Thanks Shabtay, This is very helpful and I thank you for putting in the time and effort. Brian _____ From: owner-itc@server.eda.org [mailto:owner-itc@server.eda.org] On Behalf Of Shabtay Matalon Sent: Friday, April 28, 2006 6:15 PM To: brian_bailey@acm.org; itc@server.eda.org Cc: Shabtay Matalon Subject: SCE-MI 2.0 IMs Hi Brian, ITC Techies I am providing my comments on the SCEMI 2.0 IMs by this weekend subject to my feedback not being viewed yet as Cadence's final and binding response (end of disclaimer). I hope that this email could help us to focus on the active IMs that require discussion and resolution, close those tagged Close and defer those defined Pending to the spec review time frame. I have annotated the list on the web in blue with my comments with my recommendation for the status field for each IM. Regards, Shabtay Number Description Status IM200 <http://www.eda.org/itc/open/IM200.htm> VHDL and Verilog compatibility Our resolution to this sticky issue is that Verilog 2001 and VHDL data types using DPI import/export should be deprecated from the SCE-MI 2.0 spec. VHDL/Verilog HW side will be supported using the SCEMI 1.1 use model and reconsidered in SCEMI 2.1. ACTIVE Active IM201 <http://www.eda.org/itc/open/IM201.htm> Function arg data type subset I suggest deprecating the Verilog 2001 and VHDL support data types from this IM as this is covered in IM200. This will allow us to quickly converge on this IM as we are in principle agreement about SV data types. If agreed this is pending subject to spec review. ACTIVE Pending? IM202 <http://www.eda.org/itc/open/IM202.htm> DPI call restrictions Need to converge on the recursion issue. a) What level must be supported? What (if any) should be prohibited. I think this could be resolved quite quickly if SCE-MI 2.0 will not impose restrictions on supporting more than what SCE-MI 2.0 requires. ACTIVE Active IM203 <http://www.eda.org/itc/open/IM203.htm> Time tracking Issues resolved. Addressed by spec and only requires spec review. ACTIVE Pending IM204 <http://www.eda.org/itc/open/IM204.htm> Init, Version and Shutdown in DPI-only applications Not sure. Discuss if this can be closed ACTIVE Pending? IM205 <http://www.eda.org/itc/open/IM205.htm> SceMiParamaters in DPI-only applications Issues resolved. Needs explanation in the mixed usage section and only requires spec review. ACTIVE Pending IM206 <http://www.eda.org/itc/open/IM206.htm> Mixed SCE-MI 1.1/2.0 usage Issues resolved. Needs explanation in the mixed usage section and only requires spec review. PENDING Pending IM207 <http://www.eda.org/itc/open/IM207.htm> Compliance Issues Just clarification IM. Issue resolved and no action needed. ACTIVE Close IM208 <http://www.eda.org/itc/open/IM208.htm> Streaming / Reactivity This is mostly addressed with automatic flush-on-eom mode and is quite high level. ACTIVE Close IM209 <http://www.eda.org/itc/open/IM209.htm> Pipes High level clarification exchange. Issue resolved and no action needed. ACTIVE Close IM210 <http://www.eda.org/itc/open/IM210.htm> Determinism for streaming Issue resolved. We should consider inserting this clarification in SCEMI 2.0 use model which will require spec review. ACTIVE Pending IM211 <http://www.eda.org/itc/open/IM211.htm> Transaction Pipe issues High level clarification exchange. Issue resolved and no action needed. ACTIVE Close IM212 <http://www.eda.org/itc/open/IM212.htm> Support for multi-valued logic Requires resolution but Tharas who requested support for 4 values has not showed up for quite some times. ACTIVE Active IM213 <http://www.eda.org/itc/open/IM213.htm> Support of SystemVerilog deprecated types: will support P1800 types only Resolved and subject to spec review PENDING Pending IM214 <http://www.eda.org/itc/open/IM214.htm> Error Handlers Dependency on IM212 ACTIVE Active IM215 <http://www.eda.org/itc/open/IM215.htm> Implicit calls to SCE-MI setup macros Suggest to close that as duplicate with IM202 ACTIVE Close IM216 <http://www.eda.org/itc/open/IM216.htm> Upgrade to clock port macro Resolved and subject to spec review ACTIVE Pending IM217 <http://www.eda.org/itc/open/IM217.htm> Ability to query time on hardware side Requires resolution. SUBMITTED Active IM218 <http://www.eda.org/itc/open/IM218> Creating a Pipe object instead of using Pipe ID as an argument to a function Cadence to submit enhancements/modifications to existing pipes proposal based on Jason's initial email. ACTIVE Active IM219 <http://www.eda.org/itc/open/IM219.htm> Removing any calls to SV DPI utility functions and DPI exported functions outside an imported context function Dependency on SV working group SUBMITTED Active IM220 <http://www.eda.org/itc/open/IM220.htm> Implementing implicit synchronization on EOM and the default during bring-up mode. Bring up mode not resolved and some areas in the spec need to be enhanced. SUBMITTED Active IM221 <http://www.eda.org/itc/open/IM221> Defining clearly buffer length (HW only or HW SW) Mostly resolved with the exception if the standard will address user level knob for controlling HW side buffer depth ACTIVE Pending? IM222 <http://www.eda.org/itc/open/IM222> HW/SW side synchronization when buffer is smaller than a transaction. Resolved subject to review of the reference implementation. ACTIVE Pending IM223 <http://www.eda.org/itc/open/IM223> Defining SCE-MI 2.0 terminology (transaction/message and element including EOM/EOT). Mostly resolved. But may need to be reflected in the spec and subject to spec review ACTIVE Pending IM224 <http://www.eda.org/itc/open/IM224> Synchronization semantics of FLUSH for non-blocking. Ok with partial acceptance of the message with eom_auto_flush disable. Recommend that we don't allow this with eom_auto_flush enable. The term eom_auto_flush is quite of cumbersome. ACTIVE Active IM225 <http://www.eda.org/itc/open/IM225> Rename pipes function calls with sceme as opposed to dpi prefix. Resolved. Pending spec review PENDING Pending IM226 <http://www.eda.org/itc/open/IM226> Flush semantics Requires posting a response and capturing in pipes section. ACTIVE Active IM227 <http://www.eda.org/itc/open/IM227> Pipe Scope Requires resolution. One resolution is that Verilog 2001 and VHDL support should be deprecated from the SCE-MI 2.0 for pipes as well and we'll deal with this in SCEMI 2.1. ACTIVE Active IM228 <http://www.eda.org/itc/open/IM228.htm> Granularity Requires resolution. We will present our position with IM 218, but suggest keeping this IM separated as the issue is orthogonal. ACTIVE Active IM229 <http://www.eda.org/itc/open/IM229.htm> Removal of old Appendix D - Event based support We agreed. ACTIVE Close IM230 <http://www.eda.org/itc/open/IM230.htm> Inclusion of reference implementation Converged, but not closed ACTIVE Active ------------------------------------- Shabtay Matalon Solution Architect R&D, CVA Phone: (408) 428 5081 email: shabtay@cadence.comReceived on Fri Apr 28 18:49:33 2006
This archive was generated by hypermail 2.1.8 : Fri Apr 28 2006 - 18:49:35 PDT