You may be able to reverse engineer it, but there is one problem this guy is running into an error when trying to build the latest radeontool (1.6.2): https://build.opensuse.org/package/l...enSUSE_Factory
Any idea how to fix this?Code:radeontool.c:(.text+0x1238): undefined reference to `pci_system_init' radeontool.c:(.text+0x1274): undefined reference to `pci_slot_match_iterator_create' radeontool.c:(.text+0x1284): undefined reference to `pci_device_next' radeontool.c:(.text+0x1296): undefined reference to `pci_device_probe' radeontool.c:(.text+0x13ad): undefined reference to `pci_device_map_region' radeontool.c:(.text+0x13e5): undefined reference to `pci_device_map_region' radeontool.c:(.text+0x1407): undefined reference to `pci_iterator_destroy' /tmp/ccjP1HAm.o: In function `fatal': radeontool.c:(.text.unlikely+0x1d): undefined reference to `pci_system_cleanup' collect2: ld returned 1 exit status error: Bad exit status from /var/tmp/rpm-tmp.FBFokV (%build)
Well, since they missed the 3.2 merge window (@ rc2 now) or we'd have heard about it, I guess practically the next chance is the 3.3 merge window in January or so. Of course it would be good to get information that they'll actually make that release window, but it won't enter any mainstream kernel faster. Hopefully this first release of HDMI audio information (the R600/700 support was reverse engineered AFAIK) is the tough one and that it'll get easier after this.
September, October, November gone... any progress made? Or has it hit some "can't be done" roadblock?
The way IP release works is that we try to release, we hit "can't be done" roadblocks, then the proposed release sits until one of us has an idea for either a different set of information to release, a different approach to how we trace the IP back to original licenses/standards, or a different way to "make the release safe" in the case of things where DRM lurks nearby, then we try again and usually hit a different roadblock, rinse and repeat. Sometimes we get lucky and are able to release the info, other times we are not.
We obviously try to release the most broadly useful info first, but in practice what gets released is a function of both our perception of importance and our ability to make a compelling case for being able to release... so to a certain extent we "pushed the easy stuff out first" and are dealing with the harder stuff now.
HDMI audio is one of those things that seems "stupidly simple" at first glance but in practice is fairly complicated. I think this will be our fourth attempt.
EDIT -- wierd; the word "either" appears twice in the first sentence when viewing the thread, but when I go into "Edit" it only appears once in the edit box, but if I adjust the width of my window I can make the second occurance of "either" appear and disappear. The word doubles-up only when the last letter of "either" is the last letter of the first row of text in thread view.
Last edited by bridgman; 12-02-2011 at 03:46 PM.
I don't want to buy a new card without support for open-source-hdmi-audio.
i think the market share for Linux in "desktop hardware" is only so low because Linux users always be punished if they buy new hardware.
for example i do have a very old notebook 600mhz/192mb ram and i want buy a new amd-fusion based notebook but I'm sure i will be punished if i buy it. because this old notebook work perfect with open-source drivers the new one will not.
hey its Christmas and there will be no computer hardware as a gift below the Christmas tree because i don't want punished my family or myself.
AMD just lose money because of this.