Bug 274814 - graphics/drm-515-kmod System locks up when i915kms loaded on 14.0-RC3
Summary: graphics/drm-515-kmod System locks up when i915kms loaded on 14.0-RC3
Status: New
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: amd64 Any
: --- Affects Only Me
Assignee: freebsd-x11 (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2023-10-30 15:09 UTC by David Fiander
Modified: 2024-03-06 15:03 UTC (History)
1 user (show)

See Also:
bugzilla: maintainer-feedback? (x11)


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description David Fiander 2023-10-30 15:09:24 UTC
I've got a Lenovo ThinkCentre with an Alder Lake i7-12700T. This processor includes integrated Intel® UHD Graphics 770. When I run 'fwget' it reports

Needed firmware packages: 'gpu-firmware-intel-kmod-alderlake gpu-firmware-intel-kmod-tigerlake'

I've installed drm-kmod, and it pulled in drm-515-kmod.

The machine boots fine, but when I run

$ kldload i915kms

the system locks up and I have to power cycle it. /var/log messages provides these messages from before the crash:

Oct 28 16:32:41 victoria kernel: iic0: <I2C generic I/O> on iicbus0
Oct 28 16:32:42 victoria kernel: drmn0: <drmn> on vgapci0
Oct 28 16:32:42 victoria kernel: vgapci0: child drmn0 requested pci_enable_io
Oct 28 16:32:42 victoria syslogd: last message repeated 1 times
Oct 28 16:32:42 victoria kernel: drmn0: [drm] Incompatible option enable_guc=-1 - undocumented flag
Oct 28 16:32:42 victoria kernel: [drm] Unable to create a private tmpfs mount, hugepage support will be disabled(-19).
Oct 28 16:32:42 victoria kernel: [drm] Got stolen memory base rx4b800000, size 0x4000000
Oct 28 16:32:42 victoria kernel: lkpi_iic0: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus1: <Philips I2C bus> on lkpi_iic0
Oct 28 16:32:42 victoria kernel: iic1: <I2C generic I/O> on iicbus1
Oct 28 16:32:42 victoria kernel: lkpi_iic1: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus2: <Philips I2C bus> on lkpi_iic1
Oct 28 16:32:42 victoria kernel: iic2: <I2C generic I/O> on iicbus2
Oct 28 16:32:42 victoria kernel: lkpi_iic2: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus3: <Philips I2C bus> on lkpi_iic2
Oct 28 16:32:42 victoria kernel: iic3: <I2C generic I/O> on iicbus3
Oct 28 16:32:42 victoria kernel: lkpi_iic3: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus4: <Philips I2C bus> on lkpi_iic3
Oct 28 16:32:42 victoria kernel: iic4: <I2C generic I/O> on iicbus4
Oct 28 16:32:42 victoria kernel: lkpi_iic4: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus5: <Philips I2C bus> on lkpi_iic4
Oct 28 16:32:42 victoria kernel: iic5: <I2C generic I/O> on iicbus5
Oct 28 16:32:42 victoria kernel: lkpi_iic5: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus6: <Philips I2C bus> on lkpi_iic5
Oct 28 16:32:42 victoria kernel: iic6: <I2C generic I/O> on iicbus6
Oct 28 16:32:42 victoria kernel: lkpi_iic6: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus7: <Philips I2C bus> on lkpi_iic6
Oct 28 16:32:42 victoria kernel: iic7: <I2C generic I/O> on iicbus7
Oct 28 16:32:42 victoria kernel: lkpi_iic7: <LinuxKPI I2C> on drmn0
Oct 28 16:32:42 victoria kernel: iicbus8: <Philips I2C bus> on lkpi_iic7
Oct 28 16:32:42 victoria kernel: iic8: <I2C generic I/O> on iicbus8
Oct 28 16:32:42 victoria kernel: lkpi_iic8: <LinuxKPI I2C>
Oct 28 16:32:42 victoria kernel:  on drmn0
Oct 28 16:32:42 victoria kernel: iicbus9: <Philips I2C bus> on lkpi_iic8
Oct 28 16:32:42 victoria kernel: iic9: <I2C generic I/O> on iicbus9
Oct 28 16:32:42 victoria kernel: drmn0: successfully loaded firmware image 'i915/adls_dmc_ver2_01.bin'
Oct 28 16:32:42 victoria kernel: drmn0: [drm] Finished loading DMC firmware i915/adls_dmc_ver2_01.bin (v2.1)
Oct 28 16:32:42 victoria kernel: drmn0: successfully loaded firmware image 'i915/tgl_guc_62.0.0.bin'
Oct 28 16:32:42 victoria kernel: drmn0: successfully loaded firmware image 'i915/tgl_huc_7.9.3.bin'

Here's the CPU information reported at boot time:

Oct 28 10:28:53 victoria kernel: CPU: 12th Gen Intel(R) Core(TM) i7-12700T (1382.40-MHz K8-class CPU)
Oct 28 10:28:53 victoria kernel:   Origin="GenuineIntel"  Id=0x90672  Family=0x6  Model=0x97  Stepping=2
Oct 28 10:28:53 victoria kernel:   Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
Oct 28 10:28:53 victoria kernel:   Features2=0x7ffafbff<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,SMX,EST,TM2,SSSE3,SDBG,FMA,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,MOVBE,POPCNT,TSCDLT,AESNI,XSAVE,OSXSAVE,AVX,F16C,RDRAND>
Oct 28 10:28:53 victoria kernel:   AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
Oct 28 10:28:53 victoria kernel:   AMD Features2=0x121<LAHF,ABM,Prefetch>
Oct 28 10:28:53 victoria kernel:   Structured Extended Features=0x239c27eb<FSGSBASE,TSCADJ,BMI1,AVX2,FDPEXC,SMEP,BMI2,ERMS,INVPCID,NFPUSG,RDSEED,ADX,SMAP,CLFLUSHOPT,CLWB,PROCTRACE,SHA>
Oct 28 10:28:53 victoria kernel:   Structured Extended Features2=0x98c027bc<UMIP,PKU,OSPKE,WAITPKG,GFNI,VAES,VPCLMULQDQ,TME,RDPID,MOVDIRI,MOVDIR64B>
Oct 28 10:28:53 victoria kernel:   Structured Extended Features3=0xfc1cc410<FSRM,MD_CLEAR,PCONFIG,IBT,IBPB,STIBP,L1DFL,ARCH_CAP,CORE_CAP,SSBD>
Oct 28 10:28:53 victoria kernel:   XSAVE Features=0xf<XSAVEOPT,XSAVEC,XINUSE,XSAVES>
Oct 28 10:28:53 victoria kernel:   IA32_ARCH_CAPS=0xfd6b<RDCL_NO,IBRS_ALL,SKIP_L1DFL_VME,MDS_NO,TAA_NO>
Oct 28 10:28:53 victoria kernel:   VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID,VID,PostIntr
Oct 28 10:28:53 victoria kernel:   TSC: P-state invariant, performance statistics
Comment 1 rkoberman 2023-10-30 23:13:25 UTC
I reported this issue in bug #270888, though it properly belongs on the FreeBSD github drm-kmod page at https://github.com/freebsd/drm-kmod/issues/252. While I dislike the idea of s separate bug reporting system outside of normal FreeBSD systems, that is how the graphics people insist on operating. You might want to add your report there. 

If you want to get graphics working at all, just delete the firmware from you system. The load will fail and, if you system behaves like mine, you won't get any hardware acceleration, but normal graphics will work using software rendering. Not great and keeps the CPUs rather busy for things like 1920x1080p video, but my CPUs do work and do not quite eat the entire capacity of my 12 threads. Other features like backlight control do work.
Comment 2 David Fiander 2023-10-31 13:41:51 UTC
(In reply to rkoberman from comment #1)
Thanks for following up. That's how I've been running for the past year or so. I'll check in on the issue you mentioned above.