diff options
| author | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-09-22 23:32:44 -0400 |
|---|---|---|
| committer | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-09-22 23:32:44 -0400 |
| commit | 14283fe73fb5e432a7af45de94961655714910f6 (patch) | |
| tree | e993bf56f3e2ef93195019ac0fd996d82204384a /kernel/main.c | |
| parent | b34e4531d8f0b984dc15660d3ab406ae48822931 (diff) | |
| download | open-iov-14283fe73fb5e432a7af45de94961655714910f6.tar.gz open-iov-14283fe73fb5e432a7af45de94961655714910f6.zip | |
kernel: FB/BUS info-list controls + PCIe link caps for the UVM path
CUDA (cuInit) now reaches UVM's device create: libcuda opens nvidia-uvm,
UVM calls nvGpuOpsDeviceCreate/gpuGetGpuInfo, and the driver fails at
getPCIELinkRateMBps -> calculatePCIELinkRateMBps ("Unknown PCIe speed",
NV_ERR_INVALID_STATE) -> nv_open_device aborts -> cuInit returns
CUDA_ERROR_NOT_INITIALIZED (3).
Implemented along the way:
- FB_GET_INFO_V2 (0x20801303): echo the requested FB info indices and
answer TU104 values (256MB RAM/heap/BAR1, GDDR6, 8 LTCs, 4MB L2, ...).
Fixes the kern_mem_sys_ctrl.c:911 asserts and gives CUDA/UVM real
memory sizes.
- BUS_GET_INFO_V2 (0x20801823): echo indices; PCIe Gen3 x16 link caps /
GEN info / link status, SYSMEM_CONNECTION_TYPE=PCIE. UVM's
gpuGetSysmemLinkInfo reads SYSMEM_CONNECTION_TYPE and the nominal link
bandwidth from here.
- PCIe Link Capabilities in the emulated config space (0x8C/0x92) and in
the BAR0 PCFG mirror (gpuReadBusConfigReg -> NV_PCFG+0x84 = BAR0+
0x88084): MAX_SPEED=3 (8GT/s Gen3), MAX_WIDTH=0x10 (x16); link status
mirror at 0x88088. Also mirrored under the VF-window/low aliases.
- Temporary change-watch on PCFG[0x88084] to see if the guest overwrites
it (it does not - stays 0x103).
Verified: guest nvidia-smi -L still works and reports PCIe Generation
Max/Current = 3. cuInit still fails; the remaining question is which
address the driver's kbifGetGpuLinkCapabilities read sees as zero (the
PCFG shadow holds 0x103). Guest-side kprobe tracing of that function
panicked the VM (free_swap_cache page fault on a corrupted pointer) -
possible guest memory corruption to investigate separately.
Diffstat (limited to 'kernel/main.c')
| -rw-r--r-- | kernel/main.c | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/kernel/main.c b/kernel/main.c index 0fea09e..209942b 100644 --- a/kernel/main.c +++ b/kernel/main.c @@ -1723,14 +1723,14 @@ static int nv_vgpu_config_space_init(struct nv_vgpu_device *vgpu, struct pci_dev cfg[0x89] = 0x00; cfg[0x8a] = 0x10; /* Device Status */ cfg[0x8b] = 0x00; - cfg[0x8c] = 0x00; /* Link Capabilities */ - cfg[0x8d] = 0x00; + cfg[0x8c] = 0x03; /* Link Capabilities: MAX_SPEED 3:0 = 3 (8 GT/s) */ + cfg[0x8d] = 0x01; /* ... MAX_WIDTH 9:4 = 0x10 (x16) -> 0x103 */ cfg[0x8e] = 0x00; cfg[0x8f] = 0x00; cfg[0x90] = 0x00; /* Link Control */ cfg[0x91] = 0x00; - cfg[0x92] = 0x00; /* Link Status */ - cfg[0x93] = 0x00; + cfg[0x92] = 0x03; /* Link Status: speed 3:0 = 3 (Gen3) */ + cfg[0x93] = 0x01; /* ... width 9:4 = 0x10 (x16) */ cfg[0x94] = 0x00; /* Slot Capabilities */ cfg[0x95] = 0x00; cfg[0x96] = 0x00; |