On Mon, May 22, 2000, Donald Becker <becker@scyld.com> wrote:
; > ;
; > ; The very old code used eeprom_delay(100), which mean 100ns, *not* 100usec.
; >
; > If I greep the driver that was in 2.2.6 kernel I get
; >
; > #define eeprom_delay(nanosec) udelay(1);
; > eprom_delay(100);
; > eprom_delay(150);
; >
; > So the "nanosec" parameter gets trashed and you delay
; > for 1 us in all of the cases.
;
; Well, there isn't a nanodelay(nsec) function.
; And, as you know, udelay() isn't reliable on some kernel versions.
; The usual maximum clock rate is 2.5Mhz or 4Mhz.
; That's 250 or 400ns per cycle, or 125/200ns per half cycle.
; A 33Mhz PCI bus transaction will take a minimum of 120ns, likely more.
OK now it's all clear where does the confucsion come from,
the specs that I have define a bit higher delays...
Now I understand that eeprom_delay(150) was more of a reminder
code...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed May 31 2000 - 21:00:26 EST