1. Forum
  2. Usenet
  3. LINUX.DEBIAN.PORTS.POWERP

  • Re: PowerMac7,3 DualG5 with [AMD/ATI] RV360 [Radeon 9600/X1050 Series]

    From Ken Cunningham@21:1/5 to All on Sat Nov 6 08:50:01 2021
    --Apple-Mail-EA1BDFE9-7DE1-47AC-8D2A-9ED550D28104
    Content-Type: text/plain;
    charset=us-ascii
    Content-Transfer-Encoding: quoted-printable

    I suspect I might have to create a new kernel with this firmware embedded in it?

    https://www.linuxquestions.org/questions/slackware-14/loading-firmware-into-kernel-at-boot-4175495624/



    --Apple-Mail-EA1BDFE9-7DE1-47AC-8D2A-9ED550D28104
    Content-Type: text/html;
    charset=utf-8
    Content-Transfer-Encoding: 7bit

    <html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto">I suspect I might have to create a new kernel with this firmware embedded in it?<div></div><div><br></div><div><a href="https://www.linuxquestions.org/
    questions/slackware-14/loading-firmware-into-kernel-at-boot-4175495624/">https://www.linuxquestions.org/questions/slackware-14/loading-firmware-into-kernel-at-boot-4175495624/</a></div><div><br></div><div><br></div></body></html>
    --Apple-Mail-EA1BDFE9-7DE1-47AC-8D2A-9ED550D28104--

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From John Paul Adrian Glaubitz@21:1/5 to Ken Cunningham on Sat Nov 6 09:40:01 2021
    Hello Ken!

    On 11/6/21 01:47, Ken Cunningham wrote:
    $ ls -la /lib/firmware/radeon/R3*
    -rw-r--r-- 1 root root 2048 Aug 24 08:46 /lib/firmware/radeon/R300_cp.bin

    (...)
    [ 2.627772] radeon 0000:f0:10.0: firmware: failed to load radeon/R300_cp.bin (-2)
    [ 2.627802] radeon 0000:f0:10.0: Direct firmware load for radeon/R300_cp.bin failed with error -2
    [ 2.627810] [drm:.r100_cp_init [radeon]] *ERROR* Failed to load firmware! [ 2.628128] radeon 0000:f0:10.0: failed initializing CP (-2).
    [ 2.628145] radeon 0000:f0:10.0: Disabling GPU acceleration
    [ 2.628158] [drm] radeon: cp finalized
    [ 2.794433] radeon 0000:f0:10.0: [drm] fb0: radeon frame buffer device
    [ 2.810380] [drm] Initialized radeon 2.50.0 20080528 for 0000:f0:10.0 on minor 0

    Error 2 means that the file could not be found. Did you make sure that the firmware
    is actually part of your initramfs image? Looking at the timestamp, just 2.62 seconds
    after the start of the kernel is too early for the hard drive to be available yet.

    Try using the lsinitramfs command to check whether your initramfs actually contains
    the firmware. I don't remember exactly, but there might be a switch for initramfs-tools
    to control which firmware to include.

    Adrian

    --
    .''`. John Paul Adrian Glaubitz
    : :' : Debian Developer - glaubitz@debian.org
    `. `' Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
    `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ken Cunningham@21:1/5 to All on Sat Nov 6 17:20:02 2021
    Thank you for the clue! My knowledge of the inner workings of debian is nascent but growing.

    With that clue, I noticed that after installing the firmware packages a new initrd.img was automatically helpfully generated, and the new image does contain the needed firmware as you point out.

    old:

    :/boot$ lsinitramfs initrd.img-5.14.0-3-powerpc64 | grep R300*
    :/boot$

    new:

    :/boot$ lsinitramfs initrd.img-5.14.0-4-powerpc64 | grep R300* usr/lib/firmware/radeon/R300_cp.bin


    However, updating the grub.cfg to actually use it is a manual step that I didn’t know about/realize I would need to do, so after doing that:

    sudo /usr/sbin/grub-mkconfig -o /boot/grub/grub.cfg

    The system now boots from the new, proper, image and all is well, the radeon driver is found and used:

    $ sudo dmesg | grep radeon
    [ 3.106881] [drm] radeon kernel modesetting enabled.
    [ 3.107020] fb0: switching to radeondrmfb from OFfb ATY,Whelk_
    [ 3.107675] fb1: switching to radeondrmfb from OFfb ATY,Whelk_
    [ 3.107901] radeon 0000:f0:10.0: vgaarb: deactivate vga console
    [ 3.108119] radeon 0000:f0:10.0: enabling device (0006 -> 0007)
    [ 3.108738] radeon 0000:f0:10.0: Invalid PCI ROM header signature: expecting 0xaa55, got 0x0000
    [ 3.196056] radeon 0000:f0:10.0: VRAM: 256M 0x00000000A0000000 - 0x00000000AFFFFFFF (128M used)
    [ 3.196065] radeon 0000:f0:10.0: GTT: 512M 0x0000000080000000 - 0x000000009FFFFFFF
    [ 3.196151] radeon 0000:f0:10.0: dma_iommu_get_required_mask: returning bypass mask 0x1ffffffff
    [ 3.196351] [drm] radeon: 128M of VRAM memory ready
    [ 3.196358] [drm] radeon: 512M of GTT memory ready.
    [ 3.198547] [drm] radeon: 1 quad pipes, 1 Z pipes initialized
    [ 3.198710] radeon 0000:f0:10.0: WB enabled
    [ 3.198724] radeon 0000:f0:10.0: fence driver on ring 0 use gpu addr 0x0000000080000000
    [ 3.199225] [drm] radeon: irq initialized.
    [ 3.199414] radeon 0000:f0:10.0: firmware: direct-loading firmware radeon/R300_cp.bin
    [ 3.200789] [drm] radeon: ring at 0x0000000080010000
    [ 3.366523] radeon 0000:f0:10.0: [drm] fb0: radeon frame buffer device
    [ 3.378931] [drm] Initialized radeon 2.50.0 20080528 for 0000:f0:10.0 on minor 0


    Thanks!

    My apologies that people coming into this mailing list are probably often like me, perhaps not deeply versed in debian/linux as yet, and finding themselves having to do a bit more than is needed compared to just booting Ubuntu onto an Intel system (which
    I have four of at present, for the past several years, and had to do almost nothing to get them to work).

    Your patience is much appreciated!

    That is now two PowerPC systems converted to debian, one 32bit, and this one 64bit.

    I have several more to go!

    Best,

    Ken


    On Nov 6, 2021, at 1:31 AM, John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> wrote:

    Hello Ken!

    On 11/6/21 01:47, Ken Cunningham wrote:
    $ ls -la /lib/firmware/radeon/R3*
    -rw-r--r-- 1 root root 2048 Aug 24 08:46 /lib/firmware/radeon/R300_cp.bin

    (...)
    [ 2.627772] radeon 0000:f0:10.0: firmware: failed to load
    radeon/R300_cp.bin (-2)
    [ 2.627802] radeon 0000:f0:10.0: Direct firmware load for
    radeon/R300_cp.bin failed with error -2
    [ 2.627810] [drm:.r100_cp_init [radeon]] *ERROR* Failed to load firmware! >> [ 2.628128] radeon 0000:f0:10.0: failed initializing CP (-2).
    [ 2.628145] radeon 0000:f0:10.0: Disabling GPU acceleration
    [ 2.628158] [drm] radeon: cp finalized
    [ 2.794433] radeon 0000:f0:10.0: [drm] fb0: radeon frame buffer device
    [ 2.810380] [drm] Initialized radeon 2.50.0 20080528 for 0000:f0:10.0 on >> minor 0

    Error 2 means that the file could not be found. Did you make sure that the firmware
    is actually part of your initramfs image? Looking at the timestamp, just 2.62 seconds
    after the start of the kernel is too early for the hard drive to be available yet.

    Try using the lsinitramfs command to check whether your initramfs actually contains
    the firmware. I don't remember exactly, but there might be a switch for initramfs-tools
    to control which firmware to include.

    Adrian

    --
    .''`. John Paul Adrian Glaubitz
    : :' : Debian Developer - glaubitz@debian.org
    `. `' Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
    `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913


    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From John Paul Adrian Glaubitz@21:1/5 to Ken Cunningham on Sat Nov 6 18:00:02 2021
    Hello!

    On 11/6/21 17:11, Ken Cunningham wrote:
    However, updating the grub.cfg to actually use it is a manual step that I didn’t
    know about/realize I would need to do, so after doing that:

    sudo /usr/sbin/grub-mkconfig -o /boot/grub/grub.cfg

    The system now boots from the new, proper, image and all is well, the radeon driver is found and used:

    This should happen automatically by update-grub which is called after update-initramfs.

    Not sure why that didn't work for you, it always worked fine for me when I tested it.

    My apologies that people coming into this mailing list are probably often like me, perhaps
    not deeply versed in debian/linux as yet, and finding themselves having to do a bit more

    No worries.

    than is needed compared to just booting Ubuntu onto an Intel system (which I have four of
    at present, for the past several years, and had to do almost nothing to get them to work).

    It involves a bit more work because I'm effectively maintaining this port alone and ironing
    out all these small issues takes quite a lot of time and effort. We're getting there.

    Your patience is much appreciated!

    Thanks.

    That is now two PowerPC systems converted to debian, one 32bit, and this one 64bit.

    Sounds good.

    I have several more to go!

    Great.

    Adrian

    --
    .''`. John Paul Adrian Glaubitz
    : :' : Debian Developer - glaubitz@debian.org
    `. `' Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
    `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ken Cunningham@21:1/5 to All on Sat Nov 6 20:50:01 2021
    On Nov 6, 2021, at 9:53 AM, John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> wrote:

    Not sure why that didn't work for you, it always worked fine for me when I tested it.

    I suspect I know what might have happened.

    When I first installed debian 64b from the 2021-04-17 image onto the DualG5, it would not boot after installation… got just past grub and died with a black screen and no ssh login available.

    I used rescue mode from the CD to mount the root filesystem on the drive, and noticed in the syslog there were many errors relating to libffi8 symbols (as we know), so I manually built and installed libffi8 on the DualG5 using rescue mode from the CD.
    After that, it would boot.

    During that process I also (for the first time) installed the various firmware packages. So likely something that was supposed to run did not (locked /boot from the CD I suspect).

    If I had much experience at all with troubleshooting debian boots I would have known what to do — I will for next time (at least for that issue :> ).

    BTW I also noticed that an Apple monitor attached to the Apple Display Connector on that ATI 9600 card would not bring up an image when X came up — just went black screen. Changing to a DVI monitor attached to the other connector on the card did work.

    I have not retested that Apple monitor as yet -- now that the AMD firmware is actually working it might work now.

    Best,

    Ken
    <html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><br class=""><div><br class=""><blockquote type="cite" class=""><
    div class="">On Nov 6, 2021, at 9:53 AM, John Paul Adrian Glaubitz &lt;<a href="mailto:glaubitz@physik.fu-berlin.de" class="">glaubitz@physik.fu-berlin.de</a>&gt; wrote:</div><br class="Apple-interchange-newline"><div class=""><span style="caret-color:
    rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-
    text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Not sure why that didn't work for you, it always worked fine for me when I tested it.</span></div></blockquote></div><br class=""><div class="">I suspect I
    know what might have happened.</div><div class=""><br class=""></div><div class="">When I first installed debian 64b from the 2021-04-17 image onto the DualG5, it would not boot after installation… got just past grub and died with a black screen and no
    ssh login available.</div><div class=""><br class=""></div><div class="">I used rescue mode from the CD to mount the root filesystem on the drive, and noticed in the syslog there were many errors relating to libffi8 symbols (as we know), so I manually
    built and installed libffi8 on the DualG5 using rescue mode from the CD. After that, it would boot.</div><div class=""><br class=""></div><div class="">During that process I also (for the first time) installed the various firmware packages. So likely
    something that was supposed to run did not (locked /boot from the CD I suspect).</div><div class=""><br class=""></div><div class="">If I had much experience at all with troubleshooting debian boots I would have known what to do — I will for next time (
    at least for that issue :&gt; ).</div><div class=""><br class=""></div><div class="">BTW I also noticed that an Apple monitor attached to the Apple Display Connector on that ATI 9600 card would not bring up an image when X came up — just went black
    screen. Changing to a DVI monitor attached to the other connector on the card did work.</div><div class=""><br class=""></div><div class="">I have not retested that Apple monitor as yet -- now that the AMD firmware is actually working it might work now.</
    <div class=""><br class=""></div><div class="">Best,</div><div class=""><br class=""></div><div class="">Ken</div></body></html>

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)