Unable to run square example generated by HIP on gfx902

[mbozdogan-SwissVault]

I am trying to convert a simple CUDA code to HIP to be able to run it on V1000. I can compile and run cl code on the CPU&GPU. Actually GPU is not recognised since the GPU is embedded, but I can see the difference(with or without GPU support) when I use benchmark tests on it. I tried to compile it via ROCM 4.1, the gfx902 is not even in the supported device list(gputable.txt). So, I could generate HIP code but I could not compile it. I tried different versions of ROCM. I could generate executable via version 2.6, but I got the error below. I do not know what I am doing wrong. Thank you for your advice.

Device Info :

https://www.ibase.com.tw/english/download/Embedded_Computing/Disk-Size_SBCs/IB918_Datasheet.pdf AMD Ryzen Embedded V1000 AMD Raden Vega integrated Device name: pthread-AMD Ryzen Embedded V1605B with Radeon Vega Gfx

ERROR

$ ./square.out GPU device 0 doesn't not exist Aborted (core dumped)

Info

$ /opt/rocm/bin/rocm_agent_enumerator gfx000

sudo dmesg | grep gfx [ 5.872901] [drm] add ip block number 4 <gfx_v9_0>

$ /opt/rocm/bin/rocminfo ROCk module is loaded user is member of video group

HSA System Attributes

Runtime Version: 1.1 System Timestamp Freq.: 1000.000000MHz Sig. Max Wait Duration: 18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count) Machine Model: LARGE
System Endianness: LITTLE

HSA Agents

Agent 1

Name: AMD Ryzen Embedded V1605B with Radeon Vega Gfx Marketing Name: AMD Ryzen Embedded V1605B with Radeon Vega Gfx Vendor Name: CPU
Feature: None specified
Profile: FULL_PROFILE
Float Round Mode: NEAR
Max Queue Number: 0(0x0)
Queue Min Size: 0(0x0)
Queue Max Size: 0(0x0)
Queue Type: MULTI
Node: 0
Device Type: CPU
Cache Info:
L1: 32(0x20) KB
Chip ID: 0(0x0)
Cacheline Size: 64(0x40)
Max Clock Freq. (MHz): 2000
BDFID: 0
Internal Node ID: 0
Compute Unit: 8
SIMDs per CU: 4
Shader Engines: 1
Shader Arrs. per Eng.: 0
WatchPts on Addr. Ranges:1
Features: None Pool Info:
Pool 1
Segment: GLOBAL; FLAGS: KERNARG, FINE GRAINED Size: 16776832(0xfffe80) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Acessible by all: TRUE
Pool 2
Segment: GLOBAL; FLAGS: COARSE GRAINED
Size: 16776832(0xfffe80) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Alignment: 4KB
Acessible by all: TRUE
ISA Info:
N/A
*** Done ***

[mysoreanoop]

Did you figure out what was wrong?

[ppanchad-amd]

@mbozdogan-SwissVault , Sorry for the lack of response. Please try latest ROCm 6.0.2 (HIP 6.0.32831) to see if your issue still exists? If resolved, please close the ticket. Thanks.