Announcement

Collapse
No announcement yet.

AMD Publishes Open-Source HD 7000, Trinity Code

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by d2kx View Post
    Yeah, unfortunately I can't use these drivers other than for testing purposes due to missing hardware acceleration support.
    FTFY

    (note: I'm talking about the Trinity/Southern Island drivers, NOT Northern Islands and earlier. I agree with your sentiment in the latter case, although I used (past tense; I no longer have my Evergreen card) the r600g driver on my desktop for several months before I got my 7970.)

    Comment


    • #32
      Originally posted by allquixotic View Post
      FTFY
      (note: I'm talking about the Trinity/Southern Island drivers, NOT Northern Islands and earlier. I agree with your sentiment in the latter case, although I used (past tense; I no longer have my Evergreen card) the r600g driver on my desktop for several months before I got my 7970.)
      I really hope it's happy ending for us. As for me and other that have notebook using AMD HD 6xxx model, using linux + radeon open source driver is a no go. Lack Hardware Acc, lack power management, etc. Most of the time I use windows 7 on my Asus 1215B 'cause of that.

      My next lappie's Intel one cause theirs better in that [HW acc, Pow. Manag.] front. Will go AMD route if they have same driver capability over Intel.

      Comment


      • #33
        Originally posted by t.s. View Post
        I really hope it's happy ending for us. As for me and other that have notebook using AMD HD 6xxx model, using linux + radeon open source driver is a no go. Lack Hardware Acc, lack power management, etc. Most of the time I use windows 7 on my Asus 1215B 'cause of that.

        My next lappie's Intel one cause theirs better in that [HW acc, Pow. Manag.] front. Will go AMD route if they have same driver capability over Intel.
        Yeah; the Linux AMD driver support is why I've been using Intel IGPs in my laptops for years.

        Comment


        • #34
          Originally posted by t.s. View Post
          As for me and other that have notebook using AMD HD 6xxx model, using linux + radeon open source driver is a no go. Lack Hardware Acc, lack power management, etc.
          re: Lack of hardware accel, are you just talking about the UVD block or something more ?

          Comment


          • #35
            Originally posted by t.s. View Post
            I really hope it's happy ending for us. As for me and other that have notebook using AMD HD 6xxx model, using linux + radeon open source driver is a no go. Lack Hardware Acc, lack power management, etc.
            Since AMD has joined the bullshit game of renaming cards into higher series this isn't true anymore.
            For example the HD 6550 M is a renamed HD 5650 M: it's a Redwood. And it works quite well.

            Comment


            • #36
              Yeah; the Linux AMD driver support is why I've been using Intel IGPs in my laptops for years.
              Yep, for now I plan to use intel on mobile (netbook/notebook) and AMD for desktop.

              Originally posted by bridgman View Post
              re: Lack of hardware accel, are you just talking about the UVD block or something more ?
              Yep, video acceleration & 3D acceleration with open source radeon driver. 3D is quite good now

              Since AMD has joined the bullshit game of renaming cards into higher series this isn't true anymore. For example the HD 6550 M is a renamed HD 5650 M: it's a Redwood. And it works quite well.
              Yes, true. But the power management, one important thing if we use netbook/notebook still in bad shape, IMO. My Asus 1215B kernel 3.2.11 still can't go to sleep/ turn off the display with open source radeon driver. IIRC, my former netbook (atom) can do that with no problem ~1 year ago.

              Comment


              • #37
                Originally posted by t.s. View Post
                My Asus 1215B kernel 3.2.11 still can't go to sleep/ turn off the display with open source radeon driver. IIRC, my former netbook (atom) can do that with no problem ~1 year ago.
                That must be from some customization Asus did, because my MSI Wind U270 (with AMD E-450 APU) suspended without any problem even with the 2.6 kernels, currently running 3.2.6 or 3.2.9 under Gentoo. Or it could also be from the distribution you're using? I tried various live CDs on my laptop, but I don't think I ever tried suspend on any of them.

                Comment


                • #38
                  Originally posted by Ansla View Post
                  That must be from some customization Asus did, because my MSI Wind U270 (with AMD E-450 APU) suspended without any problem even with the 2.6 kernels, currently running 3.2.6 or 3.2.9 under Gentoo. Or it could also be from the distribution you're using? I tried various live CDs on my laptop, but I don't think I ever tried suspend on any of them.
                  Glad it working for you. As for the distro, IIRC, I've tried ubuntu, opensuse, archlinux. Same result: can't suspended/sleep mode and can't switch off the display.

                  Comment


                  • #39
                    Has anyone tested this?

                    I'm looking for a new card these days and GCN seems to hold more promise for performance on the open stack due to the new design (halving the gap to Catalyst was mentioned in the other thread...)

                    It would be nice to hear if anyone has got it working with real hardware.

                    Comment


                    • #40
                      Originally posted by pingufunkybeat View Post
                      Has anyone tested this?

                      I'm looking for a new card these days and GCN seems to hold more promise for performance on the open stack due to the new design (halving the gap to Catalyst was mentioned in the other thread...)

                      It would be nice to hear if anyone has got it working with real hardware.
                      It's not currently possible to "test" this, because the userspace parts of the driver aren't released yet.

                      The article is (as usual; what's new?) misleading: only the kernel and DDX bits are released. The really complex part -- the Gallium3d driver -- is not available in any form. According to bridgman they don't really test their initial code drop very much in-house (except for on whatever hardware they can get their hands on), and they release it pretty much when it's in a nominal releasable state -- so, with those two facts in mind, my conclusion is that the gallium3d support is somewhere between 0% and 99% complete, where "complete" is defined as "the minimal releasable quality that AMD is willing to accept before pushing the code out to the public" (which, if past releases are any indication, is not actually a viable driver for end-users until the community has spent at least 3 months with the code to fix bugs and so on).

                      If you want to know a more accurate estimate of how close the g3d driver is to being publicly unveiled, that's a question for bridgman. I'm not even going to wager a guess this time, because bridgman is the master of rhetoric and will make me look like an idiot regardless of what estimate I make.

                      TL;DR: I live on the edge when it comes to testing/using the open drivers, but even if the gallium3d code were published today, I probably wouldn't even download the code until June/July timeframe. Not worth the disappointment of "oh, X doesn't start, whee... back to Catalyst".
                      Last edited by allquixotic; 03-28-2012, 07:01 PM.

                      Comment


                      • #41
                        I remember reading something about the userspace bits also being there, but on a closer look, it seems to only affect the Trinity APUs, not the 7000 series.

                        If that's the case, I'll wait a bit longer.

                        Comment


                        • #42
                          That's true. Ubuntu 12.10 will probably have open-source Radeon HD 7000 support but at the moment you're stuck with the vesa driver when running X.

                          Comment


                          • #43
                            Originally posted by AlbertP View Post
                            That's true. Ubuntu 12.10 will probably have open-source Radeon HD 7000 support but at the moment you're stuck with the vesa driver when running X.
                            Actually, that's not a foregone conclusion. I wouldn't say it's flat out not going to happen, but I'd at least say that there should be significant doubt about whether 12.10 will have out of the box support for HD7000.

                            The wait between the initial push of the kernel code (which is done, now, for HD7000) and the initial push of the Gallium3d code (which is yet to come) was 201 days for classic mesa, and an additional 21 (so, 222 days total) for Gallium3d. And that was just absolute bare bones -- this "bare bones" state is not stable/usable enough for Ubuntu to turn it on by default, mainly because it won't run something like Unity 3d with appreciable results, and they'd rather have people running Unity 2D stably than running Unity 3d that's crash-prone.

                            Since Ubuntu likes to cut off pulling in kernel updates and 3d stack updates about 2 months into a release development cycle, that means the Gallium3d code for HD7000 would have to be in a usable state no later than about June or July. If it landed in August, due to Canonical's strict time-based release schedule and ostrich-like (head-in-sand) attitude towards upstream updates, I'm betting 99% chance they won't pull it. September, 100% guarantee they won't pull it. October, and hell would freeze over before they'd pull it.

                            222 days from March 20, 2012 works out to be October 28, 2012, which is more or less going to be the week of release of Ubuntu 12.10. So if it hit that late, there's no WAY it'd ever be pulled into Ubuntu 12.10. 13.04 it is.

                            But let's be charitable and say that, through some miracle, AMD managed to cut the time between the kernel push and the mesa push by a whopping 60%. 222 * 0.40 = 88 days. 88 days from March 20, 2012 works out to be June 16, 2012. Even I'm optimistic enough to say that, indeed, an initial push of Gallium3d on June 16, 2012 would probably be enough to stabilize the driver in June-July and get it into Ubuntu 12.10 and turned on by default by some miracle.

                            But the question is, will AMD be able to cut their time to release by that much? Or will it sit behind closed doors for a period closer to the track record of 222 days for Evergreen? Every day that tick-tick-ticks by is slimming the chances that the G3D driver will land in distros being developed in 2012. And without the G3D bits, you have absolutely nothing.

                            Comment


                            • #44
                              Originally posted by allquixotic View Post
                              But let's be charitable and say that, through some miracle, AMD managed to cut the time between the kernel push and the mesa push by a whopping 60%. 222 * 0.40 = 88 days. 88 days from March 20, 2012 works out to be June 16, 2012. Even I'm optimistic enough to say that, indeed, an initial push of Gallium3d on June 16, 2012 would probably be enough to stabilize the driver in June-July and get it into Ubuntu 12.10 and turned on by default by some miracle.

                              But the question is, will AMD be able to cut their time to release by that much? Or will it sit behind closed doors for a period closer to the track record of 222 days for Evergreen? Every day that tick-tick-ticks by is slimming the chances that the G3D driver will land in distros being developed in 2012. And without the G3D bits, you have absolutely nothing.
                              Mesa push happened today (16 days ?). Is that soon enough ?

                              The code is in a public branch on Alex's freedesktop.org repo until Tom comes back from vacation next week and we work out where the llvm bits should go -- a big chunk of the 114 KSLOC is LLVM back-end, including both common-across-all-AMD-GPU bits and SI-specific bits.
                              Last edited by bridgman; 04-05-2012, 01:07 PM.

                              Comment


                              • #45
                                Right, let's see if I can cover all the responses:
                                • Hurrah! Congratulations AMD!
                                • I'm really glad about this - I'm going to order a new card.
                                • The release->driver support time is coming down very quickly
                                • I already own a 7XXX - I can't believe it took this long!
                                • Does it have UVD support?
                                • Why don't you work on UVD?
                                • AMD should just open up Catalyst
                                • No open source EVERYTHING, right NOW means AMD don't even support linux - go to Nvidia!

                                Have I missed anything out? *grin*

                                Incidentally, I'm all of the first 3.

                                Comment

                                Working...
                                X