It Looks Like AMDGPU DC (DAL) Will Not Be Accepted In The Linux Kernel

[pal666]

At this state a lot of users are using it and will be angry if it's ripped out because it was not improved.

but will that "a lot" be larger than a lot of users who are already angry because it is not merged?

[pal666]

AMD is a huge company, they make profit from Linux

can you show some evidence of this claim?

[pal666]

Personally I believe that AMD can only benefit from the OS efforts. I'm considering switching to full AMD as soon as I can have everything I need for my professional needs (3D graphics development).

so amd had zero benefit from you and it will stay like that in foreseeable future?

[JordanL]

I have read Dave Airlie's response (https://lists.freedesktop.org/archiv...er/126516.html) and it sounds like he is rejecting all of the new source code.

From from you wrote it however sounds like you believe he is rejecting just some parts of the new source code.

If the latter is true there will be a reduction in DAL/DC LOC (loc means "lines of code") after Linux kernel 4.10 and 4.11 are released.

However, from the overall historical record of actually merging DAL into Linux 4.7, 4.8 and 4.9 it seems that there has not been a sizeable LOC reduction so far. Note that a complete rewrite isn't a LOC reduction caused by inclusion of DAL code into Linux kernel. (This may be incorrect because I didn't examine DAL LOC evolution over time).

The question is: Why should we expect the 4.10 and 4.11 situation to turn out differently from 4.7-4.9?

DC actually went from 93k lines of code to 66k lines of code

[bridgman]

I have read Dave Airlie's response (https://lists.freedesktop.org/archiv...er/126516.html) and it sounds like he is rejecting all of the new source code.

From from you wrote it however sounds like you believe he is rejecting just some parts of the new source code.

If the latter is true there will be a reduction in DAL/DC LOC (loc means "lines of code") after Linux kernel 4.10 and 4.11 are released.

However, from the overall historical record of actually merging DAL into Linux 4.7, 4.8 and 4.9 it seems that there has not been a sizeable LOC reduction so far. Note that a complete rewrite isn't a LOC reduction caused by inclusion of DAL code into Linux kernel. (This may be incorrect because I didn't examine DAL LOC evolution over time).

The question is: Why should we expect the 4.10 and 4.11 situation to turn out differently from 4.7-4.9?

Well it's not just rejecting "some parts of the new source code" (which could result in it getting smaller although it would probably stop working as well) he is pushing back on the interface we currently use to call into that source code. I don't see how you got "rejecting all of the new source code" from his comments though - as long as he is rejecting *any* part of the code it can't go upstream because it won't work without all of the code.

DAL code has gone from 90-something KSLOC to 60-something, which I would call sizeable.

You should expect future kernel version situation to turn out differently from past kernel version situation because at some point we will have finished enough rework for the code to be accepted upstream. That seems too obvious though, maybe I didn't understand your question.

[atomsymbol]

A lot of the confusion here is that DC has two meanings - it's the big chunk of developer effort & code we want to re-use across platforms for all kinds of good reasons, and it is also a very specific abstraction layer inside that code. It's the second meaning that Dave and Daniel have concerns about, because it internalizes things that they feel should be part of the driver rather than part of what is effectively a blob (despite being publicly developed open source) to them. We understand that.

Again, this was primarily a miscommunication, so other than making for some good reading it doesn't mean much.

Michael really needs to change the article - seems like everyone is posting and getting all worried based on the article and not reading any of the comments before going ahead and making things even worse. I'm trying to be on vacation and can't spend all day responding to basically the same "I didn't read the other comments so this looks bad what do I do" posts over and over again.

I have read Dave Airlie's response (https://lists.freedesktop.org/archiv...er/126516.html) and it sounds like he is rejecting all of the new source code.

From from you wrote it however sounds like you believe he is rejecting just some parts of the new source code.

If the latter is true there will be a reduction in DAL/DC LOC (loc means "lines of code") after Linux kernel 4.10 and 4.11 are released.

However, from the overall historical record of actually merging DAL into Linux 4.7, 4.8 and 4.9 it seems that there has not been a sizeable LOC reduction so far. Note that a complete rewrite isn't a LOC reduction caused by inclusion of DAL code into Linux kernel. (This may be incorrect because I didn't examine DAL LOC evolution over time).

The question is: Why should we expect the 4.10 and 4.11 situation to turn out differently from 4.7-4.9?

[pal666]

I guess on this point that DAL (i wouldn't call that AC nor DC, when it is DAL) is not high priority for AMD , opensource driver customers seems don't wanna HDMI 2.0, HDMI Audio, FreeSync...

your list misses vega

[pal666]

Dave says "NO! NEVER!"

actually, he didn't say that. he asked "what would happen if someone rewrites half of your code" and the answer is "everyone would be happy"

[pal666]

There are quality borders that are more important than the development budget of AMD

this thread is starting to look like a census of people who never heard of staging

[DonQ]

David megalomania Airlie...

To be honest I don't understand this harsh judgement (and I think the unfriendly wording is simply unnescessary) - all AMD want is to have this code *shipped* with the linux kernel, so that users do not have to download/compile/install 3rd party kernel drivers. I guess nobody seriously expects anybody else than AMD putting serious effort and maintenance into this codebase (development of the open-source radeonsi-driver is also almost exclusivly done by paid AMD OSS developers).

And for AMD's efforts, after I upgraded from nvidia 7600GT, I have bought a 3850, 5850, 7950, 380x, and 390 (I may be off on card numbers, I gave away the 390 because of bad drivers, though). I am also eagerly waiting to spend $600+ on a Vega card when it comes out. AMD, your efforts don't go unnoticed.