walki

6 ms·

1.
by walki 2y ago
I feel like the NT kernel is in maintenance only mode and will eventually be replaced by the Linux kernel. I submitted a Windows kernel bug to Microsoft a few years ago and even though they acknowledged the bug the issue was closed as a &qu
2.
by walki 3y ago
Microsoft's %Appdata% directory is a security nightmare in my opinion. Ideally applications should only have access to their own directories in %Appdata% by default. I recently came across a python script on GitHub that allows to decry
3.
by walki 5y ago
> For performance, can’t you “just” have a swap-endianness instruction in your CPU Yes, most CPUs have special instructions for swapping between little and big endian byte arrangement. The GCC compiler has the __builtin_bswap64(x) for ac
4.
by walki 5y ago
> not sure if this is still the case though No this is not the case anymore. Nowadays support for unaligned memory accesses is very good on ARM and most other CPU architectures. On x86 aligned memory used to be very important for SIMD bu
5.
by walki 5y ago
> it’s good for software robustness that file formats and hardware use different endianness, as it forces you to read things byte-by-byte rather than lazily assuming you can just read 4 bytes and cast them directly to an int32. Except th
6.
by walki 6y ago
What's interesting to note though is that I tried marking pointers __restrict__ in the performance critical sections in 2 of my C++ projects and the assembly generated by Clang was identical in all cases! So while it is true that by de
7.
by walki 6y ago
OK thanks, indeed Clang is able to generate better assembly using __restrict__. And -O3 generates the same assembly as -O3 -fstrict-aliasing (which is not as good as __restrict__). I wish there was a C/C++ compiler flag for treating al
8.
by walki 6y ago
> C compilers have to assume that pointers to memory locations can overlap, unless you mark them __restrict... What I don't fully understand is: "GCC has the option -fstrict-aliasing which enables aliasing optimizations globall
9.
by walki 6y ago
> Has anyone tried machine learning on this or the Graviton2? I have not done any machine learning on AWS Graviton2 CPUs but I ran many other CPU benchmarks on Graviton2 CPUs and overall I have been disappointed by their performance. The
10.
by walki 6y ago
They also asked for the VAT number of the company...
11.
by walki 6y ago
Yes, I am based in Europe. If this was EU's fault then why didn't any of the other cloud vendors have this requirement?
12.
by walki 6y ago
I have personally used a large number of cloud compute services: AWS, packet, Alibaba, Digital Ocean, nimbix, Azure, Oracle, ... I still use all of these services from time to time. If you carefully look at my list of cloud vendors you real
13.
by walki 6y ago
> The facts: we gain almost nothing by having tiktok around. We lose nothing by banning it, and gain a little bit of buffer against possible threats like election meddling, data mining for nefarious purposes and other things. Completely
14.
by walki 6y ago
> It's worth pointing out how extremely far ahead Apple seems to be in terms of CPU power... I agree that Apple's ARM CPUs are very competitive on simple scalar instructions and memory latency/bandwidth. However x86/x
15.
by walki 6y ago
> Net result: they do not. For my use case (Clojure/JVM and ClojureScript compilation), compile times did not get shorter. There seemed to be a slight improvement, but it was below the level of measuring noise (which was around 8%).
16.
by walki 7y ago
I absolutely agree with this. The author has spent an insane amount of time to build all this stuff and instead of getting recognition and/or being payed adequately he is getting shitstorm after shitstrom. As an outsider I have to say
17.
by walki 7y ago
I absolutely think you made the right decision! I look up to you because you are not just another random math professor producing papers that no one ever reads. Instead you take a risk and try to succeed at what you are most passionate abou
18.
by walki 8y ago
For many packages it is difficult to get packaged for all Linux distributions. Hence if your package does not get packaged for a particular Linux distribution users can still install your package using homebrew.
19.
by walki 8y ago
> #aargh Many system programmers will think this line is commented out. Indeed what a nightmare!
20.
by walki 8y ago
No garbage collection.
21.
by walki 8y ago
> I've recently dropped out of highschool and I've been considering two options: promptly returning to school and finishing my degree or dedicating time to self-educate via the internet. There is a 3rd option (if you want to wo
22.
by walki 8y ago
Gnome has never been hosted on GitHub!
23.
by walki 8y ago
Agreed, this is ridiculous.
24.
by walki 9y ago
> ... Glibc are evidently not the best The GNU GCC compiler and the glibc are actually the 2 GNU projects I like most. If you are interested in high performance you won't find any better, most Linux servers (which means 99% of cloud
25.
by walki 9y ago
> How does, e.g., POWER8 compare? (And POWER9, soon?) POWER8 is not well suited for HPC workloads as their vector instructions (VMX/AltiVec) are only 128-bit wide vs Intel's latest Skylake Xeon CPUs have vector instructions (AV
26.
by walki 10y ago
> Just out of curiosity, why would tax competition be considered bad? This is not about tax competition, this is about abusing tax loop holes and trying to avoid paying any tax at all. E.g. before 2015 Amazon declared their entire Europe
27.
by walki 10y ago
> This is ongoing - enabled by Junker himself ... No, it is not. Before 2015 many companies (one prominent example being Amazon) declared their entire European income in Luxembourg and payed about 1 percent tax on that money. This tax lo
28.
by walki 10y ago
I am a Luxembourger and I fully agree that the EU must close these kinds of tax loop holes. In the past my country made a little money out of these tax loop holes, this was immoral money that my country did not even need. Doing this, other
29.
by walki 10y ago
I compiled my code using the latest GCC version available, I even made sure I used the right compiler flags to enable e.g. hardware POPCNT support. I know that IBM claimed that POWER8 CPUs were faster than Intel Xeon CPUs but I found the op
30.
by walki 10y ago
When POWER8 came out in 2014 I ran a couple of numerical benchmarks against Intel Xeon CPUs. In my benchmarks the POWER8 CPUs where slightly faster than Intel Xeon CPUs when the data fitted into the CPU's cache, so far so good (the spe

More ›