Technically speaking, all of the available GPU benchmarking tools in Linux only test under the OpenGL renderer of course.

While your GPU may be compatible with some versions of Direct3D, it is not possible to test this renderer under Linux.

• GLX-Gears
• GL Mark 2
• Hashcat
• Unigine Benchmarks

hashcat -I

returns:

hashcat (v6.2.5) starting in backend information mode
 
OpenCL Info:
============
 
OpenCL Platform ID #1
  Vendor..: Advanced Micro Devices, Inc.
  Name....: AMD Accelerated Parallel Processing
  Version.: OpenCL 2.1 AMD-APP (3513.0)
 
  Backend Device ID #1
    Type...........: GPU
    Vendor.ID......: 1
    Vendor.........: Advanced Micro Devices, Inc.
    Name...........: Radeon RX 7900 XTX
    Version........: OpenCL 2.0 
    Processor(s)...: 48
    Clock..........: 3220
    Memory.Total...: 24560 MB (limited to 20876 MB allocatable in one block)
    Memory.Free....: 24448 MB
    OpenCL.Version.: OpenCL C 2.0 
    Driver.Version.: 3513.0 (HSA1.1,LC)
    PCI.Addr.BDF...: 0c:00.0

hashcat -b | uniq

returns:

hashcat (v6.2.5) starting in benchmark mode
 
Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.
 
OpenCL API (OpenCL 2.1 AMD-APP (3513.0)) - Platform #1 [Advanced Micro Devices, Inc.]
=====================================================================================
* Device #1: Radeon RX 7900 XTX, 24448/24560 MB (20876 MB allocatable), 48MCU
 
Benchmark relevant options:
===========================
* --optimized-kernel-enable
 
-------------------
* Hash-Mode 0 (MD5)
-------------------
 
Speed.#1.........: 70140.4 MH/s (11.02ms) @ Accel:512 Loops:1024 Thr:32 Vec:1
 
----------------------
* Hash-Mode 100 (SHA1)
----------------------
 
Speed.#1.........: 30426.9 MH/s (26.05ms) @ Accel:128 Loops:1024 Thr:128 Vec:1
...