AMD Radeon RX 480 On Linux

[Nille_kungen]

Michael it would probably help twriter get a bit more recognition if he had a forum title like 'AMD Big Cheese'.

I think AMD open source il capo dei capi.

[bridgman]

Michael it would probably help twriter get a bit more recognition if he had a forum title like 'AMD Big Cheese'.

Excellent idea

[pszilard]

That's pretty much the big hairy question of the compiler world AFAICS. Winning with toolchains seems to be simple in principle - identify the perfect IR from both code and dev tool perspective, implement toolchains around that IR, profit - but in practice there seem to be conflicting pressures on the IR from the code (complex & non-flat is better) and dev tool (simple and flat is better) perspective and the representations on both sides of the IR keep changing over time.

Having two levels of IR (one source-oriented and one target-oriented) is one solution but even that gets less than ideal when either source or target are changing rapidly. It's probably fair to say that source changes less (some variant of C++) and target changes more (OMG) these days. The other complication is the usual huge gap between processor speeds and memory latencies, which is managed pretty well in single-thread environments (caches tend to be big enough for working sets these days) but which becomes more complex in highly parallel implementations where you need to fit (#threads x working set of registers & heavily used variables) into RF+cache or performance plummets.

Thanks for this and the other explanations!

I hope the new compiler gets better with register allocation and that it will allow/take hints form the programmer.

[Herem]

Michael it would probably help twriter get a bit more recognition if he had a forum title like 'AMD Big Cheese'.

[dungeon]

Tim has been managing the open source team since 2012 or so... and came to the job with a very solid Linux background... unlike me

Don't worry, after 10K posts on phoronix you will learn something abot Linux

[bridgman]

Tim has been managing the open source team since 2012 or so... and came to the job with a very solid Linux background... unlike me

http://www.phoronix.com/scan.php?pag...tem&px=MTE4ODU

[juno]

Sorry, I don't know him but I think this is included in "the whole team"

[GreatEmerald]

bridgman, agd5f, etc (the whole team): thanks for making this happen!

Poor twriter always gets forgotten

[Kano]

I would like to know how many kernel crashes YOU reported! You can be sure that unbootable systems because of a binary module are very rare if you work correctly. The main kernel error messages i remember that I reported have been i965 (Intel) related. Standard desktop systems can be debugged relatively easy, i don't get your problem. I developed on the fly GFX detection with possible mesa override too, i can switch cards/drivers very fast to debug issues you could not report

[SystemCrasher]

I see no reason to focus on OSS drivers for gaming. I would like to see vdpauinfo and Kodi/mpv test in case of HEVC Main 10 active. I would use the best possible driver for the purpose i want to use a card.

Then you should probably change your badge to Windows Something Developer, don't you? You're so good at praising proprietary or patented crap. If you do not really care whether things are opensource, windows got more games and they tend to perform better. At least because most engines were created with DX in mind and GL happens via wrappers or some code bolted in really odd ways. I do not really get how you could "develop" something if you can't even report Linux kernel bug. Linux kernel devs would show you middle finger once they spot tainted kernels. Then you're really on your own and when one thinks its okay, they taking a bit too much on themselves, assuming they could properly support whole Linux kernel single-handed.