Hi,

I just saw your 38059770 commit. Thanks for the Windows TLS work and 
the DEF parser follow-up.

Indeed, I over-scoped Richard's report and pushed too quickly.
I should have preserved his authorship, and the unwrapped commit 
messages are my mistake.

The wider work was prompted by concrete cases I noticed while tracing
the code: `_end` is created before copy relocations can extend `.bss`.
Separately named allocatable sections can lie beyond the canonical
sections, and a symbol can be visible only through an input DSO.

`char end = 7` was deliberately testing ELF symbol preemption, and the
non-writable-section scan was an attempt to derive the boundary from
the final layout. Thus, these additions were not random, but I
agree that they are separate linker-policy questions and didn't
belong in Richard's focused FreeBSD fix.

So yes: I think Richard's original three-line BSD fix should land under
his authorship.

I can prepare a normal cleanup commit that preserves your changes and 
removes my broader ELF work, unless you prefer to do the cleanup on 
your side.

Regards,

Mounir IDRASSI



From: grischka via Tinycc-devel <[email protected]>
To: <[email protected]>
Cc: "grischka"<[email protected]>
Date: Wed, 29 Jul 2026 07:20:49 +0900
Subject: Re: [Tinycc-devel] [PATCH] tccelf: define plain etext/edata/end linker 
symbols on BSD targets

 > On 28.07.2026 14:38, Mounir IDRASSI via Tinycc-devel wrote:
 > > Hi Richard,
 > >
 > > Thanks for the detailed report and, especially, for testing the fix with a 
 > > real libgc build on FreeBSD. The missing plain symbols were indeed the 
 > > root cause.
 > >
 > > I pushed the resulting work to mob as two commits:
 > 
 > Hi,
 > 
 > I'd normally recommend not to push other people's work, for several
 > reasons, such as authorship or precision wrt. addressing a real problem.
 > 
 > Btw. convention with commit messages is to break lines at length
 > of 40 or so.  Otherwise hard to read them in gitk or web interfaces.
 > 
 > As always, the more lines a patch has the more more problems it tends
 > to have.  For example
 > 
 > >  char end = 7;
 > 
 > Obviously under no circumstances this can mean the end of something
 > and have value '7' at the same time.
 > 
 > >         if (!(s->sh_flags & SHF_WRITE))
 > 
 > Not sure whether tcc would want .plt or .note.ABI-tag count as text
 > sections.
 > 
 > So, before we throw even more code at that maybe we should just go
 > with Richards original simple 3 line patch?  WDY think?
 > 
 > ...etc
 > 




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

Reply via email to