Repository navigation
Conversation
Step: test-qwen-serving-host. Old pins: e166f3f. New base: 063d881 (origin/main). Sets the five kernel hashes (experts.cu, experts_pack.cu, experts_prefill.cu, qmm_group.cu, qmm_prefill.cu), the qmm_group prefix and the launch-sequence hash to 063d881's bytes. The prefix is now group[0..16095]: line 1 lost 37 bytes and the rest is byte-equal, so it covers the same text e166f3f's group[0..16132] did. No code changes; 2/2 tests pass. Recorded, as merged and reviewed: - #2229 (3d4f6aa): a prompt chunk's last layer runs q, attention, gate, o and the MLP over its last 128 rows (the launch-sequence pin). - efdf28f, fcb1529, f3bcb78, ecef17d, 2ebc3c2, 42d3461 (lean, no PR number in the history): line-1 comments of the five kernel files only.
Step: test-gradient-serving. Old pins: e166f3f. New base: 063d881 (origin/main). Sets the hash of the 46 changed entries in gradient-serving.tsv to 063d881's bytes; paths, modes (all raw) and note columns are unchanged. pocket-serving-approved.tsv needs no edit: its entries for these files have prefix 0, so they never read the manifest hash. No code changes; 1/1 test passes. Recorded, as merged and reviewed: - #1784 (cf63ff6): Flash Next at --tp 2 splits dense output columns with exact FP32 gathers. - #1784 (cde4fe7): the BF16 prompt tile folds a pinned (cut) weight's whole K-slice order. - #1942 (15f118a): a shared round's DeltaNet steps run as one launch over its lanes. - #1792 (52b9f90): the whole shared expert's down writes straight into its output slot. - #2160 (a54748a): beside another open stream a greedy chain stops at 0.6 a draft; wide rounds draft 6 deep. - #2246 (4294752): at --parallel 1 on one GPU a prompt pass takes 8,192 rows. - #2214 (e4d7f13): kept prompt states with drafts across ranks. - #2230 (1f2c993): Gemma 4's PLE kernel in its own image; controls and blob constant. - #2103 (a3962b4, 194115d): synchronous downloads stage through a pinned bounce, on integrated devices only, and the bounce counts in memory use. - #2289 (e9ee738): the remaining synchronous readbacks use that staging; close() releases it. - #2127 (b2c277a): the lane host's KV ledger counts the paged pool's promises. - 4fb2f65, 60d80da, 8cc6ac4, 36a87df, 41ff25d, efdf28f, fcb1529, f3bcb78, ecef17d, 2ebc3c2, 42d3461, a86c31c, 1913a46, 80218cd: comments, verbatim moves, test-block imports, exports, and opt-in or off-by-default paths only.
files: prefill_attention.cu (new constexpr P256 and IDS flags, false in every existing kernel), prefill_attention_rows.cuh (the 256-position page items and their page lookup) and cuda/abi.zig (memory_type_host). Its review and device runs were done before it landed, so the pins take main's bytes.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Part of #2031.
test-gradient-servingandtest-qwen-serving-hostpin the bytes of the serving kernels and launches, so nobody changes them without someone reading the change. Both fail on main: the pins were right ate166f3f79, and main has moved since. CI doesn't run either step, so nothing caught it.We read every pinned file that differs since then, and every commit that touched it: 28 commits, from 612 in the range plus tonight's landings. Each is in one of two groups:
flashnext_batch.cu's verbatim move intoflashnext_gdn_tpr.cuh, the opt-inTF_FP_NATIVEdecode and theTF_GLM_L2PFblobs.The reading these pins ask for was done in each of those reviews, so the pins take main's bytes. One commit for each step names the PRs it records, and a third takes #2011, which landed tonight after the first two were written. No code changes.
qmm_group.cu's pinned prefix moves from 16,132 to 16,095 bytes because line 1 is 37 bytes shorter. The prefix covers the same text.Tests
Both steps fail on main and pass here.
zig build testis otherwise unchanged, and the landing gate's host checks pass.