Talk:Start Input/Output

What is the issue, or your concerns?

I wrote the page, and I am a "subject matter expert" in this topic.

I am also the author of the referenced "white paper".

I have been a specialist in this field, and in this OS, for 45 years.

Expertise acknowledged but...
There are no references or citations Greenmaven (talk) 22:45, 15 June 2011 (UTC)

Reference was uploaded and citation was provided.

Extensible
I'm aware of no IBM documentation to suggest that EXCP[VR] cannot be used with new I/O devices, and, in fact, new devices are supported by it. Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:27, 13 September 2012 (UTC)

Unexplained removal of reference to DDT
A recent edit by user:Peterh5322 removed the paragraph The Unit Control Block (UCB) for each device points to a Device Descriptor Table (DDT), which identified routines needed for device dependent processing, e.g., handling of Attention interrupts, appending of RESERVE CCWs, appending of Set File Mask and Seek. with no explanation. Is there any reason not to revert it? And why the scare quotes?Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:19, 2 July 2017 (UTC)


 * I noticed that user:Peterh5322 seems to be a dead link, although there is a talk page. Shmuel (Seymour J.) Metz Username:Chatul (talk) 18:09, 19 July 2017 (UTC)

Yr edit removed his new material on "trapcode" - was it incorrect, not particularly relevent, untimely, ...? The absence of a user talk page is not particularly significant; Peterh5322 has been editing since 2011. Tom94022 (talk) 19:27, 19 July 2017 (UTC)


 * It's the user page that's missing, not the talk page.


 * I reverted the edit because it removed the references to the UCB and DDT.


 * Note that the descriptions in of most of the device-dependent exit addresses do not include the term trap. Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:48, 20 July 2017 (UTC)


 * In the process u of course deleted the new added material. Unless you had a reason to believe it was incorrect, shouldn't u have modified the article to contain both?  I don't think the "trapcode" addition is appropriate for a SIO article but I'm not certian and therefore haven't tried editing it back in.  That's why I asked the question?  Are u sure it is appropriate to delete it? Tom94022 (talk) 01:53, 21 July 2017 (UTC)


 * Well, the term trapcode certainly doesn't appear in the data area mapping; I'll have to dig out my dead tree IOS logic manual and check whether it exists there. Shmuel (Seymour J.) Metz Username:Chatul (talk) 16:03, 24 July 2017 (UTC)


 * I checked the index of with   and  There was no index entry for either trap or trapcode. There was an entry for device dependent trap routine, but that is only one of many device dependent routines pointed to by the DDT.


 * So the use of trapcode is inappropriate and unless someone wants to add descriptions of all of the device dependent routines, the use of trap is also inappropriate.
 * DDTSIO     SIO exit address
 * DDTTRAP    TRAP code exit
 * DDTUNIN    Unsolicited interrupt exit address
 * DDTSENSE   Sense routine exit address
 * DDTEOS     End-Of-Sense exit address
 * DDTMIH     MIH exit address
 * DDTERASE   Erase Exit Address
 * DDTDSE     Device service exit address
 * DDTDDR     DDR exit ID or address
 * DDTCPS     Channel Program Scan exit address
 * DDTRERPA   Resident ERP address or zero.  Shmuel (Seymour J.) Metz Username:Chatul (talk) 15:44, 25 July 2017 (UTC)


 * Thanks for the analysis. I agree with you conclusion.  Tom94022 (talk) 00:54, 26 July 2017 (UTC)