linux/drivers/net/wan/z85230.c

1794 lines
39 KiB
C
Raw Normal View History

/*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
* (c) Copyright 1998 Alan Cox <alan@lxorguk.ukuu.org.uk>
* (c) Copyright 2000, 2001 Red Hat Inc
*
* Development of this driver was funded by Equiinet Ltd
* http://www.equiinet.com
*
* ChangeLog:
*
* Asynchronous mode dropped for 2.2. For 2.5 we will attempt the
* unification of all the Z85x30 asynchronous drivers for real.
*
* DMA now uses get_free_page as kmalloc buffers may span a 64K
* boundary.
*
* Modified for SMP safety and SMP locking by Alan Cox
* <alan@lxorguk.ukuu.org.uk>
*
* Performance
*
* Z85230:
* Non DMA you want a 486DX50 or better to do 64Kbits. 9600 baud
* X.25 is not unrealistic on all machines. DMA mode can in theory
* handle T1/E1 quite nicely. In practice the limit seems to be about
* 512Kbit->1Mbit depending on motherboard.
*
* Z85C30:
* 64K will take DMA, 9600 baud X.25 should be ok.
*
* Z8530:
* Synchronous mode without DMA is unlikely to pass about 2400 baud.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/net.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/delay.h>
#include <linux/hdlc.h>
#include <linux/ioport.h>
#include <linux/init.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/gfp.h>
#include <asm/dma.h>
#include <asm/io.h>
#define RT_LOCK
#define RT_UNLOCK
#include <linux/spinlock.h>
#include "z85230.h"
/**
* z8530_read_port - Architecture specific interface function
* @p: port to read
*
* Provided port access methods. The Comtrol SV11 requires no delays
* between accesses and uses PC I/O. Some drivers may need a 5uS delay
*
* In the longer term this should become an architecture specific
* section so that this can become a generic driver interface for all
* platforms. For now we only handle PC I/O ports with or without the
* dread 5uS sanity delay.
*
* The caller must hold sufficient locks to avoid violating the horrible
* 5uS delay rule.
*/
static inline int z8530_read_port(unsigned long p)
{
u8 r=inb(Z8530_PORT_OF(p));
if(p&Z8530_PORT_SLEEP) /* gcc should figure this out efficiently ! */
udelay(5);
return r;
}
/**
* z8530_write_port - Architecture specific interface function
* @p: port to write
* @d: value to write
*
* Write a value to a port with delays if need be. Note that the
* caller must hold locks to avoid read/writes from other contexts
* violating the 5uS rule
*
* In the longer term this should become an architecture specific
* section so that this can become a generic driver interface for all
* platforms. For now we only handle PC I/O ports with or without the
* dread 5uS sanity delay.
*/
static inline void z8530_write_port(unsigned long p, u8 d)
{
outb(d,Z8530_PORT_OF(p));
if(p&Z8530_PORT_SLEEP)
udelay(5);
}
static void z8530_rx_done(struct z8530_channel *c);
static void z8530_tx_done(struct z8530_channel *c);
/**
* read_zsreg - Read a register from a Z85230
* @c: Z8530 channel to read from (2 per chip)
* @reg: Register to read
* FIXME: Use a spinlock.
*
* Most of the Z8530 registers are indexed off the control registers.
* A read is done by writing to the control register and reading the
* register back. The caller must hold the lock
*/
static inline u8 read_zsreg(struct z8530_channel *c, u8 reg)
{
if(reg)
z8530_write_port(c->ctrlio, reg);
return z8530_read_port(c->ctrlio);
}
/**
* read_zsdata - Read the data port of a Z8530 channel
* @c: The Z8530 channel to read the data port from
*
* The data port provides fast access to some things. We still
* have all the 5uS delays to worry about.
*/
static inline u8 read_zsdata(struct z8530_channel *c)
{
u8 r;
r=z8530_read_port(c->dataio);
return r;
}
/**
* write_zsreg - Write to a Z8530 channel register
* @c: The Z8530 channel
* @reg: Register number
* @val: Value to write
*
* Write a value to an indexed register. The caller must hold the lock
* to honour the irritating delay rules. We know about register 0
* being fast to access.
*
* Assumes c->lock is held.
*/
static inline void write_zsreg(struct z8530_channel *c, u8 reg, u8 val)
{
if(reg)
z8530_write_port(c->ctrlio, reg);
z8530_write_port(c->ctrlio, val);
}
/**
* write_zsctrl - Write to a Z8530 control register
* @c: The Z8530 channel
* @val: Value to write
*
* Write directly to the control register on the Z8530
*/
static inline void write_zsctrl(struct z8530_channel *c, u8 val)
{
z8530_write_port(c->ctrlio, val);
}
/**
* write_zsdata - Write to a Z8530 control register
* @c: The Z8530 channel
* @val: Value to write
*
* Write directly to the data register on the Z8530
*/
static inline void write_zsdata(struct z8530_channel *c, u8 val)
{
z8530_write_port(c->dataio, val);
}
/*
* Register loading parameters for a dead port
*/
u8 z8530_dead_port[]=
{
255
};
EXPORT_SYMBOL(z8530_dead_port);
/*
* Register loading parameters for currently supported circuit types
*/
/*
* Data clocked by telco end. This is the correct data for the UK
* "kilostream" service, and most other similar services.
*/
u8 z8530_hdlc_kilostream[]=
{
4, SYNC_ENAB|SDLC|X1CLK,
2, 0, /* No vector */
1, 0,
3, ENT_HM|RxCRC_ENAB|Rx8,
5, TxCRC_ENAB|RTS|TxENAB|Tx8|DTR,
9, 0, /* Disable interrupts */
6, 0xFF,
7, FLAG,
10, ABUNDER|NRZ|CRCPS,/*MARKIDLE ??*/
11, TCTRxCP,
14, DISDPLL,
15, DCDIE|SYNCIE|CTSIE|TxUIE|BRKIE,
1, EXT_INT_ENAB|TxINT_ENAB|INT_ALL_Rx,
9, NV|MIE|NORESET,
255
};
EXPORT_SYMBOL(z8530_hdlc_kilostream);
/*
* As above but for enhanced chips.
*/
u8 z8530_hdlc_kilostream_85230[]=
{
4, SYNC_ENAB|SDLC|X1CLK,
2, 0, /* No vector */
1, 0,
3, ENT_HM|RxCRC_ENAB|Rx8,
5, TxCRC_ENAB|RTS|TxENAB|Tx8|DTR,
9, 0, /* Disable interrupts */
6, 0xFF,
7, FLAG,
10, ABUNDER|NRZ|CRCPS, /* MARKIDLE?? */
11, TCTRxCP,
14, DISDPLL,
15, DCDIE|SYNCIE|CTSIE|TxUIE|BRKIE,
1, EXT_INT_ENAB|TxINT_ENAB|INT_ALL_Rx,
9, NV|MIE|NORESET,
23, 3, /* Extended mode AUTO TX and EOM*/
255
};
EXPORT_SYMBOL(z8530_hdlc_kilostream_85230);
/**
* z8530_flush_fifo - Flush on chip RX FIFO
* @c: Channel to flush
*
* Flush the receive FIFO. There is no specific option for this, we
* blindly read bytes and discard them. Reading when there is no data
* is harmless. The 8530 has a 4 byte FIFO, the 85230 has 8 bytes.
*
* All locking is handled for the caller. On return data may still be
* present if it arrived during the flush.
*/
static void z8530_flush_fifo(struct z8530_channel *c)
{
read_zsreg(c, R1);
read_zsreg(c, R1);
read_zsreg(c, R1);
read_zsreg(c, R1);
if(c->dev->type==Z85230)
{
read_zsreg(c, R1);
read_zsreg(c, R1);
read_zsreg(c, R1);
read_zsreg(c, R1);
}
}
/**
* z8530_rtsdtr - Control the outgoing DTS/RTS line
* @c: The Z8530 channel to control;
* @set: 1 to set, 0 to clear
*
* Sets or clears DTR/RTS on the requested line. All locking is handled
* by the caller. For now we assume all boards use the actual RTS/DTR
* on the chip. Apparently one or two don't. We'll scream about them
* later.
*/
static void z8530_rtsdtr(struct z8530_channel *c, int set)
{
if (set)
c->regs[5] |= (RTS | DTR);
else
c->regs[5] &= ~(RTS | DTR);
write_zsreg(c, R5, c->regs[5]);
}
/**
* z8530_rx - Handle a PIO receive event
* @c: Z8530 channel to process
*
* Receive handler for receiving in PIO mode. This is much like the
* async one but not quite the same or as complex
*
* Note: Its intended that this handler can easily be separated from
* the main code to run realtime. That'll be needed for some machines
* (eg to ever clock 64kbits on a sparc ;)).
*
* The RT_LOCK macros don't do anything now. Keep the code covered
* by them as short as possible in all circumstances - clocks cost
* baud. The interrupt handler is assumed to be atomic w.r.t. to
* other code - this is true in the RT case too.
*
* We only cover the sync cases for this. If you want 2Mbit async
* do it yourself but consider medical assistance first. This non DMA
* synchronous mode is portable code. The DMA mode assumes PCI like
* ISA DMA
*
* Called with the device lock held
*/
static void z8530_rx(struct z8530_channel *c)
{
u8 ch,stat;
while(1)
{
/* FIFO empty ? */
if(!(read_zsreg(c, R0)&1))
break;
ch=read_zsdata(c);
stat=read_zsreg(c, R1);
/*
* Overrun ?
*/
if(c->count < c->max)
{
*c->dptr++=ch;
c->count++;
}
if(stat&END_FR)
{
/*
* Error ?
*/
if(stat&(Rx_OVR|CRC_ERR))
{
/* Rewind the buffer and return */
if(c->skb)
c->dptr=c->skb->data;
c->count=0;
if(stat&Rx_OVR)
{
pr_warn("%s: overrun\n", c->dev->name);
c->rx_overrun++;
}
if(stat&CRC_ERR)
{
c->rx_crc_err++;
/* printk("crc error\n"); */
}
/* Shove the frame upstream */
}
else
{
/*
* Drop the lock for RX processing, or
* there are deadlocks
*/
z8530_rx_done(c);
write_zsctrl(c, RES_Rx_CRC);
}
}
}
/*
* Clear irq
*/
write_zsctrl(c, ERR_RES);
write_zsctrl(c, RES_H_IUS);
}
/**
* z8530_tx - Handle a PIO transmit event
* @c: Z8530 channel to process
*
* Z8530 transmit interrupt handler for the PIO mode. The basic
* idea is to attempt to keep the FIFO fed. We fill as many bytes
* in as possible, its quite possible that we won't keep up with the
* data rate otherwise.
*/
static void z8530_tx(struct z8530_channel *c)
{
while(c->txcount) {
/* FIFO full ? */
if(!(read_zsreg(c, R0)&4))
return;
c->txcount--;
/*
* Shovel out the byte
*/
write_zsreg(c, R8, *c->tx_ptr++);
write_zsctrl(c, RES_H_IUS);
/* We are about to underflow */
if(c->txcount==0)
{
write_zsctrl(c, RES_EOM_L);
write_zsreg(c, R10, c->regs[10]&~ABUNDER);
}
}
/*
* End of frame TX - fire another one
*/
write_zsctrl(c, RES_Tx_P);
z8530_tx_done(c);
write_zsctrl(c, RES_H_IUS);
}
/**
* z8530_status - Handle a PIO status exception
* @chan: Z8530 channel to process
*
* A status event occurred in PIO synchronous mode. There are several
* reasons the chip will bother us here. A transmit underrun means we
* failed to feed the chip fast enough and just broke a packet. A DCD
* change is a line up or down.
*/
static void z8530_status(struct z8530_channel *chan)
{
u8 status, altered;
status = read_zsreg(chan, R0);
altered = chan->status ^ status;
chan->status = status;
if (status & TxEOM) {
/* printk("%s: Tx underrun.\n", chan->dev->name); */
chan->netdevice->stats.tx_fifo_errors++;
write_zsctrl(chan, ERR_RES);
z8530_tx_done(chan);
}
if (altered & chan->dcdcheck)
{
if (status & chan->dcdcheck) {
pr_info("%s: DCD raised\n", chan->dev->name);
write_zsreg(chan, R3, chan->regs[3] | RxENABLE);
if (chan->netdevice)
netif_carrier_on(chan->netdevice);
} else {
pr_info("%s: DCD lost\n", chan->dev->name);
write_zsreg(chan, R3, chan->regs[3] & ~RxENABLE);
z8530_flush_fifo(chan);
if (chan->netdevice)
netif_carrier_off(chan->netdevice);
}
}
write_zsctrl(chan, RES_EXT_INT);
write_zsctrl(chan, RES_H_IUS);
}
struct z8530_irqhandler z8530_sync = {
.rx = z8530_rx,
.tx = z8530_tx,
.status = z8530_status,
};
EXPORT_SYMBOL(z8530_sync);
/**
* z8530_dma_rx - Handle a DMA RX event
* @chan: Channel to handle
*
* Non bus mastering DMA interfaces for the Z8x30 devices. This
* is really pretty PC specific. The DMA mode means that most receive
* events are handled by the DMA hardware. We get a kick here only if
* a frame ended.
*/
static void z8530_dma_rx(struct z8530_channel *chan)
{
if(chan->rxdma_on)
{
/* Special condition check only */
u8 status;
read_zsreg(chan, R7);
read_zsreg(chan, R6);
status=read_zsreg(chan, R1);
if(status&END_FR)
{
z8530_rx_done(chan); /* Fire up the next one */
}
write_zsctrl(chan, ERR_RES);
write_zsctrl(chan, RES_H_IUS);
}
else
{
/* DMA is off right now, drain the slow way */
z8530_rx(chan);
}
}
/**
* z8530_dma_tx - Handle a DMA TX event
* @chan: The Z8530 channel to handle
*
* We have received an interrupt while doing DMA transmissions. It
* shouldn't happen. Scream loudly if it does.
*/
static void z8530_dma_tx(struct z8530_channel *chan)
{
if(!chan->dma_tx)
{
pr_warn("Hey who turned the DMA off?\n");
z8530_tx(chan);
return;
}
/* This shouldn't occur in DMA mode */
pr_err("DMA tx - bogus event!\n");
z8530_tx(chan);
}
/**
* z8530_dma_status - Handle a DMA status exception
* @chan: Z8530 channel to process
*
* A status event occurred on the Z8530. We receive these for two reasons
* when in DMA mode. Firstly if we finished a packet transfer we get one
* and kick the next packet out. Secondly we may see a DCD change.
*
*/
static void z8530_dma_status(struct z8530_channel *chan)
{
u8 status, altered;
status=read_zsreg(chan, R0);
altered=chan->status^status;
chan->status=status;
if(chan->dma_tx)
{
if(status&TxEOM)
{
unsigned long flags;
flags=claim_dma_lock();
disable_dma(chan->txdma);
clear_dma_ff(chan->txdma);
chan->txdma_on=0;
release_dma_lock(flags);
z8530_tx_done(chan);
}
}
if (altered & chan->dcdcheck)
{
if (status & chan->dcdcheck) {
pr_info("%s: DCD raised\n", chan->dev->name);
write_zsreg(chan, R3, chan->regs[3] | RxENABLE);
if (chan->netdevice)
netif_carrier_on(chan->netdevice);
} else {
pr_info("%s: DCD lost\n", chan->dev->name);
write_zsreg(chan, R3, chan->regs[3] & ~RxENABLE);
z8530_flush_fifo(chan);
if (chan->netdevice)
netif_carrier_off(chan->netdevice);
}
}
write_zsctrl(chan, RES_EXT_INT);
write_zsctrl(chan, RES_H_IUS);
}
static struct z8530_irqhandler z8530_dma_sync = {
.rx = z8530_dma_rx,
.tx = z8530_dma_tx,
.status = z8530_dma_status,
};
static struct z8530_irqhandler z8530_txdma_sync = {
.rx = z8530_rx,
.tx = z8530_dma_tx,
.status = z8530_dma_status,
};
/**
* z8530_rx_clear - Handle RX events from a stopped chip
* @c: Z8530 channel to shut up
*
* Receive interrupt vectors for a Z8530 that is in 'parked' mode.
* For machines with PCI Z85x30 cards, or level triggered interrupts
* (eg the MacII) we must clear the interrupt cause or die.
*/
static void z8530_rx_clear(struct z8530_channel *c)
{
/*
* Data and status bytes
*/
u8 stat;
read_zsdata(c);
stat=read_zsreg(c, R1);
if(stat&END_FR)
write_zsctrl(c, RES_Rx_CRC);
/*
* Clear irq
*/
write_zsctrl(c, ERR_RES);
write_zsctrl(c, RES_H_IUS);
}
/**
* z8530_tx_clear - Handle TX events from a stopped chip
* @c: Z8530 channel to shut up
*
* Transmit interrupt vectors for a Z8530 that is in 'parked' mode.
* For machines with PCI Z85x30 cards, or level triggered interrupts
* (eg the MacII) we must clear the interrupt cause or die.
*/
static void z8530_tx_clear(struct z8530_channel *c)
{
write_zsctrl(c, RES_Tx_P);
write_zsctrl(c, RES_H_IUS);
}
/**
* z8530_status_clear - Handle status events from a stopped chip
* @chan: Z8530 channel to shut up
*
* Status interrupt vectors for a Z8530 that is in 'parked' mode.
* For machines with PCI Z85x30 cards, or level triggered interrupts
* (eg the MacII) we must clear the interrupt cause or die.
*/
static void z8530_status_clear(struct z8530_channel *chan)
{
u8 status=read_zsreg(chan, R0);
if(status&TxEOM)
write_zsctrl(chan, ERR_RES);
write_zsctrl(chan, RES_EXT_INT);
write_zsctrl(chan, RES_H_IUS);
}
struct z8530_irqhandler z8530_nop = {
.rx = z8530_rx_clear,
.tx = z8530_tx_clear,
.status = z8530_status_clear,
};
EXPORT_SYMBOL(z8530_nop);
/**
* z8530_interrupt - Handle an interrupt from a Z8530
* @irq: Interrupt number
* @dev_id: The Z8530 device that is interrupting.
*
* A Z85[2]30 device has stuck its hand in the air for attention.
* We scan both the channels on the chip for events and then call
* the channel specific call backs for each channel that has events.
* We have to use callback functions because the two channels can be
* in different modes.
*
* Locking is done for the handlers. Note that locking is done
* at the chip level (the 5uS delay issue is per chip not per
* channel). c->lock for both channels points to dev->lock
*/
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
irqreturn_t z8530_interrupt(int irq, void *dev_id)
{
struct z8530_dev *dev=dev_id;
u8 uninitialized_var(intr);
static volatile int locker=0;
int work=0;
struct z8530_irqhandler *irqs;
if(locker)
{
pr_err("IRQ re-enter\n");
return IRQ_NONE;
}
locker=1;
spin_lock(&dev->lock);
while(++work<5000)
{
intr = read_zsreg(&dev->chanA, R3);
if(!(intr & (CHARxIP|CHATxIP|CHAEXT|CHBRxIP|CHBTxIP|CHBEXT)))
break;
/* This holds the IRQ status. On the 8530 you must read it from chan
A even though it applies to the whole chip */
/* Now walk the chip and see what it is wanting - it may be
an IRQ for someone else remember */
irqs=dev->chanA.irqs;
if(intr & (CHARxIP|CHATxIP|CHAEXT))
{
if(intr&CHARxIP)
irqs->rx(&dev->chanA);
if(intr&CHATxIP)
irqs->tx(&dev->chanA);
if(intr&CHAEXT)
irqs->status(&dev->chanA);
}
irqs=dev->chanB.irqs;
if(intr & (CHBRxIP|CHBTxIP|CHBEXT))
{
if(intr&CHBRxIP)
irqs->rx(&dev->chanB);
if(intr&CHBTxIP)
irqs->tx(&dev->chanB);
if(intr&CHBEXT)
irqs->status(&dev->chanB);
}
}
spin_unlock(&dev->lock);
if(work==5000)
pr_err("%s: interrupt jammed - abort(0x%X)!\n",
dev->name, intr);
/* Ok all done */
locker=0;
return IRQ_HANDLED;
}
EXPORT_SYMBOL(z8530_interrupt);
static const u8 reg_init[16]=
{
0,0,0,0,
0,0,0,0,
0,0,0,0,
0x55,0,0,0
};
/**
* z8530_sync_open - Open a Z8530 channel for PIO
* @dev: The network interface we are using
* @c: The Z8530 channel to open in synchronous PIO mode
*
* Switch a Z8530 into synchronous mode without DMA assist. We
* raise the RTS/DTR and commence network operation.
*/
int z8530_sync_open(struct net_device *dev, struct z8530_channel *c)
{
unsigned long flags;
spin_lock_irqsave(c->lock, flags);
c->sync = 1;
c->mtu = dev->mtu+64;
c->count = 0;
c->skb = NULL;
c->skb2 = NULL;
c->irqs = &z8530_sync;
/* This loads the double buffer up */
z8530_rx_done(c); /* Load the frame ring */
z8530_rx_done(c); /* Load the backup frame */
z8530_rtsdtr(c,1);
c->dma_tx = 0;
c->regs[R1]|=TxINT_ENAB;
write_zsreg(c, R1, c->regs[R1]);
write_zsreg(c, R3, c->regs[R3]|RxENABLE);
spin_unlock_irqrestore(c->lock, flags);
return 0;
}
EXPORT_SYMBOL(z8530_sync_open);
/**
* z8530_sync_close - Close a PIO Z8530 channel
* @dev: Network device to close
* @c: Z8530 channel to disassociate and move to idle
*
* Close down a Z8530 interface and switch its interrupt handlers
* to discard future events.
*/
int z8530_sync_close(struct net_device *dev, struct z8530_channel *c)
{
u8 chk;
unsigned long flags;
spin_lock_irqsave(c->lock, flags);
c->irqs = &z8530_nop;
c->max = 0;
c->sync = 0;
chk=read_zsreg(c,R0);
write_zsreg(c, R3, c->regs[R3]);
z8530_rtsdtr(c,0);
spin_unlock_irqrestore(c->lock, flags);
return 0;
}
EXPORT_SYMBOL(z8530_sync_close);
/**
* z8530_sync_dma_open - Open a Z8530 for DMA I/O
* @dev: The network device to attach
* @c: The Z8530 channel to configure in sync DMA mode.
*
* Set up a Z85x30 device for synchronous DMA in both directions. Two
* ISA DMA channels must be available for this to work. We assume ISA
* DMA driven I/O and PC limits on access.
*/
int z8530_sync_dma_open(struct net_device *dev, struct z8530_channel *c)
{
unsigned long cflags, dflags;
c->sync = 1;
c->mtu = dev->mtu+64;
c->count = 0;
c->skb = NULL;
c->skb2 = NULL;
/*
* Load the DMA interfaces up
*/
c->rxdma_on = 0;
c->txdma_on = 0;
/*
* Allocate the DMA flip buffers. Limit by page size.
* Everyone runs 1500 mtu or less on wan links so this
* should be fine.
*/
if(c->mtu > PAGE_SIZE/2)
return -EMSGSIZE;
c->rx_buf[0]=(void *)get_zeroed_page(GFP_KERNEL|GFP_DMA);
if(c->rx_buf[0]==NULL)
return -ENOBUFS;
c->rx_buf[1]=c->rx_buf[0]+PAGE_SIZE/2;
c->tx_dma_buf[0]=(void *)get_zeroed_page(GFP_KERNEL|GFP_DMA);
if(c->tx_dma_buf[0]==NULL)
{
free_page((unsigned long)c->rx_buf[0]);
c->rx_buf[0]=NULL;
return -ENOBUFS;
}
c->tx_dma_buf[1]=c->tx_dma_buf[0]+PAGE_SIZE/2;
c->tx_dma_used=0;
c->dma_tx = 1;
c->dma_num=0;
c->dma_ready=1;
/*
* Enable DMA control mode
*/
spin_lock_irqsave(c->lock, cflags);
/*
* TX DMA via DIR/REQ
*/
c->regs[R14]|= DTRREQ;
write_zsreg(c, R14, c->regs[R14]);
c->regs[R1]&= ~TxINT_ENAB;
write_zsreg(c, R1, c->regs[R1]);
/*
* RX DMA via W/Req
*/
c->regs[R1]|= WT_FN_RDYFN;
c->regs[R1]|= WT_RDY_RT;
c->regs[R1]|= INT_ERR_Rx;
c->regs[R1]&= ~TxINT_ENAB;
write_zsreg(c, R1, c->regs[R1]);
c->regs[R1]|= WT_RDY_ENAB;
write_zsreg(c, R1, c->regs[R1]);
/*
* DMA interrupts
*/
/*
* Set up the DMA configuration
*/
dflags=claim_dma_lock();
disable_dma(c->rxdma);
clear_dma_ff(c->rxdma);
set_dma_mode(c->rxdma, DMA_MODE_READ|0x10);
set_dma_addr(c->rxdma, virt_to_bus(c->rx_buf[0]));
set_dma_count(c->rxdma, c->mtu);
enable_dma(c->rxdma);
disable_dma(c->txdma);
clear_dma_ff(c->txdma);
set_dma_mode(c->txdma, DMA_MODE_WRITE);
disable_dma(c->txdma);
release_dma_lock(dflags);
/*
* Select the DMA interrupt handlers
*/
c->rxdma_on = 1;
c->txdma_on = 1;
c->tx_dma_used = 1;
c->irqs = &z8530_dma_sync;
z8530_rtsdtr(c,1);
write_zsreg(c, R3, c->regs[R3]|RxENABLE);
spin_unlock_irqrestore(c->lock, cflags);
return 0;
}
EXPORT_SYMBOL(z8530_sync_dma_open);
/**
* z8530_sync_dma_close - Close down DMA I/O
* @dev: Network device to detach
* @c: Z8530 channel to move into discard mode
*
* Shut down a DMA mode synchronous interface. Halt the DMA, and
* free the buffers.
*/
int z8530_sync_dma_close(struct net_device *dev, struct z8530_channel *c)
{
u8 chk;
unsigned long flags;
c->irqs = &z8530_nop;
c->max = 0;
c->sync = 0;
/*
* Disable the PC DMA channels
*/
flags=claim_dma_lock();
disable_dma(c->rxdma);
clear_dma_ff(c->rxdma);
c->rxdma_on = 0;
disable_dma(c->txdma);
clear_dma_ff(c->txdma);
release_dma_lock(flags);
c->txdma_on = 0;
c->tx_dma_used = 0;
spin_lock_irqsave(c->lock, flags);
/*
* Disable DMA control mode
*/
c->regs[R1]&= ~WT_RDY_ENAB;
write_zsreg(c, R1, c->regs[R1]);
c->regs[R1]&= ~(WT_RDY_RT|WT_FN_RDYFN|INT_ERR_Rx);
c->regs[R1]|= INT_ALL_Rx;
write_zsreg(c, R1, c->regs[R1]);
c->regs[R14]&= ~DTRREQ;
write_zsreg(c, R14, c->regs[R14]);
if(c->rx_buf[0])
{
free_page((unsigned long)c->rx_buf[0]);
c->rx_buf[0]=NULL;
}
if(c->tx_dma_buf[0])
{
free_page((unsigned long)c->tx_dma_buf[0]);
c->tx_dma_buf[0]=NULL;
}
chk=read_zsreg(c,R0);
write_zsreg(c, R3, c->regs[R3]);
z8530_rtsdtr(c,0);
spin_unlock_irqrestore(c->lock, flags);
return 0;
}
EXPORT_SYMBOL(z8530_sync_dma_close);
/**
* z8530_sync_txdma_open - Open a Z8530 for TX driven DMA
* @dev: The network device to attach
* @c: The Z8530 channel to configure in sync DMA mode.
*
* Set up a Z85x30 device for synchronous DMA transmission. One
* ISA DMA channel must be available for this to work. The receive
* side is run in PIO mode, but then it has the bigger FIFO.
*/
int z8530_sync_txdma_open(struct net_device *dev, struct z8530_channel *c)
{
unsigned long cflags, dflags;
printk("Opening sync interface for TX-DMA\n");
c->sync = 1;
c->mtu = dev->mtu+64;
c->count = 0;
c->skb = NULL;
c->skb2 = NULL;
/*
* Allocate the DMA flip buffers. Limit by page size.
* Everyone runs 1500 mtu or less on wan links so this
* should be fine.
*/
if(c->mtu > PAGE_SIZE/2)
return -EMSGSIZE;
c->tx_dma_buf[0]=(void *)get_zeroed_page(GFP_KERNEL|GFP_DMA);
if(c->tx_dma_buf[0]==NULL)
return -ENOBUFS;
c->tx_dma_buf[1] = c->tx_dma_buf[0] + PAGE_SIZE/2;
spin_lock_irqsave(c->lock, cflags);
/*
* Load the PIO receive ring
*/
z8530_rx_done(c);
z8530_rx_done(c);
/*
* Load the DMA interfaces up
*/
c->rxdma_on = 0;
c->txdma_on = 0;
c->tx_dma_used=0;
c->dma_num=0;
c->dma_ready=1;
c->dma_tx = 1;
/*
* Enable DMA control mode
*/
/*
* TX DMA via DIR/REQ
*/
c->regs[R14]|= DTRREQ;
write_zsreg(c, R14, c->regs[R14]);
c->regs[R1]&= ~TxINT_ENAB;
write_zsreg(c, R1, c->regs[R1]);
/*
* Set up the DMA configuration
*/
dflags = claim_dma_lock();
disable_dma(c->txdma);
clear_dma_ff(c->txdma);
set_dma_mode(c->txdma, DMA_MODE_WRITE);
disable_dma(c->txdma);
release_dma_lock(dflags);
/*
* Select the DMA interrupt handlers
*/
c->rxdma_on = 0;
c->txdma_on = 1;
c->tx_dma_used = 1;
c->irqs = &z8530_txdma_sync;
z8530_rtsdtr(c,1);
write_zsreg(c, R3, c->regs[R3]|RxENABLE);
spin_unlock_irqrestore(c->lock, cflags);
return 0;
}
EXPORT_SYMBOL(z8530_sync_txdma_open);
/**
* z8530_sync_txdma_close - Close down a TX driven DMA channel
* @dev: Network device to detach
* @c: Z8530 channel to move into discard mode
*
* Shut down a DMA/PIO split mode synchronous interface. Halt the DMA,
* and free the buffers.
*/
int z8530_sync_txdma_close(struct net_device *dev, struct z8530_channel *c)
{
unsigned long dflags, cflags;
u8 chk;
spin_lock_irqsave(c->lock, cflags);
c->irqs = &z8530_nop;
c->max = 0;
c->sync = 0;
/*
* Disable the PC DMA channels
*/
dflags = claim_dma_lock();
disable_dma(c->txdma);
clear_dma_ff(c->txdma);
c->txdma_on = 0;
c->tx_dma_used = 0;
release_dma_lock(dflags);
/*
* Disable DMA control mode
*/
c->regs[R1]&= ~WT_RDY_ENAB;
write_zsreg(c, R1, c->regs[R1]);
c->regs[R1]&= ~(WT_RDY_RT|WT_FN_RDYFN|INT_ERR_Rx);
c->regs[R1]|= INT_ALL_Rx;
write_zsreg(c, R1, c->regs[R1]);
c->regs[R14]&= ~DTRREQ;
write_zsreg(c, R14, c->regs[R14]);
if(c->tx_dma_buf[0])
{
free_page((unsigned long)c->tx_dma_buf[0]);
c->tx_dma_buf[0]=NULL;
}
chk=read_zsreg(c,R0);
write_zsreg(c, R3, c->regs[R3]);
z8530_rtsdtr(c,0);
spin_unlock_irqrestore(c->lock, cflags);
return 0;
}
EXPORT_SYMBOL(z8530_sync_txdma_close);
/*
* Name strings for Z8530 chips. SGI claim to have a 130, Zilog deny
* it exists...
*/
static const char *z8530_type_name[]={
"Z8530",
"Z85C30",
"Z85230"
};
/**
* z8530_describe - Uniformly describe a Z8530 port
* @dev: Z8530 device to describe
* @mapping: string holding mapping type (eg "I/O" or "Mem")
* @io: the port value in question
*
* Describe a Z8530 in a standard format. We must pass the I/O as
* the port offset isn't predictable. The main reason for this function
* is to try and get a common format of report.
*/
void z8530_describe(struct z8530_dev *dev, char *mapping, unsigned long io)
{
pr_info("%s: %s found at %s 0x%lX, IRQ %d\n",
dev->name,
z8530_type_name[dev->type],
mapping,
Z8530_PORT_OF(io),
dev->irq);
}
EXPORT_SYMBOL(z8530_describe);
/*
* Locked operation part of the z8530 init code
*/
static inline int do_z8530_init(struct z8530_dev *dev)
{
/* NOP the interrupt handlers first - we might get a
floating IRQ transition when we reset the chip */
dev->chanA.irqs=&z8530_nop;
dev->chanB.irqs=&z8530_nop;
dev->chanA.dcdcheck=DCD;
dev->chanB.dcdcheck=DCD;
/* Reset the chip */
write_zsreg(&dev->chanA, R9, 0xC0);
udelay(200);
/* Now check its valid */
write_zsreg(&dev->chanA, R12, 0xAA);
if(read_zsreg(&dev->chanA, R12)!=0xAA)
return -ENODEV;
write_zsreg(&dev->chanA, R12, 0x55);
if(read_zsreg(&dev->chanA, R12)!=0x55)
return -ENODEV;
dev->type=Z8530;
/*
* See the application note.
*/
write_zsreg(&dev->chanA, R15, 0x01);
/*
* If we can set the low bit of R15 then
* the chip is enhanced.
*/
if(read_zsreg(&dev->chanA, R15)==0x01)
{
/* This C30 versus 230 detect is from Klaus Kudielka's dmascc */
/* Put a char in the fifo */
write_zsreg(&dev->chanA, R8, 0);
if(read_zsreg(&dev->chanA, R0)&Tx_BUF_EMP)
dev->type = Z85230; /* Has a FIFO */
else
dev->type = Z85C30; /* Z85C30, 1 byte FIFO */
}
/*
* The code assumes R7' and friends are
* off. Use write_zsext() for these and keep
* this bit clear.
*/
write_zsreg(&dev->chanA, R15, 0);
/*
* At this point it looks like the chip is behaving
*/
memcpy(dev->chanA.regs, reg_init, 16);
memcpy(dev->chanB.regs, reg_init ,16);
return 0;
}
/**
* z8530_init - Initialise a Z8530 device
* @dev: Z8530 device to initialise.
*
* Configure up a Z8530/Z85C30 or Z85230 chip. We check the device
* is present, identify the type and then program it to hopefully
* keep quite and behave. This matters a lot, a Z8530 in the wrong
* state will sometimes get into stupid modes generating 10Khz
* interrupt streams and the like.
*
* We set the interrupt handler up to discard any events, in case
* we get them during reset or setp.
*
* Return 0 for success, or a negative value indicating the problem
* in errno form.
*/
int z8530_init(struct z8530_dev *dev)
{
unsigned long flags;
int ret;
/* Set up the chip level lock */
spin_lock_init(&dev->lock);
dev->chanA.lock = &dev->lock;
dev->chanB.lock = &dev->lock;
spin_lock_irqsave(&dev->lock, flags);
ret = do_z8530_init(dev);
spin_unlock_irqrestore(&dev->lock, flags);
return ret;
}
EXPORT_SYMBOL(z8530_init);
/**
* z8530_shutdown - Shutdown a Z8530 device
* @dev: The Z8530 chip to shutdown
*
* We set the interrupt handlers to silence any interrupts. We then
* reset the chip and wait 100uS to be sure the reset completed. Just
* in case the caller then tries to do stuff.
*
* This is called without the lock held
*/
int z8530_shutdown(struct z8530_dev *dev)
{
unsigned long flags;
/* Reset the chip */
spin_lock_irqsave(&dev->lock, flags);
dev->chanA.irqs=&z8530_nop;
dev->chanB.irqs=&z8530_nop;
write_zsreg(&dev->chanA, R9, 0xC0);
/* We must lock the udelay, the chip is offlimits here */
udelay(100);
spin_unlock_irqrestore(&dev->lock, flags);
return 0;
}
EXPORT_SYMBOL(z8530_shutdown);
/**
* z8530_channel_load - Load channel data
* @c: Z8530 channel to configure
* @rtable: table of register, value pairs
* FIXME: ioctl to allow user uploaded tables
*
* Load a Z8530 channel up from the system data. We use +16 to
* indicate the "prime" registers. The value 255 terminates the
* table.
*/
int z8530_channel_load(struct z8530_channel *c, u8 *rtable)
{
unsigned long flags;
spin_lock_irqsave(c->lock, flags);
while(*rtable!=255)
{
int reg=*rtable++;
if(reg>0x0F)
write_zsreg(c, R15, c->regs[15]|1);
write_zsreg(c, reg&0x0F, *rtable);
if(reg>0x0F)
write_zsreg(c, R15, c->regs[15]&~1);
c->regs[reg]=*rtable++;
}
c->rx_function=z8530_null_rx;
c->skb=NULL;
c->tx_skb=NULL;
c->tx_next_skb=NULL;
c->mtu=1500;
c->max=0;
c->count=0;
c->status=read_zsreg(c, R0);
c->sync=1;
write_zsreg(c, R3, c->regs[R3]|RxENABLE);
spin_unlock_irqrestore(c->lock, flags);
return 0;
}
EXPORT_SYMBOL(z8530_channel_load);
/**
* z8530_tx_begin - Begin packet transmission
* @c: The Z8530 channel to kick
*
* This is the speed sensitive side of transmission. If we are called
* and no buffer is being transmitted we commence the next buffer. If
* nothing is queued we idle the sync.
*
* Note: We are handling this code path in the interrupt path, keep it
* fast or bad things will happen.
*
* Called with the lock held.
*/
static void z8530_tx_begin(struct z8530_channel *c)
{
unsigned long flags;
if(c->tx_skb)
return;
c->tx_skb=c->tx_next_skb;
c->tx_next_skb=NULL;
c->tx_ptr=c->tx_next_ptr;
if(c->tx_skb==NULL)
{
/* Idle on */
if(c->dma_tx)
{
flags=claim_dma_lock();
disable_dma(c->txdma);
/*
* Check if we crapped out.
*/
if (get_dma_residue(c->txdma))
{
c->netdevice->stats.tx_dropped++;
c->netdevice->stats.tx_fifo_errors++;
}
release_dma_lock(flags);
}
c->txcount=0;
}
else
{
c->txcount=c->tx_skb->len;
if(c->dma_tx)
{
/*
* FIXME. DMA is broken for the original 8530,
* on the older parts we need to set a flag and
* wait for a further TX interrupt to fire this
* stage off
*/
flags=claim_dma_lock();
disable_dma(c->txdma);
/*
* These two are needed by the 8530/85C30
* and must be issued when idling.
*/
if(c->dev->type!=Z85230)
{
write_zsctrl(c, RES_Tx_CRC);
write_zsctrl(c, RES_EOM_L);
}
write_zsreg(c, R10, c->regs[10]&~ABUNDER);
clear_dma_ff(c->txdma);
set_dma_addr(c->txdma, virt_to_bus(c->tx_ptr));
set_dma_count(c->txdma, c->txcount);
enable_dma(c->txdma);
release_dma_lock(flags);
write_zsctrl(c, RES_EOM_L);
write_zsreg(c, R5, c->regs[R5]|TxENAB);
}
else
{
/* ABUNDER off */
write_zsreg(c, R10, c->regs[10]);
write_zsctrl(c, RES_Tx_CRC);
while(c->txcount && (read_zsreg(c,R0)&Tx_BUF_EMP))
{
write_zsreg(c, R8, *c->tx_ptr++);
c->txcount--;
}
}
}
/*
* Since we emptied tx_skb we can ask for more
*/
netif_wake_queue(c->netdevice);
}
/**
* z8530_tx_done - TX complete callback
* @c: The channel that completed a transmit.
*
* This is called when we complete a packet send. We wake the queue,
* start the next packet going and then free the buffer of the existing
* packet. This code is fairly timing sensitive.
*
* Called with the register lock held.
*/
static void z8530_tx_done(struct z8530_channel *c)
{
struct sk_buff *skb;
/* Actually this can happen.*/
if (c->tx_skb == NULL)
return;
skb = c->tx_skb;
c->tx_skb = NULL;
z8530_tx_begin(c);
c->netdevice->stats.tx_packets++;
c->netdevice->stats.tx_bytes += skb->len;
dev_kfree_skb_irq(skb);
}
/**
* z8530_null_rx - Discard a packet
* @c: The channel the packet arrived on
* @skb: The buffer
*
* We point the receive handler at this function when idle. Instead
* of processing the frames we get to throw them away.
*/
void z8530_null_rx(struct z8530_channel *c, struct sk_buff *skb)
{
dev_kfree_skb_any(skb);
}
EXPORT_SYMBOL(z8530_null_rx);
/**
* z8530_rx_done - Receive completion callback
* @c: The channel that completed a receive
*
* A new packet is complete. Our goal here is to get back into receive
* mode as fast as possible. On the Z85230 we could change to using
* ESCC mode, but on the older chips we have no choice. We flip to the
* new buffer immediately in DMA mode so that the DMA of the next
* frame can occur while we are copying the previous buffer to an sk_buff
*
* Called with the lock held
*/
static void z8530_rx_done(struct z8530_channel *c)
{
struct sk_buff *skb;
int ct;
/*
* Is our receive engine in DMA mode
*/
if(c->rxdma_on)
{
/*
* Save the ready state and the buffer currently
* being used as the DMA target
*/
int ready=c->dma_ready;
unsigned char *rxb=c->rx_buf[c->dma_num];
unsigned long flags;
/*
* Complete this DMA. Necessary to find the length
*/
flags=claim_dma_lock();
disable_dma(c->rxdma);
clear_dma_ff(c->rxdma);
c->rxdma_on=0;
ct=c->mtu-get_dma_residue(c->rxdma);
if(ct<0)
ct=2; /* Shit happens.. */
c->dma_ready=0;
/*
* Normal case: the other slot is free, start the next DMA
* into it immediately.
*/
if(ready)
{
c->dma_num^=1;
set_dma_mode(c->rxdma, DMA_MODE_READ|0x10);
set_dma_addr(c->rxdma, virt_to_bus(c->rx_buf[c->dma_num]));
set_dma_count(c->rxdma, c->mtu);
c->rxdma_on = 1;
enable_dma(c->rxdma);
/* Stop any frames that we missed the head of
from passing */
write_zsreg(c, R0, RES_Rx_CRC);
}
else
/* Can't occur as we dont reenable the DMA irq until
after the flip is done */
netdev_warn(c->netdevice, "DMA flip overrun!\n");
release_dma_lock(flags);
/*
* Shove the old buffer into an sk_buff. We can't DMA
* directly into one on a PC - it might be above the 16Mb
* boundary. Optimisation - we could check to see if we
* can avoid the copy. Optimisation 2 - make the memcpy
* a copychecksum.
*/
skb = dev_alloc_skb(ct);
if (skb == NULL) {
c->netdevice->stats.rx_dropped++;
netdev_warn(c->netdevice, "Memory squeeze\n");
} else {
skb_put(skb, ct);
skb_copy_to_linear_data(skb, rxb, ct);
c->netdevice->stats.rx_packets++;
c->netdevice->stats.rx_bytes += ct;
}
c->dma_ready = 1;
} else {
RT_LOCK;
skb = c->skb;
/*
* The game we play for non DMA is similar. We want to
* get the controller set up for the next packet as fast
* as possible. We potentially only have one byte + the
* fifo length for this. Thus we want to flip to the new
* buffer and then mess around copying and allocating
* things. For the current case it doesn't matter but
* if you build a system where the sync irq isn't blocked
* by the kernel IRQ disable then you need only block the
* sync IRQ for the RT_LOCK area.
*
*/
ct=c->count;
c->skb = c->skb2;
c->count = 0;
c->max = c->mtu;
if (c->skb) {
c->dptr = c->skb->data;
c->max = c->mtu;
} else {
c->count = 0;
c->max = 0;
}
RT_UNLOCK;
c->skb2 = dev_alloc_skb(c->mtu);
if (c->skb2 == NULL)
netdev_warn(c->netdevice, "memory squeeze\n");
else
skb_put(c->skb2, c->mtu);
c->netdevice->stats.rx_packets++;
c->netdevice->stats.rx_bytes += ct;
}
/*
* If we received a frame we must now process it.
*/
if (skb) {
skb_trim(skb, ct);
c->rx_function(c, skb);
} else {
c->netdevice->stats.rx_dropped++;
netdev_err(c->netdevice, "Lost a frame\n");
}
}
/**
* spans_boundary - Check a packet can be ISA DMA'd
* @skb: The buffer to check
*
* Returns true if the buffer cross a DMA boundary on a PC. The poor
* thing can only DMA within a 64K block not across the edges of it.
*/
static inline int spans_boundary(struct sk_buff *skb)
{
unsigned long a=(unsigned long)skb->data;
a^=(a+skb->len);
if(a&0x00010000) /* If the 64K bit is different.. */
return 1;
return 0;
}
/**
* z8530_queue_xmit - Queue a packet
* @c: The channel to use
* @skb: The packet to kick down the channel
*
* Queue a packet for transmission. Because we have rather
* hard to hit interrupt latencies for the Z85230 per packet
* even in DMA mode we do the flip to DMA buffer if needed here
* not in the IRQ.
*
* Called from the network code. The lock is not held at this
* point.
*/
netdev_tx_t z8530_queue_xmit(struct z8530_channel *c, struct sk_buff *skb)
{
unsigned long flags;
netif_stop_queue(c->netdevice);
if(c->tx_next_skb)
return NETDEV_TX_BUSY;
/* PC SPECIFIC - DMA limits */
/*
* If we will DMA the transmit and its gone over the ISA bus
* limit, then copy to the flip buffer
*/
if(c->dma_tx && ((unsigned long)(virt_to_bus(skb->data+skb->len))>=16*1024*1024 || spans_boundary(skb)))
{
/*
* Send the flip buffer, and flip the flippy bit.
* We don't care which is used when just so long as
* we never use the same buffer twice in a row. Since
* only one buffer can be going out at a time the other
* has to be safe.
*/
c->tx_next_ptr=c->tx_dma_buf[c->tx_dma_used];
c->tx_dma_used^=1; /* Flip temp buffer */
skb_copy_from_linear_data(skb, c->tx_next_ptr, skb->len);
}
else
c->tx_next_ptr=skb->data;
RT_LOCK;
c->tx_next_skb=skb;
RT_UNLOCK;
spin_lock_irqsave(c->lock, flags);
z8530_tx_begin(c);
spin_unlock_irqrestore(c->lock, flags);
return NETDEV_TX_OK;
}
EXPORT_SYMBOL(z8530_queue_xmit);
/*
* Module support
*/
static const char banner[] __initconst =
KERN_INFO "Generic Z85C30/Z85230 interface driver v0.02\n";
static int __init z85230_init_driver(void)
{
printk(banner);
return 0;
}
module_init(z85230_init_driver);
static void __exit z85230_cleanup_driver(void)
{
}
module_exit(z85230_cleanup_driver);
MODULE_AUTHOR("Red Hat Inc.");
MODULE_DESCRIPTION("Z85x30 synchronous driver core");
MODULE_LICENSE("GPL");