Can we drop upstream Linux x32 support?
I like the idea of x32 in theory: any machine with 4GB or less of memory for sure does not really need to waste any precious space storing 64-bit pointers; and even if more memory is available, software that needs to access more than 4GB per process is somewhat rare (as it should be). I used a laptop every day with 4GB until recently, so was generally trying to run relatively lean software. I have played around with the Gentoo x32 variant on another machine, just to see how that would go. I haven't used it for much... but tried out some basic things; and as long as it is powered on, it functions fine as a node in an icecream cluster, too. The icecream statistics shown in icemon tell me that this machine seems to be performing faster than it should be, for its age. So I think there is some worthwhile average speedup (without being more scientific about it).
As a Qt developer, I have some interest that Qt and Qt Quick ought to work well on this architecture. (When it breaks, I write up bugs like QTBUG-52658 and QTBUG-42759.) There is still no JIT support for QML, but perhaps there could be (I could try, if I ever find time to learn how the existing QML/V4 JIT works on other architectures well enough to add another one); and meanwhile, at least the interpreter works AFAIK. I was just thinking of testing this again, maybe running gentoo or debian x32 on that old 4GB laptop; so I did a quick google search, and found this thread right away.
Linux supports a lot of rarely-used hardware and features; so as long as the burden of supporting x32 is not too great, I'd like to put in my vote to keep it working. And I will keep testing from time to time and do my bit to keep Qt working on this architecture, as long as I'm not wasting my time (if it's going to disappear, it would be a waste of time).
I can't explain why there aren't more users, and more distros. It seems to me it would be a good choice for the compact or small-is-beautiful distros, like Puppy.
(CC me on this thread if you like, since I'm not currently subscribed to LKML)
Quoted 4 lines · Shawn Rutledge
I like the idea of x32 in theory: any machine with 4GB or less of memory for sure does not really need to waste any precious space storing 64-bit pointers; and even if more memory is available, software that needs to access more than 4GB per process is somewhat rare (as it should be).
You can't mmap(2) files the size of which exceeds the size of the 32-bit
user address space with an ILP32 ABI though, regardless of how much physical memory a given system supports or has actually installed.
Also you can have a process's VM size exceed the 32-bit user address
space where swap files are used, e.g. I have a system with 3GiB of DRAM installed and as much swap space available, where the amount of memory consumed by a process sometimes exceeds 4GiB -- it's enough to open a couple of web pages at once in a browser these days to get there (don't ask me what it uses all that memory for).
So there are still use cases for an LP64 ABI even for systems the whole
physical memory of which can be handled with 32-bit addressing.
Maciej