compute-runtime/FAQ.md

183 lines
8.2 KiB
Markdown
Raw Normal View History

# Frequently asked questions
## OS support
### Which Linux versions does NEO support?
NEO should work on any modern Linux distribution (i.e. Ubuntu, Fedora, etc.)
with default / stock configuration (no kernel patches), assuming the underlying
kernel's drm subsystem is 4.7 or higher. Newer platforms will require a kernel version
that provides support for that platform (e.g. Coffee Lake requires kernel 4.14 or higher).
Our default (most frequent) validation config is currently (as of Q4'18) Ubuntu 18.04 LTS.
Note: Neo will not work correctly on legacy kernels with Intel patches (e.g. 4.7.0.intel.r4.0).
### Does NEO support Microsoft Windows?
Our closed-source driver for Windows is using the same codebase. At this time,
we do not support compilation of the OpenCL stack for Windows. It is our long-term
intention to offer that option.
### Why is the feature set different in latest Windows driver vs. latest NEO on github?
Our Windows release process takes up to several weeks before drivers are available through intel.com
and/or Windows update. Features available in github will be available on Windows later.
Note: Older platforms (e.g. Broadwell) are considered to be in maintenance mode for Windows.
### Does NEO support Android?
The code can be ported and cross-compiled for Android systems. It is our long-term intention to offer that option.
## Platform support
### Which Intel platforms are supported by the driver?
See [README.md](https://github.com/intel/compute-runtime/blob/master/README.md).
### How can I check that my specific device is supported by the driver?
To check support for any device, you can follow these steps:
1. Go to [Ark]( https://ark.intel.com) and find your Device ID
1. Find the corresponding device ID label in [GMM]( https://github.com/intel/gmmlib/blob/master/Source/inc/common/igfxfmid.h)
1. Check if this device ID label is enumerated in the [supported device list](https://github.com/intel/compute-runtime/blob/master/runtime/dll/linux/devices/devices_base.inl)
### When will support for platform X be added?
We will start adding platform support after platform is disclosed by Intel.
It is our intention to offer full support ahead of platform's market availability.
## OpenCL version
### Which version of OpenCL is supported?
See [README.md](https://github.com/intel/compute-runtime/blob/master/README.md).
### Which platforms will receive OpenCL 2.2 support?
Any platforms supporting OpenCL 2.1 are eligible for move to OpenCL 2.2.
### How can I enable OpenCL 2.1 on the Apollo Lake / Broxton platform?
You can enable a higher version of OpenCL using the ForceOCLVersion debug flag.
## Feature: cl_cache
### What is cl_cache?
This is a mechanism to cache binary representations of OpenCL kernels provided in text form by the application.
By storing the binary representations, compiling is required only the first time, which improves performance.
### How can cl_cache be enabled?
In the working directory, manually create *cl_cache* directory.
The driver will use this directory to store the binary representations of the compiled kernels.
Note: This will work on all supported OSes.
### Configuring cl_cache location
Cached kernels can be stored in a different directory than the default one. This is useful when the application is installed into a directory for which the user doesn't have permissions.
#### Linux configuration
Set the environment variable named `cl_cache_dir` to new location of cl_cache directory.
#### Example:
If the application's directory is `/home/user/Document`, by default cl_cache will be stored in `/home/user/Document/cl_cache`.
If the new path should be `/home/user/Desktop/cl_cache_place`, set environment variable `cl_cache_dir` to `/home/user/Desktop/cl_cache_place`.
```bash
export cl_cache_dir=/home/user/Desktop/cl_cache_place
```
Subsequent application runs with passed source code and `cl_cache_dir` environment variable set will reuse previously cached kernel binaries instead of compiling kernels from source.
#### Windows configuration
To set the new location of cl_cache directory - in the registry `HKEY_LOCAL_MACHINE\SOFTWARE\Intel\IGFX\OCL`:
1. add key `cl_cache_dir`
1. add string value named <path_to_app> to `cl_cache_dir` key
1. set data of added value to desired location of cl_cache
#### Example:
If application is located in `C:\Program Files\application\app.exe`, by default cl_cache will be stored in `C:\Program Files\application\cl_cache`.
If the new path should be `C:\Users\USER\Documents\application\cl_cache`, to subkey `HKEY_LOCAL_MACHINE\SOFTWARE\Intel\IGFX\OCL\cl_cache_dir` add string value named `C:\Program Files\application\app.exe` with data `C:\Users\USER\Documents\application\cl_cache`.
e.g.
string value : `HKEY_LOCAL_MACHINE\SOFTWARE\Intel\IGFX\OCL\cl_cache_dir\C:\Program Files\application\app.exe`
data : `C:\Users\USER\Documents\application\cl_cache`
Neo will look for string value (REG_SZ) `C:\Program Files\application\app.exe` in key `HKEY_LOCAL_MACHINE\SOFTWARE\Intel\IGFX\OCL\cl_cache_dir`. Data of this string value will be used as new cl_cache dump directory for this specific application.
### What are the known limitations of cl_cache?
1. Not thread safe. (Workaround: Make sure your clBuildProgram calls are executed in thread safe fashion.)
1. Binary representation may not be compatible between various versions of NEO and IGC drivers. (Workaround: Manually empty *cl_cache* directory prior to update)
1. Cache is not automatically cleaned. (Workaround: Manually empty *cl_cache* directory)
1. Cache may exhaust disk space and cause further failures. (Workaround: Monitor and manually empty *cl_cache* directory)
1. Cache is not process safe.
## Feature: Out of order queues
### Implementation details of out of order queues implementation
Current implementation of out of order queues allows multiple kernels to be run concurently. This allows for better device utilization in scenarios where single kernel
doesn't fill whole device.
More details can be found here:
https://github.com/intel/compute-samples/tree/master/compute_samples/applications/commands_aggregation
https://www.iwocl.org/wp-content/uploads/iwocl-2019-michal-mrozek-intel-breaking-the-last-line-of-performance-border.pdf
### Known issues and limitations
1. Turning on profiling on out of order command queue serializes kernel execution.
1. Blocking command queue with user events blocks all further submissions until event is unblocked.
1. Commands blocked by user events, when unblocked are serialized as well.
## Feature: Double-precision emulation (FP64)
By default NEO driver enables double precision operations only on platforms with supporting hardware. This is signified by exposing the "cl_khr_fp64" extension in the extension string. For other platforms, this support can be emulated by the compiler (IGC).
### How do I enable emulation?
FP64 emulation can only be enabled on Linux. There are two settings that have to be set.
#### Runtime setting:
There are two ways you can enable this feature in NEO:
* Set an environment variable **OverrideDefaultFP64Settings** to **1**:
`OverrideDefaultFP64Settings=1`
* In **igdrcl.config** configuration file in the same directory as application binary (you may have to create this file) add a line as such:
`OverrideDefaultFP64Settings = 1`
#### Compiler setting:
IGC reads flags only from environment, so set **IGC_EnableDPEmulation** to **1** as such:
`IGC_EnableDPEmulation=1`
After both settings have been set you can run the application normally.
### Known issues and limitations
Intel does not claim full specification conformance when using emulated mode. We reserve the right to not fix issues that appear only in emulation mode. Performance degradation is to be expected and has not been measured by Intel.
## Who are we?
NEO OpenCL team is part of VTT (Visual Technologies Team).
Historically, our team was responsible for delivery of the closed source OpenCL driver for Windows, Linux, and Android.
Most of the developers are located in Poland (UTC+1 timezone).
### I would like to join your team. How can I apply for a job?
1. Go to: https://jobs.intel.com/ListJobs/All/Search/city/Gdansk/
1. Filter for OpenCL in the job title
1. Browse the descriptions looking for positions mentioning this GitHub repository