mozilla::SignalTrampoline is designed to work around a bug in older ARM
kernels; it constructs a trampoline function with a NOP slide and then
calls a specified function. This feat is accomplished using inline
assembly and naked functions, which is a GCC extension where you get to
write the entire body of your function using GCC inline assembly.
Unfortunately, the particular implementation that it uses requires the
specified function's address to be loaded into a register. GCC permits
this and we use input arguments to the assembly statement to ensure that
GCC knows it shouldn't clobber the incoming argument registers when
trying to load the function's address.
clang, however, complains about the use of input parameters in naked
functions. So we need to find something that will work on both GCC and
clang.
The trick is to realize that we're a) tail-calling the specified
function and b) we don't have to worry about calling a fully-general
function. We just have to worry about calling a function inside libxul,
and we can therefore "assume" that the offset between the branch and the
called function fits into the immediate field of a Thumb (or ARM) branch
instruction. (This assumption is not strictly true; the branch range is
+/-16MB or so and libxul is actually quite a bit bigger than that. But
it works in practice, and the linker will insert branch stubs if
necessary to make things work out OK.)
The upshot is that we can use a "b" instruction instead of a "bx"
instruction, and this makes clang much happier. As a small bonus, the
stub gets ever-so-much-more efficient, which is probably the
least-significant micro-optimization ever.
This removes the unnecessary setting of c-basic-offset from all
python-mode files.
This was automatically generated using
perl -pi -e 's/; *c-basic-offset: *[0-9]+//'
... on the affected files.
The bulk of these files are moz.build files but there a few others as
well.
MozReview-Commit-ID: 2pPf3DEiZqx
--HG--
extra : rebase_source : 0a7dcac80b924174a2c429b093791148ea6ac204
These new constructor accepts two RangedPtr<T> arguments.
MozReview-Commit-ID: 8a3bYserLMr
--HG--
extra : rebase_source : 216de17b7a51783fe48d604b432d4dc7df6ad6eb
This means we can return already_AddRefed<T> for any RefPtr<T>s
being held as instance variables easier.
MozReview-Commit-ID: HFHdkF8EUsK
--HG--
extra : rebase_source : df650d39c010386afcb8cb2dd48292c26fbc6501
This patch implements mozilla::NotNull, which is similar but not identicial to
gsl::not_null.
The current draft(?) implementation of gsl::not_null is at
https://github.com/Microsoft/GSL/blob/master/include/gsl.h.
The main difference is that not_null allows implicit conversion from T to
not_null<T>. In contrast, NotNull only allows explicit conversion from T to
NotNull<T> via WrapNotNull().
The rationale for this is that when converting from a less-constrained type to
a more constrained type, implicit conversion is undesirable. For example, if I
changed a function f from this:
f(int* aPtr);
to this:
f(gsl::not_null<int*> aPtr);
no call sites would have to be modified. But if I changed it to this:
f(mozilla::NotNull<int*> aPtr);
call sites *would* need to be modified. This is a good thing! It forces the
author to audit the call sites for non-nullness, and encourages them to
back-propagate NotNull throughout the code.
The other difference between not_null and NotNull is that not_null disables
pointer arithmetic, which means it cannot be used with array pointers. I have
not implemented this restriction for NotNull because it seems arbitrary and
unnecessary.
This patch:
- Adds MOZ_MUST_USE to AllocPolicy::checkSimulatedOOM().
- Adds MOZ_MUST_USE to LZ4::decompress() (both variants) and fixes their
comments.
- Changes the return type of SplayTree::insert() from bool to void, because it
always returns true and its callers don't check the return value.
- Changes the return type of SplayTree::finishInsertion() from T* to void,
because it makes things clearer -- it was just returning the aNew argument.
- Adds MOZ_MUST_USE to a Vector::growTo() (both variants).
--HG--
extra : rebase_source : 1547cdeb9ee71d0ecec608ab474ab5e75bfc4b42
It's an annotation that is used a lot, and should be used even more, so a
shorter name is better.
MozReview-Commit-ID: 1VS4Dney4WX
--HG--
extra : rebase_source : b26919c1b0fcb32e5339adeef5be5becae6032cf
This will allow MOZ_MUST_USE to be used for a different and more common case.
MozReview-Commit-ID: 4dQsdWjJfc6
--HG--
extra : rebase_source : 390ab56ef83d71eb6d28759a0195a79a78b153bd
The prohibition against returning a UniquePtr was based on
bugs in older compilers which required an explicit Move()
to properly transfer ownership. This is no longer the case
since we now have code returning this type directly.
--HG--
extra : rebase_source : 6190b3f14b650d448f18f7611a22576701657795
If the image load is from the same document that cached the image we are required to use the cached version. Otherwise we should be free to ignore the cached version.
Writing PopLastN in this way is probably a bit of a micro-optimization,
but at least it comes with tests and some comments for verifying the
goodness of the code.
The scheme in TestSegmentedVectors to use manually-annotated points and
magic numbers corresponding to those annotations works OK for small
numbers of operations. But for testing bulk push and pop, we're going
to be doing many more operations, so let's move to recording explicitly
in code the operations we expect to see, and checking those instead.
CLOSED TREE
Backed out changeset dbadb8fe5803 (bug 1216001)
Backed out changeset a30593ebd58e (bug 1216001)
Backed out changeset c1646ffa71b4 (bug 1216001)
Atomic's constructor is marked as constexpr, but it calls a
non-constexpr function, ToStorageTypeArgument::convert. For compilers
which require constexpr-ness on constructors to inline away the actual
constructor call, the call to ToStorageTypeArgument::convert completely
disables the constexpr-ness of the constructor. Let's fix this by
marking all relevant instances of ToStorageTypeArgument::convert as
MOZ_CONSTEXPR, thus satisfying the compiler once again.
constexpr-ness of things can change the ABI requirements, and MSVC 2013
doesn't support constexpr, so choosing it for clang-cl when emulating
MSVC 2013 will cause ABI incomap issues between object files compiled
with the two compilers.
This reverts part of the commit for the original bug.
On B2G/Android we get warnings if the begin/size parameters to the constructor
match the method names begin()/size(): declaration of 'size/begin' shadows
a member of 'this'.
The current implementation of HierarchicalStringList allows the following:
FOO.bar = [
'foo',
'bar',
]
while
FOO.bar += [
'foo',
'bar',
]
would be invalid because of the StrictOrderingOnAppendList enforcement.
It also allows to overwrite the entire list with a subsequent
FOO.bar = [
'baz',
]
while we've explicitly forbidden such things for every other list.
While in the vicinity, fix HierarchicalStringList._get_export_variable to not
call the HierarchicalStringList constructor uselessly.
The bulk of this commit was generated with a script, executed at the top
level of a typical source code checkout. The only non-machine-generated
part was modifying MFBT's moz.build to reflect the new naming.
CLOSED TREE makes big refactorings like this a piece of cake.
# The main substitution.
find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \
xargs perl -p -i -e '
s/nsRefPtr\.h/RefPtr\.h/g; # handle includes
s/nsRefPtr ?</RefPtr</g; # handle declarations and variables
'
# Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h.
perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h
# Handle nsRefPtr.h itself, a couple places that define constructors
# from nsRefPtr, and code generators specially. We do this here, rather
# than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename
# things like nsRefPtrHashtable.
perl -p -i -e 's/nsRefPtr/RefPtr/g' \
mfbt/nsRefPtr.h \
xpcom/glue/nsCOMPtr.h \
xpcom/base/OwningNonNull.h \
ipc/ipdl/ipdl/lower.py \
ipc/ipdl/ipdl/builtin.py \
dom/bindings/Codegen.py \
python/lldbutils/lldbutils/utils.py
# In our indiscriminate substitution above, we renamed
# nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up.
find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \
xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g'
if [ -d .git ]; then
git mv mfbt/nsRefPtr.h mfbt/RefPtr.h
else
hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h
fi
--HG--
rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
This commit was generated using the following script, executed at the
top level of a typical source code checkout.
# Don't modify select files in mfbt/ because it's not worth trying to
# tease out the dependencies currently.
#
# Don't modify anything in media/gmp-clearkey/0.1/ because those files
# use their own RefPtr, defined in their own RefCounted.h.
find . -name '*.cpp' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \
grep -v 'mfbt/RefPtr.h' | \
grep -v 'mfbt/nsRefPtr.h' | \
grep -v 'mfbt/RefCounted.h' | \
grep -v 'media/gmp-clearkey/0.1/' | \
xargs perl -p -i -e '
s/mozilla::RefPtr/nsRefPtr/g; # handle declarations in headers
s/\bRefPtr</nsRefPtr</g; # handle local variables in functions
s#mozilla/RefPtr.h#mozilla/nsRefPtr.h#; # handle #includes
s#mfbt/RefPtr.h#mfbt/nsRefPtr.h#; # handle strange #includes
'
# |using mozilla::RefPtr;| is OK; |using nsRefPtr;| is invalid syntax.
find . -name '*.cpp' -o -name '*.mm' | xargs sed -i -e '/using nsRefPtr/d'
# RefPtr.h used |byRef| for dealing with COM-style outparams.
# nsRefPtr.h uses |getter_AddRefs|.
# Fixup that mismatch.
find . -name '*.cpp' -o -name '*.h'| \
xargs perl -p -i -e 's/byRef/getter_AddRefs/g'
The problematic class is AtomicRefCountedWithFinalize, from gfx/layers/;
I don't know all the details for why that class is written the way it
is, but it is simpler to make nsRefPtr work the same way as
mozilla::RefPtr in this case.
A number of places depend on RefPtr.h providing this function. When we
s/RefPtr/nsRefPtr/, such places still need to be able to see this
function. Moving it to nsRefPtr.h makes it still visible before we
switch (since RefPtr.h includes nsRefPtr.h), and after we switch (since
every place that #includes RefPtr.h will now be #including nsRefPtr.h).
Various bits depend on RefPtr.h to provide RefCounted<T> and RefPtr<T>.
It will be easier to manage an automatic conversion from RefPtr<T> to
nsRefPtr<T> if we split out the dependency on RefCounted<T> first.
RefPtr.h's byref permits callees to see the incoming value of the
outparam; XPCOM's getter_AddRefs zeros outparams prior to the call, so
information doesn't leak through inadvertently. Given this difference,
we need to eliminate tests that depended on this (arguably dangerous)
behavior. The numerous assertion fixups are required because we're
removing construction and destruction of objects along the way.
This prevents operator overload resolution from failing when this file
is included in a file that uses Chromium IPC logging.
--HG--
extra : rebase_source : 661efe98060d405e87f828655b13c0c0a8a428e0
The patch removes 455 occurrences of FAIL_ON_WARNINGS from moz.build files, and
adds 78 instances of ALLOW_COMPILER_WARNINGS. About half of those 78 are in
code we control and which should be removable with a little effort.
--HG--
extra : rebase_source : 82e3387abfbd5f1471e953961d301d3d97ed2973
Since Char16.h is included everywhere, and MSVC 2015 uses the char16ptr_t trick
it contains, we include <string> everywhere, but that has the side effect of
breaking the build in subtle ways. One way around this would be to avoid including
Char16.h in the first place, but that requires more work than I was ready to put
in. So instead, just avoid including <string> by removing the conversion operator
for std::wstring.
Chris Peterson
Cameron McCormack
Andi-Bogdan Postelnicu
Wes Kocher
Nathan Froyd
Carsten "Tomcat" Book
Tom Tromey
Andrew McCreight
Jeff Walden
Terrence Cole
Boris Chiou
Chris Pearce
Seth Fowler
Luke Wagner
L. David Baron
Nicholas Nethercote
Bill McCloskey
Jan de Mooij
Michael Layzell
Benjamin Bouvier
James Cheng
Benoit Girard
Jon Coppeard
Dan Gohman
Ehsan Akhgari
Ralph Giles
Lee Salzman
Timothy Nikkel
Bobby Holley
David Keeler
Markus Stange
Mike Hommey
Nick Fitzgerald
Jonathan Watt
Nathan Froyd
Eric Rahm
Xidorn Quan
Sebastian Hengst
Trevor Saunders
Mats Palmgren
Phil Ringnalda
Thomas Zimmermann
Morgan Phillips
Jim Blandy
Brian Birtles
Andrea Marchesini
Jeff Muizelaar
Botond Ballo
Sean Stangl
Birunthan Mohanathas
Steve Fink
Jared Wein
Olli Pettay
Robert O'Callahan
Jeff Gilbert
Kan-Ru Chen
Makoto Kato