P1500 Meeting at VTS'2002
5/2/2002, 4:30pm to 8:00pm

Attendees

Scribe

Review of Sunday 4/28 CTAG meeting

• Much discussion occurred with regard to the PIL (Parallel Interface Layer). This is especially critical given the goal of being ready for ballot by Sep 2002 and that it is anticipated to take another 6 months to iron out all the issues for the PIL.

• The SIL (Serial Interface Layer) is defined and agreed to. It incorporates a single scan-in and scan-out for connecting to internal scan chains and wrapper chains.

• Yervant stated that a parallel interface is necessary for initial version of the standard.

• CTAG has defined "model-1" which contains the essential parts of the parallel interface.

• CTAG is looking at "model-1-plus-plus" which would add some parallel capabilities. Note: Yervant suggests a better name would be "PIL-minus". It is defined as:
  1. the functionality illustrated in "slide 19" (from the CTAG presentation at DATE'02).
  2. NOT to include bandwidth conversion. If user requires bandwidth conversion, then it should be done outside of the P1500 wrapper (in the TAM). Bandwidth conversion is defined as the case where there are multiple SI/SO connections to the wrapper, but a different number of PI/PO signals.
  3. NOT to include definition of parallel bypass
  4. Lee Whetsel has AI to define and produce a diagram
  5. Teresa says that only "WEXTEST-S" instruction is mandatory. No instructions should be removed from the document.

• Brion Keller - suggests that there needs to be a way to specify a connection of a special signal (e.g., analog) directly to a PI/PO. It is believed that this can already be specified in CTL. Brion has the AI to verify that it is covered in the CTL document. The CTAG team has the AI to define all "special care" signals and their handling.

CTL Issues

• Remaining work to be done in P1450.6 is to complete the explanations, add code examples, and add usage examples. This is to be split up amongst the working group and to be ballot ready by sep 2002.

• Yervant questions whether CTL can support a partially wrapped core. Maurice answered that the working groups believe the answer is YES, however, a real example is needed in order to verify it. See later comments on CTL examples.

• The following test cases were identified for the purpose of generating full CTL definitions:
  1. Grady's example - partly complete verilog wrapper, CTL-wg to generate full CTL.
  2. Teresa's example - verilog wrapper complete, Rohit/Teresa to generate full CTL.
  3. Mohsen/BCU example - VHDL wrapper complete, Mike Ricchetti to verify the wrapper, Mohsen to look into a resource to generate the complete CTL, CTL-wg will verify.
  4. Virage example - verilog wrapper complete, Mike Ricchetti to verify the wrapper, Rohit to generate complete CTL (or possibly someone to be identified by Samvel in Armenia).

Review P1500 Compliance effort

• The compliance-wg has generated 10 compliance rules for the information model (CTAG and CTL related). No written documentation (other than in the minutes) is available yet. The compliance-wg has the AI to complete this work and include in the P1500 document. Another AI is to review this rule set with the main P1500 working group.

Other issues

• Brion Keller raised an issue with regard to WC-SISU type of WBR cell. There was general discussion, but no specific issue identified.

• Alan raised the issue of allowing an asynchronous reset in P1500. This issue is to be taken up by the CTAG working group.

Action Item summary

AI-1:

Lee Whetsel - define and produce a diagram of "PIL-minus" for inclusion into the document in time for September 2002 ballot.

AI-2:

Brion Keller - verify that CTL allows to connect a special signal directly to a PI/PO.

AI-3:

CTAG team - define all "special care" signals and their handling.

AI-4:

compliance-wg - complete the definition of the compliance rules, include in the P1500 document, and review at a full P1500-wg meeting.

Next meeting plans

• There will be an "awareness presentation" meeting at DAC. Date/time TBD.

• There will be a video conference meeting in June or July. Date/time TBD.