Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Revision M+ support #461

Merged
merged 31 commits into from
Jan 8, 2024
Merged

Revision M+ support #461

merged 31 commits into from
Jan 8, 2024

Conversation

calc84maniac
Copy link
Contributor

@calc84maniac calc84maniac commented Mar 18, 2023 •
edited
Loading

Makes the following changes to support new hardware on rev M+ and lay groundwork for future Python support and LCD improvements:

  • Added serial Flash chip support (commands used by the OS are supported)
  • Added Flash cache support with appropriate wait states
  • Fixed some aspects of parallel Flash support, like resetting it properly to avoid unrecoverable state when misconfigured
  • Improved SPI emulation and added UART emulation, both with accurate transfer timing, loopback support, and FIFO/interrupt behavior
  • Added SPI chip select support (allows deselecting the LCD on Python editions)
  • Added basic support for interlaced scan and MCU/VSYNC display modes (inaccurate timing but should display all pixels now)
  • Revision selection in settings rather than previously scattered options like IM 2 enable
  • Automatic revision selection based on the detected boot code version, also restricts usable revisions unless a debug option is set
  • Automatic Python edition detection based on the certificate, can be overridden with the same debug option
  • Flash statistics (average cycles per access and number of cache misses) added to debug cycle counting view

@calc84maniac
Copy link
Contributor Author

I've discovered an unusual bug in rev M emulation, which I think is a blocker on this PR.

Basically, the issue is that if you turn off the calculator while emulating rev M, linking stops working. This issue persists through resets and state saves/loads, and even when resetting into another revision. The only way to restore linking is to reload the ROM.

I did a git bisect, and found that this issue exists even back in @jacobly0's first commit, which leads me to believe it's something related to different code paths (in the boot code?) executed on rev M. But I don't understand why the problem persists through resets and other revisions. Perhaps there's some USB register that's never getting reset? I would appreciate if an expert on linking could look into this issue.

@calc84maniac calc84maniac force-pushed the feature/rev_m branch 5 times, most recently from 29fb4ee to 88536ab Compare September 4, 2023 01:33
jacobly0 and others added 15 commits October 1, 2023 22:40
@jacobly0 @calc84maniac
Make HW Rev M boot.
679cbee
@jacobly0 @calc84maniac
Implement remaining used flash instructions, and make flash self-test…
38645c1 
… look reasonable.
@jacobly0 @calc84maniac
Support python version checking.
836ec43
@jacobly0 @calc84maniac
Revision M change to privileged code behavior.
9ebf6bd
@jacobly0 @calc84maniac
Maybe optimize rawPC.
5939adb
@jacobly0 @calc84maniac
Fix missing mirror.
8e44ca4
@calc84maniac
Add ASIC revision selection options
02288ca
@calc84maniac
Disallow parallel flash commands on serial flash
09237b8
@calc84maniac
Fix signal connection
c057ef2
@calc84maniac
Improve revision selection interface, ensure consistency with settings
4c028b1
@calc84maniac
Implement improved SPI transfer support with scheduler-based timing
0c31eb3
@calc84maniac
Implement Flash cache timing, add Flash reset handler, improve Flash …
3dca1e5 
…mapping
@calc84maniac
Disallow Flash port writes when locked, fix Flash port mirrors on pre-M
f123ee6
@calc84maniac
Ignore SPI select bit in write value when entering sleep mode
53444f9
@calc84maniac
Add UART implementation with loopback support
c701ab4
@calc84maniac
Copy link
Contributor Author

Finally fixed the outstanding USB issue, last call for reviews / sanity testing

@adriweb
Copy link
Member

adriweb commented Jan 8, 2024

@calc84maniac said it's ready and people have tested it without issues so far, let's merge :)

@adriweb adriweb merged commit dd34079 into master Jan 8, 2024
5 checks passed
@adriweb adriweb deleted the feature/rev_m branch January 8, 2024 21:13
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@adriweb adriweb Awaiting requested review from adriweb

@runer112 runer112 Awaiting requested review from runer112

@drdnar drdnar Awaiting requested review from drdnar

@mateoconlechuga mateoconlechuga Awaiting requested review from mateoconlechuga

@jacobly0 jacobly0 Awaiting requested review from jacobly0

Assignees
No one assigned
Labels
core enhancement
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@calc84maniac @adriweb @jacobly0