[snapshot] Refactor the FixedRawData bytecode
From recent discussions on https://crrev.com/c/2057355 it became clear that the FixedRawData serializer bytecodes need to be refactored. This CL - updates documentation to correctly specify that data length is given in kTaggedSize unit, and - changes the way smi roots are serialized s.t. deserialization no longer needs hacks to be correct on big-endian architectures and when (theoretically) deserializing sequences of smi roots. We do this by serializing the full object slot for smi roots instead of just the smi contents. Bug: v8:10374 Change-Id: I190172d8030888e1950413e436e43b3d6e3b8fd5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2130276 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Reviewed-by: Igor Sheludko <ishell@chromium.org> Reviewed-by: Junliang Yan <jyan@ca.ibm.com> Cr-Commit-Position: refs/heads/master@{#66966}
Showing
Please
register
or
sign in
to comment