On Mon, 19 Feb 2001 14:49:35 -0800, Andrey Savochkin <saw@saw.sw.com.sg> wrote:
> On Tue, Feb 20, 2001 at 09:21:06AM +1100, CaT wrote:
>>
>> It happened again. Same deal. Once was after a reboot and this time
>> was after a resume. :/
>
> In my experiments wait_for_cmd timeouts almost always were related to
> DumpStats command.
> I think, we need to investigate what time constraints are related to this
> command.
Nothing documented...
CaT, can you apply this debugging patch and let us know what you get in the
logs? It should allow us to pinpoint the error a bit more precisely.
Thanks,
Ion
--
It is better to keep your mouth shut and be thought a fool,
than to open it and remove all doubt.
------------------------------------------
--- /mnt/3/linux-2.2.19pre/drivers/net/eepro100.c Thu Feb 15 16:09:19 2001
+++ linux-2.2.18/drivers/net/eepro100.c Mon Feb 19 16:13:45 2001
@@ -368,9 +368,16 @@
#define outl writel
#endif
+static char *cmdwait_string = "wait_for_cmd_done timed out at %s:%d\n";
+#define wait_for_cmd_done(ioaddr) \
+ do { \
+ if (do_wait_for_cmd_done(ioaddr)) \
+ printk(cmdwait_string, __FUNCTION__, __LINE__); \
+ } while (0)
+
/* How to wait for the command unit to accept a command.
Typically this takes 0 ticks. */
-static inline void wait_for_cmd_done(long cmd_ioaddr)
+static inline int do_wait_for_cmd_done(long cmd_ioaddr)
{
int wait = 20000;
char cmd_reg1, cmd_reg2;
@@ -383,10 +390,10 @@
if(cmd_reg2){
printk(KERN_ALERT "eepro100: cmd_wait for(%#2.2x) timedout with(%#2.2x)!\n",
cmd_reg1, cmd_reg2);
-
+ return 1;
}
}
-
+ return 0;
}
/* Offsets to the various registers.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Feb 23 2001 - 21:00:21 EST