search

dirtyjtag / DirtyJTAG

JTAG probe firmware
MIT License
464 stars 70 forks source link

Up the max frequency in SPI mode and fix a bug #88

Closed phdussud closed 3 years ago

phdussud commented 3 years ago

in SPI mode fix the omission to set TMS low before the transfer TDI

jeanthom commented 3 years ago

Thanks for your contribution! Should I backport this fix to DirtyJTAG 1?

phdussud commented 3 years ago

Hello, The bug is only present in Jtag2 in the SPI implementation. The main branch does not have a SPI implementation. Thanks!

phdussud commented 3 years ago

Another question.... I finally noticed that we should not support setting tms, tdi and tck independently because tms and tdi are clocked by tck. It would make things faster if we could support one command that would set tms, tdi, then clock tck and optionally rreturn the value tdo read when tck was high. We send only one USB packet and maybe read one instead of sending one, reading one and sending another one. What do you think?

jeanthom commented 3 years ago

Are you talking about CMD_SETSIG? If so this command is here just to support UrJTAG.

phdussud commented 3 years ago

Yes, I am taking about the combination of CMD_SETSIG, CMD_GETSIG, CMD_SETSIG. It also used in openFPGALoader, and I bet that the OpenOCD plugin uses it as well.

From: Jean THOMAS @.> Sent: Monday, March 29, 2021 2:14 PM To: jeanthom/DirtyJTAG @.> Cc: phdussud @.>; Author @.> Subject: Re: [jeanthom/DirtyJTAG] Up the max frequency in SPI mode and fix a bug (#88)

Are you talking about CMD_SETSIG? If so this command is here just to support UrJTAG.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fjeanthom%2FDirtyJTAG%2Fpull%2F88%23issuecomment-809718758&data=04%7C01%7C%7C411de84a18354406694b08d8f2f7a476%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637526492713601500%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=upx%2BLlqP3XMLCvaZ9ruADCadRrMNiwDk6gUJUClivqM%3D&reserved=0, or unsubscribehttps://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FABVQKX3UJCWIWTIXPZJ4GO3TGD3UHANCNFSM4Z6J2KLA&data=04%7C01%7C%7C411de84a18354406694b08d8f2f7a476%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637526492713601500%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=2QGbXjVK5djmktkOZ88Zbr8j%2FN9ToQ0YxapSJMtAeqU%3D&reserved=0.

jeanthom commented 3 years ago

I insist on keeping CMD_SETSIG/CMD_GETSIG as is this part of what UrJTAG requires. Maybe we could add a command modifier to CMD_CLK to return TDO when we need it?

May I suggest you continuing this discussion in #77? That would help keep track of all the discussion related to the DirtyJTAG 2.0 revision.