[Info-vax] Calling standards, was: Re: Byte range locking - was Re: Oracle

[VAXman- at SendSpamHere.ORG]

In article <o15cvh$4oj$1 at dont-email.me>, "Craig A. Berry" <craigberry at nospam.mac.com> writes:
>On 11/23/16 12:59 PM, Bob Koehler wrote:
>> In article <e9m017F7ab9U1 at mid.individual.net>, Bill Gunshannon <bill.gunshannon at gmail.com> writes:
>>>
>>> As I said, I don't have a manual handy.  I am aware that the VAX
>>> had descriptors, but I would be extremely surprised if the .ASCIZ
>>> and .PRINT weren't still there again, primarily for backwards
>>> compatibility.
>>
>>    VAX macro has an .ASCIZ directive.  Never had any use for it, even
>>    after we started doing C.  Can't seem to find a .PRINT directive, but
>>    I haven't done an exhaustive search.
>
>30 seconds with a search engine reveals that .PRINT, .WARNING, and
>..ERROR in VAX MACRO are assembly-time directives that print the comment,
>i.e., everything after the semicolon. As far as I can see, the strings
>are always constants of known length and the only thing that could crash
>is assembling your program, not running it. I've never really been a
>MACRO person but I think I've usually seen people use LIB$PUT_OUTPUT for
>printing, and of course that is a descriptor-based interface.

	$DSCDEF

	.PSECT	DATA,WRT,NOEXE,5

ZTSTR:	.ASCIZ	"This string ends with a null."

	.ALIGN	QUAD
AZFMT:	.ASCID	"!AZ"

	.PSECT	CODE,NOWRT,EXE,5
	.ENTRY	HOW_TO,0

	CLRL    -(SP)					;\
	PUSHL	#<DSC$K_CLASS_D@<DSC$B_CLASS at 3>>!-	;- dynamic descriptor on stack
		 <DSC$K_DTYPE_T@<DSC$B_DTYPE at 3>>	;/

	PUSHAL	ZTSTR
	PUSHAL	4(SP)			; dynamic descriptor
	PUSHL	#0
	PUSHAL	AZFMT
	CALLS	#4,G^LIB$SYS_FAO	; format .ASCIZ string

	PUSHAL	(SP)			; dynmaic descriptor
	CALLS	#1,G^LIB$PUT_OUTPUT	; let's see it.

	PUSHAL	(SP)			; dynamic descriptor
	CALLS	#1,G^STR$FREE1_DX	; free it
	ADDL2	#8,SP			; clear from stack

	RET
	.END	HOW_TO

Note, I would have used LIB$ROUTINESDEF but its $LIB_SYS_FAO_S macro has been
hopelessly broken for EONS!!!



>I don't recall whether there are any native interfaces that use .ASCIZ
>but there are certainly MACRO programs that use it. And when coding in
>assembler, surely null-terminated strings are the least of your worries
>as far as accessing memory that doesn't belong to you. Which makes all
>the dumping on C kind of silly considering a third of VMS is written in
>a far more dangerous language. Dunno about BLISS. Maybe it's two thirds.

What's more *dangerous*?


-- 
VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.