dwc_eqos - jumbo packets up to 4088 #35
Merged
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Outdated suggestions cannot be applied.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.
Support jumbo packets up to 4088 bytes (MTU 4074). This means raising the receive buffer size up to 4096, but since it appears that NetAdapterCx allocates buffers at page granularity anyway, I don't think this actually costs anything.
I tried to enable 9014-byte packets, but with system-managed buffering, I can't see how to specify that I want a small buffer (e.g. 2KB or 4KB) while also indicating that I might receive a large packet (e.g. 9014 bytes) -- the two sizes appear to be tied together. I think if we want 9014-byte jumbo frames, we either use 12KB receive buffers (very inefficient use of memory) or we use driver-managed receive buffers (a bit more work that I can't finish tonight). So stick with a 4088 byte limit for now.