[j-nsp] Sandisk 1G CF card compatibility

[Dave Diller]

>
>> You're best off getting a regular classic SanDisk with PIO support,  
>> there are no
>> benefits to be had by putting nice CFs into old and crappy hosts. '
>
> ..."standard" 1G Sandisk [SDCFB-1024-A10] is $18.95 on Amazon...

...and they don't work either.  Different point for the same TIMEOUT  
failure, at least.

Are you using the 1G ones successfully?  What's ad0 ID itself as?


This one IDs itself as
ad0: 977MB <SanDisk SDCFB-1024 HDX 4.03> at ata0-master PIO4
and the code on the side is BB061004971B.

The Extreme III that *worked* was also ver 4.03 and the same code  
except for the last few digits, which are instead 765B - probably a  
part identifier.  So I had high hopes up front, given the seeming match.


Snapshots fine, which is an improvement over the v4.08 Extremes:

dave at re0-lab-t640> request system snapshot partition
Clearing current label...
Partitioning compact-flash media (ad0) ...

   Partition  Mountpoint  Size    Snapshot argument
       a      /           879MB   root-size
       e      /config     97MB    config-size

Running newfs (879MB) on compact-flash media / partition (ad0s1a)...
Running newfs (97MB) on compact-flash media /config partition  
(ad0s1e)...
Copying '/dev/ad1s1a' to '/dev/ad0s1a' .. (this may take a few minutes)
Copying '/dev/ad1s1e' to '/dev/ad0s1e' .. (this may take a few minutes)
The following filesystems were archived: / /config



but on reboot it crashes and burns:

Trying to mount root from ufs:/dev/ad0s1a
Attaching /packages/jbase via /dev/mdctl...
Mounted jbase package on /dev/md0...

Verified manifest signed by PackageProduction_8_5_0
Verified jboot signed by PackageProduction_8_5_0
ad0: TIMEOUT - READ_MUL retrying (2 retries left) LBA=493887
ad0: TIMEOUT - READ_MUL retrying (2 retries left) LBA=494783
panic: Loss of soft watchdog
db_log_stack_trace_cmd(c0d4e5c0) at 0
panic(c0cdc731,f1765cb8,c0d418c0,c51b7b00,f1765c74) at 0
statclock(f1765cb8) at 0
rtcintr(f1765cb8) at 0
intr_execute_handlers(c0d418c0,f1765cb8,c51bda8c,c052d359,0) at 0
atpic_handle_intr(8) at 0
Xatpic_intr8() at 0
--- interrupt, eip = 0xc09219c0, esp = 0xf1765cfc, ebp = 0xf1765cfc ---
cpu_idle_default(f1765d10,c052d36a,c51bda8c,f1765d24,c052c734) at 0
cpu_idle(c51bda8c,f1765d24,c052c734,0,f1765d38) at 0
idle_proc(0,f1765d38) at 0
fork_exit(c052d359,0,f1765d38) at 0
fork_trampoline() at 0
--- trap 0x1, eip = 0, esp = 0xf1765d6c, ebp = 0 ---
KDB: enter: panic
[thread pid 11 tid 100005 ]
Stopped at      kdb_enter+0x37: pushl   $-0x1
db>

These are the only other 1G ones that SanDisk makes, so it seems  
they're right out of the game.

Maybe this is why Juniper switched to SiliconSystems?  I was quoted  
4-5 week leadtime though, bah.

-dd