len);
w_ctr(ppb, 0x4);
w_ctr(ppb, 0xc);
break;
case PPA_EPP_32:
case PPA_EPP_16:
case PPA_EPP_8:
epp_reset(ppb);
w_ctr(ppb, 0x24);
#ifdef CONFIG_SCSI_IZIP_EPP16
if (!(((long) buffer | len) & 0x01))
insw(ppb + 4, buffer, len >> 1);
#else
if (!(((long) buffer | len) & 0x03))
insl(ppb + 4, buffer, len >> 2);
#endif
else
insb(ppb + 4, buffer, len);
w_ctr(ppb, 0x2c);
r = !(r_str(ppb) & 0x01);
w_ctr(ppb, 0x2c);
ecp_sync(dev);
break;
default:
printk(KERN_ERR "PPA: bug in ppa_ins()\n");
r = 0;
break;
}
return r;
}
/* end of ppa_io.h */
static inline void ppa_d_pulse(unsigned short ppb, unsigned char b)
{
w_dtr(ppb, b);
w_ctr(ppb, 0xc);
w_ctr(ppb, 0xe);
w_ctr(ppb, 0xc);
w_ctr(ppb, 0x4);
w_ctr(ppb, 0xc);
}
static void ppa_disconnect(ppa_struct *dev)
{
unsigned short ppb = dev->base;
ppa_d_pulse(ppb, 0);
ppa_d_pulse(ppb, 0x3c);
ppa_d_pulse(ppb, 0x20);
ppa_d_pulse(ppb, 0xf);
}
static inline void ppa_c_pulse(unsigned short ppb, unsigned char b)
{
w_dtr(ppb, b);
w_ctr(ppb, 0x4);
w_ctr(ppb, 0x6);
w_ctr(ppb, 0x4);
w_ctr(ppb, 0xc);
}
static inline void ppa_connect(ppa_struct *dev, int flag)
{
unsigned short ppb = dev->base;
ppa_c_pulse(ppb, 0);
ppa_c_pulse(ppb, 0x3c);
ppa_c_pulse(ppb, 0x20);
if ((flag == CONNECT_EPP_MAYBE) && IN_EPP_MODE(dev->mode))
ppa_c_pulse(ppb, 0xcf);
else
ppa_c_pulse(ppb, 0x8f);
}
static int ppa_select(ppa_struct *dev, int target)
{
int k;
unsigned short ppb = dev->base;
/*
* Bit 6 (0x40) is the device selected bit.
* First we must wait till the current device goes off line...
*/
k = PPA_SELECT_TMO;
do {
k--;
udelay(1);
} while ((r_str(ppb) & 0x40) && (k));
if (!k)
return 0;
w_dtr(ppb, (1 << target));
w_ctr(ppb, 0xe);
w_ctr(ppb, 0xc);
w_dtr(ppb, 0x80); /* This is NOT the initator */
w_ctr(ppb, 0x8);
k = PPA_SELECT_TMO;
do {
k--;
udelay(1);
}
while (!(r_str(ppb) & 0x40) && (k));
if (!k)
return 0;
return 1;
}
/*
* This is based on a trace of what the Iomega DOS 'guest' driver does.
* I've tried several different kinds of parallel ports with guest and
* coded this to react in the same ways that it does.
*
* The return value from this function is just a hint about where the
* handshaking failed.
*
*/
static int ppa_init(ppa_struct *dev)
{
int retv;
unsigned short ppb = dev->base;
ppa_disconnect(dev);
ppa_connect(dev, CONNECT_NORMAL);
retv = 2; /* Failed */
w_ctr(ppb, 0xe);
if ((r_str(ppb) & 0x08) == 0x08)
retv--;
w_ctr(ppb, 0xc);
if ((r_str(ppb) & 0x08) == 0x00)
retv--;
if (!retv)
ppa_reset_pulse(ppb);
udelay(1000); /* Allow devices to settle down */
ppa_disconnect(dev);
udelay(1000); /* Another delay to allow devices to settle */
if (retv)
return -EIO;
return device_check(dev);
}
static inline int ppa_send_command(struct scsi_cmnd *cmd)
{
ppa_struct *dev = ppa_dev(cmd->device->host);
int k;
w_ctr(dev->base, 0x0c);
for (k = 0; k < cmd->cmd_len; k++)
if (!ppa_out(dev, &cmd->cmnd[k], 1))
return 0;
return 1;
}
/*
* The bulk flag enables some optimisations in the data transfer loops,
* it should be true for any command that transfers data in integral
* numbers of sectors.
*
* The driver appears to remain stable if we speed up the parallel port
* i/o in this function, but not elsewhere.
*/
static int ppa_completion(struct scsi_cmnd *cmd)
{
/* Return codes:
* -1 Error
* 0 Told to schedule
* 1 Finished data transfer
*/
ppa_struct *dev = ppa_dev(cmd->device->host);
unsigned short ppb = dev->base;
unsigned long start_jiffies = jiffies;
unsigned char r, v;
int fast, bulk, status;
v = cmd->cmnd[0];
bulk = ((v == READ_6) ||
(v == READ_10) || (v == WRITE_6) || (v == WRITE_10));
/*
* We only get here if the drive is ready to comunicate,
* hence no need for a full ppa_wait.
*/
r = (r_str(ppb) & 0xf0);
while (r != (unsigned char) 0xf0) {
/*
* If we have been running for more than a full timer tick
* then take a rest.
*/
if (time_after(jiffies, start_jiffies + 1))
return 0;
if ((cmd->SCp.this_residual <= 0)) {
ppa_fail(dev, DID_ERROR);
return -1; /* ERROR_RETURN */
}
/* On some hardware we have SCSI disconnected (6th bit low)
* for about 100usecs. It is too expensive to wait a
* tick on every loop so we busy wait for no more than
* 500usecs to give the drive a chance first. We do not
* change things for "normal" hardware since generally
* the 6th bit is always high.
* This makes the CPU load higher on some hardware
* but otherwise we can not get more than 50K/secs
* on this problem hardware.
*/
if ((r & 0xc0) != 0xc0) {
/* Wait for reconnection should be no more than
* jiffy/2 = 5ms = 5000 loops
*/
unsigned long k = dev->recon_tmo;
for (; k && ((r = (r_str(ppb) & 0xf0)) & 0xc0) != 0xc0;
k--)
udelay(1);
if (!k)
return 0;
}
/* determine if we should use burst I/O */
fast = (bulk && (cmd->SCp.this_residual >= PPA_BURST_SIZE))
? PPA_BURST_SIZE : 1;
if (r == (unsigned char) 0xc0)
status = ppa_out(dev, cmd->SCp.ptr, fast);
else
status = ppa_in(dev, cmd->SCp.ptr, fast);
cmd->SCp.ptr += fast;
cmd->SCp.this_residual -= fast;
if (!status) {
ppa_fail(dev, DID_BUS_BUSY);
return -1; /* ERROR_RETURN */
}
if (cmd->SCp.buffer && !cmd->SCp.this_residual) {
/* if scatter/gather, advance to the next segment */
if (cmd->SCp.buffers_residual--) {
cmd->SCp.buffer = sg_next(cmd->SCp.buffer);
cmd->SCp.this_residual =
cmd->SCp.buffer->length;
cmd->SCp.ptr = sg_virt(cmd->SCp.buffer);
}
}
/* Now check to see if the drive is ready to comunicate */
r = (r_str(ppb) & 0xf0);
/* If not, drop back down to the scheduler and wait a timer tick */
if (!(r & 0x80))
return 0;
}
return 1; /* FINISH_RETURN */
}
/*
* Since the PPA itself doesn't generate interrupts, we use
* the scheduler's task queue to generate a stream of call-backs and
* complete the request when the drive is ready.
*/
static void ppa_interrupt(struct work_struct *work)
{
ppa_struct *dev = container_of(work, ppa_struct, ppa_tq.work);
struct scsi_cmnd *cmd = dev->cur_cmd;
if (!cmd) {
printk(KERN_ERR "PPA: bug in ppa_interrupt\n");
return;
}
if (ppa_engine(dev, cmd)) {
schedule_delayed_work(&dev->ppa_tq, 1);
return;
}
/* Command must of completed hence it is safe to let go... */
#if PPA_DEBUG > 0
switch ((cmd->result >> 16) & 0xff) {
case DID_OK:
break;
case DID_NO_CONNECT:
printk(KERN_DEBUG "ppa: no device at SCSI ID %i\n", cmd->device->target);
break;
case DID_BUS_BUSY:
printk(KERN_DEBUG "ppa: BUS BUSY - EPP timeout detected\n");
break;
case DID_TIME_OUT:
printk(KERN_DEBUG "ppa: unknown timeout\n");
break;
case DID_ABORT:
printk(KERN_DEBUG "ppa: told to abort\n");
break;
case DID_PARITY:
printk(KERN_DEBUG "ppa: parity error (???)\n");
break;
case DID_ERROR:
printk(KERN_DEBUG "ppa: internal driver error\n");
break;
case DID_RESET:
printk(KERN_DEBUG "ppa: told to reset device\n");
break;
case DID_BAD_INTR:
printk(KERN_WARNING "ppa: bad interrupt (???)\n");
break;
default:
printk(KERN_WARNING "ppa: bad return code (%02x)\n",
(cmd->result >> 16) & 0xff);
}
#endif
if (cmd->SCp.phase > 1)
ppa_disconnect(dev);
ppa_pb_dismiss(dev);
dev->cur_cmd = NULL;
cmd->scsi_done(cmd);
}
static int ppa_engine(ppa_struct *dev, struct scsi_cmnd *cmd)
{
unsigned short ppb = dev->base;
unsigned char l = 0, h = 0;
int retv;
/* First check for any errors that may of occurred
* Here we check for internal errors
*/
if (dev->failed)
return 0;
switch (cmd->SCp.phase) {
case 0: /* Phase 0 - Waiting for parport */
if (time_after(jiffies, dev->jstart + HZ)) {
/*
* We waited more than a second
* for parport to call us
*/
ppa_fail(dev, DID_BUS_BUSY);
return 0;
}
return 1; /* wait until ppa_wakeup claims parport */
case 1: /* Phase 1 - Connected */
{ /* Perform a sanity check for cable unplugged */
int retv = 2; /* Failed */
ppa_connect(dev, CONNECT_EPP_MAYBE);
w_ctr(ppb, 0xe);
if ((r_str(ppb) & 0x08) == 0x08)
retv--;
w_ctr(ppb, 0xc);
if ((r_str(ppb) & 0x08) == 0x00)
retv--;
if (retv) {
if (time_after(jiffies, dev->jstart + (1 * HZ))) {
printk(KERN_ERR "ppa: Parallel port cable is unplugged.\n");
ppa_fail(dev, DID_BUS_BUSY);
return 0;
} else {
ppa_disconnect(dev);
return 1; /* Try again in a jiffy */
}
}
cmd->SCp.phase++;
}
/* fall through */
case 2: /* Phase 2 - We are now talking to the scsi bus */
if (!ppa_select(dev, scmd_id(cmd))) {
ppa_fail(dev, DID_NO_CONNECT);
return 0;
}
cmd->SCp.phase++;
/* fall through */
case 3: /* Phase 3 - Ready to accept a command */
w_ctr(ppb, 0x0c);
if (!(r_str(ppb) & 0x80))
return 1;
if (!ppa_send_command(cmd))
return 0;
cmd->SCp.phase++;
/* fall through */
case 4: /* Phase 4 - Setup scatter/gather buffers */
if (scsi_bufflen(cmd)) {
cmd->SCp.buffer = scsi_sglist(cmd);
cmd->SCp.this_residual = cmd->SCp.buffer->length;
cmd->SCp.ptr = sg_virt(cmd->SCp.buffer);
} else {
cmd->SCp.buffer = NULL;
cmd->SCp.this_residual = 0;
cmd->SCp.ptr = NULL;
}
cmd->SCp.buffers_residual = scsi_sg_count(cmd) - 1;
cmd->SCp.phase++;
/* fall through */
case 5: /* Phase 5 - Data transfer stage */
w_ctr(ppb, 0x0c);
if (!(r_str(ppb) & 0x80))
return 1;
retv = ppa_completion(cmd);
if (retv == -1)
return 0;
if (retv == 0)
return 1;
cmd->SCp.phase++;
/* fall through */
case 6: /* Phase 6 - Read status/message */
cmd->result = DID_OK << 16;
/* Check for data overrun */
if (ppa_wait(dev) != (unsigned char) 0xf0) {
ppa_fail(dev, DID_ERROR);
return 0;
}
if (ppa_in(dev, &l, 1)) { /* read status byte */
/* Check for optional message byte */
if (ppa_wait(dev) == (unsigned char) 0xf0)
ppa_in(dev, &h, 1);
cmd->result =
(DID_OK << 16) + (h << 8) + (l & STATUS_MASK);
}
return 0; /* Finished */
break;
default:
printk(KERN_ERR "ppa: Invalid scsi phase\n");
}
return 0;
}
static int ppa_queuecommand_lck(struct scsi_cmnd *cmd,
void (*done) (struct scsi_cmnd *))
{
ppa_struct *dev = ppa_dev(cmd->device->host);
if (dev->cur_cmd) {
printk(KERN_ERR "PPA: bug in ppa_queuecommand\n");
return 0;
}
dev->failed = 0;
dev->jstart = jiffies;
dev->cur_cmd = cmd;
cmd->scsi_done = done;
cmd->result = DID_ERROR << 16; /* default return code */
cmd->SCp.phase = 0; /* bus free */
schedule_delayed_work(&dev->ppa_tq, 0);
ppa_pb_claim(dev);
return 0;
}
static DEF_SCSI_QCMD(ppa_queuecommand)
/*
* Apparently the disk->capacity attribute is off by 1 sector
* for all disk drives. We add the one here, but it should really
* be done in sd.c. Even if it gets fixed there, this will still
* work.
*/
static int ppa_biosparam(struct scsi_device *sdev, struct block_device *dev,
sector_t capacity, int ip[])
{
ip[0] = 0x40;
ip[1] = 0x20;
ip[2] = ((unsigned long) capacity + 1) / (ip[0] * ip[1]);
if (ip[2] > 1024) {
ip[0] = 0xff;
ip[1] = 0x3f;
ip[2] = ((unsigned long) capacity + 1) / (ip[0] * ip[1]);
if (ip[2] > 1023)
ip[2] = 1023;
}
return 0;
}
static int ppa_abort(struct scsi_cmnd *cmd)
{
ppa_struct *dev = ppa_dev(cmd->device->host);
/*
* There is no method for aborting commands since Iomega
* have tied the SCSI_MESSAGE line high in the interface
*/
switch (cmd->SCp.phase) {
case 0: /* Do not have access to parport */
case 1: /* Have not connected to interface */
dev->cur_cmd = NULL; /* Forget the problem */
return SUCCESS;
break;
default: /* SCSI command sent, can not abort */
return FAILED;
break;
}
}
static void ppa_reset_pulse(unsigned int base)
{
w_dtr(base, 0x40);
w_ctr(base, 0x8);
udelay(30);
w_ctr(base, 0xc);
}
static int ppa_reset(struct scsi_cmnd *cmd)
{
ppa_struct *dev = ppa_dev(cmd->device->host);
if (cmd->SCp.phase)
ppa_disconnect(dev);
dev->cur_cmd = NULL; /* Forget the problem */
ppa_connect(dev, CONNECT_NORMAL);
ppa_reset_pulse(dev->base);
mdelay(1); /* device settle delay */
ppa_disconnect(dev);
mdelay(1); /* device settle delay */
return SUCCESS;
}
static int device_check(ppa_struct *dev)
{
/* This routine looks for a device and then attempts to use EPP
to send a command. If all goes as planned then EPP is available. */
static u8 cmd[6] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
int loop, old_mode, status, k, ppb = dev->base;
unsigned char l;
old_mode = dev->mode;
for (loop = 0; loop < 8; loop++) {
/* Attempt to use EPP for Test Unit Ready */
if ((ppb & 0x0007) == 0x0000)
dev->mode = PPA_EPP_32;
second_pass:
ppa_connect(dev, CONNECT_EPP_MAYBE);
/* Select SCSI device */
if (!ppa_select(dev, loop)) {
ppa_disconnect(dev);
continue;
}
printk(KERN_INFO "ppa: Found device at ID %i, Attempting to use %s\n",
loop, PPA_MODE_STRING[dev->mode]);
/* Send SCSI command */
status = 1;
w_ctr(ppb, 0x0c);
for (l = 0; (l < 6) && (status); l++)
status = ppa_out(dev, cmd, 1);
if (!status) {
ppa_disconnect(dev);
ppa_connect(dev, CONNECT_EPP_MAYBE);
w_dtr(ppb, 0x40);
w_ctr(ppb, 0x08);
udelay(30);
w_ctr(ppb, 0x0c);
udelay(1000);
ppa_disconnect(dev);
udelay(1000);
if (dev->mode == PPA_EPP_32) {
dev->mode = old_mode;
goto second_pass;
}
return -EIO;
}
w_ctr(ppb, 0x0c);
k = 1000000; /* 1 Second */
do {
l = r_str(ppb);
k--;
udelay(1);
} while (!(l & 0x80) && (k));
l &= 0xf0;
if (l != 0xf0) {
ppa_disconnect(dev);
ppa_connect(dev, CONNECT_EPP_MAYBE);
ppa_reset_pulse(ppb);
udelay(1000);
ppa_disconnect(dev);
udelay(1000);
if (dev->mode == PPA_EPP_32) {
dev->mode = old_mode;
goto second_pass;
}
return -EIO;
}
ppa_disconnect(dev);
printk(KERN_INFO "ppa: Communication established with ID %i using %s\n",
loop, PPA_MODE_STRING[dev->mode]);
ppa_connect(dev, CONNECT_EPP_MAYBE);
ppa_reset_pulse(ppb);
udelay(1000);
ppa_disconnect(dev);
udelay(1000);
return 0;
}
return -ENODEV;
}
static int ppa_adjust_queue(struct scsi_device *device)
{
blk_queue_bounce_limit(device->request_queue, BLK_BOUNCE_HIGH);
return 0;
}
static struct scsi_host_template ppa_template = {
.module = THIS_MODULE,
.proc_name = "ppa",
.show_info = ppa_show_info,
.write_info = ppa_write_info,
.name = "Iomega VPI0 (ppa) interface",
.queuecommand = ppa_queuecommand,
.eh_abort_handler = ppa_abort,
.eh_host_reset_handler = ppa_reset,
.bios_param = ppa_biosparam,
.this_id = -1,
.sg_tablesize = SG_ALL,
.can_queue = 1,
.slave_alloc = ppa_adjust_queue,
};
/***************************************************************************
* Parallel port probing routines *
***************************************************************************/
static LIST_HEAD(ppa_hosts);
/*
* Finds the first available device number that can be alloted to the
* new ppa device and returns the address of the previous node so that
* we can add to the tail and have a list in the ascending order.
*/
static inline ppa_struct *find_parent(void)
{
ppa_struct *dev, *par = NULL;
unsigned int cnt = 0;
if (list_empty(&ppa_hosts))
return NULL;
list_for_each_entry(dev, &ppa_hosts, list) {
if (dev->dev_no != cnt)
return par;
cnt++;
par = dev;
}
return par;
}
static int __ppa_attach(struct parport *pb)
{
struct Scsi_Host *host;
DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waiting);
DEFINE_WAIT(wait);
ppa_struct *dev, *temp;
int ports;
int modes, ppb, ppb_hi;
int err = -ENOMEM;
struct pardev_cb ppa_cb;
dev = kzalloc(sizeof(ppa_struct), GFP_KERNEL);
if (!dev)
return -ENOMEM;
dev->base = -1;
dev->mode = PPA_AUTODETECT;
dev->recon_tmo = PPA_RECON_TMO;
init_waitqueue_head(&waiting);
temp = find_parent();
if (temp)
dev->dev_no = temp->dev_no + 1;
memset(&ppa_cb, 0, sizeof(ppa_cb));
ppa_cb.private = dev;
ppa_cb.wakeup = ppa_wakeup;
dev->dev = parport_register_dev_model(pb, "ppa", &ppa_cb, dev->dev_no);
if (!dev->dev)
goto out;
/* Claim the bus so it remembers what we do to the control
* registers. [ CTR and ECP ]
*/
err = -EBUSY;
dev->waiting = &waiting;
prepare_to_wait(&waiting, &wait, TASK_UNINTERRUPTIBLE);
if (ppa_pb_claim(dev))
schedule_timeout(3 * HZ);
if (dev->wanted) {
printk(KERN_ERR "ppa%d: failed to claim parport because "
"a pardevice is owning the port for too long "
"time!\n", pb->number);
ppa_pb_dismiss(dev);
dev->waiting = NULL;
finish_wait(&waiting, &wait);
goto out1;
}
dev->waiting = NULL;
finish_wait(&waiting, &wait);
ppb = dev->base = dev->dev->port->base;
ppb_hi = dev->dev->port->base_hi;
w_ctr(ppb, 0x0c);
modes = dev->dev->port->modes;
/* Mode detection works up the chain of speed
* This avoids a nasty if-then-else-if-... tree
*/
dev->mode = PPA_NIBBLE;
if (modes & PARPORT_MODE_TRISTATE)
dev->mode = PPA_PS2;
if (modes & PARPORT_MODE_ECP) {
w_ecr(ppb_hi, 0x20);
dev->mode = PPA_PS2;
}
if ((modes & PARPORT_MODE_EPP) && (modes & PARPORT_MODE_ECP))
w_ecr(ppb_hi, 0x80);
/* Done configuration */
err = ppa_init(dev);
ppa_pb_release(dev);
if (err)
goto out1;
/* now the glue ... */
if (dev->mode == PPA_NIBBLE || dev->mode == PPA_PS2)
ports = 3;
else
ports = 8;
INIT_DELAYED_WORK(&dev->ppa_tq, ppa_interrupt);
err = -ENOMEM;
host = scsi_host_alloc(&ppa_template, sizeof(ppa_struct *));
if (!host)
goto out1;
host->io_port = pb->base;
host->n_io_port = ports;
host->dma_channel = -1;
host->unique_id = pb->number;
*(ppa_struct **)&host->hostdata = dev;
dev->host = host;
list_add_tail(&dev->list, &ppa_hosts);
err = scsi_add_host(host, NULL);
if (err)
goto out2;
scsi_scan_host(host);
return 0;
out2:
list_del_init(&dev->list);
scsi_host_put(host);
out1:
parport_unregister_device(dev->dev);
out:
kfree(dev);
return err;
}
static void ppa_attach(struct parport *pb)
{
__ppa_attach(pb);
}
static void ppa_detach(struct parport *pb)
{
ppa_struct *dev;
list_for_each_entry(dev, &ppa_hosts, list) {
if (dev->dev->port == pb) {
list_del_init(&dev->list);
scsi_remove_host(dev->host);
scsi_host_put(dev->host);
parport_unregister_device(dev->dev);
kfree(dev);
break;
}
}
}
static struct parport_driver ppa_driver = {
.name = "ppa",
.match_port = ppa_attach,
.detach = ppa_detach,
.devmodel = true,
};
static int __init ppa_driver_init(void)
{
printk(KERN_INFO "ppa: Version %s\n", PPA_VERSION);
return parport_register_driver(&ppa_driver);
}
static void __exit ppa_driver_exit(void)
{
parport_unregister_driver(&ppa_driver);
}
module_init(ppa_driver_init);
module_exit(ppa_driver_exit);
MODULE_LICENSE("GPL");