GitLab

The recent cuda variant changes uncovered a bug in ByLibrary
which failed to group by common variant for GPU types.

During rebasing, the ordering was inverted causing the cuda version
selection logic to break, with driver version being evaluated as zero
incorrectly causing a downgrade to v11.

Based on compute capability and driver version, pick
v12 or v11 cuda variants.

This adds new variants for arm64 specific to Jetson platforms

This adjusts linux to follow a similar model to windows with a discrete archive
(zip/tgz) to cary the primary executable, and dependent libraries. Runners are
still carried as payloads inside the main binary

Darwin retain the payload model where the go binary is fully self contained.

If the system has multiple numa nodes, enable numa support in llama.cpp
If we detect numactl in the path, use that, else use the basic "distribute" mode.

For systems that enumerate over 10 CPUs the default lexicographical
sort order interleaves CPUs and GPUs.

The v5 hip library returns unsupported GPUs which wont enumerate at
inference time in the runner so this makes sure we align discovery.  The
gfx906 cards are no longer supported so we shouldn't compile with that
GPU type as it wont enumerate at runtime.

* llm: avoid loading model if system memory is too small

* update log

* Instrument swap free space

On linux and windows, expose how much swap space is available
so we can take that into consideration when scheduling models

* use `systemSwapFreeMemory` in check

---------
Co-authored-by: Daniel Hiltgen <daniel@ollama.com>

This also adjusts our algorithm to favor our bundled ROCm.
I've confirmed VRAM reporting still doesn't work properly so we
can't yet enable concurrency by default.

This adds logic to detect skew between the driver and
management library which can be attributed to OS overhead
and records that so we can adjust subsequent management
library free VRAM updates and avoid OOM scenarios.

Refine the way we log GPU discovery to improve the non-debug
output, and report more actionable log messages when possible
to help users troubleshoot on their own.

Until ROCm v6.2 ships, we wont be able to get accurate free memory
reporting on windows, which makes automatic concurrency too risky.
Users can still opt-in but will need to pay attention to model sizes otherwise they may thrash/page VRAM or cause OOM crashes.
All other platforms and GPUs have accurate VRAM reporting wired
up now, so we can turn on concurrency by default.

This reverts commit 755b4e4f.

pointer deref's weren't correct on a few libraries, which explains
some crashes on older systems or miswired symlinks for discovery libraries.

This reverts commit 163cd3e7.

This seems to be ROCm version, not actually driver version, but
it may be useful for toggling logic for VRAM reporting in the future

We update the PATH on windows to get the CLI mapped, but this has
an unintended side effect of causing other apps that may use our bundled
DLLs to get terminated when we upgrade.

Signed-off-by: Lei Jitang <leijitang@outlook.com>

* gpu: add env var for detecting intel oneapi gpus

* fix build error

Also removes an unused overall count variable

Signed-off-by: Lei Jitang <leijitang@outlook.com>

This should aid in troubleshooting by capturing and reporting the GPU
settings at startup in the logs along with all the other server settings.

Implement support for GPU env var workarounds, and leverage
this for the Vega RX 56 which needs
HSA_ENABLE_SDMA=0 set to work properly