Hi Joseph, > 04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. > [AMD/ATI] Ellesmere [Radeon RX 470/480/570/580] (rev e7) (prog-if 00 > [VGA controller]) >    Subsystem: Micro-Star International Co., Ltd. [MSI] Device 3418 >    Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- > ParErr- Stepping- SERR- FastB2B- DisINTx+ >    Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- > <TAbort- <MAbort- >SERR- <PERR- INTx- >    Latency: 0, Cache Line Size: 64 bytes >    Interrupt: pin A routed to IRQ 141 >    Region 0: Memory at e0000000 (64-bit, prefetchable) [size=256M] >    Region 2: Memory at f0000000 (64-bit, prefetchable) [size=2M] >    Region 4: I/O ports at 9000 [size=256] >    Region 5: Memory at f7800000 (32-bit, non-prefetchable) [size=256K] >    Expansion ROM at 000c0000 [disabled] [size=128K] >    Capabilities: <access denied> Please provide the lspci output as root. Going to work with this for now, but I'm not sure if there is everything in there I need. >    Kernel driver in use: amdgpu >    Kernel modules: amdgpu It would be good to get the configuration without amdgpu loaded, e.g. what the BIOS programmed for the BARs. Additional to that please provide the output of "sudo cat /proc/iomem". Regards, Christian. Am 16.02.2018 um 18:11 schrieb Joseph Wang: > Thanks. > > Here are three logs. > > failed is the dmesg with 9 cards > good is the boot with 8 cards > lspci.log is the pci log with 8 cards > > The 9th card is a copy of > > 0f:00.0 VGA compatible controller: Advanced Micro Devices, Inc. > [AMD/ATI] Ellesm > ere [Radeon RX 470/480/570/580] (rev ef) (prog-if 00 [VGA controller]) >        Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Device 0b31 > > One other thing is that if I put in 7 or 8 cards, I don't get x11 > video. All of the diagnostics > are the same, but I don't get x11 video, but I can switch to text console. > > Also with nine cards, I don't get the grub boot screen. I get this > graphics display, but then > the boot process seems to work and then it switches to text before the > kernel panic. > > One final thing is that I'm working with a group of scientists that > would like to use ROCm for > opencl supercomputing, and I'm trying to get resources so that they > can do development work. >
