[Lldb-commits] [PATCH] D77759: [lldb/Reproducers] Fix inline replay for functions returning string as (char*, size_t).

Jonas Devlieghere via Phabricator via lldb-commits lldb-commits at lists.llvm.org
Wed Apr 8 15:46:23 PDT 2020

JDevlieghere created this revision.
JDevlieghere added a reviewer: labath.
Herald added a subscriber: abidh.
JDevlieghere added a parent revision: D77602: [lldb/Reproducers] Support inline replay.

Several SB API functions return strings a `(char*, size_t)` pair. During capture, we serialize an empty string for the `char*` because the memory can be uninitialized.  During replay, we have custom replay redirects that ensure that we don't override the buffer from which we're reading, but rather a buffer on the heap with the given length. This is sufficient for the "regular" reproducer use case, where we only care about the side effects of the API calls, not the actual return values.

This is not sufficient for inline replay. For inline replay, we ignore all the incoming arguments, and re-execute the current function with the arguments serialized in the reproducer. This means that these function will update the deserialized copy of the arguments, rather than whatever was passed in by the SWIG wrapper. To solve this problem, I've extended the reproducer instrumentation with special case replayers and corresponding macros. They ignore the replayer in the registry and the incoming char pointer, and instead reinvoke the current method on the deserialized class, and populate the output argument. It's unfortunate that this needs to be special cased, but I don't see a better solution.




-------------- next part --------------
A non-text attachment was scrubbed...
Name: D77759.256124.patch
Type: text/x-patch
Size: 13463 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/lldb-commits/attachments/20200408/ca1cd003/attachment-0001.bin>

More information about the lldb-commits mailing list