But this is not "backwards compatibility". No one promises this type of "forward compatibility" that you are asking for . Even win32 only does it exceptionally... maybe today you can still build a win10 binary with a win11 toolchain, but you cannot build a win98 binary with it for sure.

And this has nothing to do with 1996, or 2004 glibc at all. In fact, glibc makes this otherwise impossible task actually possible: you can force to link with older symbols, but that solves only a fraction of the problem of what you're trying to achieve. Statically linking / musl does not solve this either. At some point musl is going to use a newer syscall, or any other newer feature, and you're broke again.

Also, what is so hard about building your software in your "security updates only" server? Or a chroot of it at least ? As I was saying below, I have a Debian 2006-ish chroot for this purpose....

▲

marcosdumay 5 hours ago | parent [-]

Windows dlls are forward compatible in that sense. If you use the Linux kernel directly, it is forward compatible in that sense. And, of course, there is no issue at all with statically linked code.

The problem is with the Linux dynamic linking, and the idea that you must not statically link the glibc code. And you can circumvent it by freezing your glibc abstraction interface, so that if you need to add another function, you do so by making another library entirely. But I don't know if musl does that.

▲

AshamedCaptain 5 hours ago | parent [-]

> Windows dlls are forward compatible in that sense.

If you want to go to such level, ELF is also forward compatible in that sense.

This is completely irrelevant, because what the developer is going to see is the binaries he builts in XP SP3 no longer work in XP SP2 because of a link error: the _statically linked_ runtime is going to call symbols that are not in XP SP2 DLLs (e.g. the DecodePointer debacle).

> If you use the Linux kernel directly, it is forward compatible in that sense.

Or not, because there will be a note in the ELF headers with the minimum kernel version required, which is going to be set to a recent version even if you do not use any newer feature. (unless you play with the toolchain) (PE has similar field too, leading to the "not a valid win32 executable" messages).

> And, of course, there is no issue at all with statically linked code.

I would say statically linked code is precisely the root of all these problems.

In addition to bring more problems of its own. E.g. games that dynamically link with SDL can be patched to have any other SDL version, including one with bugfixes for X support, audio, etc. Games that statically link with SDL? Sorry..

> And you can circumvent it by freezing your glibc abstraction interface, so that if you need to add another function, you do so by making another library entirely. But I don't know if musl does that.

Funnily, I think that is exactly the same as the solution I'm proposing for this conundrum: just (dynamically) link with the older glibc ! Voila: your binary now works with glibc from 1996 and glibc from 2026.

Frankly, glibc is already the project with the best binary compatibility of the entire Linux desktop , if not the only one with a binary compatibility story at all . The kernel is _not_ better in this regard (e.g. /dev/dsp).

▲

marcosdumay 4 hours ago | parent [-]

If you use only features available on the older version, for sure, you can compile your software in Win-7 and have it run in Win-2000. Without following any special procedure.

I know, I've done that.

> just (dynamically) link with the older glibc!

Except that the older glibc is unmaintained and very hard to get a hold of and use. If you solve that, yeah, it's the same.

▲

AshamedCaptain 3 hours ago | parent [-]

> If you use only features available on the older version, for sure, you can compile your software in Win-7 and have it run in Win-2000. Without following any special procedure.

No, you can't. When you use 7-era toolchain (e.g. VS 2012) it sets the minimum client version in PE header to Vista, not XP much less 2k.

If you use VC++6 in 7, then yes, you can; but that's not really that different from me using a Debian Etch chroot to build.

Even within XP era this happens, since there are VS versions that target XP _SP2_ and produce binaries that are not compatible with XP _SP1_. That's the "DecodePointer" debacle I was mentioning. _Even_ if you do not use any "SP2" feature (as few as they are), the runtime (the statically linked part; not MSVCRT) is going to call DecodePointer, so even the smallest hello world will catastrophically fail in older win32 version.

Just Google around for hundreds of confused developers.

> Except that the older glibc is unmaintained and very hard to get a hold of and use.

"unmaintained" is another way of saying "frozen" or "security updates only" I guess. But ... hard to get a hold of ? You are literally running it on your the "security updates only" server that you wanted to target in the first place!

▲

CSm1n 2 hours ago | parent [-]

> No, you can't. When you use 7-era toolchain (e.g. VS 2012) it sets the minimum client version in PE header to Vista, not XP much less 2k.

Yes, you can! There are even multiple Windows 10 era toolchains that officially support XP. VS 2017 was the last release that could build XP binaries.

	▲	AshamedCaptain 2 hours ago \| parent [-]
		"Without following any special procedure". I know you can install older toolchains and then build using those, but I can do as much on any platform (e.g. by using a chroot). The default on VS2012 is Vista-only binaries.