HNHacker News
TopNewBestAskShowJobs

mattst88

517 karma · joined April 28, 2012

https://mattst88.com/

Netflix. ex-Google. ex-Intel. Gentoo Developer.

submissionscomments
mattst88··on Speeding up Linux disk encryption
Yeah, the time frame is somewhat unclear. The patch they link to in their tree is dated December 2019 however [1], so I assume this blog post is about stuff they've completed recently.

[1] https://github.com/cloudflare/linux/blob/master/patches/0023...

mattst88··on Clang vs. other open-source compilers
Could you provide a few examples of what information is out of date?
mattst88··on Richard M. Stallman resigns
Maybe I'm just dense, but I don't understand what the "implications" are. Someone clear up the meaning for me?
mattst88··on China is increasingly using exit bans to bar Americans from leaving
For those like me that have never heard of this: https://en.wikipedia.org/wiki/Dikko_affair
mattst88··on J-Core Open Processor
Oh, the website is back up. It's been down for months and the project seemed to have gone into stealth mode (or maybe died). Still no news for about three years though. Unsure why this is being posted now, unless I've missed some news.

Interesting project, if it is still alive.

mattst88··on New OpenGL Driver for Intel Gen8 GPUs Merged into Mesa
> Meltdown/Spectre turned out not to be a reason big enough to upgrade from Sandy/Ivy Bridge, so they had to find another way to "discontinue" current users.

I sit beside the author of the Iris driver and work on the same team at Intel. The reasoning for only supporting BDW+ has absolutely nothing to do with spectre/meltdown or trying to force people to upgrade their hardware. Such a statement really indicates that you're not familiar with the people on Intel's Mesa team.

FYI the reasoning for only supporting BDW+ is that BDW+ supports 48-bit addressing which allows the CPU and GPU to share addresses trivially. This means no relocations are required and that's a great driver simplification and performance improvement.

BDW+ also supports executing all shader stages in the "scalar" (vs the vec4) mode, which allows much better opportunity for shader compiler optimization, which is the area I spend most of my time.

mattst88··on AMD Open Source Driver for Vulkan
> floating point co-coprocessor and the "drivers" just rename mesa-software with intelgfx

That is hilariously false. I work on the drivers. They're open source. Go take a look.

mattst88··on Brand New ODroid Single Board Computer
> ia64

I wouldn't normally comment on something like this, but I noticed this mistake twice in the linked post as well. ia64 refers to Itanium, while amd64/x86-64/x64 refer to the 64-bit version of x86.

mattst88··on AMD once again have a top performance crown
The Phoronix test suite is open source. If patches were provided, I feel confident the result would be new benchmarks included in his articles.
mattst88··on Intel Ships (hopefully Stable) Microcode for Skylake, Kaby Lake, Coffee Lake
They release a microcode package for Linux. The kernel will load it from an initramfs, so no need to flash your BIOS. First lines in my dmesg are:

   [    0.000000] microcode: microcode updated early to revision 0xc2, date = 2017-11-16
   [    0.000000] Linux version 4.14.14+ (mattst88@p50-ethernet) (gcc version 6.4.0 (Gentoo 6.4.0-r1 p1.3)) #19 SMP Sun Feb 4 00:20:02 PST 2018
mattst88··on Intel Ships (hopefully Stable) Microcode for Skylake, Kaby Lake, Coffee Lake
Here's the full PDF from Intel, showing the status of microcode updates for what looks like CPUs going back to what looks like the first 64-bit Core CPUs -- https://newsroom.intel.com/wp-content/uploads/sites/11/2018/...

(Disclaimer: I work for Intel, but not on anything related to this and have no knowledge on the topic outside of this PDF)

mattst88··on Slack: Other browsers require significant effort so we're focused on Chrome
I've noticed the same thing -- often people refer to the "hive mind" feeling one way or another.

Rather than assuming individuals are just fickle, I think it makes much more sense that individuals are often only vocal about things they disagree with. We as readers don't know those individuals and lump them into our "hive mind" concept, as mistakenly assume that people's minds have changed.

mattst88··on Ask HN: Is big-endian dead?
I'm interested to know what you worked on on Alpha.
mattst88··on Linux Graphics Demystified (2014) [pdf]
This is an extremely simplistic view that misunderstands the situation.

For one, those people working on Wayland made enormous contributions to the X Server long before Wayland existed. The creator of Wayland, Kristian Høgsberg did AIGLX, just to name one example.

mattst88··on The Loyalty to AMD’s GPU Product Among AMD CPU Buyers Is Decreasing
FWIW, it appears that all of the i3, i5, and i7 Skylakes support both VT-x and VT-d.

I agree that this has been confusing in the past (as evidenced by the mix of Atom, Celeron, Pentium, i3, i5, i7, Xeon in your link), but I think at least in VT-* it's been greatly simplified.

mattst88··on The Loyalty to AMD’s GPU Product Among AMD CPU Buyers Is Decreasing
> whereas Intel only allows virtualization on Xeon and K-series processors.

I'm not an expert on Intel CPU market segmentation, but I do work for Intel on the i965 Mesa driver. I don't think your statement is true.

See http://ark.intel.com/search/advanced?s=t&VTX=true

That's a list of all Intel CPUs with VT-x. It's really easy to narrow down the selection from there. FWIW, I see lots of Skylake i3, i5, and i7s with VT-x.

mattst88··on GPU folks: we need to talk about control flow
I see "Next stop: Intel."

I sincerely hope that you test our Linux drivers (I work for Intel on them). We'd be excited to learn about anything you find. Please feel free to file bugs here (https://bugs.freedesktop.org/enter_bug.cgi?product=Mesa&comp...) and ping us on #intel-gfx on Freenode.

mattst88··on Fedora 25: With Wayland, Linux has never been easier (or more handsome)
Three times throughout the comments on this article, you have referred to the freedesktop/DE/container "boys".

I could be wrong for doing so, but I read that as somewhat derogatory, in case you didn't mean it that way.

mattst88··on NSA Said to Use Manhattan Tower as Listening Post
> The judiciary committee was led by none other than Strom Thurmond. Thurmond and company thought he was too racist. Thurmond! Let that sink in.

Mentioning Thurmond is simply misleading and detracts from the main point that Sessions was rejected from a federal judgeship because of concerns that he was a racist.

The fact of the matter is that Thurmond voted for Sessions.

> On June 5, 1986, the Committee voted 10–8 against recommending the nomination to the Senate floor, with Republican Senators Charles Mathias of Maryland and Arlen Specter of Pennsylvania voting with the Democrats.

https://en.wikipedia.org/wiki/Jeff_Sessions

mattst88··on Realistic alternatives to Apple computers
Ubuntu 14.04 is very likely not going to be a good experience on a 4k display. A newer release, however, should be significantly better. I'm not sure if it's possible to work around Dell's configuration constraint system.
mattst88··on Gem in the Linux source code
Another discussion: https://news.ycombinator.com/item?id=12363012
mattst88··on What New York Can Learn from Barcelona’s ‘Superblocks’
In Portland, a far too frequent cause of MAX (the light rail) service interruptions is due to car/train collisions. See the following links for a small taste:

http://koin.com/2014/04/01/1-hurt-car-max-crash-portland/ http://www.oregonlive.com/portland/index.ssf/2013/06/max_col... http://www.oregonlive.com/portland/index.ssf/2015/05/max_blu... https://www.reddit.com/r/Portland/comments/3d5i9g/saw_an_ora...

There may be more to it in Portland than "at grade", but that certainly seems to be a large component of the problem.

mattst88··on OpenSSL CVE-2016-0799: heap corruption via BIO_printf
The project I work on, Mesa3D, routinely ignores malloc failures for a couple of reasons: no one knows how to reasonably test the additional code, and often no one has any idea what sort of graceful failure is even possible when malloc fails.

We've had some cargo culted patches to call _mesa_error_no_memory() when malloc fails, but it was recently noted that this can't possibly work because it internally calls fprintf, and there's no guarantee that fprintf doesn't call malloc! In fact, glibc's implementation does.

Suggestions for best practices welcome.

mattst88··on You Compiled This, Driver – Trust Me
The article is very impressive, and the author is an experienced graphics developer, which makes me wonder why not do these experiments using the free software graphics drivers in Mesa?

We (the team that writes the Intel driver in Mesa) would love to be involved in things like that and would be happy to answer questions where possible -- #intel-gfx on Freenode or mesa-dev@lists.freedesktop.org.

The documentation doesn't contain much if anything in the way of performance data like instruction latency, so we've implemented similar methods to what the author used to measure the number of cycles a piece of code takes (environment variable INTEL_DEBUG=shader_time). We'd likely even learn something about the hardware ourselves by being involved in experiments like these.

mattst88··on Introduction to GEN Assembly in OpenCL
That's a good question.

SIMD8/SIMD16 refers to the number of fragments processed per thread invocation. There is some overhead to spawning a thread, and so processing 16 fragments at a time is typically faster even though the shader itself is doing more work.

The driver provides both versions because it's the GPU that decides which version is use and where, even using both versions to shade the same primitive. For instance, on a triangle boundary maybe only 4 or 8 fragments are "lit", so the hardware spawns a SIMD8 thread and saves itself a little work.

SIMD16 shaders typically use twice the number of registers as SIMD8, and if they require registers to be spilled to memory it's likely the SIMD8 shader is faster even with the additional thread-spawning overhead. Lots of compiler optimization revolves around trying to squeak in under the register limit to get the program compiled as SIMD16 without spilling. :)

mattst88··on Introduction to GEN Assembly in OpenCL
Another Intel employee here (don't know the author). I work on Intel's Open Source OpenGL driver, which is part of Mesa. Most of the work I do is on our GLSL compiler and Gen hardware backend.

I've never used VTune (but it looks very cool) and some of the notation is different from what I'm used to.

If you're using an Intel GPU made in the last 10 years on Linux, the environment variable INTEL_DEBUG will allow you to see the disassembled shaders for a given program. Try "INTEL_DEBUG=fs,vs glxgears" to see the fragment and vertex shaders' assembly and the dumps of some intermediate representations along the way.

The Gen (or i965, as we call it) instruction set is really powerful, but does take a while to understand fully.

I've actually been trying to finish up an article about some tricks we do in the i965_dri.so driver (think bit twiddling hacks, but each using some interesting features of the instruction set). I'm curious if there's interest in such a thing. I'd probably be more motivated to finish it. :)

mattst88··on Stop using gzip
I don't believe that's true, though there are multiple projects that offer that feature. lbzip2.org and compression.ca/pbzip2 to name a couple.
mattst88··on Remove the ++ and –- operators
Compiling

  #include <stdint.h>
  uint64_t inc(uint64_t x) { return x + 1; }
With -O0 gives me:

  48 83 c0 01          	add    $0x1,%rax
and with -O2:

  48 8d 47 01          	lea    0x1(%rdi),%rax
That is, both are four bytes.
mattst88··on Remove the ++ and –- operators
That's quite a claim. I agree that most uses of these operators as an expression are not wise. Some are though. As a counter-claim, I offer:

  array[i++] = this;
  array[i++] = that;
  array[i++] = the_other;
which also extends nicely when some of those elements are optional.
mattst88··on Comparison Of Nvidia and AMD Graphics Cards On SteamOS For Steam Linux Gaming
Would be nice to indicate in the title that the article is about AMD's proprietary drivers, rather than the free software drivers in Mesa.
← PreviousPage 3 of 4Next →