Mem-Apr 19, 2007

From STDFGroup

STDF Memory Datalog meeting Notes

Date*: 19th Apr 2007

Attendees:

Kochen Liao -- Qualcomm

Liang Lai -- Mentor Graphics

Sauro Landini -- ARM

Ajay Khoche -- Verigy

Ping Wen -- Yield Dynamics

Agenda:

1. Discuss scan fail log datamodel
2. Discuss need to long meeting along the line of scan group

Minutes:

• Ajay presented the scan fail log data model to the group.
• Group decided that we need to prepare a similar one for the memory and perhaps merger the scan and memory into one.
• Following points came up during the discussions
  • The standard shodul be able to handle multi-site and multi-core situations
    • multiple scan and capture frequencies should be allowed in the environment block
  • The optional field in the datamodel should be an integer instead of boolean help tools skips a known no. of bytes if the optional data is not interesting to it.
  • expected data is not needed for embedded memories
  • The formats that need to supported for memory fail data should be
    • address & bit/word
    • fail cycle and Pin no (translation to bit/word is done by the downstream tools.

Action Items:

• get PDF to present the requirement of their tool (Ajay)
• check on other environment information (ping)
• Prepare starting requirements document (Ajay)
• Review ST' proposal (All)