This message is from the T13 list server.
(B
(B
(BHi,
(B
(BSorry to be replying to an old thread.
(B
(BI have always thought tJ was a minimum timing, but,
(BI have overlooked that small note.
(B
(BI have come across a commercial ATAPI Host Adapter simulator
(Bmodel that has tJ=0ns (because it thinks that DMACK- should be
(Bnegated within tJ).
(B
(B> The tj minimum time insures that IOR- or IOW- is deassert *before*
(B> DMACK- is deasserted. The statement in Fig. 70 is not worded
(B
(BIf this is the case, I think the note should be corrected
(Bin the future drafts to prevent people from wrong understanding
(Bof the specificaion.
(B
(BI have looked at the current and previous documents on this.
(BThe contradiction began appearing in ATA/ATAPI-5.
(B
(BThe note for Multiword DMA first appeared in the following document
(Bhttp://www.t13.org/technical/d99132r0.pdf
(B
(BWhich was one of the action items in the following minutes
(Bhttp://www.t13.org/meetings/d99129r0.pdf
(B
(BAt the time it was phrased:-
(B "To terminate the transmission of a data burst, the host shall
(B negate DMACK- within the specified time after a DIOR- or DIOW-
(B pulse."
(B
(BIt was in ATAPI-6 Revision 6 that the phrase "within the specified
(Btime" was replaced by "within tJ". The wording has not changed
(Bsince then.
(B
(BBest regards,
(B
(B-------------------------------------------------
(BSony Corporation,
(BSemiconductor Solutions Network Company
(BSoC Solution Center
$BBh([EMAIL PROTECTED](B1$BIt(B2$B2](B
(BTel. +81-3-6834-5252 / Fax. +81-3-6834-5256
(BTel. 9-30-4-5252 / Fax. 9-30-4-5256 (Ext)
(BTadashi Nakamura ([EMAIL PROTECTED])
