Re: TLS
Finn Thain dixit:
>Probably adding syscalls could fix the 32-bit time_t without breaking
>anything (given symbol versioning in libc).
No. You need to rebuild everything that depends on libc, too.
>Fixing everything (as in a new ABI) means simultaneous changes to kernel,
[…]
Yes, I know. Which is why OpenBSD piled on a few such changes
before doing them all at once. I did the 64-bit time_t dance
in MirBSD/i386 almost a decade ago, too. It does indeed invalidate
all old binaries (including .o .a .so).
And that may not be a bad thing.
Besides, even with libc symbol versioning, old and new stuff
could not coexist without major version bumping either, and
shlib versions are usually set upstream, not by packagers,
even.
>Better perhaps to add stuff to the existing ABI/API where possible. A new
At some point: not really, any more.
>> >get/set_thread_area to a VDSO (as well as the atomic ops). As I
>> >understand it, this need not break the ABI and should be faster than
>> >the TLS system calls.
>>
>> It will not break the ABI, but how'd it work?
>>
>> - One kernel/user shared, readonly, page global (for time, atomic)
>> - One kernel/user shared page per thread (instead of process), for TLS
>> (and possibly random)?
>
>Whatever is easiest to cut and paste from some other architecture, I guess
Really?
I’d wager that something that’s about to become kernel/userspace ABI
and has such an impact should be designed with m68k (including Coldfire)
in mind, and optimised for it. (Of course, looking at the other arches
is not bad.)
Of course, this is all just talk, nobody is doing the work.
bye,
//mirabilos (still BSD’er by heart, not understanding why Linux
wants their binary compatibility… after all, everything
is Open Source and can just be recompiled)
⇒ https://lwn.net/Articles/563285/
--
<gcc> ncal.c: In function 'parsemonth': warning: comparison between pointer
and integer • <mirabilos> ↑ hab da „in function parselmouth“ gelesen
<Natureshadow> ICH AUCH! • <Natureshadow> Ich hab gerade gedacht "Häh? Wie,
hab da parselmouth gelesen ... steht da doch auch :o?" -- too much fanfic…
Reply to: