mirror of
https://github.com/deepseek-ai/DeepEP.git
synced 2026-11-05 00:47:32 +00:00
[GH-ISSUE #395] A bug about 'sizeof(int4)' in 'deep_ep.cpp' #269
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepEP#269
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @arminzhu on GitHub (Sep 8, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/395
Today, I want to test my script, but I got:
Failed: Assertion error /root/DeepEP-main/csrc/deep_ep.cpp:341 '(x.size(1) * x.element_size()) % sizeof(int4) == 0'Then, I started a debug journey. I surprisingly found a strange phenomenon:
‘int4‘ should be measured in bytes. As far as I know, it takes 1 byte (although 0.5 bytes is enough), but here it is 16.
So, this check seems...
@hexclover commented on GitHub (Sep 8, 2025):
int4here is a pack of 4 32-bit integers, not a 4-bit integer.@arminzhu commented on GitHub (Sep 8, 2025):
Okay. The shape asserts may encounter some failures. I will close this issue.