Open wmat opened 11 months ago
Hi Bill,
The trickiest part is probably dealing with the way we handle registers. Their definitions are in https://github.com/riscv/riscv-debug-spec/tree/master/xml and there is some LaTeX markup mixed in there as well. Then https://github.com/riscv/riscv-debug-spec/blob/master/registers.py takes that XML and turns it into LaTeX figures/tables. Modifying that script is probably not too bad, to change the output to asciidoc + wavedrom.
The way this project has been presented to me is that you'll do the straightforward 80% of this project, leaving the remaining 80% to debug group. Is that right?
Tim
That's correct, I'll do the bulk of the conversion but hope that things like tweaking the scripts to output asciidoc instead of latex can be done by the debug spec authors.
On Wed, Sep 13, 2023 at 6:33 PM Tim Newsome @.***> wrote:
Hi Bill,
The trickiest part is probably dealing with the way we handle registers. Their definitions are in https://github.com/riscv/riscv-debug-spec/tree/master/xml and there is some LaTeX markup mixed in there as well. Then https://github.com/riscv/riscv-debug-spec/blob/master/registers.py takes that XML and turns it into LaTeX figures/tables. Modifying that script is probably not too bad, to change the output to asciidoc + wavedrom.
The way this project has been presented to me is that you'll do the straightforward 80% of this project, leaving the remaining 80% to debug group. Is that right?
Tim
— Reply to this email directly, view it on GitHub https://github.com/riscv/riscv-debug-spec/issues/873#issuecomment-1718400119, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAN6ZCKU3F7IITLC6VMDJTX2IYFDANCNFSM6AAAAAA4VLJW7Y . You are receiving this because you authored the thread.Message ID: @.***>
If you can convert some of the LaTeX output for one file into the relevant asciidoc and wavedrom so I have something to go on, I can change the script to output that format.
E.g. make dm_registers.tex
and then show me what that file should look like up to (not including) the dmcontrol subsection.
Hi Tim,
I'm just getting back to your spec now. In looking at the src tree, I don't see a dm_registers.tex file. I see it referenced in the Makefile but don't see it in the src. Am I looking in the wrong place?
Thanks Bill
On Thu, Sep 14, 2023 at 2:23 PM Tim Newsome @.***> wrote:
If you can convert some of the LaTeX output for one file into the relevant asciidoc and wavedrom so I have something to go on, I can change the script to output that format.
E.g. make dm_registers.tex and then show me what that file should look like up to (not including) the dmcontrol subsection.
— Reply to this email directly, view it on GitHub https://github.com/riscv/riscv-debug-spec/issues/873#issuecomment-1719935138, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAN6ZASNYBZCL5DM2YTLUDX2NDQNANCNFSM6AAAAAA4VLJW7Y . You are receiving this because you authored the thread.Message ID: @.***>
make dm_registers.tex
should get you the file. I'll attach it here as well, renamed to .txt otherwise github won't let me.
dm_registers.txt
FWIW, when I do make dm_registers.tex the result is an empty file even though it appears to have worked. I can just use your file you attached.
make dm_registers.tex make: 'dm_registers.tex' is up to date.
On Mon, Sep 25, 2023 at 12:03 PM Tim Newsome @.***> wrote:
make dm_registers.tex should get you the file. I'll attach it here as well, renamed to .txt otherwise github won't let me. dm_registers.txt https://github.com/riscv/riscv-debug-spec/files/12717362/dm_registers.txt
— Reply to this email directly, view it on GitHub https://github.com/riscv/riscv-debug-spec/issues/873#issuecomment-1734045290, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAN6ZGODO6L5P47B2HBF3TX4GTNVANCNFSM6AAAAAA4VLJW7Y . You are receiving this because you authored the thread.Message ID: @.***>
The Makefile isn't very good. Probably the file was generated but there was an error while generating it. So now make
won't regenerate it because it already exists.
@rtwfroody just want to ask if there's anything left to do for the asciidoc conversion?
I think it looks good, although I look forward to bringing in a list of table and figures once you figure out the best way to do that.
I might tweak the diagrams more at some point (really wish I could left/right align the bit offsets), but they're good enough.
Whenever you open an issue around formatting, feel free to @ me so I don't miss it. If I can help, I certainly will.
Opening this issue to track the conversion of the LaTeX specification to AsciiDoc. Conversion will be done on the asciidoc branch.