[Lldb-commits] [lldb] [lldb] IRMemoryMap zero address mapping fix (PR #99045)

[Jason Molenda via lldb-commits]

jasonmolenda wrote:

I don't think this change is correct, or I don't understand what it is trying to do.

When we're connected to a stub that can allocate memory in the target, none of this code is executed.  So lldb-server/debugserver will not hit this.

We send `qMemoryRegionInfo` packets to look for a block of memory that is unallocated, or has permissions no-read+no-write+no-execute.  The memory region scan starts at address 0 and moves up by the size of each memory region, looking for `size` bytes between allocated memory regions, and returns that address if it is found.

So I'm guessing we have (1) a target that cannot allocate memory, (2) a target that supported `qMemoryRegionInfo`, a non-accessible block of memory at address 0? (did I read this right?) and in that case, we end up picking address 0 as our host-side virtual address+memory buffer for expression data storage.

Can you show what your `qMemoryRegionInfo` results are for address 0 on this target, so I can confirm I didn't misread/misunderstand the situation here?

https://github.com/llvm/llvm-project/pull/99045