trentnelson

25 ms·

91.
by trentnelson 8y ago
On the plus side, it only takes, like, 3 hours to read end-to-end.
92.
Is Prefix of String in Table? a Journey into SIMD String Processing(trent.me)
181 points by trentnelson 8y ago|60 comments
93.
by trentnelson 8y ago
You can't do Aho-Corasick or any form of tree/trie structure (practically) without using pointers, and any form of pointer-chasing algorithm is going to have a tough time competing with the lower bounds of this algorithm, when we&
94.
Is Prefix of String in Table? a Journey into SIMD String Processing(trent.me)
5 points by trentnelson 8y ago|3 comments
95.
by trentnelson 8y ago
So, basically, we need the NT-style, completion-oriented, inherently-asynchronous I/O model? ;-) (Amen!)
96.
by trentnelson 8y ago
This injection thunk I wrote in order to do facilitate remote code and context (file handles, event handles, shared memory maps etc) injection: https://github.com/tpn/tracer/blob/master/Asm/Injecti
97.
by trentnelson 9y ago
Similar concept to GVFS: https://blogs.msdn.microsoft.com/devops/2017/02/03/announcin...
98.
by trentnelson 9y ago
Dave Cutler got a call from Bill Gates to build a next generation OS when he was 47! Circa 1988-89.
99.
by trentnelson 9y ago
You can get the best of both worlds with PyParallel! Async I/O and multiple threads. (Experimental project, not intended for production use, so I say this some what facetiously.) http://pyparallel.org/
100.
by trentnelson 9y ago
It's a very different paradigm to wrap your head around, but once you grok the NT kernel's approach to I/O (packet based IRPs, inherently asynchronous, thread-agnostic), and thread scheduling, I/O completion ports are ve
101.
by trentnelson 9y ago
How does this compare to SAL, out of interest? Somewhat competing aims, no?
102.
by trentnelson 9y ago
That's a funny one. The kernel changes they needed to make in order to support transactional stuff were pervasive, and it's used extensively under the covers. The MSDN commentary makes it sound like they're going to complet
103.
by trentnelson 9y ago
Yeah that's one of the things Cutler played an important role in -- formalizing x64 prologue and epilogue constructs. The epilogue stuff is interesting, there's a small set of instructions that are allowed because the unwind code
104.
by trentnelson 9y ago
There was a third leak, circa 2003, the Windows Research Kernel, which is what you'd get if, say, you had an academic license as part of a university or something. That one you could very much build and alter.
105.
by trentnelson 9y ago
Technically three Windows source code leaks: Windows NT 4.0, Windows 2000 and the "Windows Research Kernel" (Windows 2003). All moot given that the biggest leak in history was obviously the Falcon 4.0 source code.
106.
by trentnelson 9y ago
You don't store your plaintext passwords in a modded version of Doom with a kernel scheduler hack to ensure it mostly gets dispatched on the HyperThread core neighboring the one your browser is getting dispatched onto? Amateur.
107.
by trentnelson 9y ago
It's a great book. Motivated me to switch industries and relocate to Seattle as an NT kernel programmer.
108.
by trentnelson 9y ago
> For long-running, batch-oriented programs, though, where there are no pathological corner cases, the usefulness of cProfile diminishes. It becomes hard to distinguish signal from noise with regards to the cumulative call counts and fun
109.
by trentnelson 9y ago
> Hm interesting -- what parts of it are tied to Windows? If I had to guess... I'd say... almost all of it? :-) I like systems programming in C on NT these days far more than Linux systems programming. Once you've gotten used
110.
by trentnelson 9y ago
I guess this is as good a time as any to make my post-PyParallel pet project public (holy alliteration Batman): https://github.com/tpn/tracer . I started it about 2.5 years ago because I wasn't really happy with ex
111.
by trentnelson 9y ago
NT has asynchronous I/O for disk. And network, and anything I/O-based. In UNIX land, you have read() and write(). These calls translate to system calls, and for most cases, a driver is called upon to process the I/O reques
112.
by trentnelson 9y ago
Well wasn't that a pleasant surprise to see you were quoting me :-) I was about to say it's so disappointing reading articles like this (about trying to improve AIO on *nix) because no-one ever seems to want to look at the details
113.
by trentnelson 9y ago
You're missing another possibility: providing a thin Linux shim layer that farms out to a process on NT that is able to exploit the far richer kernel offerings. (Or just wait for the virtualized NT kernel that SQL Server Linux is buil
114.
by trentnelson 9y ago
It's not buggy at all -- I've been using it extensively since it was first available and have not run into a single issue.
115.
by trentnelson 9y ago
Related: excellent article on IA64/x64 etc: https://github.com/tpn/pdfs/blob/master/A%20History%20of%20M... Features quotes from various interviews with folks like David Cutler. Fascinating (and no
116.
by trentnelson 9y ago
Arrival has an R Jupyter Notebook featured about half way through when they're doing analysis. And 24 had Chloe walking over to a terminal and doing `svn unlock xyz` once.
117.
by trentnelson 9y ago
Like NTFS has had for the last 20+ years? ;-) One of the benefits of NT's packet-driven I/O model (versus the UNIX synchronous VFS read/write blocking buffer I/O model).
118.
by trentnelson 10y ago
Sounds like you need a good mentor.
119.
by trentnelson 10y ago
PyParallel v1 was a nice checkpoint. I'm working on the next incarnation of it now.
120.
by trentnelson 10y ago
Google "davec apcobj.c"; that'll point you in the direction of some of Dave's brilliant work that dates back to 1989. (Three versions of NT have been leaked; NT 4.0, Windows 2000, an the Windows Research Kit (which is Wi

More ›