shihab

7 ms·

1.
by shihab 2mo ago
US burnt almost entire latin american down last century because "can't have communism in our backyard", even though these countries are quite far geographically than Bahrain is to Iran. Besides, under every international law
2.
by shihab 7mo ago
I brought up Israeli-American donors because that’s what is relevant in the context of the story we’re discussing. We are talking about a war many right wing Israelis wanted for decades. If it were a general discussion about Citizens Unit
3.
by shihab 7mo ago
Exactly what part of my statement was dog whistling? Can you stop throwing around this serious accusation of antisemitism without any attempt to substantiate your claim?
4.
by shihab 7mo ago
Citizens United is an existential threat for USA. You cannot have Israeli-American dual citizens pouring $200 million dollars in elections. and that’s just her alone. This is simply not sustainable.
5.
by shihab 7mo ago
Another mid east war entirely on Israel’s behalf, another war Americans will pay tax for, die for- just so Israel can keep grabbing few parcels of lands from Palestine.
6.
by shihab 7mo ago
I think there are two steps here: converting video to sensor data input, and using that sensor data to drive. Only the second step will be handled by cars on road, first one is purely for training.
7.
by shihab 7mo ago
The article strictly talks about people who were pals with him _after_ his Pedophilia conviction. And please don't do this strawman "evil person eating babies", nobody sane is claiming that.
8.
by shihab 8mo ago
I work with GPUs and I'm also trying to understand the motivations here. Side note & a hot take: that sort of abstraction never really existed for GPU and it's going to be even harder now as Nvidia et al races to put more &amp
9.
by shihab 8mo ago
To the author (or anyone from vectorware team), can you please give me, admittedly a skeptic, a motivating example of a "GPU-native" application? That is, where does it truly make a difference to dispatch non-parallel/syscall
10.
by shihab 8mo ago
> For example, NEON ... can hold up to 32 128-bit vectors to perform your operations without having to touch the "slow" memory. Something I recently learnt: the actual number of physical registers in modern x86 CPUs are signifi
11.
by shihab 8mo ago
I'm not asking an academic program first published 8 year ago (e3nn) to beat actively developed CuEquivariance library. An academic proposing new algorithms doesn't need to worry too much about performance. But any new work which
12.
by shihab 8mo ago
I should note PETSc is a big piece of software that does a lot of things. It also wraps many libraries, and those might ultimately dictate actual performance depending on what you plan on doing.
13.
by shihab 8mo ago
To be practically useful, we don't need to beat vendors, just getting close would be enough, by the virtue of being open-source (and often portable). But I found, as an example, PETSc to be ~10x slower than MKL on CPU and CUDA on GPU;
14.
by shihab 8mo ago
Hi, I just wanted to note that e3nn is more of an academic software that's a bit high-level by design. A better baseline for comparison would be Nvidia's cuEquivariance, which does pretty much the same thing as you did- take e3nn
15.
by shihab 8mo ago
Hi, I actually mentioned ISPC several times there. And although I strenuously avoided crowning one approach "better" over the other, it is worth pointing out that 1) Many of these benefits of ISPC can be had from explicit SIMD lib
16.
by shihab 8mo ago
No. Assuming `k` is small enough, which in practice often is, the arithmetic intensity of this kernel is 25-90 Flops/Byte, way above the roofline knee of any modern CPU.
17.
by shihab 8mo ago
Hi, thanks for reading. Re (b) I'm curious what that middle ground is. Is there any simple refactor to help GCC to get rid of this `if`? (Note, ISPC did fine here) (c) Just to be clear, all the codes in benchmark figures (baseline and
18.
by shihab 8mo ago
Yeah N is big enough that entire data isn't in the cache, but the memory access pattern here is the next best thing: totally linear, predictable access. I remember seeing around 94%+ L1d cache hit rate.
19.
AVX-512: First Impressions on Performance and Programmability(shihab-shahriar.github.io)
125 points by shihab 8mo ago|53 comments
20.
by shihab 8mo ago
For SIMD at least, the {mins[3], maxs[3]} representation aligns more naturally with actual instructions on x86. To compute a new bounding box: new_box.mins = _mm_min_ps(a.mins[3], b.mins[3]);
21.
by shihab 8mo ago
If you are from ML/Data science world, the analogy that finally unlocked FFT for me is feature size reduction using Principal Component Analysis. In both cases, you project data to a new "better" co-ordinate system ("tim
22.
by shihab 8mo ago
I'd love to see a breakdown of what exactly worked here, or better yet, PR to upstream Abseil that implements those ideas. AI is always good at going from 0 to 80%, it's the last 20% it struggles with. It'd be interesting to
23.
by shihab 8mo ago
I was surprised by that defiant tone there in an official page. But it's missing actual numbers, which makes it all pretty strange.
24.
by shihab 11mo ago
I'm curious what you did with the "active sorting range" after a push/pop event. Since it's a vector underneath, I don't see any option other than to sort the entire range after each event, O(N). This would sur
25.
by shihab 11mo ago
> To Koreans, they looked more like sauce bowls, leading them to conclude that the Japanese had starved themselves to stretch out the siege. As a Bengali man, that's exactly how I felt when I came to USA and first visited japanese r
26.
by shihab 11mo ago
Could you please elaborate on your example? Thanks.
27.
Mojo: MLIR-Based Performance-Portable HPC Science Kernels on GPUs(arxiv.org)
6 points by shihab 1y ago|0 comments
28.
by shihab 1y ago
One feedback from someone interested in using this about the examples: I have looked at several and they seem too high level to get a sense of the actual API (i.e. the expected benefit of using this library vs the development complexity of
29.
by shihab 1y ago
That was government of qatar, this is Abu Dhabi (UAE). They had a diplomatic crisis, with full-scale blockade not so long ago.
30.
by shihab 1y ago
I’m a big, big fan of acquired podcast, listened to every single of their episodes. But I remember their entire Microsoft episodes felt like a lengthy defense of Steve Ballmer. There were too many instances of “here’s why this bad decision

More ›