News for hackers. Stuff that matters.
Re:Systemd is bad parsing by someguydave on Saturday October 27, 2018 @12:13PM (#18316353)
The long term solution is to burn down x86 and C and start over with sane hardware, which has bounds checking in hardware.
Re:Systemd is bad parsing by chmod775 on Saturday October 27, 2018 @12:24PM (#18316419)
> bounds checking in hardware.

What. That's a horrible idea. Not only would that require your hardware to understand abstractions several layers above what it actually operates on, it would also mean it'd have to check bounds of - for example - an array on every instruction accessing it. That is something compilers can do at more strategic places.

Please keep that complexity out of my CPUs, they're buggy enough already. At least with software we can fix issues later, which isn't a given with hardware.

Re:Systemd is bad parsing by marcosdumay on Saturday October 27, 2018 @12:37PM (#18316475)
LoadAddressIfSmallerThan makes for a long opcode name, but it doesn't look much more complex than your usual load. The added complexity is negligible.

And, of course, the added complexity of doing that in two instructions is also negligible. The entire problem seems to be that arrays are on a more abstract level than C operates. It looks like entirely a language problem.

Re:Systemd is bad parsing by pjc50 on Saturday October 27, 2018 @12:38PM (#18316480)
Having the CPU do the checking within an instruction saves you instruction issue events. ARM have an interesting solution to this, pointer authentication: https://news.ycombinator.com/item?id=14479438

> several layers above what it actually operates on

The instruction set is something of an abstraction bottleneck - C is tied to machines that look something like a PDP-11, while a lot of the stuff that the machine is doing is invisible at the instruction set layer.

Re:Systemd is bad parsing by makomk on Saturday October 27, 2018 @07:04PM (#18318190)
Pointer authentication is not a form of bounds checking. It solely exists to prevent a malicious attacker using a bug to directly overwrite pointers (and in practice, it's mostly limited to return addresses right now). Code which does buggy pointer arithmetic will still create valid pointers to data outside the intended buffer.
Re:Systemd is bad parsing by DblPlusUngood on Saturday October 27, 2018 @12:33PM (#18316457)
Turns out the x86 already has some bounds checking support; see chapter 17 in volume 1 of Intel's developer manuals and the BOUND and BND* instructions.
Re:Systemd is bad parsing by hyperman1 on Saturday October 27, 2018 @05:03PM (#18317635)
In fact they were removed in 64 bit mode. Same for INTO
Re:Systemd is bad parsing by sseth on Saturday October 27, 2018 @01:30PM (#18316752)
I worked (a long time ago) on Burroughs mainframes which did bounds checking of arrays (and in fact even type checking), because the whole machine was build around ALGOL.

I am not sure whether the experience from these machines was negative. It may be that when you move beyond one-dimensional arrays, bounds checking becomes more complicated.

Perhaps performance suffers too. Languages which do bounds checking often have intelligent compilers which can skip the bounds check in a variety of conditions. So even if the hardware is faster, it may be doing unnecessary checks. See for example : https://www.ardanlabs.com/blog/2018/04/bounds-check-eliminat...

Re:Systemd is bad parsing by qznc on Saturday October 27, 2018 @01:36PM (#18316782)
I also believe that bounds checking in hardware is not worth it. Memory accesses dominate the runtime such that a few extra instructions for bounds checking is negligible.
Re:Systemd is bad parsing by sly010 on Saturday October 27, 2018 @01:10PM (#18316655)
> which has bounds checking in hardware.

x86 has bounds checking in hardware. it's called a page fault.

[edit: formatting]

Re:Systemd is bad parsing by erik_seaberg on Saturday October 27, 2018 @04:21PM (#18317484)
Segmentation was expensive (loading an 80286 segment register was very slow and the segment table was impractically small) but could express "offsets between 0 and 42 are valid". Paging is much coarser and suffers from false negatives, because x[y] might accidentally walk off the end of x, skip a guard page, and land on a mapped page from another object.
Re:Systemd is bad parsing by insertcredit on Saturday October 27, 2018 @05:32PM (#18317776)
Not quite true: There would not exist exploitable buffer overflows if that was the case.
Re:Systemd is bad parsing by archgoon on Saturday October 27, 2018 @12:23PM (#18316410)
I'm curious; what does your memory hardware and cpu instruction set look like in this scenario? What's the interface between the cpu and memory? (rough idea, not asking for a whitepaper here :) )
Re:Systemd is bad parsing by someguydave on Saturday October 27, 2018 @11:15PM (#18318993)
Consider the scheme 79 chip: http://dspace.mit.edu/handle/1721.1/6334