diku-dk / futhark

:boom::computer::boom: A data-parallel functional programming language
http://futhark-lang.org
ISC License
2.4k stars 165 forks source link

Array Slicing #297

Closed McCulloughRT closed 7 years ago

McCulloughRT commented 7 years ago

I'm just getting started with futhark, and so I have a strong suspicion I'm simply doing something wrong, but whenever I attempt to slice an array using the a[i:j:s] syntax I end up with a runtime error.

Even with toy examples like: fun main():[]int32 = let arr = [1,2,3,4] in arr[::-1]

It compiles just fine, but at runtime produces the following OpenCL error: ./slice: slice.c:1484: OpenCL call clEnqueueNDRangeKernel(fut_cl_queue, kernel_copy_162, 1, NULL, global_work_sizze_170, local_work_sizze_174, 0, NULL, NULL) failed with error code -54 (Invalid work group size)

Can anyone provide some guidance? -Thanks

RasmusWL commented 7 years ago

Hi there. It seems like you are using an old version of the compiler, as the typename int32 is not a valid anymore -- instead you should use i32.

After changing this, your example works on my version of the compiler, so you probably just try updating the compiler :)

athas commented 7 years ago

I suspect the problem is unrelated to array slicing. Are you using macOS? It looks like an error I have seen before on that platform. The issue is that Futhark (by default) uses the first available OpenCL device, which on macOS is the CPU. While Futhark interrogates the chosen device about limitations (such as the maximum workgroup size) and adapts as necessary, the macOS CPU device seems to misreport its capabilities, so the first kernel launch ends up failing. I don't have access to a Mac, so I can't debug it well enough to figure out what's going on.

The workaround is simple: specify another OpenCL device or platform using the -d or -p options on the generated executable. For example, if you have an NVIDIA GeForce card that you would like to use, you could pass either -p NVIDIA or -d GeForce. You'll have to figure out yourself which other OpenCL platforms and devices are available, and what they call themselves. I usually use the clinfo tool for this purpose.

In the future, if you suspect that your program is wrong, it's probably a good idea to run it with futhark-c and see if you get a similar error. The OpenCL compiler is still much more fragile and sensitive to its runtime environment, while futhark-c should be able to handle anything.

McCulloughRT commented 7 years ago

Thanks to both of you!

Rasmus, I inadvertantly typed int32 in the code block above, generally I'm using i32 as recommended, but that is some interesting history on the compiler!

Athas, it looks like you're correct. I am compiling on a MacBook pro, and when I compile with futhark-c everything works as expected! Thank you! Is there anything I can do or provide to help in debugging the OpenCL compiler issue?

athas commented 7 years ago

If you can run clinfo and post the output here, that would be a start. Maybe I'm misunderstanding what the device is telling Futhark.

McCulloughRT commented 7 years ago

Here is the output of clinfo: `Number of platforms 1 Platform Name Apple Platform Vendor Apple Platform Version OpenCL 1.2 (Jan 4 2017 22:35:59) Platform Profile FULL_PROFILE Platform Extensions cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event

Platform Name Apple Number of devices 3 Device Name Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz Device Vendor Intel Device Vendor ID 0xffffffff Device Version OpenCL 1.2 Driver Version 1.1 Device OpenCL C Version OpenCL C 1.2 Device Type CPU Device Available Yes Device Profile FULL_PROFILE Max compute units 8 Max clock frequency 2500MHz Device Partition (core) Max number of sub-devices 0 Supported partition types None Max work item dimensions 3 Max work item sizes 1024x1x1 Max work group size 1024 Compiler Available Yes Linker Available Yes Preferred work group size multiple 1 Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a) float 4 / 4
double 2 / 2 (cl_khr_fp64) Half-precision Floating-point support (n/a) Single-precision Floating-point support (core) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations Yes Double-precision Floating-point support (cl_khr_fp64) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations No Address bits 64, Little-Endian Global memory size 17179869184 (16GiB) Error Correction support No Max memory allocation 4294967296 (4GiB) Unified memory for Host and Device Yes Minimum alignment for any data type 128 bytes Alignment of base address 1024 bits (128 bytes) Global Memory cache type Read/Write Global Memory cache size 64 Global Memory cache line 6291456 bytes Image support Yes Max number of samplers per kernel 16 Max size for 1D images from buffer 65536 pixels Max 1D or 2D image array size 2048 images Base address alignment for 2D image buffers 1 bytes Pitch alignment for 2D image buffers 1 bytes Max 2D image size 8192x8192 pixels Max 3D image size 2048x2048x2048 pixels Max number of read image args 128 Max number of write image args 8 Local memory type Global Local memory size 32768 (32KiB) Max constant buffer size 65536 (64KiB) Max number of constant args 8 Max size of kernel argument 4096 (4KiB) Queue properties
Out-of-order execution No Profiling Yes Prefer user sync for interop Yes Profiling timer resolution 1ns Execution capabilities
Run OpenCL kernels Yes Run native kernels Yes printf() buffer size 1048576 (1024KiB) Built-in kernels
Device Extensions cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_image2d_from_buffer cl_APPLE_fp64_basic_ops cl_APPLE_fixed_alpha_channel_orders cl_APPLE_biased_fixed_point_image_formats cl_APPLE_command_queue_priority

Device Name Iris Pro Device Vendor Intel Device Vendor ID 0x1024500 Device Version OpenCL 1.2 Driver Version 1.2(Jan 4 2017 22:48:41) Device OpenCL C Version OpenCL C 1.2 Device Type GPU Device Available Yes Device Profile FULL_PROFILE Max compute units 40 Max clock frequency 1200MHz Device Partition (core) Max number of sub-devices 0 Supported partition types None Max work item dimensions 3 Max work item sizes 512x512x512 Max work group size 512 Compiler Available Yes Linker Available Yes Preferred work group size multiple 32 Preferred / native vector sizes
char 1 / 1
short 1 / 1
int 1 / 1
long 1 / 1
half 0 / 0 (n/a) float 1 / 1
double 0 / 0 (n/a) Half-precision Floating-point support (n/a) Single-precision Floating-point support (core) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations Yes Double-precision Floating-point support (n/a) Address bits 64, Little-Endian Global memory size 1610612736 (1.5GiB) Error Correction support No Max memory allocation 402653184 (384MiB) Unified memory for Host and Device Yes Minimum alignment for any data type 128 bytes Alignment of base address 1024 bits (128 bytes) Global Memory cache type None Image support Yes Max number of samplers per kernel 16 Max size for 1D images from buffer 25165824 pixels Max 1D or 2D image array size 2048 images Base address alignment for 2D image buffers 4 bytes Pitch alignment for 2D image buffers 32 bytes Max 2D image size 16384x16384 pixels Max 3D image size 2048x2048x2048 pixels Max number of read image args 128 Max number of write image args 8 Local memory type Local Local memory size 65536 (64KiB) Max constant buffer size 65536 (64KiB) Max number of constant args 8 Max size of kernel argument 1024 Queue properties
Out-of-order execution No Profiling Yes Prefer user sync for interop Yes Profiling timer resolution 80ns Execution capabilities
Run OpenCL kernels Yes Run native kernels No printf() buffer size 1048576 (1024KiB) Built-in kernels
Device Extensions cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_image2d_from_buffer cl_khr_gl_depth_images cl_khr_depth_images cl_khr_3d_image_writes

Device Name AMD Radeon R9 M370X Compute Engine Device Vendor AMD Device Vendor ID 0x1021c00 Device Version OpenCL 1.2 Driver Version 1.2 (Jan 10 2017 22:25:08) Device OpenCL C Version OpenCL C 1.2 Device Type GPU Device Available Yes Device Profile FULL_PROFILE Max compute units 10 Max clock frequency 800MHz Device Partition (core) Max number of sub-devices 0 Supported partition types None Max work item dimensions 3 Max work item sizes 256x256x256 Max work group size 256 Compiler Available Yes Linker Available Yes Preferred work group size multiple 64 Preferred / native vector sizes
char 4 / 4
short 2 / 2
int 1 / 1
long 1 / 1
half 0 / 0 (n/a) float 1 / 1
double 1 / 1 (cl_khr_fp64) Half-precision Floating-point support (n/a) Single-precision Floating-point support (core) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add No Support is emulated in software No Correctly-rounded divide and sqrt operations Yes Double-precision Floating-point support (cl_khr_fp64) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations No Address bits 32, Little-Endian Global memory size 2147483648 (2GiB) Error Correction support No Max memory allocation 536870912 (512MiB) Unified memory for Host and Device No Minimum alignment for any data type 128 bytes Alignment of base address 32768 bits (4096 bytes) Global Memory cache type None Image support Yes Max number of samplers per kernel 16 Max size for 1D images from buffer 65536 pixels Max 1D or 2D image array size 2048 images Base address alignment for 2D image buffers 256 bytes Pitch alignment for 2D image buffers 256 bytes Max 2D image size 16384x16384 pixels Max 3D image size 2048x2048x2048 pixels Max number of read image args 128 Max number of write image args 8 Local memory type Local Local memory size 32768 (32KiB) Max constant buffer size 65536 (64KiB) Max number of constant args 8 Max size of kernel argument 1024 Queue properties
Out-of-order execution No Profiling Yes Prefer user sync for interop Yes Profiling timer resolution 37ns Execution capabilities
Run OpenCL kernels Yes Run native kernels No printf() buffer size 134217728 (128MiB) Built-in kernels
Device Extensions cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_image2d_from_buffer cl_khr_depth_images cl_APPLE_command_queue_priority cl_APPLE_command_queue_select_compute_units cl_khr_fp64

NULL platform behavior clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Apple clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [P0] clCreateContext(NULL, ...) [default] Success [P0] clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) Success (1) Platform Name Apple Device Name Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (2) Platform Name Apple Device Name Iris Pro Device Name AMD Radeon R9 M370X Compute Engine clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) <checkNullCtxFromType:2350: create context from type CL_DEVICE_TYPE_CUSTOM : error -30> clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (3) Platform Name Apple Device Name Iris Pro Device Name AMD Radeon R9 M370X Compute Engine Device Name Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz`

athas commented 7 years ago

I have no idea why Futhark fails when running on the CPU device - it seems to report Max work group size: 1024, which is not less than what Futhark will use. I suspect there's a bug in Apple's CPU driver. I will take a closer look when I manage to get physical access to a Mac one day.

Anyway, your workaround would be passing -d Radeon or -d Iris to the executable generated by Futhark, depending on whether you want to use the AMD GPU or the integrated Intel GPU.