IMO one good direction for tcc improvement is to make it able to compile 
more/most recent Linux kernels. I think you might be interested in this too)
Of course it's not of high priority, but it's good showcase for tcc.

Vad Rulezz

9 aug 2025 г. 23:18:34 GMT+03:00, Michael Ackermann via Tinycc-devel 
<[email protected]> пишет:
>I would propose a trivial criteria for the required C language feature set:
>a _complete_ operating system distribution must compile and become bootable as
>first priority with whatever tinycc was capable to compile and link already.
>That's mostly C99 excluding complex data type currently
>(which can be implemented outside compiler and libc).
>
>The following package set passed completely with tinycc already:
>http://tinyfront.mooo.com/downloads/x86-embedded/666666/tinyfront-packages.list
>The userspace-fork is discussed here:
>http://tinyfront.mooo.com/docs.html#Userspace-Fork
>
>TinyCC kernel support is published already.
>
>In this regard TinyCC would not need further C11 extensions implemented,
>and i would propose to keep in mind acceptance criteria outlined with regards 
>to
>standards compliance:
>http://tinyfront.mooo.com/docs.html#Acceptance-Criteria
>
>Once temperatures cooled back down i'll get back to the tinycc-compiled
>bash.static bug to retain the whole system-integration self-hosting with the
>i586-tinycc-linux2-musl.iso system. That's one major blocker remaining to
>cope with before uploading the devdrop. It's much easier if development
>could be coordinated with a common baseline distribution and the fully forked
>portage tree maintained within git, and _all_ development utilities being
>available, self-hosting with tinycc i mean.
>
>Focus on C11 itself opens a can of worms i rather not touch myself, which was
>one reason why i've patched linux-5.9 and finally forked away everything.
>One such C11 related problem was outlined in the kernel section here:
>http://tinyfront.mooo.com/docs.html#Kernel
>
>Long story short, except for some bugs and quirks, tinycc already got all
>features implemented which were necessary to compile an extended set of
>packages of mostly most recent software versions.
>
>Anyway, that's my 2cents from the system integration perspective of a 
>complete(!)
>tinycc driven operating system.
>
>Regards
>
>On 2025-08-09 11:16, [email protected] wrote:
>> For known bugs, yeah.
>> 
>> But as stated before, first and foremost is to set a goal.
>> 
>> 1- What should be present in the 1.0.0 release ?
>> 
>> Full C99 or C11 support ? C11 is safer and more "up to date" :
>> 
>> https://web.archive.org/web/20200806193736/https://smartbear.com/blog/test-and-monitor/c11-a-new-c-standard-aiming-at-safer-programming/
>> 
>> Hence C11 should be getting way more prevalent in the existing source code 
>> out there :
>> 
>> https://www.geeksforgeeks.org/c/c-programming-language-standard/
>> 
>> 2- Then do a full analysis of the current source code (static analysis, 
>> conformance test suites, etc).
>> 
>> 3- Spread the tasks between bug fixing, feature support, etc.
>> 
>> 4- Write tickets in a VSC like GitHub (or Savannah, provided everyone stick 
>> to it).
>> 
>> 5- Set a roadmap and deliverables with expected feature support and bug 
>> fixes, not necessarily deadlines (0.9.27 has been out for a while).
>> 
>> 6- Set an easy to use cross platform working environnent (unzip and play) 
>> for volunteers to debug and code.
>> 
>> 7- Get volunteers to take charge of tickets and carry out their holy duty.
>> 
>> 8- Get their job peer reviewed, merge their commits from separate 
>> development branch into the main one (depending on the workflow used) :
>> 
>> https://nvie.com/posts/a-successful-git-branching-model/
>> 
>> https://medium.com/@sreekanth.thummala/choosing-the-right-git-branching-strategy-a-comparative-analysis-f5e635443423
>> 
>> https://dev.to/juniourrau/6-types-of-git-branching-strategy-g54
>> 
>> https://www.atlassian.com/git/tutorials/comparing-workflows/gitflow-workflow
>> 
>> https://docs.gitlab.com/user/project/repository/branches/strategies/
>> 
>> 9- Each intermediate (0.9.30, ...) and the final release (1.0.0) should be 
>> performed by grischka once he verified each ticket has been done and tested.
>> 
>> 10- Reach 1.0.0 in a timely fashion (our lifetime)
>> 
>> Deal ?
>> 
>> 
>> 
>> 
>> ----- Mail d'origine -----
>> De: Robin Rowe <[email protected]>
>> À: [email protected]
>> Envoyé: Sat, 09 Aug 2025 07:00:39 +0200 (CEST)
>> Objet: Re: [Tinycc-devel] VERSION Number 1.0
>> 
>> >> What is the tcc bug reporting/ticketing system?
>> > You can get some "tracking" there : 
>> > https://savannah.nongnu.org/projects/tinycc
>>  > So right now we're at 0.9.28rc but a roadmap should be established.
>> 
>> Thank you, David. That lists 84 bugs. For a roadmap, perhaps the bug 
>> list can be divided to prioritize C99 breakage over C11 support? As a 
>> path toward a more agile release tempo, have C99 be version 1.0 and C11 
>> to follow?
>> 
>> Robin
>> 
>> _______________________________________________
>> Tinycc-devel mailing list
>> [email protected]
>> https://lists.nongnu.org/mailman/listinfo/tinycc-devel
>> 

_______________________________________________
Tinycc-devel mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/tinycc-devel

Reply via email to