[cfe-dev] Locking files on Windows with clangd

[Dmitry.Kozhevnikov@jetbrains.com via cfe-dev]

Hi!

A short background: we at JetBrains are experimenting with using clangd with
CLion, and we're very happy with the results so far. Unfortunately, we've hit
a problem when header files are locked on Windows, so they can't be saved,
they break `git rebase` (leading to the loss of work), etc.

It happens when the file is opened using a memory mapped file. It's fairly
easy to reproduce with clangd:

1. Create a header big enough to pass the threshold in `shouldUseMmap()`
(lib/Support/MemoryBuffer.cpp)
2. Include it in a source file
3. Keep this source file opened via clangd
4. Now an attempt to modify this header will fail

The situation is especially unpleasant because the handle is locked not only
during the parse, but for all the time clangd holds the preamble containing
this header.

This issue is mitigated to an extent in libclang: non-system files are
considered "volatile" (see r160074, r334070), so memory mapped files are not
used for them. However, I feel like it's not enough: you can easily have a
`#pragma system_header` in your codebase (IIUC it affects the mentioned
behavior), locking such file and i.e. losing user's work during `git rebase`
is still unacceptable.

Also, I think the fact that the proper compiler has this behavior is also
unfortunate (forgotten build in the background might lead to the loss of data).

IIUC the main reason for having memory mapped files is to reduce the memory
footprint. However, for a regular translation unit, headers usually take several
megabytes, which IMO is tolerable, and is usually topped by other per-TU data
anyway.

Hence, what do you think about not using memory mapped files on Windows at all
for source files? Are there any implications that I'm not aware of?

The naive patch (obviously not for commit, just for the illustration):

--- a/lib/Basic/SourceManager.cpp
+++ b/lib/Basic/SourceManager.cpp
@@ -109,7 +109,12 @@ llvm::MemoryBuffer *ContentCache::getBuffer(DiagnosticsEngine &Diag,
return Buffer.getPointer();
}

+#ifndef _WIN32
bool isVolatile = SM.userFilesAreVolatile() && !IsSystemFile;
+#else
+ bool isVolatile = true;
+#endif
+
auto BufferOrError =
SM.getFileManager().getBufferForFile(ContentsEntry, isVolatile);

If I've not missed something, this would effectively disable memory mapped
files for source files on Windows, but they would still be used for
other potentially large files (including PCHs and preambles).

If this is unacceptable, at the very least, clangd should treat user files
similar to libclang (we can work on a patch in this case).

Note that there is a similar bug report for QtCreator/libclang with some
interesting comments: https://bugreports.qt.io/browse/QTCREATORBUG-15449 <https://bugreports.qt.io/browse/QTCREATORBUG-15449>

Cheers,
Dmitry.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20180722/857b7471/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2861 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20180722/857b7471/attachment.bin>