if_cpm.html

vxworks相关论文

This routine is expected to clear the interrupt for the Ethernet interface
specified by <i>unit</i>.
<p>
<dt>
STATUS sysCpmEnetAddrGet (int unit, UINT8 * addr)
<dd>

The driver expects this routine to provide the 6-byte Ethernet
hardware address that will be used by <i>unit</i>.  This routine must copy
the 6-byte address to the space provided by <i>addr</i>.  This routine is
expected to return OK on success, or ERROR.
The driver calls this routine, once per unit, from the <b><i>cpmInit</i>(&nbsp;)</b> routine. 
</dl>

<p>

</blockquote><h4>SYSTEM RESOURCE USAGE</h4><blockquote><p>
This driver requires the following system resources:
<p>
&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;one&nbsp;mutual&nbsp;exclusion&nbsp;semaphore<br>
&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;one&nbsp;interrupt&nbsp;vector<br>
&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;0&nbsp;bytes&nbsp;in&nbsp;the&nbsp;initialized&nbsp;data&nbsp;section&nbsp;(data)<br>
&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;1272&nbsp;bytes&nbsp;in&nbsp;the&nbsp;uninitialized&nbsp;data&nbsp;section&nbsp;(BSS)
<p>
The data and BSS sections are quoted for the CPU32 architecture and may vary
for other architectures.  The code size (text) varies greatly between
architectures, and is therefore not quoted here.
<p>
If the driver allocates the memory shared with the Ethernet device unit,
it does so by calling the <b><i><a href="./cacheLib.html#cacheDmaMalloc">cacheDmaMalloc</a></i>(&nbsp;)</b> routine.  For the default case
of 32 transmit buffers, 32 receive buffers, and 16 loaner buffers, the total
size requested is 121,600 bytes.  If a non-cacheable memory region is provided
by the user, the size of this region should be this amount, unless the
user has specified a different number of transmit or receive BDs.
<p>
This driver can operate only if the shared memory region is non-cacheable,
or if the hardware implements bus snooping.  The driver cannot maintain
cache coherency for the device because the buffers are asynchronously
modified by both the driver and the device, and these fields may share the
same cache line.  Additionally, the chip's dual ported RAM must be
declared as non-cacheable memory where applicable.
<p>
</blockquote><h4>SEE ALSO</h4><blockquote><p>
<b><a href="./if_cpm.html#top">if_cpm</a></b>, <b><a href="./ifLib.html#top">ifLib</a></b>,  
<i>Motorola MC68EN360 User's Manual </i>,  
<i>Motorola MPC860 User's Manual </i>,  
<i>Motorola MPC821 User's Manual </i>
<p>


<hr>
<a name="cpmattach"></a>
<p align=right>
<a href="rtnIndex.html"><i>Libraries :  Routines</i></a></p>

</blockquote><h1><i>cpmattach</i>(&nbsp;)</h1> <blockquote></a></blockquote><h4>NAME</h4><blockquote>  
<p><strong><i>cpmattach</i>(&nbsp;)</strong> - publish the <b>cpm</b> network interface and initialize the driver</p>

</blockquote><h4>SYNOPSIS</h4><blockquote><p>
<pre>STATUS cpmattach
    (
    int           unit,     /* unit number */
    SCC *         pScc,     /* address of SCC parameter RAM */
    SCC_REG *     pSccReg,  /* address of SCC registers */
    VOIDFUNCPTR * ivec,     /* interrupt vector offset */
    SCC_BUF *     txBdBase, /* transmit buffer descriptor base address */
    SCC_BUF *     rxBdBase, /* receive buffer descriptor base address */
    int           txBdNum,  /* number of transmit buffer descriptors */
    int           rxBdNum,  /* number of receive buffer descriptors */
    UINT8 *       bufBase   /* address of memory pool; NONE = malloc it */
    )</pre>

</blockquote><h4>DESCRIPTION</h4><blockquote><p>
The routine publishes the <b>cpm</b> interface by filling in a network Interface
Data Record (IDR) and adding this record to the system's interface list.
<p>
The SCC shares a region of memory with the driver.  The caller of this
routine can specify the address of a shared, non-cacheable memory region
with <i>bufBase</i>.  If this parameter is NONE, the driver obtains this
memory region by calling <b><i><a href="./cacheLib.html#cacheDmaMalloc">cacheDmaMalloc</a></i>(&nbsp;)</b>.  Non-cacheable memory space
is important for cases where the SCC is operating with a processor
that has a data cache.
<p>
Once non-cacheable memory is obtained, this routine divides up the
memory between the various buffer descriptors (BDs).  The number
of BDs can be specified by <i>txBdNum</i> and <i>rxBdNum</i>, or if NULL, a
default value of 32 BDs will be used.  Additional buffers are
reserved as receive loaner buffers.  The number of loaner buffers is the
lesser of <i>rxBdNum</i> and a default value of 16.
<p>
The user must specify the location of the transmit and receive BDs in
the CPU's dual-ported RAM.  <i>txBdBase</i> and <i>rxBdBase</i> give the
base address of the BD rings.  Each BD uses 8 bytes. Care must be taken so
that the specified locations for Ethernet BDs do not conflict with other
dual-ported RAM structures. 
<p>
Up to four individual device units are supported by this driver.  Device
units may reside on different processor chips, or may be on different SCCs
within a single CPU.
<p>
Before this routine returns, it calls <b><i>cpmReset</i>(&nbsp;)</b> and <b><i>cpmInit</i>(&nbsp;)</b> to 
configure the Ethernet controller, and connects the interrupt vector <i>ivec</i>.
<p>
</blockquote><h4>RETURNS</h4><blockquote><p>
OK or ERROR.
<p>
</blockquote><h4>SEE ALSO</h4><blockquote><p>
<b><a href="./if_cpm.html#top">if_cpm</a></b>, <b><a href="./ifLib.html#top">ifLib</a></b>,  
<i>Motorola MC68360 User's Manual </i>,  
<i>Motorola MPC821 and MPC860 User's Manual </i>

<hr>
<a name="cpmStartOutput"></a>
<p align=right>
<a href="rtnIndex.html"><i>Libraries :  Routines</i></a></p>

</blockquote><h1><i>cpmStartOutput</i>(&nbsp;)</h1> <blockquote></a></blockquote><h4>NAME</h4><blockquote>  
<p><strong><i>cpmStartOutput</i>(&nbsp;)</strong> - output packet to network interface device</p>

</blockquote><h4>SYNOPSIS</h4><blockquote><p>
<pre>#ifdef BSD43_DRIVER LOCAL void cpmStartOutput
    (
    int unit /* unit number */
    )</pre>

</blockquote><h4>DESCRIPTION</h4><blockquote><p>
<b><i><a href="./if_cpm.html#cpmStartOutput">cpmStartOutput</a></i>(&nbsp;)</b> takes a packet from the network interface output queue,
copies the mbuf chain into an interface buffer, and sends the packet over
the interface.  etherOutputHookRtns are supported.
<p>
Collision stats are collected in this routine from previously sent BDs.
These BDs will not be examined until after the transmitter has cycled the
ring, coming upon the BD after it has been sent. Thus, collision stat
collection will be delayed a full cycle through the Tx ring.
<p>
This routine is called from several possible threads.  Each one will be
described below.
<p>
The first, and most common thread, is when a user task requests the
transmission of data. Under BSD 4.3, this will cause <b><i>cpmOutput</i>(&nbsp;)</b> to be
called, which calls <b><i>ether_output</i>(&nbsp;)</b>, which usually calls this routine. 
This routine will not be called if <b><i>ether_output</i>(&nbsp;)</b> finds that our interface
output queue is full. In this very rare case, the outgoing data will be 
thrown out. BSD 4.4 uses a slightly different model in which the generic
<b><i>ether_output</i>(&nbsp;)</b> routine is called directly, followed by a call to this
routine.
<p>
The second thread is when a transmitter error occurs that causes a
TXE event interrupt.  This happens for the following errors: transmitter
underrun, retry limit reached, late collision, and heartbeat error.
The ISR sets the txStop flag to stop the transmitter until the errors are
serviced. These events require a RESTART command of the transmitter, which
occurs in the <b><i>cpmTxRestart</i>(&nbsp;)</b> routine.  After the transmitter is restarted,
<b><i>cpmTxRestart</i>(&nbsp;)</b> does a netJobAdd of <b><i><a href="./if_cpm.html#cpmStartOutput">cpmStartOutput</a></i>(&nbsp;)</b> to send any packets
left in the interface output queue.  Thus, the second thread executes
in the context of <b><i><a href="./netLib.html#netTask">netTask</a></i>(&nbsp;)</b>.
<p>
The third, and most unlikely, thread occurs when this routine is executing
and it runs out of free Tx BDs.  In this case, this routine turns on
transmit interrupt and exits.  When the next BD is actually sent, an interrupt
occurs.  The ISR does a netJobAdd of <b><i><a href="./if_cpm.html#cpmStartOutput">cpmStartOutput</a></i>(&nbsp;)</b> to continue sending
packets left in the interface output queue.  Once again, we find ourselves
executing in the context of <b><i><a href="./netLib.html#netTask">netTask</a></i>(&nbsp;)</b>.
<p>
</blockquote><h4>RETURNS</h4><blockquote><p>
N/A
</blockquote><h4>SEE ALSO</h4><blockquote><p>
<b><a href="./if_cpm.html#top">if_cpm</a></b>

</body>
</html>