[NTOSKRNL]: Kill a bunch of now-useless DPRINT1s in Mm, significantly cleaning up the log.
[HIVES/MSI]: .NET needs InstallRoot, not InstallDir. Kills millions of DPRINT1s, thus cleaning up the log.
svn path=/trunk/; revision=57221
[NTOSKRNL]: No longer support non-ARM3 sections in NtProtectVirtualMemory, as the only OS calls were already NO-OPS.
[NTOSKRNL]: Always use ARM3 sections unless SEC_PHYSICAL_MEMORY is used, and make the check explicit.
[NTOSKRNL]: No longer support allocating memory on top of non-ARM3 sections.
[NTOSKRNL]: No longer ASSERT when certain features are not yet implemented, instead return an error code.
[NTOSKRNL]: Add another check in NtFreevirtualMemory when invalid memory is being freed, insert of ASSERTing.
[NTOSKRNL]: Implement and use MiIsEntireRangeCommitted when protecting memory to make sure the entire range is committed.
This patch removes multiple hacks, ASSERTs, and evil mixing of ARM3 and non-ARM3 code/memory.
svn path=/trunk/; revision=57215
Mark intrinsics on gcc with __attribute__((artificial)) so that they appear as one unit in the debug info. This way traces don't point inside the implementations, when they cause an eception, but rather to the line where they are used.
svn path=/trunk/; revision=57214
[NTOSKRNL]: Remove an ASSERT(FALSE) that was only there for testing.
[NTOSKRNL]: Support transition pages during prototype PTE faults, which is our first try at soft faults! Should fix ASSERTs that were seen in the previous attempts in ole32, corrupting the registry.
[NTOSKRNL]: It's fine for MiCreatePagingFileMap to fail in MmCreateSection -- don't assert and simply return failure. Should fix the ASSERTs taht were seen in KmTest.
[NTOSKRNL]: Enable richard's ARM3 section code unconditionally for all non-file backed sections. Works4me. Let's see what Testbot says.
Nobody has showed me how to use/where is PatchBot, and google founds 0 relevant results, so this is going into main again. However I'm actually home this week to revert if something goes wrong :)
svn path=/trunk/; revision=57209
Freeldr size is currently limited to 448 KB. On MSVC it was already at 442 KB, before wine's wctype table was used. The new wctype table is itself 37 KB. This lead to freeldr overflowing into memory regions that were used for the filesystem buffer, causing bootfailures. Fix this by giving freeldr it's own using _isctype(), since freeldr casts WCHAR to CHAR anyway.
svn path=/trunk/; revision=57205
Move imagesoft to applications and add it to build.
Convert lang files to UTF-8.
Fix casts and x64 warnings.
Patch by Hermes Belusca.
See bug 7185 for details.
svn path=/trunk/; revision=57201
Add vgafontedit to build.
Convert the language resource files into UTF-8.
Fix x64 warnings.
Patch by Hermes Belusca.
Bug 7152.
svn path=/trunk/; revision=57200
- Fix a memory overwrite bug in some crappy WINE code that leads to nasty memory corruption of the service cache entry. This corrupts the first field of the cache, causing it to always reallocate memory because it thinks the size is 0. This would be even worse if the code didn't overwrite the other 12 bytes of corruption with proper data on the next lines. This also caused the aliases value to be garbage because it pointed at uninitialized data.
svn path=/trunk/; revision=57199
- Implement proper support for partial sends on TCP sockets
- This prevents a deadlock when a very very large packet is queued to be sent
svn path=/trunk/; revision=57186
* Prefer STREQUAL over MATCHES since we're comparing with strings here. Brought to you by Hermès Bélusca.
See issue #7306 for more details.
svn path=/trunk/; revision=57181
- Add some tests for lstrncpyW
- This function is fine, I wrote that when chasing some bug thinking this function was at fault.
svn path=/trunk/; revision=57179
- Fix one of the worst bugs in AFD of all time. Our AFD code operated under the assumption that none of the input parameters would change once we called LockRequest. This assumption is completely false. The only guarantee that made was that the pages never disappeared from us, not they they couldn't be modified. There are frequent cases where the user-mode buffer was modified from underneath us (WSPRecv allocates the struct on stack which makes it invalid during overlapped operations that complete later). When this happened, we would bugcheck when we tried to unlock the buffers since we accessed this in a member of the struct the caller passed us.
- I've fixed this by adding a parameter to LockRequest which specifies whether the buffer should be copied back when it is unlocked.
- This bug has been around for ages and I was never able to figure out why we just freed garbage sometimes. Now that the ws2_32_winetest exposed it reliably, I was finally able to fix it.
svn path=/trunk/; revision=57175
