Repository navigation
config/gpu-kernels.config: pin to 60b8eef (gk#97: kqt_mm_q4k -7-8 % at 4 rows, bit-exact) - #2572
Conversation
PR Reviewer Guide 🔍(Review updated until commit 9cea2d8)Here are some key observations to aid the review process:
|
|
VERDICT: READY I enumerated the range rather than reading the table, So the pin is exactly the eight I expect, and two of them are behavioural, which is the part that needs a sentence in the description:
And one thing to keep as written: |
|
VERDICT at 9cea2d8: READY — the audit is green at the new sha and, more usefully, it is unchanged despite 298 new lines of kernel code. The gate, measured at the sha main would pinIdentical totals to the outgoing pin ( So +298 lines of kernel code introduce no new f16→f32 narrowing and no f16 accumulator. That is the claim a Q4_K/DP_Q4_0 change has to make and rarely states: the pipelined body keeps f32 accumulation, which is what "bit-exact against the old body" implies and what this audit independently confirms. 62 narrowings before and after is the specific number — a rewrite that had introduced a The one thing to run after this lands
On the tree measured in What this pin does not touchThe Loom corpus and its ratchet are in the fork, not here: a And the provenanceThe body names GEAK run |
|
VERDICT: READY on the pin at The pin identity, checked against the repo rather than the descriptionThe pin is gpu-kernels#97's merge commit and the branch head — not a commit that merely contains the fix, and not one 1–5 commits short of it. That is the case I have had to reject three times tonight, so it is worth stating plainly when it holds: this pin is exactly the change it claims to pin. And the sha lives in one place. The diff is the single line it says it is: What this PR cannot tell us, and the row that canA pin bump is the one change whose correctness cannot be judged from its own diff: the kernel-level evidence lives in gpu-kernels (#97's 40/40 against a CPU reference at T=1..8, the bit-identical guards on the neighbouring kernels, the 5 ABBA rounds), and what remains is whether this tree still passes with the new kernels. That is So: merge on green One note on the pairing
|
bong-water-water-bong
left a comment
There was a problem hiding this comment.
VERDICT at 9cea2d8: READY
Pin checked the way a pin has to be: it is the gk main tip, the bump is 8 commits, the sha is well-formed, and nothing else in the tree still names the old one.
config/gpu-kernels.config: repo=1bit-MONSTER/gpu-kernels
sha=baabfc3d6e… -> 60b8eef5848…
git ls-remote … refs/heads/main -> 60b8eef58487a774109284f6942a53dfa7fcd12c ← the pin IS the tip
git rev-list --count baabfc3d6..60b8eef58 -> 8
python3 tools/ci/config_pin_check.py --check -> 4 sha pin(s) are 40 lowercase hex characters
grep -rn <old sha> -> only the config line this PR replaces
What the 8 commits carry, and why that matters for a review
60b8eef58 kquant_tokens: a pipelined 4-token Q4_K path (GEAK) ← #97, reviewed
4d695e0f6 README item 3: the bench returns rc 4 now, not rc 0 ← #96
ddb4a3253 dense-prefill-bench: a run that prints no timing line fails ← #94, reviewed
263e44f09 hip-hrx: K >= 8192 batch under 256 tokens takes the plain m64 tile ← #93
82ec45d8a tools: check every object's kernel ids against the registry ← #92, reviewed
de38cd20e hip-hrx: Q4_0 in the dense-prefill kernels (x2.95 on a 27B, KLD 1…) ← #91
074b76146 docs: the four things that break a kernel measurement ← #82
4cc4d8d52 hip-hrx: why hip.fa_decode_d256 has two n_kv thresholds ← #88, reviewed
So this pin is the first time #91's Q4_0 prefill (claimed ×2.95 on a 27B) enters the engine tree — it is the one item in the list I have not reviewed, and a pin bump is where an unreviewed kernel change arrives. That is not a blocker for this PR (the pin is a deliberate pointer, and the change was reviewed in gk), but it is worth naming: the pin's performance claim now affects a number a lane could quote from this repo.
The rest are the ones I did review and they arrive as a set: the rc 4 bench guard (#94/#96), the per-object id check (#92), and #97's −7–8 % at 4 rows with the bit-exactness argument. A pin bump's review is mostly "do I know what is in it", and here the answer is yes for seven of eight.
Two checks the config pin does not do, and why that is fine
- No gitlink moves with it —
config/gpu-kernels.configis the single source (grepfinds the old sha only on the line this PR replaces), so there is no second place to forget. Pins that exist in two files are the ones that drift. --onlinestill reports the fork tip rather than upstream's, which is the right default for this repo: the config names1bit-MONSTER/gpu-kernels, and our fork is what the engine builds against. If upstreamrocm/...ever needs to be the reference, that is a different config.
One note, not a request: git ls-remote <url> main printed two lines for this repo (1356edde… and 60b8eef…), and 1356edde is a different commit (bench: merged-fork decode row on strixhalo…). Querying refs/heads/main explicitly returns only the real tip. I could not explain the double line — the ref filter should be unambiguous — but it is a property of what the remote advertises, not of this pin, and I am recording it rather than assuming it away, because a pin check that parses the first line of ls-remote would have taken the wrong sha.
|
Persistent review updated to latest commit 9cea2d8 |
Moves
config/gpu-kernels.configfrombaabfc3to60b8eef(gpu-kernels main, which ends with #97).What the pin brings
kqt_mm_q4k/kqt_swiglu_q4k_q4k. Bit-exact;kqt_mm_q4kis −7–8 % per dispatch in-graph at 4 rows on Qwen3.8-27B (C4-STREAMS); swiglu flat. GEAK runwf_a078f173-1c0The pinned tree's own suite
Engine main
21a186d9+ fork8aa770fbd(main's pin) + gk60b8eef, built clean. Run in one gpu-hold ticket (cl1051-gkpin97-suite, 18:19–18:37Z, no device fault):test-hrx-loader-mapped, was Not Run. It is new at this fork pin (serve: /health follows the backend; progress watchdog for a hung backend (P3) #394), and the engine build does not build its target. Built by name and run in its own ticket: PASS. Ctest skips it with no model (exit 77), so it was run directly onQwen3-0.6B-Q4_K_M.ggufin ticketcl1051-gkpin97-loader2: "2 weight buffer lines, 1 HRX0_Mapped, 0 failure(s)", rc 0, no device fault.ssm_conv|gdn: 18/18.No new failure. The kernel code embedded in this build's library is #97's (
kquant_tokens_q8.text77a716e8af70).🤖 Generated with Claude Code