mshockwave

4 ms·

1.
RISC-V Extension Landscape(rpsene.github.io)
4 points by mshockwave 1mo ago|0 comments
2.
by mshockwave 1mo ago
I disagree, compiler optimization and software ecosystem in general already take a really really really long time to build, and I think it'll take longer without a standardized low-level binary interface i.e. ISA to enable rapid distri
3.
RVV Benchmark SiFive P870 (Lanxin LX500, Epic Semi Contrail AIx)(camel-cdr.github.io)
1 points by mshockwave 1mo ago|0 comments
4.
by mshockwave 2mo ago
or have little vision to begin with
5.
by mshockwave 3mo ago
one of the reasons I rarely read press releases is that I don't believe in promises -- I believe in _incentives_. In this case, what will Qualcomm be incentivized to do? What are in their interests?
6.
by mshockwave 3mo ago
indeed, open sourcing is only half (or even less) of the picture: who is driving the open source community and how it is driven (i.e. governing structure) are probably more important IMHO. There are countless of cases where an open source p
7.
by mshockwave 5mo ago
it's a book excerpt
8.
by mshockwave 5mo ago
> Does gcc use LLVM anywhere under the hood No
9.
by mshockwave 6mo ago
One thing I observed is that RVV code is usually slower in QEMU
10.
by mshockwave 6mo ago
LLVM now has another way to implement RTTI using the `CastInfo` trait instead of `classof`: https://llvm.org/doxygen/structllvm_1_1CastInfo.html But it's really just an implementation difference, the idea is still
11.
by mshockwave 7mo ago
how did it do regalloc before instruction selection? How do you select the correct register class without knowing which instruction you're gonna use?
12.
RISC-V Vector Primer(github.com)
4 points by mshockwave 7mo ago|0 comments
13.
by mshockwave 8mo ago
> I don’t know many good reasons for extrusive linked lists for one, its iterator won't be invalidated
14.
by mshockwave 10mo ago
Is it normal to spend 10minutes on tuning nowadays? Do we need to spend another 10 minutes upon changing the code?
15.
by mshockwave 10mo ago
It's likely that Swift compiler is using LLVM LIT ( https://llvm.org/docs/CommandGuide/lit.html ), which is implemented in python, as the test driver
16.
Machine Scheduler in LLVM – Part II(myhsu.xyz)
31 points by mshockwave 11mo ago|0 comments
17.
by mshockwave 1y ago
> In the end, programs will want probably to stay conservative and will implement only the core ISA Unlikely, as pointed out in sibling comments the core ISA is too limited. What might prevail is profiles, specifically profiles for appli
18.
by mshockwave 1y ago
yes, it has been done for at least a decade if not more > Even more of a wild idea is to pair up two cores and have them work together this way I don't think that'll be profitable, because... > When you have a core that woul
19.
by mshockwave 1y ago
The article is easy to follow but I think the author missed the e point: branchless programming (a subset of the more known constant time programming) is almost exclusively used in cryptography only nowadays. As shown by the benchmarks in t
20.
by mshockwave 1y ago
yes, the short answer is LLVM uses RegPressureTracker ( https://llvm.org/doxygen/classllvm_1_1RegPressureTracker.htm... ) to do all those calculations. Slightly longer answer: I should probably be a little more specific
21.
Machine Scheduler in LLVM(myhsu.xyz)
1 points by mshockwave 1y ago|0 comments
22.
by mshockwave 1y ago
The scenario suitable for TBM is surprisingly limited so even nowadays many tunnels are still dig using the good’o way, mostly with explosives
23.
by mshockwave 1y ago
Came to say Apple also did a great job on tagging my bois who are both grey-ish cats, even in pictures they faced backward, no idea how they did that
24.
by mshockwave 1y ago
I thought the original comment meant “_for_ who doesn’t use X or Discord, here is the github mirror link”. There’s a “for” missing, and thus I think they agree with you
25.
by mshockwave 1y ago
second this, it didn't show anything when I hovered over North America
26.
Jelu: Self hosted read and to-read list book tracker(github.com)
11 points by mshockwave 1y ago|0 comments
27.
by mshockwave 1y ago
and grep / ripgrep. Or did ripgrep migrate to using io_uring already?
28.
by mshockwave 1y ago
tangent: I believe Air Force and Navy have loosen the vision requirement quite a bit. IIRC the new rule only measures the corrected vision, regardless how you corrected it, including wearing glasses.
29.
by mshockwave 1y ago
> RVV is the hardest though (20k intrinsics) A bit late to this comment but most of these intrinsics are overloads of different LMUL and SEW on a single instruction. I'm pretty sure the actual number of RVV instructions is way less.
30.
GlobalFoundries to Acquire MIPS(mips.com)
243 points by mshockwave 1y ago|148 comments

More ›