Module 1 — The HIP compute model across RDNA and CDNA
1–2 weeks · Part of Kernel Ladder · Prev: module-0-setup · Next: module-2-memory-bound
What this module is for
The same HIP source compiles for a 5700 XT and an MI300X, but the two run it very differently. This module builds the mental model you need to predict those differences before you measure them: wave size, LDS, the register file, and occupancy.
What changes between the rungs
| Topic | RDNA1 (RX 5700 XT) | RDNA4 (RX 9070 XT) | CDNA3 / CDNA4 (MI300X / MI355X) |
|---|---|---|---|
| Wave size | wave32 native, wave64 supported | wave32 native, wave64 supported | wave64 only |
| Matrix units | None | [[rocm-kernels-course/concepts#WMMA | WMMA]], including FP8 |
| Occupancy limited by | VGPRs per wave, LDS per workgroup | Same | VGPRs plus [[rocm-kernels-course/concepts#AGPR |
| Memory | GDDR6 | GDDR6 + Infinity Cache | HBM + Infinity Cache, chiplet layout |
| HIP target | gfx1010 | gfx1201 | gfx942 / gfx950 |
Exact per-card figures (CU counts, LDS size, cache sizes) are in the GPU hardware specifications table in the official ROCm documentation. Look them up rather than trusting a blog.
Concepts to own: grid, block and thread indexing; why warpSize must never be hard-coded to 32; LDS and bank conflicts; occupancy; __syncthreads, fences and atomics; reading the ISA the compiler emits (s_waitcnt, buffer_load, ds_read, v_mfma).
1.1 Three kernels from memory
Goal
Write vector add, SAXPY and 2D transpose in HIP without copying from anywhere.
Run each on the 5700 XT and the 9070 XT, log GB/s per card, and add MI300X numbers once you have Instinct time.
1.2 Transpose and bank conflicts
Goal
Make the LDS-tiled transpose reach 80%+ of peak bandwidth.
- Write a naive transpose. Note that either the reads or the writes are uncoalesced.
- Stage a tile through LDS so both global reads and writes coalesce.
- Measure LDS bank conflicts with
rocprofv3counters. Pad the tile by one element and measure again.
Done when
The padded LDS transpose reaches 80%+ of peak bandwidth on the 9070 XT, and you can explain every instruction in its ISA hot loop.
1.3 Find a race
ROCm has no race detector equivalent to CUDA’s racecheck. So:
- Write a histogram with a deliberate race (non-atomic increments to shared bins).
- Write a deterministic test that exposes it: a skewed input where many threads hit one bin, compared against a CPU count.
- Fix it twice: global atomics, then per-block LDS histograms merged at the end. Compare the speed of both.
1.4 Wave size
On the 9070 XT, compile one warp-level reduction for wave32 and again with -mwavefrontsize64, and explain the throughput difference. Then make the code correct for wave64-only CDNA without #ifdef per target.
1.5 Occupancy is not the goal
Sweep block size from 64 to 1024 and vary __launch_bounds__ on one kernel. Record VGPRs per thread (from -Rpass-analysis=kernel-resource-usage), waves per SIMD and throughput.
Done when
You have a plot of occupancy vs throughput and can explain why the fastest point is not the highest-occupancy point.
Check your understanding
- A kernel uses 128 VGPRs per thread. What happens to occupancy if you add 8 more, and why can that make it faster or slower?
- Why does a reduction written for 32 lanes give wrong answers on CDNA rather than just running slowly?