[DRAFT] Changes fp8 implementation to more closely match NCCL, and added logi… #1619
+647
−488
Conversation
This file contains 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
corey-derochie-amd
force-pushed
the
fp8-fnuz
branch
2 times, most recently
from
25f9475 to
e4fcde4
Compare
corey-derochie-amd
force-pushed
the
fp8-fnuz
branch
from
27a6182 to
da94a6f
Compare
corey-derochie-amd
requested review from
wenkaidu,
gilbertlee-amd,
akolliasAMD,
PedramAlizadeh,
nusislam,
nileshnegi,
KawtharShafie,
AtlantaPepsi,
mberenjk,
mustafabar,
thananon,
JhaShweta1,
BertanDogancay,
rahulvaidya20,
jee7s,
isaki001 and
PJAvinash
as code owners
corey-derochie-amd
requested review from
AbandiGa,
Nikhil-Nunna and
haripriya-amd
as code owners
corey-derochie-amd
changed the title
| @@ -277,11 +277,19 @@ static ncclResult_t hostToDevRedOp( | |||
| #if defined(RCCL_FLOAT8) | |||
| case ncclFloat8e4m3: | |||
| opFull->op = ncclDevPreMulSum; | |||
| fp8_e4m3 = (rccl_float8)(float(1.0/comm->nRanks)); | |||
| if (rccl_float8_useFnuz) { | |||
| fp8_e4m3_fnuz = (rccl_float8_fnuz)(float(1.0/comm->nRanks)); | |||
There was a problem hiding this comment.
Why do we cast from double to float and then again to the 8-bit float type? Are casts from double to rccl_float8_fnuz not possible?
I'm also curious why we're not using static_cast instead of c-style casts.
test/common/PtrUnion.cpp
Outdated
| case ncclFloat8e4m3: F1[idx] = rccl_float8(ReduceOp(op, float(F1[idx]), float(inputCpu.F1[idx]))); break; | ||
| case ncclFloat8e4m3: | ||
| { | ||
| if (PtrUnion_Float8UseFnuz) { |
There was a problem hiding this comment.
Does this slow down rccl unit testing? If so, by how much?
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Outdated suggestions cannot be applied.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.
…c to handle FNUZ types.
Details
Do not mention proprietary info or link to internal work items in this PR.
Work item: "Internal", or link to GitHub issue (if applicable).
What were the changes?
One sentence describing the work done.
Why were the changes made?
Explain the motivation behind the work. Provide any publicly-available historical context.
How was the outcome achieved?
Technical details behind the work. Explain any publicly-available hardware peculiarities.
Additional Documentation:
What else should the reviewer know?
Approval Checklist
Do not approve until these items are satisfied.