Thread (1 message) 1 message, 1 author, 2005-02-15

Re: [PATCH] natsemi long cable fix

From: Gary Spiess <hidden>
Date: 2005-02-15 21:00:15

This is a minor modification to the previous patch submission that does 
not assume the default contents of the DSPCFG register are zero.

When used with Revision D of the DP83815, the "Recommended Registers 
Configuration" from page 78 of the DP83815 data sheet is not entirely 
compatible with the driver's "short cable patch".  When the DSPCFG 
register is written with the value suggested in the document, then 
do_cable_magic() can't read the DSP coefficient and determines that all 
cables attached to the DP83815D are 'short', regardless of actual 
length.  Short cables (< 30m) cause do_cable_magic to enable additional 
attenuation to reduce CRC and idle errors.  If the extra attenuation is 
unintentionally enabled for long cables (> 50m?), they will not operate 
properly.  The National Semiconductor driver, 'dp83815.c' from 
http://www.national.com/appinfo/networks/files/linux_2_4.tar.gz was used 
as a basis for this modification.  This patch applies to the driver 
distributed in linux-2.6.10.
--- linux-2.6.10.orig/drivers/net/natsemi.c	2004-12-24 15:33:50.000000000 -0600
+++ linux-2.6.10/drivers/net/natsemi.c	2005-02-14 16:45:46.000000000 -0600
@@ -441,6 +441,7 @@
 #define DSPCFG_VAL	0x5040
 #define SDCFG_VAL	0x008c	/* set voltage thresholds for Signal Detect */
 #define DSPCFG_LOCK	0x20	/* coefficient lock bit in DSPCFG */
+#define DSPCFG_COEF	0x1000	/* see coefficient (in TSTDAT) bit in DSPCFG */
 #define TSTDAT_FIXED	0xe8	/* magic number for bad coefficients */
 
 /* misc PCI space registers */
@@ -1243,7 +1244,8 @@
 		writew(1, ioaddr + PGSEL);
 		writew(PMDCSR_VAL, ioaddr + PMDCSR);
 		writew(TSTDAT_VAL, ioaddr + TSTDAT);
-		np->dspcfg = DSPCFG_VAL;
+		np->dspcfg = (np->srr <= SRR_DP83815_C)?
+			DSPCFG_VAL : (DSPCFG_COEF | readw(ioaddr + DSPCFG));
 		writew(np->dspcfg, ioaddr + DSPCFG);
 		writew(SDCFG_VAL, ioaddr + SDCFG);
 		writew(0, ioaddr + PGSEL);
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help