On Wed, Jun 20, 2018 at 03:16:28PM -0400, Steve Litt wrote: > There was a discussion of whether to retain 32 bit, and that brought up > another question in my mind: When will we have 128 bit computing? > > > 1971: 4bit: Intel 4004 > 1974: 8bit: Intel 8080 > 1978: 16bit: Intel 8086 > 1985: 32bit: Intel 20386 > 2003: 64bit: AMD Opteron / Pentium 4 EO revision > > When are we going to have 128 bit microprocessors?
riscv128. There's only an emulator for it yet, and neither gcc nor Linux implemented support. It turns out gcc has maximum word width hard-coded in a lot of places, thus adding 128-bit there will be a lot of work. No one is insane enough to try to extend x86 this way. Interesting that your list includes only x86 and predecessors, which didn't win in any of the categories. For physical memory, 64TB is already not enough for everyone, and we had a bump to 5-level paging in the kernel recently. Someone raised the question when the next bump will happen, and IIRC 4-level was 12 years before. The old limit was 46 bit physical 48 bit virtual, new is 52 physical 57 virtual. For disk storage, there already are server farms with more than 2⁶⁴ bytes, but AFAIK no one put them into a single address space. Meow! -- ⢀⣴⠾⠻⢶⣦⠀ There's an easy way to tell toy operating systems from real ones. ⣾⠁⢰⠒⠀⣿⡁ Just look at how their shipped fonts display U+1F52B, this makes ⢿⡄⠘⠷⠚⠋⠀ the intended audience obvious. It's also interesting to see OSes ⠈⠳⣄⠀⠀⠀⠀ go back and forth wrt their intended target. _______________________________________________ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
