Commit Graph

650 Commits

Author SHA1 Message Date
Richard Smith 87bb569844 [modules] Fix some visibility issues with default template arguments.
There are still problems here, but this is a better starting point.

The main part of the change is: when doing a lookup that would accept visible
or hidden declarations, prefer to produce the latest visible declaration if
there are any visible declarations, rather than always producing the latest
declaration.

Thus, when we inherit default arguments (and other properties) from a previous
declaration, we inherit them from the previous visible declaration; if the
previous declaration is hidden, we already suppress inheritance of default
arguments.

There are a couple of other changes here that fix latent bugs exposed by this
change.

llvm-svn: 239371
2015-06-09 00:35:49 +00:00
Richard Smith 04765ae01e [modules] If we re-enter a submodule from within itself (when submodule
visibility is enabled) or leave and re-enter it, restore the macro and module
visibility state from last time we were in that submodule.

This allows mutually-#including header files to stand a chance at being
modularized with local visibility enabled.

llvm-svn: 237871
2015-05-21 01:20:10 +00:00
Richard Smith fd8b64eea0 [modules] Support merging a parsed default argument with an imported hidden one for non-type and template template parameters too.
llvm-svn: 237815
2015-05-20 18:24:21 +00:00
Richard Smith c7d48d1a16 [modules] Support merging a parsed default function/template argument with an imported but hidden one.
llvm-svn: 237814
2015-05-20 17:50:35 +00:00
Richard Smith 54ef4c3bb3 Revert r237609 for now.
glibc's headers use __need_* macros to selectively export parts of themselves
to each other. This requires us to enter those files repeatedly when building
a glibc module.

This can be unreverted once we have a better mechanism to deal with that
non-modular aspect of glibc (possibly some way to mark a header as "textual if
this macro is defined").

llvm-svn: 237718
2015-05-19 19:58:11 +00:00
Richard Smith 82e57fb1e8 [modules] Support for merging a parsed definition of a static data member of a class template into an imported but hidden definition.
llvm-svn: 237647
2015-05-19 00:49:29 +00:00
Richard Smith c7e6ff02d5 [modules] Support for merging a parsed class template specialization definition into an imported but hidden definition.
llvm-svn: 237612
2015-05-18 20:36:47 +00:00
Richard Smith 2d4a11fba6 [modules] When a file is listed as a non-textual header in a module map, don't
enter it more than once, even if it doesn't have #include guards -- we already
know that it is intended to have the same effect every time it's included, and
it's already had that effect. This particularly helps with local submodule
visibility builds, where the include guard macro may not be visible in the
includer, but will become visible the moment we enter the included file.

llvm-svn: 237609
2015-05-18 20:02:41 +00:00
Richard Smith a0aafa3853 [modules] If we see a #include that maps to a module, but use of precompiled modules is disabled, track submodule visibility anyway if -fmodules-local-submodule-visibility is enabled. This, in effect, gives modules semantics but without precompilation.
llvm-svn: 237550
2015-05-18 03:52:30 +00:00
Richard Smith 4241314164 [modules] Add local submodule visibility support for declarations.
With this change, enabling -fmodules-local-submodule-visibility results in name
visibility rules being applied to submodules of the current module in addition
to imported modules (that is, names no longer "leak" between submodules of the
same top-level module). This also makes it much safer to textually include a
non-modular library into a module: each submodule that textually includes that
library will get its own "copy" of that library, and so the library becomes
visible no matter which including submodule you import.

llvm-svn: 237473
2015-05-15 20:05:43 +00:00
Richard Smith d9ba224f66 [modules] Suport for merging a parsed enum definition into an existing imported but not visible definition.
llvm-svn: 236690
2015-05-07 03:54:19 +00:00
Peter Collingbourne b2ed4d219b Fix public-private.modulemap test case to clear the cache on every run.
llvm-svn: 236655
2015-05-06 22:31:13 +00:00
Richard Smith 36bd40df72 Switch PPCallbacks to take the new MacroDefinition instead of MacroDirective*, in order to preserve full information on module macro expansion.
llvm-svn: 236404
2015-05-04 03:15:40 +00:00
Richard Smith 38477db8c4 [modules] If a module #includes a modular header that #undef's its macro, it
should not export the macro.

... at least, not unless we have local submodule visibility enabled.

llvm-svn: 236369
2015-05-02 00:45:56 +00:00
Richard Smith ee977933f7 [modules] Add -fmodules-local-submodule-visibility flag.
This flag specifies that the normal visibility rules should be used even for
local submodules (submodules of the currently-being-built module). Thus names
will only be visible if a header / module that declares them has actually been
included / imported, and not merely because a submodule that happened to be
built earlier declared those names. This also removes the need to modularize
bottom-up: textually-included headers will be included into every submodule
that includes them, since their include guards will not leak between modules.

So far, this only governs visibility of macros, not of declarations, so is not
ready for real use yet.

llvm-svn: 236350
2015-05-01 21:22:17 +00:00
Richard Smith 20e883e59b [modules] Stop trying to fake up a linear MacroDirective history.
Modules builds fundamentally have a non-linear macro history. In the interest
of better source fidelity, represent the macro definition information
faithfully: we have a linear macro directive history within each module, and at
any point we have a unique "latest" local macro directive and a collection of
visible imported directives. This also removes the attendent complexity of
attempting to create a correct MacroDirective history (which we got wrong
in the general case).

No functionality change intended.

llvm-svn: 236176
2015-04-29 23:20:19 +00:00
Richard Smith 2a553089c3 [modules] Properly attribute macros to modules if they're in a file textually included into a file in the module.
llvm-svn: 235661
2015-04-23 22:58:06 +00:00
Richard Smith d6aab59ad1 [modules] Allow a function template definition if we have a pre-existing but not visible definition of the same template.
llvm-svn: 233430
2015-03-27 21:57:41 +00:00
Richard Smith df5bf99d0b Add file forgotten from r233420.
llvm-svn: 233425
2015-03-27 21:40:57 +00:00
Richard Smith 0279375836 [modules] When merging class definitions, make the retained definition visible
if the merged definition is visible, and perform lookups into all merged copies
of the definition (not just for special members) so that we can complete the
redecl chains for members of the class.

llvm-svn: 233420
2015-03-27 21:16:39 +00:00
Richard Smith 9a71c99844 [modules] Allow a function to be redefined if the old definition is not visible.
llvm-svn: 233407
2015-03-27 20:16:58 +00:00
Richard Smith a523022b53 [modules] Handle defining a tag with a typedef name for linkage purposes on top of an existing imported-but-not-visible definition.
llvm-svn: 233345
2015-03-27 01:37:43 +00:00
Richard Smith be3980b73c [modules] Handle defining a class template on top of an existing imported-but-not-visible definition.
llvm-svn: 233341
2015-03-27 00:41:57 +00:00
Chandler Carruth ffbf705cc3 [Modules] Fix a sneaky bug in r233249 where we would look for implicit
constructors in the current lexical context even though name lookup
found them via some other context merged into the redecl chain.

This can only happen for implicit constructors which can only have the
name of the type of the current context, so we can fix this by simply
*always* merging those names first. This also has the advantage of
removing the walk of the current lexical context from the common case
when this is the only constructor name we need to deal with (implicit or
otherwise).

I've enhanced the tests to cover this case (and uncovered an unrelated
bug which I fixed in r233325).

llvm-svn: 233327
2015-03-26 22:27:09 +00:00
Richard Smith 8f4d3ff146 [modules] Restrict the module use-declaration to only appear in top-level
modules, and allow sub-modules of a module with a use-declaration to make use
of the nominated modules.

llvm-svn: 233323
2015-03-26 22:10:01 +00:00
Chandler Carruth 52cee4dad2 [Modules] Preserve source order for the map of late parsed templates.
Clang was inserting these into a dense map. While it never iterated the
dense map during normal compilation, it did when emitting a module. Fix
this by using a standard MapVector to preserve the order in which we
encounter the late parsed templates.

I suspect this still isn't ideal, as we don't seem to remove things from
this map even when we mark the templates as no longer late parsed. But
I don't know enough about this particular extension to craft a nice,
subtle test case covering this. I've managed to get the stress test to
at least do some late parsing and demonstrate the core problem here.
This patch fixes the test and provides deterministic behavior which is
a strict improvement over the prior state.

I've cleaned up some of the code here as well to be explicit about
inserting when that is what is actually going on.

llvm-svn: 233264
2015-03-26 09:08:15 +00:00
Chandler Carruth 70a1369b34 [Modules] Add some more fun code to my modules stress test, this time
templates. Turns out all of this works correctly (so far). But it should
cover more code paths and will let me test some things that don't
actually work next.

llvm-svn: 233263
2015-03-26 08:49:55 +00:00
Chandler Carruth f85d98285e [Modules] Make "#pragma weak" undeclared identifiers be tracked
deterministically.

This fixes a latent issue where even Clang's Sema (and diagnostics) were
non-deterministic in the face of this pragma. The fix is super simple --
just use a MapVector so we track the order in which these are parsed (or
imported). Especially considering how rare they are, this seems like the
perfect tradeoff. I've also simplified the client code with judicious
use of auto and range based for loops.

I've added some pretty hilarious code to my stress test which now
survives the binary diff without issue.

llvm-svn: 233261
2015-03-26 08:32:49 +00:00
Chandler Carruth 8a3d24dcf8 [Modules] Delete a bunch of complex code for ensuring visible decls in
updated decl contexts get emitted.

Since this code was added, we have newer vastly simpler code for
handling this. The code I'm removing was very expensive and also
generated unstable order of declarations which made module outputs
non-deterministic.

All of the tests continue to pass for me and I'm able to check the
difference between the .pcm files after merging modules together.

llvm-svn: 233251
2015-03-26 04:27:10 +00:00
Richard Smith 65ebb4ac8a [modules] If we reach a definition of a class for which we already have a
non-visible definition, skip the new definition and make the old one visible
instead of trying to parse it again and failing horribly. C++'s ODR allows
us to assume that the two definitions are identical.

llvm-svn: 233250
2015-03-26 04:09:53 +00:00
Chandler Carruth e972c36221 [Modules] A second attempt at writing out on-disk hash tables for the
decl context lookup tables.

The first attepmt at this caused problems. We had significantly more
sources of non-determinism that I realized at first, and my change
essentially turned them from non-deterministic output into
use-after-free. Except that they weren't necessarily caught by tools
because the data wasn't really freed.

The new approach is much simpler. The first big simplification is to
inline the "visit" code and handle this directly. That works much
better, and I'll try to go and clean up the other caller of the visit
logic similarly.

The second key to the entire approach is that we need to *only* collect
names into a stable order at first. We then need to issue all of the
actual 'lookup()' calls in the stable order of the names so that we load
external results in a stable order. Once we have loaded all the results,
the table of results will stop being invalidated and we can walk all of
the names again and use the cheap 'noload_lookup()' method to quickly
get the results and serialize them.

To handle constructors and conversion functions (whose names can't be
stably ordered) in this approach, what we do is record only the visible
constructor and conversion function names at first. Then, if we have
any, we walk the decls of the class and add those names in the order
they occur in the AST. The rest falls out naturally.

This actually ends up simpler than the previous approach and seems much
more robust.

It uncovered a latent issue where we were building on-disk hash tables
for lookup results when the context was a linkage spec! This happened to
dodge all of the assert by some miracle. Instead, add a proper predicate
to the DeclContext class and use that which tests both for function
contexts and linkage specs.

It also uncovered PR23030 where we are forming somewhat bizarre negative
lookup results. I've just worked around this with a FIXME in place
because fixing this particular Clang bug seems quite hard.

I've flipped the first part of the test case I added for stability back
on in this commit. I'm taking it gradually to try and make sure the
build bots are happy this time.

llvm-svn: 233249
2015-03-26 03:11:40 +00:00
Daniel Jasper cb47d8d17b Temporarily disable one more non-determinism flag.
This turned the bots to red after r233172 which reverted r233156.

llvm-svn: 233173
2015-03-25 06:59:19 +00:00
Chandler Carruth d5f1db6b36 [Modules] Disable the diff of the merged module, there is still some
non-determinism here, I just got lucky a bunch of times on my system.

llvm-svn: 233163
2015-03-25 01:30:02 +00:00
Chandler Carruth 75fc6a7f61 [Modules] Make the DeclUpdates map be processed in insertion order.
This fixes my stress tests non-determinism so far. However, I've not
started playing with templates, friends, or terrible macros. I've found
at least two more seeming instabilities and am just waiting for a test
case to actually trigger them.

llvm-svn: 233162
2015-03-25 01:02:12 +00:00
Chandler Carruth 580dd296fa [Modules] Stop creating timestamps for the modules cache and trying to
prune it when we have disabled implicit module generation and thus are
not using any cached modules.

Also update a test of explicitly generated modules to pass this CC1 flag
correctly.

This fixes an issue where Clang was dropping files into the source tree
while running its tests.

llvm-svn: 233117
2015-03-24 21:44:25 +00:00
Chandler Carruth 885e78cb22 [Modules] Start making explicit modules produce deterministic output.
There are two aspects of non-determinism fixed here, which was the
minimum required to cause at least an empty module to be deterministic.

First, the random number signature is only inserted into the module when
we are building modules implicitly. The use case for these random
signatures is to work around the very fact that modules are not
deterministic in their output when working with the implicitly built and
populated module cache. Eventually this should go away entirely when
we're confident that Clang is producing deterministic output.

Second, the on-disk hash table is populated based on the order of
iteration over a DenseMap. Instead, use a MapVector so that we can walk
it in insertion order.

I've added a test that an empty module, when built twice, produces the
same binary PCM file.

llvm-svn: 233115
2015-03-24 21:18:10 +00:00
Richard Smith 9e2341d093 [modules] Remove redundant import of lexical decls when building a lookup table
for a DeclContext, and fix propagation of exception specifications along
redeclaration chains.

This reverts r232905, r232907, and r232907, which reverted r232793, r232853,
and r232853.

One additional change is present here to resolve issues with LLDB: distinguish
between whether lexical decls missing from the lookup table are local or are
provided by the external AST source, and still look in the external source if
that's where they came from.

llvm-svn: 232928
2015-03-23 03:25:59 +00:00
Benjamin Kramer b7e4a1a0b6 Remove empty files left behind by r232907.
llvm-svn: 232909
2015-03-22 10:08:31 +00:00
Vince Harron 08dcf60295 Reverting 232853 and 232870 because they depend on 232793,
which was reverted because it was causing LLDB test failures

llvm-svn: 232907
2015-03-22 08:47:07 +00:00
Vince Harron a3ea9a4e09 Reverting r232793 until some new LLDB failures are discussed
llvm-svn: 232905
2015-03-22 05:59:59 +00:00
Richard Smith decef8007f [modules] When either redecl chain merging or an update record causes us to
give an exception specification to a declaration that didn't have an exception
specification in any of our imported modules, emit an update record ourselves.
Without this, code importing the current module would not see an exception
specification that we could see and might have relied on.

llvm-svn: 232870
2015-03-21 00:58:54 +00:00
Richard Smith 5a07892dec [modules] Remove temporary IdentifierInfo lookup results when we're done with them.
llvm-svn: 232853
2015-03-20 21:56:50 +00:00
Richard Smith 625ccb3f78 [modules] Remove some redundant work when building a lookup table for a DeclContext.
When we need to build the lookup table for a DeclContext, we used to pull in
all lexical declarations for the context; instead, just build a lookup table
for the local lexical declarations. We previously didn't guarantee that the
imported declarations would be in the returned map, but in some cases we'd
happen to put them all in there regardless. Now we're even lazier about this.

This unnecessary work was papering over some other bugs:

 - LookupVisibleDecls would use the DC for name lookups in the TU in C, and
   this was not guaranteed to find all imported names (generally, the DC for
   the TU in C is not a reliable place to perform lookups). We now use an
   identifier-based lookup mechanism for this.

 - We didn't actually load in the list of eagerly-deserialized declarations
   when importing a module (so external definitions in a module wouldn't be
   emitted by users of those modules unless they happened to be deserialized
   by the user of the module).

llvm-svn: 232793
2015-03-20 02:17:21 +00:00
Richard Smith 7f330cdb31 Make module files passed to a module build via -fmodule-file= available to
consumers of that module.

Previously, such a file would only be available if the module happened to
actually import something from that module.

llvm-svn: 232583
2015-03-18 01:42:29 +00:00
Richard Smith 15e32fd215 [modules] Fix bug where an anonymous namespace could cause the containing
namespace to not merge properly.

We have an invariant here: after a declaration reads its canonical declaration,
it can assume the canonical declaration is fully merged. This invariant can be
violated if deserializing some declaration triggers the deserialization of a
later declaration, because that later declaration can in turn deserialize a
redeclaration of that first declaration before it is fully merged.

The anonymous namespace for a namespace gets stored with the first declaration
of that namespace, which may be before its parent namespace, so defer loading
it until after we've finished merging the surrounding namespace.

llvm-svn: 232455
2015-03-17 02:23:11 +00:00
Richard Smith e687bf83c0 [modules] If we find more formerly-canonical declarations of an entity while
building its redecl chains, make sure we pull in the redeclarations of those
canonical declarations.

It's pretty difficult to reach a situation where we can find more canonical
declarations of an entity while building its redecl chains; I think the
provided testcase (4 modules and 7 declarations) cannot be reduced further.

llvm-svn: 232411
2015-03-16 20:54:07 +00:00
Chandler Carruth 0d745bcf66 [modules] Teach the AST reader to handle the case of importing a module
with a subset of the existing target CPU features or mismatched CPU
names.

While we can't check that the CPU name used to build the module will end
up being able to codegen correctly for the translation unit, we actually
check that the imported features are a subset of the existing features.

While here, rewrite the code to use std::set_difference and have it
diagnose all of the differences found.

Test case added which walks the set relationships and ensures we
diagnose all the right cases and accept the others.

No functional change for implicit modules here, just better diagnostics.

llvm-svn: 232248
2015-03-14 04:47:43 +00:00
David Blaikie bdf40a62a7 Test case updates for explicit type parameter to the gep operator
llvm-svn: 232187
2015-03-13 18:21:46 +00:00
Daniel Jasper 0ab544f1a5 When building a module, all headers of submodules can be used.
This extends r232159.

llvm-svn: 232168
2015-03-13 14:29:39 +00:00
Daniel Jasper b6ba46ef30 Make a module "use" also count as use of all its submodules
llvm-svn: 232159
2015-03-13 11:26:16 +00:00