Marek Safar [Sat, 25 Sep 2010 09:45:14 +0000 (10:45 +0100)]
Implemented more combinations of nullable boolean logical operators
Marek Safar [Sat, 25 Sep 2010 09:38:44 +0000 (10:38 +0100)]
Better error message for syntax error in namespace name
Miguel de Icaza [Sat, 25 Sep 2010 04:04:24 +0000 (00:04 -0400)]
Improve the Uri parsing speed
This affects MonoTouch startup as our dictionaries are a little bit
slow to be created now that we use Dictionary<> hashes for switch
statements.
Some timings:
Using a simple block of if's is twice as slow as the compiler generated
switch statement. But using this tuned code is faster than the
compiler generated code, with a million loops on x86-64:
With "http": .10 vs .51 (first check)
with "https": .16 vs .51 (second check)
with "foo": .22 vs .31 (never found)
with "mailto": .12 vs .51 (last check)
Andreia Gaita [Fri, 24 Sep 2010 17:06:18 +0000 (18:06 +0100)]
Fix UriParser.GetComponents for the PathAndQuery case
PathAndQuery should always return the path starting with /,
even when it's empty
Zoltan Varga [Fri, 24 Sep 2010 15:27:14 +0000 (17:27 +0200)]
Fix AOT support for OP_GC_CARD_TABLE on amd64.
Zoltan Varga [Fri, 24 Sep 2010 14:31:01 +0000 (16:31 +0200)]
Fix the setting of pointer type fields using reflection. Fixes #641590.
Rodrigo Kumpera [Fri, 24 Sep 2010 14:28:46 +0000 (11:28 -0300)]
Implement amd64 support for OP_CARDTABLE.
* amd64-codegen.h (amd64_alu_reg_membase_size): Add support
for RIP based addressing.
* cpu-amd64.md: Add card_table_wbarrier.
* mini-amd64.c (mono_arch_output_basic_block): Emit the
new OP.
* mini-amd64.c (mono_arch_emit_exceptions): Handle another
kind of patch-info - GC_CARD_TABLE_ADDR. This is required
because we can neither have 64bits immediates with amd64
or 2 scratch regiters with current regalloc.
* mini-amd64.h: Define MONO_ARCH_HAVE_CARD_TABLE_WBARRIER.
Atsushi Eno [Fri, 24 Sep 2010 12:26:53 +0000 (21:26 +0900)]
Assign identical operation name for each contract version so that they don't conflict each other.
Atsushi Eno [Fri, 24 Sep 2010 12:26:27 +0000 (21:26 +0900)]
Fix duplex proxy constructor arg names.
Rolf Bjarne Kvinge [Fri, 24 Sep 2010 10:02:00 +0000 (12:02 +0200)]
JsonSerializationWriter: handle char values
Atsushi Eno [Fri, 24 Sep 2010 07:57:38 +0000 (16:57 +0900)]
Filter requests only by paths, excluding host and port.
Atsushi Eno [Fri, 24 Sep 2010 07:26:28 +0000 (16:26 +0900)]
Override IsContentTypeSupported() in per-encoding-binding-element implementation to work as expected.
Zoltan Varga [Fri, 24 Sep 2010 02:27:01 +0000 (04:27 +0200)]
Avoid making the small_id table precisely GC tracked, some runtime code can't handle MonoInternalThread's being moved.
Jackson Harper [Fri, 24 Sep 2010 02:17:43 +0000 (22:17 -0400)]
On 64bit this is turning into a 16 byte struct, this works around the issue.
Zoltan Varga [Fri, 24 Sep 2010 01:21:36 +0000 (03:21 +0200)]
Disable make_root_descr_all_refs () for now since it seems to cause crashes.
Zoltan Varga [Thu, 23 Sep 2010 23:07:12 +0000 (01:07 +0200)]
Use g_malloc0 in mono_gc_make_root-descr_all_refs ().
Zoltan Varga [Thu, 23 Sep 2010 22:38:50 +0000 (00:38 +0200)]
Remove a debug printf.
Zoltan Varga [Thu, 23 Sep 2010 22:36:51 +0000 (00:36 +0200)]
Add a few variants of the MONO_GC_REGISTER_ROOT () to make most roots precisely tracked and to avoid registering locations pointing to memory allocated by mono_gc_alloc_fixed () as roots under SGEN.
Zoltan Varga [Thu, 23 Sep 2010 22:29:24 +0000 (00:29 +0200)]
Disable the wrapped exception code for now, it break some of our tests.
Rodrigo Kumpera [Thu, 23 Sep 2010 21:32:00 +0000 (18:32 -0300)]
Handle verification of types in non canonical forms.
* verify.c (mono_type_is_valid_type_in_context): Handle
types such as VAR and MVAR encoded as CLASS or VALUETYPE.
This encoding is not the canonical one, but works and
must be properly verified.
Fixes #601970
Rodrigo Kumpera [Thu, 23 Sep 2010 21:09:25 +0000 (18:09 -0300)]
Remove unwanted debug spew.
Sebastien Pouliot [Thu, 23 Sep 2010 20:05:47 +0000 (16:05 -0400)]
Fix WebRequest.RegisterPrefix / Create to match SL in some corner cases
* System.Net/WebRequest_2_1.cs: (Create) Find the prefix that is the
closest match to the specified Uri and use it's associated creator.
(RegisterPrefix) We cannot register 'http[s]:' even if registering
'http[s]' is possible (but does not work as expected) as are the (fully)
valid 'http[s]:/[/]' values.
Sebastien Pouliot [Thu, 23 Sep 2010 14:57:12 +0000 (10:57 -0400)]
Add test case to show a different behavior (from Windows) wrt Socket buffers
* Test/System.Net.Sockets/SocketTest.cs: Add test case where buffers can
be set to 0 under Windows (but has a 256 bytes minimum on Linux)
Sebastien Pouliot [Thu, 23 Sep 2010 14:30:03 +0000 (10:30 -0400)]
[moonlight] Fully honor WebResponse.SupportsHeaders
* System.Net.Policy/CrossDomainPolicyManager.cs: Use InternalHeaders,
instead of Headers, since we always use the browser stack to download
the security policies.
* System.Net/WebClient_2_1.cs: Throw a NotImplementedException from
ResponseHeaders if the response does not support them.
* System.Net/WebResponse_2_1.cs: Make Headers property smart to throw a
NotImplementedException if SupportsHeaders return false. Add a new
InternalHeaders property since the response headers are needed,
internally (e.g. policies), even if not exposed to the outside world.
Andrew Jorgensen [Thu, 23 Sep 2010 20:15:01 +0000 (14:15 -0600)]
move DynamicData to mvc because it depends on Extensions
Zoltan Varga [Thu, 23 Sep 2010 19:14:54 +0000 (21:14 +0200)]
Fix the default exception message of RuntimeWrapperException.
Zoltan Varga [Thu, 23 Sep 2010 19:13:36 +0000 (21:13 +0200)]
Wrap thrown objects which do not derive from Exception into a RuntimeWrappedException. Fixes #640687.
Gonzalo Paniagua Javier [Thu, 23 Sep 2010 19:02:05 +0000 (15:02 -0400)]
epoll_ctl takes an int, not IntPtr
Ditto.
Marek Safar [Thu, 23 Sep 2010 15:19:09 +0000 (16:19 +0100)]
[641146] Implemented IStructuralComparable.CompareTo
Marek Habersack [Thu, 23 Sep 2010 12:34:01 +0000 (14:34 +0200)]
[WCF] System.Data.Services.Client must be signed with the ECMA public key
Marek Habersack [Thu, 23 Sep 2010 08:50:35 +0000 (10:50 +0200)]
[WCF] Changes to System.Data.Services.Client to compile it for 4.0
It appears Microsoft didn't make any changes to the assembly in .NET 4.0 - they simply
changed its version number to 4.0.0.0 and added few assembly attributes.
Sebastien Pouliot [Thu, 23 Sep 2010 01:22:18 +0000 (21:22 -0400)]
Fix exceptions wrt IPAddress.ScopeId
* System.Net/IPAddress.cs: ScopeId can only be set on IPv6 or it throws
a SocketException (not an Exception). ScopeId is also limited from 0 to
2^32.
Sebastien Pouliot [Wed, 22 Sep 2010 20:28:45 +0000 (16:28 -0400)]
Add some missing validations to Socket
* System.Net.Sockets/Socket_2_1.cs: Add checks for invalid AddressFamily,
SocketType and ProtocolType (to throw ArgumentException). Add negative
check for Ttl property. Add AddressFamily mismatch checks to ConnectAsync
methods
Sebastien Pouliot [Wed, 22 Sep 2010 20:25:42 +0000 (16:25 -0400)]
Add missing SocketOptionName_HopLimit definition for IPv6
* socket-io.c|h: Add definition/case for SocketOptionName_HopLimit.
Note that this is already being used by managed code (to set the TTL
value on IPv6 sockets)
Sebastien Pouliot [Wed, 22 Sep 2010 13:45:17 +0000 (09:45 -0400)]
Fix support of non-ASCII host names in System.Uri
* System/Uri.cs: Fix escaping non-ascii host names. Avoid memory
allocations (UTF8.GetBytes) when it is not necessary while escaping.
* Test/System/UriTest2.cs: Add test case where host name is β (beta)
Zoltan Varga [Thu, 23 Sep 2010 00:52:25 +0000 (02:52 +0200)]
Don't allow EventType.Breakpoint in vm.EnableEvents ().
Zoltan Varga [Wed, 22 Sep 2010 22:23:53 +0000 (00:23 +0200)]
Avoid extracting tar.gz files in the get-cygwin-deps.sh script, we download only zip files now.
Zoltan Varga [Wed, 22 Sep 2010 22:23:25 +0000 (00:23 +0200)]
Convert ERROR_PATH_NOT_FOUND to ENOENT too.
Hib Eris [Wed, 22 Sep 2010 22:18:08 +0000 (00:18 +0200)]
Don't link against missing libiconv on windows.
* configure.ac: iconv check creates variable ICONV_LIBS
* src/Makefile.am: only link eglib to iconv when iconv is found
Zoltan Varga [Wed, 22 Sep 2010 21:55:25 +0000 (23:55 +0200)]
Set errno correctly when loading an image fails on windows. Fixes #362607.
Gonzalo Paniagua Javier [Wed, 22 Sep 2010 22:14:34 +0000 (18:14 -0400)]
EpollEvent size adjustment
EpollEvent has to be one uint32 plus one int64, i.e. 12 bytes long.
Manos libres now.
Jonathan Pryor [Wed, 22 Sep 2010 20:15:19 +0000 (16:15 -0400)]
Do not build mcs/docs if the 4.0 build is disabled.
Jonathan Pryor [Wed, 22 Sep 2010 20:06:16 +0000 (16:06 -0400)]
Revert "mdoc 2.0-only profile build support, take 3!"
We'll instead just disable the docs build entirely if --with-profile4=no
is specified [future commit, not this one].
Jonathan Pryor [Wed, 22 Sep 2010 17:46:40 +0000 (13:46 -0400)]
mdoc 2.0-only profile build support, take 3!
Further discussion suggested that, even if mdoc 2.0 works, it hasn't gotten
widespread testing (as it's been a 4.0 app for months), so it would be
preferable to keep it as a 4.0 app.
However, it still needs to be built under 2.0 so that a --with-profile4=no
build can...build...without errors.
Thus, the compromise: build mdoc twice, once for each profile. This will
ensure that *some* version of mdoc is around when mcs/docs is built.
Unfortunately we can't place mdoc into $(per_profile_dirs), as these are
built before $(net_2_0_dirs) (and thus before monodoc, which needs to be
built before mdoc). Thus the compromise of listing mdoc in *both*
$(net_2_0_dirs) and $(net_4_0_dirs), allowing build dependencies to be met.
Gonzalo Paniagua Javier [Wed, 22 Sep 2010 16:39:07 +0000 (12:39 -0400)]
LocalEndPoint works now for non-blocking connects
We didn't keep the endpoint around for non-blocking connects and later,
when Poll() was called, we didn't have the information. Also 'isbound'
wasn't being set in a few places and UDP sockets needed to be marked as
not connected after a call to Connect() with an Any address.
Removed an old test for Connect(Any) with UDP sockets, since this is
used to disconnect them now.
Fixes bug #640749.
Gonzalo Paniagua Javier [Wed, 22 Sep 2010 15:42:35 +0000 (11:42 -0400)]
The Hashtable is back
But this time it is using a hash code provider and a comparer that
don't invoke the object's methods.
Thanks to Paolo for pointing this out.
Zoltan Varga [Wed, 22 Sep 2010 15:01:36 +0000 (17:01 +0200)]
Call the thread cleanup at the end as it was done before, call it even if handle_remove () fails.
Jérémie Laval [Wed, 22 Sep 2010 10:33:55 +0000 (11:33 +0100)]
Add tidbits of documentation to ReaderWriterLockSlim.
Jérémie Laval [Wed, 22 Sep 2010 10:01:10 +0000 (11:01 +0100)]
Also set readerDoneEvent when exiting an upgradeable lock and reader condition is correct.
It could cause a deadlock when a writer tested the rwlock value before RwRead was removed and another Upgradeable operation happened at the same time.
Marek Habersack [Wed, 22 Sep 2010 09:14:26 +0000 (11:14 +0200)]
Revert "mdoc 2.0-only profile build support, take 2!"
This reverts commit
313633fc67cf1479514019a56143e5ad1643d28f because it breaks the build.
Jonathan Pryor [Wed, 22 Sep 2010 01:38:09 +0000 (21:38 -0400)]
mdoc 2.0-only profile build support, take 2!
Further discussion suggested that, even if mdoc 2.0 works, it hasn't gotten
widespread testing (as it's been a 4.0 app for months), so it would be
preferable to keep it as a 4.0 app.
However, it still needs to be built under 2.0 so that a --with-profile4=no
build can...build...without errors.
Thus, the compromise: build mdoc twice, once for each profile. This will
ensure that *some* version of mdoc is around when mcs/docs is built.
Sebastien Pouliot [Wed, 22 Sep 2010 01:18:45 +0000 (21:18 -0400)]
NET_4_0 fixes for System.Uri also apply to Moonlight (SL4)
* System/Uri.cs: Update some NET_4_0 to include MOONLIGHT too
Sebastien Pouliot [Tue, 21 Sep 2010 19:51:00 +0000 (15:51 -0400)]
Do not allow + sign when specifying the port of an URI
* System/Uri.cs: A bit more strict when parsing the port number
* Test/System/UriTest2.cs: Add test cases with different, mostly
invalid, ports (e.g. negative, float, too big...)
Sebastien Pouliot [Tue, 21 Sep 2010 18:50:29 +0000 (14:50 -0400)]
Fix some host name checks, delay port check and fix some exceptions messages
* System/Uri.cs: Fix host name checks. Delay port check until the host
name is checked (compatibility). Fix exception messages to match the
expected values from the NCL tests.
* Test/System/UriTest2.cs: Add some test cases for host names
Gonzalo Paniagua Javier [Tue, 21 Sep 2010 21:26:28 +0000 (17:26 -0400)]
Don't store objects in a Hashtable.
Storing the objects in Hashtable causes the GetHashCode() and may be
Equals() methods to be called on the object, which MS does not do.
I'm using a list holding WeakReferences. Not sure if hard references
would be ok.
Fixes bug #640688.
Andrés G. Aragoneses [Tue, 21 Sep 2010 15:20:34 +0000 (23:20 +0800)]
[corlib] Prevent BinaryFormatter to throw on missing fields
This commit includes unit tests that were run in MS.NET > 2.0.
The old behaviour of MS.NET 1.1 is still included in the patch
as an ONLY_1_1 define, to be backport-friendly to other branches
(but can be removed from now on in next commits, as mono master
doesn't include this profile anymore). Fixes BNC#324144.
Rodrigo Kumpera [Tue, 21 Sep 2010 13:58:35 +0000 (10:58 -0300)]
Enable cardtable for parallel mark-sweep.
* sgen-marksweep.c (mono_sgen_marksweep_init): Enable
cardtable for the parallel collector.
This closes by 50% the gap in binary-tree against the
serial collector.
Rodrigo Kumpera [Tue, 21 Sep 2010 13:57:30 +0000 (10:57 -0300)]
Add support for break on unverifiable for verifier failure.
* mini.c (mini_method_compile): Break on verification
failure.
Marek Safar [Tue, 21 Sep 2010 09:49:31 +0000 (10:49 +0100)]
New tests.
Marek Safar [Tue, 21 Sep 2010 09:42:28 +0000 (10:42 +0100)]
Don't restore unused flags
Marek Safar [Tue, 21 Sep 2010 09:33:31 +0000 (10:33 +0100)]
[640722] Consider delegate parameters when checking member hiding
Marek Safar [Tue, 21 Sep 2010 08:09:50 +0000 (09:09 +0100)]
[640709] Don't reuse method parameters for base proxy, they can be hoisted
Marek Safar [Tue, 21 Sep 2010 07:42:48 +0000 (08:42 +0100)]
[640699] Reduced compound expressions need explicit conversion too.
Marek Safar [Mon, 20 Sep 2010 19:12:46 +0000 (20:12 +0100)]
Implement postfix operations involving an implicit user operator.
Atsushi Eno [Tue, 21 Sep 2010 07:46:07 +0000 (16:46 +0900)]
Fix (isolate) internal namespace for Udp transport.
Atsushi Eno [Tue, 21 Sep 2010 06:23:35 +0000 (15:23 +0900)]
Add missing Properties property.
Carlos Alberto Cortez [Tue, 21 Sep 2010 01:00:04 +0000 (03:00 +0200)]
Implement IDesignerLoaderHost2 interface.
* IDesignerLoaderHost2.cs: New 4.0 interface.
Carlos Alberto Cortez [Tue, 21 Sep 2010 00:58:53 +0000 (02:58 +0200)]
Add IDesignerLoaderHost2.cs to our build.
* System.dll.sources: Add IDesignerLoaderHost2.cs to our build.
Carlos Alberto Cortez [Mon, 20 Sep 2010 19:34:35 +0000 (21:34 +0200)]
Add IDesignerHostTransactionState.cs and TypeDescriptionProviderService.cs to our build.
* System.dll.sources: Add IDesignerHostTransactionState.cs and
TypeDescriptionProviderService.cs to our build.
Carlos Alberto Cortez [Mon, 20 Sep 2010 19:32:20 +0000 (21:32 +0200)]
Implement IDesignerHostTransactionState.cs
* IDesignerHostTransactionState.cs: New 4.0 interface.
Carlos Alberto Cortez [Mon, 20 Sep 2010 19:20:15 +0000 (21:20 +0200)]
Implement TypeDescriptionProviderService.
* TypeDescriptionProviderService.cs: New class.
Carlos Alberto Cortez [Mon, 20 Sep 2010 11:42:29 +0000 (13:42 +0200)]
Implement AddProviderTransparent and RemoveProviderTransparent for TypeDescriptor.
* TypeDescriptor.cs: Implement AddProviderTransparent and
RemoveProviderTransparent - we only call to the already existing
AddProvider and RemoveProvider methods respectively.
Carsten Schlote [Sun, 22 Aug 2010 00:31:54 +0000 (02:31 +0200)]
[System.Configuration] Added check for SettingProviderAttribute on settings class
There is no checkfor the SettingProviderAttribute attached to a Settings class.
.NET uses this to replace the default LocalSettingsProvider with your
own SettingsProvider.
The patch adds this feature.
Signed-off-by: Carsten Schlote <schlote@vahanus.net>
Carsten Schlote [Sun, 22 Aug 2010 13:54:44 +0000 (15:54 +0200)]
[System.Configuration] Fixed GetProductName() function
Added missing check for empty strongname. This case is now handled the
same way as a missing strongname.
GetProductName() now correctly uses _Url_ and assembly.EscapedCodeBase
for its output.
Signed-off-by: Carsten Schlote <schlote@vahanus.net>
Carsten Schlote [Sun, 22 Aug 2010 13:50:52 +0000 (15:50 +0200)]
[System.Configuration] Fixed GetEvidenceHash() function to produce valid filenames
The previous implementation creates eventually invalid filenames in a
random fashion on linux, mac, windows by having '/' or '\' chars
in the base64 output. As a result you can't save or retrieve your settings
due to filesystem errors.
This patch changes the output to plain hex values. This is a bit longer
then a base64 string, but also 100% filesystem safe.
Might
Signed-off-by: Carsten Schlote <schlote@vahanus.net>
acken [Mon, 20 Sep 2010 21:08:39 +0000 (05:08 +0800)]
Had an issue where the linklabel was clicked, the form was redrawn with some changes and some internal behavior called the OnGotFocus where sorted_links now was null and focused_index was 0. Don't know wheter that is wrong but solved the issue with a nullcheck
Andrew Jorgensen [Mon, 20 Sep 2010 23:23:51 +0000 (17:23 -0600)]
put the x11 devel package back in buildrequires and a couple other spec fixes
Rodrigo Kumpera [Mon, 20 Sep 2010 22:26:12 +0000 (19:26 -0300)]
Use new verifier function to lazily check.
* class.c: The typeref table is now lazily verified thanks
to some awesome obfuscators that love to produce junk.
So it has to be checked everytime before usage instead of
at loading time.
Rodrigo Kumpera [Mon, 20 Sep 2010 22:22:54 +0000 (19:22 -0300)]
Add function to validate an individual row in the typeref table.
* metadata-verify.c: Add mono_verifier_verify_typeref_row
that checks the metadata of a single row. Fix code around
to couple with this new function.
* verify-internals.h: Export it.
Sebastien Pouliot [Mon, 20 Sep 2010 20:52:19 +0000 (16:52 -0400)]
Uri.LocalPath is unescaped - even if original URI is escaped
* System/Uri.cs: Unescape LocalPath
* Test/System/UriTest2.cs: Add test case where an original URI contains
an escaped path.
Sebastien Pouliot [Mon, 20 Sep 2010 20:23:31 +0000 (16:23 -0400)]
Uri.OriginalString is a bit misleading since it's computed for merged URI
* System/Uri.cs: Compute the expected OriginalString when an absolute
URI is merged with a (String or Uri) relative Uri to create a new Uri
instance. Keep original fragment around for ToString
* Test/System/UriTest2.cs: Add test cases for Fragment and Query when
special characters are used. Most of them are not escaped in ToString
* Test/System/UriTest3.cs: Uncomment an Assert that is now fixed.
Sebastien Pouliot [Mon, 20 Sep 2010 18:35:43 +0000 (14:35 -0400)]
Actual exception message vary wrt scheme
* System/Uri.cs: Exception message does not specify Authority when a
file:// URI is parsed
Sebastien Pouliot [Mon, 20 Sep 2010 18:08:12 +0000 (14:08 -0400)]
Fix different between an empty Uri.UserInfo and the absence of username/password
* System/Uri.cs: An empty username/password is treated differently than
having none, i.e. the '@' character is kept present in the URI
* Test/System/UriTest2.cs: Add test case with an empty UserInfo
Sebastien Pouliot [Mon, 20 Sep 2010 17:36:59 +0000 (13:36 -0400)]
Only return true on Windows for Uri.IsUnc on file:// URI
* System/Uri.cs: IsUnc returns true for file:/// URI on Windows only.
This makes us closer to Mac-behavior (see NCL tests) since the other
properties, e.g. LocalPath, match this. Also fix some exception text
where the values are checked by some test cases.
* Test/System/UriTest.cs: Adjust some test cases wrt Win32
* Test/System/UriTest2.cs: Adjust a test case wrt Win32
Sebastien Pouliot [Mon, 20 Sep 2010 15:27:36 +0000 (11:27 -0400)]
Remove NET_2_0 and ONLY_1_1 from UriTest.cs (its already confusing enough ;-)
* Test/System/UriTest.cs: Remove NET_2_0 and ONLY_1_1
Sebastien Pouliot [Mon, 20 Sep 2010 14:43:19 +0000 (10:43 -0400)]
Fix merging relative URI that looks like complete URI (e.g. scheme)
* System/Uri.cs: Fix merging URI where a relative URI contains a ':'
but no valid scheme
* Test/System/UriTest2.cs: Add test cases where some bad (if absolute)
URI are valid when merged (as relative) with an absolute URI
Rodrigo Kumpera [Mon, 20 Sep 2010 17:53:39 +0000 (14:53 -0300)]
Report the broken type name instead of assembly's.
Rodrigo Kumpera [Mon, 20 Sep 2010 17:44:20 +0000 (14:44 -0300)]
Use List<Exception> instead of ArrayList to make things prettier.
Rodrigo Kumpera [Mon, 20 Sep 2010 17:29:50 +0000 (14:29 -0300)]
Fix AssemblyBuilder::GetTypes () with unfinished types.
* AssemblyBuilder.cs (GetTypes): Raise ReflectionTypeLoadException
if any type was not finished.
* AssemblyBuilderTest.cs: Add test for GetTypes () and incomplete
TypeBuilders'.
Fixes #640288
Paolo Molaro [Mon, 20 Sep 2010 16:58:42 +0000 (18:58 +0200)]
First pass at profiler hooks in sgen.
* sgen-gc.c, sgen-los.c: added hooks for the important GC events.
Update some of the GC-related performance counters, too.
Geoff Norton [Mon, 20 Sep 2010 16:31:42 +0000 (12:31 -0400)]
Include TargetFrameworkAttribute in monotouch and monodroid as well
Marek Safar [Mon, 20 Sep 2010 15:08:46 +0000 (16:08 +0100)]
Don't check inferred dynamic type arguments against best candidate constraints.
Marek Safar [Mon, 20 Sep 2010 11:22:57 +0000 (12:22 +0100)]
More missing conversions between object and dynamic
Jb Evain [Mon, 20 Sep 2010 15:12:04 +0000 (17:12 +0200)]
Ignor test since its fix has been reverted
Jb Evain [Mon, 20 Sep 2010 15:04:58 +0000 (17:04 +0200)]
Revert "[Fix] Fix the default binder when selecting params methods."
This reverts commit
6a7443d8089c92f722d87c81f6c7d22c2c0deafb.
Alan McGovern [Mon, 20 Sep 2010 13:03:42 +0000 (14:03 +0100)]
[monodis] Ensure we don't try to free uninitialized memory.
Marek Safar [Mon, 20 Sep 2010 08:34:07 +0000 (09:34 +0100)]
[640285] Initialize this in current block instead of original block
Marek Safar [Mon, 20 Sep 2010 08:04:25 +0000 (09:04 +0100)]
Parse secondary static like expression as a non-cast
Rodrigo Kumpera [Thu, 16 Sep 2010 20:44:13 +0000 (17:44 -0300)]
Avoid a superfluous copy.
* sgen-marksweep.c (major_scan_card_table): Avoid an extra
copy under overlapping card.
Rodrigo Kumpera [Thu, 16 Sep 2010 20:33:35 +0000 (17:33 -0300)]
Cleanup definitions and clearly states what are the tunables.
* sgen-cardtable.h: Auto detect the need for
SGEN_HAVE_OVERLAPPING_CARDS.
* sgen-cardtable.h: Add CARD_TABLE_BITS and make CARD_COUNT_BITS
depend on it.
* sgen-cardtable.h: Document which and how each tunable affect
runtime of sgen.