11.07.2015 Views

Open Core Protocol International Partnership Governing ... - OCP-IP

Open Core Protocol International Partnership Governing ... - OCP-IP

Open Core Protocol International Partnership Governing ... - OCP-IP

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

member review. Bob agreed that this was the goal. Ian asked Bob how far off the group wasfrom reaching this goal. Ian said he just wanted to make sure that the inclusion of the exampledid not change anything.Bob told the GSC that Sprint’s donations could be ready in mid-2008. Bob said that he spokewith Alan at ITRI and they would be able to publish something by the end of the year. Bob saidthat RTL core examples were included in the text, which showed how to involve registers anddefine transactions. Bob said that Infineon’s donation included support for Sprint’s solution onmulticore debugging. Bob said that the DWG had originally thought this was on the hardwarelevel, but it turned out it to be a standard on the software side, where a debug server wasrunning. Bob said that they would like to standardize on the software. Bob said that the solutionwas to match up with ARM and companies in Europe.Bob said that the DWG’s goal was to publish future debug socket signals. Bob said they wantedto have the specification ready for member review by November 7, 2007. Bob said that this goalwas missed, but the DWG was not far away. Bob said that he had written a signal proposal, buthe needed Albrecht, Neal, and Texas Instruments to agree on it. Bob said that the example wasplanned to be done together, but that it was taken over by the Sprint group. Bob said that whenfinished at the end of 2008, it would have two cores and would connect two different debuggers atthe same time. Bob said <strong>IP</strong> Extreme was already offering MCDS, but the DWG has made noprogress on the development of <strong>OCP</strong> wrappers because of a lack of funds, so a first customerhas to be waited upon.Ian stated that the GSC wanted feedback from Bob and Neal on what to do in regards to the workbeing done at Sprint. Bob suggested that pointers be included to the Sprint work from the debugspecification. Bob added that Sprint would have to support the software side while <strong>OCP</strong>-<strong>IP</strong>supported the hardware side. Ian mentioned that <strong>OCP</strong>-<strong>IP</strong> should stay away from endorsing onesolution and instead it should be referred to as a good example. Drew stated that the debug workneeded to work with Sprint and wondered what compelling event would cause people to beginusing the <strong>OCP</strong>-<strong>IP</strong> debug solution. Bob replied that <strong>OCP</strong>-<strong>IP</strong> members would have to vote on thespecification. Drew said that the linkage with Spirit was likely to either persuade or dissuadeusage and Bob would know better than others which was more likely. Ian added that the APIneeded to be understood better and this would not happen until Christmas. Drew suggestedincluding a cover letter with the debug specification when it goes to member review that askedusers if they thought it was beneficial to link to complimentary technology within Sprint.Vesa said he understood that the Sprint project was complete and Sprint was currently looking forplaces to donate their work. Bob disagreed and said the Sprint work was scheduled to becomplete by the end of 2008. Ian reiterated that the Sprint work should be pointed to, but notincluded as part of the specification. Ian asked Bob when he thought he would have a final draft.Bob said it would take another two to three weeks.Bob left the meeting at 11:13NoC BWG Working Group StatusIan reviewed the NoC BWG’s status. Ian said the NoC BWG had a specification in reviewexplaining how to define a NoC. James asked what the fundamental motivation was for an <strong>OCP</strong>-<strong>IP</strong> NoC BWG. Ian said <strong>OCP</strong>-<strong>IP</strong> has three working groups orientated to the emergence ofheterogeneous, multiprocessor SoCs: the Cache Coherence Subgroup, the NoC BWG and theFVWG. Ian said the need had arisen to compare and evaluate NoCs and subsequently the ideaof a NoC BWG was formed. Ian added that the NoC BWG liked <strong>OCP</strong> because it was acomprehensive specification. Ian said the NoC BWG evolved by creating a white paper, whichhas had a huge number of downloads from the <strong>OCP</strong>-<strong>IP</strong> Web site. Ian stated that Axel Jantsch ofKTH in Sweden had volunteered another person to enhance the update on how to specifybenchmarks. Ian added that the NoC BWG has the calories to get this kind of work done and hadbeen involved with publishing papers and attending conferences. Ian said that the group had yet<strong>OCP</strong>-<strong>IP</strong> CONFIDENTIAL<strong>Governing</strong> Steering Committee Meeting Minutes 11/8/2007Page 13 of 22

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!