This does _not_ fix any potential uses of uninitialized memory. Without
this change, gcc issues false-positive -Wuninitialized warnings
I'm hoping this does not have a noticeable impact on performance
When fdbcli change storeType for storage engines,
we switch the store type of storage servers one by one gracefully.
This avoids recruiting multiple storage servers on the same process,
which can cause OOM error.
Before this fix, the encoding of Void happens to 'work' in that this
implementation can both read and write it, but it is not a flatbuffers
message since it contains a invalid offset (the implementation 'works'
because it does not dereference that offset).
If a actor returns something of a different type that is
assignable, it will work because the actor compiler will
compile this to an assignment. However, if you compile
without the actor compiler, this will fail.
* Rename Optional/ErrorOr cast_to to castTo.
* Make printable(Optional<T>) templated rather than restricted to StringRef types.
* Fixes bug in (unused) ErrorOr.castTo where an ErrorOr that was not set would lose its error.
There are several missing includes for cmath in the code, I added those.
Next, Coro returns a reference to a stack variable and this causes a
warning. As this is probably ok for Coro, I disabled the warning in
that file for GCC. I want to have this warning in the build system as
it is generally a very useful warning to have.
Another change is that major and minor are deprecated for a while now.
I replaced those with gnu_dev_major and gnu_dev_minor.
ErrorOr currently implements operators ==, !=, and <. These do not
compile because Error does not implement ==. This compiles on older
versions of gcc and clang because ErrorOr<T>::operator== is not used
anywhere. It is still wrong though and newer gcc versions complain.
I simply removed these methods.
The most interesting fix is that TraceEvent::~TraceEvent is currently
throwing exceptions. This is illegal behavior in C++11 and a idea in
older versions of C++. For now I simply removed the throw, but this
might need some more thought.
So the code as written works well on older gcc, but does not like newer gcc. I
don't fully know why member operators seem to be more in conflict with different
compiler versions, but experience has taught me to always use non-member
operators for maximum portability across compilers.
Make a small change to that effect here.
Signed-off-by: Robert Escriva <rescriva@dropbox.com>
Remove the use of relative paths. A header at foo/bar.h could be included by
files under foo/ with "bar.h", but would be included everywhere else as
"foo/bar.h". Adjust so that every include references such a header with the
latter form.
Signed-off-by: Robert Escriva <rescriva@dropbox.com>
tclinken