Replied in thread
@endali remember this? 
Recent searches
Search options
#gpu
4 posts4 participants0 posts today
After my #wake_word_detection #research has delievered fruits, I have plans to continue works in the voice domain. I would love if I could train a #TTS model which has #British accent so I would use it to practice.
I was wondering if I could do the inference on #A311D #NPU. However, as I am skimming papers of different models, having inference on A311D with reasonable performance seems unlikely. Even training of these models on my entry level #IntelArc #GPU would be painful.
Maybe I could just finetune an already existing models. I am also thinking about using #GeneticProgramming for some components of these TTS models to see if there will be better inference performance.
There are #FastSpeech2 and #SpeedySpeech which look promising. I wonder how much natural their accents will be. But they would be good starting points.
BTW, if anyone needs opensource models, I would love to work as a freelancer and have an #opensource job. Even if someone can just provide access to computation resources, that would be good.
Replied in thread
@TheZeldaZone and it sadly ends in,a cliffhanger with no resolution to the plot or a sequel due to poor performance in retail.
- I mean it's a good game but not a €60 title at launch. People would've enjoyed it more for €50 and they would've sold 10x the amount for €40 or as a €20 bundle as download title.
Since the end of the 7th generation there hasn't been any good game cataloge IMHO to justify paying €500+ for a console.
- Only the absurd #GPU prices in comparison make them "reasonable" but I'd say unless one loves console gaming and it's restrictions that's still a nope to me.
I'm still busy going through the #Xbox360 games I sniped before #GameStop closed in #Germany.
Amazing AMD publication of Open-Sourced GIM Driver For GPU Virtualization!
Marvelous!
Linux folks, update yo shit!
NVIDIA disclose new security flaw in their Linux GPU drivers
https://www.gamingonlinux.com/2025/04/nvidia-disclose-new-security-flaw-in-their-linux-gpu-drivers/
Replied in thread
@otherdog Unfortunately ridiculous seems like the new norm.
I was really hoping with Intel entering the GPU Market that there would be more competition and pricing adjustments, but that doesn't seem to be happening.
Having said that, if you can wait, not purchase right away, and if everyone does that, then the market will adjust down, because at some point they need to actually sell the hardware.
Quick HomeLab Update
As a follow up to this morning's iGPU 512M GART debugging (re: drm-61-kmod memory leak on amdgpu.ko), here are some pics. Reluctantly, the "ok fine" method for testing the memory leak allocation aspects involves using a non-integrated GPU which requires the same driver, for which this low-TDP single-slot AMD W7500 was acquired. Sometime tomorrow it will be installed in my workstation, which is the third chassis from the bottom (4U) situated above the two 5U chassis (HCI private cloud for GPU compute VMs).
More to follow...
Shader Programmers:
Are vector ops more efficient than scalar (ie if I add two float4s is that the same cost as adding two floats) or did that stop being a thing 15 years ago?
I just uploaded the 5000th #OpenCL hardware report to @sascha's gpuinfo.org database! 
And guess what #GPU I reserved the spot for: #Intel Arc B580 #Battlemage 
https://opencl.gpuinfo.org/displayreport.php?id=5000
I have contributed 4.2% (211) of all entries. 
opencl.gpuinfo.orgIntel(R) Arc(TM) B580 Graphics - OpenCL Hardware Database by Sascha Willems
Dynamic Register Allocation on AMD's RDNA 4 GPU Architecture
https://chipsandcheese.com/p/dynamic-register-allocation-on-amds
Chips and Cheese · Dynamic Register Allocation on AMD's RDNA 4 GPU Architecture
Continued thread
Panic most recently used by lkpikmalloc ...
Well, that was fast... didn't even get a mouse cursor of a full MATE Desktop menu system load. Was yet to connect kgdb to COM1 (need to swap from minicom to do so)... makes me want a PCIe RS232 card (for "comconsole_pcidev") so that I have a few more COMs to play with on redirects. Gotta love these iGPU tash-bins eh? "It's better than not having a GPU right?" ... not really.''
Closed bug report from the drm-515-kmod, discussing amdgpu memory leak. so, maybe a new one in drm-61-kmod, would not be surprised.
- https://github.com/freebsd/drm-kmod/issues/258
Short term revision of approach:
----
1. Today via post arrives, an AMD Radeon Pro W7500 (single slot 8GB, Navi-whatever gen)
2. I'll block off the iGPU during loader.conf sequence, using a "pptdev" blackhole (not for VM pt, but maybe an experiment for a 14.1 VM with the known-good amdgpu version).
3. Known as: throw money at the problem?
Some hardware notes:
----
1. This is not a Nvidia GPU situation; there are several generations of cards in the room which have been cycled through the workstation during "hardware isolation" and "process of elimination" sequences. I know those are stable, and which gen cards require which nvidia driver versions for stability purposes.
2. This is not a FreeBSD kernel issue, nor a Xorg "Plain Jane FrameBuffer" situation. The kernel (14.0, 14.1, 14.2) is stable and fine, and the basic vt driver for non-4K display-port functionality works fine. I can work all day in a series of tmux windows with some fifty or so panes, but that's not quite the optimal experience.
3. The AMD iGPU (Raphael) maxes out default to 512MB GART VRAM, and it can handle 240Hz @ 4K all day with no issues as long as that 512M doesn't get used up... that is until the latest amdgpu kmod drm, which crashes whenever it feels like it.
Michael... yes yes, I do have a lot of hardware, but this issue has surpassed the Sunk Cost Fallacy and has become a consumate knowledge-requirement process. I must know where this is failing so horrendously, otherwise the operating rule of "if it doesn't fulfill its hardware destiny, it will get the hammer and flames"... and the hardware is too nice for that - plus I could involve Supermicro support since it's still in warranty, but a replacement motherboard or CPU for the iGPU isn't going to solve a kernel module issue.
In the interim, laptop life and tablet meetings are getting me by, mostly decently.
Debug items of interest:
----
intsmb0: <AMD FCH SMBus Controller> at device 20.0 on pci0
intsmb0: Could not allocate I/O space
device_attach: intsmb0 attach returned 6
drmn0: Fetched VBIOS from VFCT
amdgpu: ATOM BIOS: 102-RAPHAEL-008
drmn0: Trusted Memory Zone (TMZ) feature not supported
drmn0: PCIE atomic ops is not supported
drmn0: VRAM: 512M 0x000000F400000000 - 0x000000F41FFFFFFF (512M used)
[drm ERROR :amdgpu_bo_init] Unable to set WC memtype for the aperture base
Loader items of usage:
----
# Multi-Console Output
# boot output primary: TTY, standard monitor via UEFI
# boot output secondary: COM1 RS232 Redirect (physical)
# boot output tertiary: COM2 RS232 Redirect (BMC SoL)
ipmi_load="YES"
boot_mute="NO"
boot_verbose="YES"
verbose_loading="YES"
boot_multicons="YES"
boot_serial="YES"
console="efi,comconsole,comconsole"
comconsole_port1="0x3F8"
comconsole_speed1="115200"
comconsole_port2="0x2F8"
comconsole_speed2="115200"
hw.uart.console="io:0x3f8,br:115200 io:0x2f8,br:115200"
New GPU meta just dropped!
Nvidia announces RTX 5060 Ti and 5060 mainstream graphics cards - but VRAM counts are concerning
#AMD splits #ROCm toolkit into two parts – ROCm #AMDGPU drivers get their own branch under Instinct #datacenter #GPU moniker
The new #datacenter Instinct driver is a renamed version of the #Linux AMDGPU driver packages that are already distributed and documented with ROCm. Previously, everything related to ROCm (including the amdgpu driver) existed as part of the ROCm software stack.
https://www.tomshardware.com/pc-components/gpus/amd-splits-rocm-toolkit-into-two-parts-rocm-amdgpu-drivers-get-their-own-branch-under-instinct-datacenter-gpu-moniker
Tom's Hardware · AMD splits ROCm toolkit into two parts – ROCm AMDGPU drivers get their own branch under Instinct datacenter GPU moniker
Magnetic field of a cow.
(Simulated magnetic field of a kitchen magnet sized and magnetized cow model)
Available now in Ionsolver: https://github.com/VioletSpace/IonSolver
#physics #gpu #hpc
FreeBSD CUDA drm-61-kmod
"Just going to test the current pkg driver, this will only take a second...", the old refrain goes. Surely, it will not punt away an hour or so of messing about in loader.conf on this EPYC system...
- Here are some notes to back-track a botched/crashing driver kernel panic situation.
- Standard stuff, nothing new over the years here with loader prompt.
- A few directives are specific to this system, though may provide a useful general reference.
- The server has an integrated GPU in addition to nvidia pcie, so a module blacklist for the "amdgpu" driver is necessary (EPYC 4564P).
Step 1: during boot-up, "exit to loader prompt"
Step 2: set/unset the values as needed at the loader prompt
unset nvidia_load
unset nvidia_modeset_load
unset hw.nvidiadrm.modeset
set module_blacklist=amdgpu,nvidia,nvidia_modeset
set machdep.hyperthreading_intr_allowed=0
set verbose_loading=YES
set boot_verbose=YES
set acpi_dsdt_load=YES
set audit_event_load=YES
kern.consmsgbuf_size=1048576
set loader_menu_title=waffenschwester
boot
Step 3: login to standard tty shell
Step 4: edit /boot/loader.conf (and maybe .local)
Step 5: edit /etc/rc.conf (and maybe .local)
Step 6: debug the vast output from kern.consmsgbuf logs
Using Integer Addition to Approximate Float Multiplication - Once the domain of esoteric scientific and business computing, floating point calc... - https://hackaday.com/2025/04/10/using-integer-addition-to-approximate-float-multiplication/ #softwaredevelopment #softwarehacks #floatingpoint #fpu #gpu
Hackaday · Using Integer Addition To Approximate Float MultiplicationOnce the domain of esoteric scientific and business computing, floating point calculations are now practically everywhere. From video games to large language models and kin, it would seem that a pr…
What an honor to start the #IWOCL conference with my keynote talk! Nowhere else you get to talk to so many #OpenCL and #SYCL experts in one room! I shared some updates on my #FluidX3D #CFD solver, how I optimized it at the smallest level of a single grid cell, to scale it up on the largest #Intel #Xeon6 #HPC systems that provide more memory capacity than any #GPU server. 
Can't wait to install my new GPU
Blog Post Forthcoming for GPUs
ok, it's morning nap time starting 5min ago... but I'm decently pleased that this morning I've wrangled the (yes, linux) kernel into behaving more decently (a little like FreeBSD's sanity) when allocating similarly ID'd PCIe devices to VMs for GPU passthrough.
Short version:
- many GPUs of same PCI ID type in host
- want certain VMs to get certain GPUs
- want main host to get one GPU for compute (server, no display)
FreeBSD:
- the above requires two loader.conf lines and one rc.conf line
Linux:
- oh lordy, how many modprobe and modules-load.d files, and grub entries, and kmod blacklists, rebuild initramfs, and rebuild grub menu, and don't forget to purge the entire nouveau driver from the kernel so it doens't try to backstab nvidia and steal your first born...