search

StanfordVLSI / dragonphy2

Open Source PHY v2
Apache License 2.0
24 stars 2 forks source link

Where does JTAG fit into the reset sequence? #60

Closed sgherbst closed 4 years ago

sgherbst commented 4 years ago

During a meeting a few days ago, we defined the reset sequence as:

  1. De-assert ext_rstb
  2. De-assert int_rstb (over JTAG)
  3. Assert en_inbuf
  4. Assert en_gf
  5. Assert en_v2t

But I realized this doesn't include the JTAG reset pin, phy_trst_n. Where does that belong in this sequence?

zamyers commented 4 years ago

the JTAG reset pin is controlled by the JTAG controller.

-Zach

On May 1, 2020, at 9:42 AM, Steven Herbst notifications@github.com wrote:



During a meeting a few days ago, we defined the reset sequence as:

  1. De-assert ext_rstb
  2. De-assert int_rstb (over JTAG)
  3. Assert en_inbuf
  4. Assert en_gf
  5. Assert en_v2t But I realized this doesn't include the JTAG reset pin, phy_trst_n. Where does that belong in this sequence?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/StanfordVLSI/dragonphy2/issues/60, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACY6NPAMSXZB2CQLWL4NDJ3RPL3YTANCNFSM4MXGZ7AQ.

sgherbst commented 4 years ago

Sorry, to clarify I mean when this should happen, rather than how. It definitely needs to happen before step 2 (since that involves JTAG communication), so I think my question can be simplified to "which should be de-asserted first, ext_rstb or phy_trst_n"

zamyers commented 4 years ago

It happens after ext_rstb is de-asserted. I can double check the logic but I believe the ext_rstb acts as a standard reset inside the JTAG and the JTAG's trst_n's primary function is to reset JTAG state machine into a known state.

-Zach

From: Steven Herbst notifications@github.com Sent: Friday, May 1, 2020 10:26 AM To: StanfordVLSI/dragonphy2 dragonphy2@noreply.github.com Cc: Zachary Alexander Myers zamyers@stanford.edu; Comment comment@noreply.github.com Subject: Re: [StanfordVLSI/dragonphy2] Where does JTAG fit into the reset sequence? (#60)

Sorry, to clarify I mean when this should happen, rather than how. It definitely needs to happen before step 2 (since that involves JTAG communication), so I think my question can be simplified to "which should be de-asserted first, ext_rstb or phy_trst_n"

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/StanfordVLSI/dragonphy2/issues/60#issuecomment-622481394, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACY6NPCKT27ZJ67EWPS3CIDRPMA4NANCNFSM4MXGZ7AQ.

sgherbst commented 4 years ago

Great, thanks. In that case I'll update the reset sequence to

  1. De-assert ext_rstb
  2. De-assert phy_trst_n
  3. De-assert int_rstb (over JTAG)
  4. Assert en_inbuf
  5. Assert en_gf
  6. Assert en_v2t