Repository navigation
VABSDIFF (sad) translates (chain change) - #2053
Open
Know-It-All-Brother wants to merge 3 commits into
Open
Know-It-All-Brother wants to merge 3 commits into
Know-It-All-Brother wants to merge 3 commits into
Conversation
…re_orin.py runs about 190 operations, each the GPU compiler's own machine code for one PTX operation, through the translator, spirv-cross, the chain's MSL rewrites and Metal on this Mac's GPU, and compares every output word with what a Jetson Orin (sm_87) computed from the same 1,024 input rows. The translator tests checked the MSL a few encodings give, not the values; this one finds the FSETP.NE, SHFL lane, F2FP, rounding-mode and signed-zero differences the tracking issue lists, which KNOWN names until each fix takes its entry out. dm/gfx/tools/ampere_orin/capture.py records the fixture on any Linux machine with an Ampere GPU and NVIDIA's driver, with no toolkit or compiler on either side
…ules with its functions, so a rewrite that uses one (struct in PR seedds#1902's hfma_msl) runs as it does in a build instead of raising NameError
…ad (|a - b| + c) compiles to VABSDIFF (opcode 0x214), which the Ampere opcode table had no entry for, so a shader using it failed to translate. The emitter takes max(a, b) - min(a, b) + c, wrapping at 32 bits, comparing a and b unsigned when bit 73 is clear (.U32, as IMNMX) and signed when it is set, and it is registered for every ALU form, whose operands the shared reader takes; a predicate output other than PT or any other modifier bit still fails the translation, since none has been seen. test_dmsh_ampere_orin.py's isad (sad.u32) failed to translate before and now matches the Jetson Orin in all 1,024 lanes, and the other translator tests give what main gives; chain d136f647d7b7
github-actions
Bot
added
nvn
Know-It-All-Brother
added a commit
to Know-It-All-Brother/nxnative
that referenced
this pull request
Nov 9, 2026
…ator rebuilt from its own source; chain 44916a290a03
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 the Ampere accuracy series (#2040). It depends on #2041 for the test: the commits before this PR's own are #2041's and drop out once it lands, so this PR's own change is the last commit.
What and why
sad.u32compiles to VABSDIFF (opcode 0x214): |a − b| + c. The translator had no entry for it in any form, so a shader with one failed to translate.AVabsdiffcomputesmax(a, b) − min(a, b) + cwith 32-bit wraparound, comparing signed when bit 73 is set and unsigned when it's clear, the same convention IMNMX uses. The fixture's encoding has a predicate output at bits 81–83 set to PT. Anything else there, or in the other modifier bits, fails the shader rather than guessing, since I couldn't find a description of those fields (Mesa NAK has no VABSDIFF). Each one costs a min, a max, a subtract and an add.Reference entry: "VABSDIFF, the absolute difference with an add" in
docs/reference/shaders.md.How to verify
isadfailed to translate on main and now matches the Orin in all 1,024 lanes, so it comes out ofKNOWN. Not checked against the GPU: the signed form (the fixture only has.U32) and the non-register operand forms.Checked
test_dmsh_ampere_orin.py; every othertest_dmsh_*gives the same result as on main (test_dmsh_ampere_mem.py,test_dmsh_meshstage.pyandtest_dmsh_pairs.pyfail on main the same way);test_nsp_chain.py,test_xlate_anycpu.py,test_docs.py.tool_lock.py --update-lockanddmsh_chain.py --headerwere run (chain d136f647d7b7).regress.py: not run. My only local Switch 2 build's shader package comes from an older shader split than main's map, so main's chain can't build it. Ocarina of Time's build already translates every shader, so it can't contain this instruction.Docs
docs/reference/shaders.mdhas the entry, andpython3 dm/tools/docs_symptoms.pywas rundmsh_chain.py --headerwas run, anddmsh_chain.h, the translator dlls and the lock are part of the change