At 12:45 PM -0400 8/17/01, Dan Sugalski wrote:
>At 06:41 AM 8/17/2001 -0400, Carl Friedberg wrote:
>>$ write sys$output f$getsyi("thermal_vector")
>>FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
>>$
>>
>>Neither of these seems useful...
>
>They aren't. You've apparently got a set of systems without the on-chip temperature
>thingies.
It's actually quite useful to have the software fill in a
not-a-temperature value rather than go belly-up. It appears there
are three categories of systems:
1.) those that have the appropriate hardware support and return good values.
2.) those that do not have the hardware support but still return
successfully from $getsyi, having filled in values that make clear
the information is not available.
3.) those that do not have the hardware support and also do not have
software smart enough to know that they don't have hardware support.
These return a failure code from $getsyi.
It is only category #3 that needs help because the failed return
completely prevents use of VMS::System::get_all_sys_info_items().
My current thinking is that we should call $getsyi with SYI$_CPUTYPE,
and if it returns a value indicating a 21264 or later (>= 8), then
it's safe to request the power and temperature stuff. Otherwise,
since all that stuff is at the end of the item list, we can just zero
it out. What I'm not sure of yet is whether a 21264 or later
processor covers everything in categories 1 and 2 above.
--
____________________________________________
Craig A. Berry
mailto:[EMAIL PROTECTED]
"Literary critics usually know what they're
talking about. Even if they're wrong."
-- Perl creator Larry Wall
