- Fix parameter types of NtQuery/SetSystemInformation, ExInitialize(N)PagedLookasideList, KeBugCheckWithTf, IoGetstackLimits
- Use ALIGN_UP_POINTER for a pointer
- Fix pointer to ULONG cast
svn path=/trunk/; revision=35715
- Enable WFI during idle loop.
- Implement undefined exception handler, and get rid of reserved exception code -- this will never happen except for an SoC bug.
- Request an APC interrupt if the new thread we context switched to has APCs pending and enabled.
- Perform DPC watchdog debugging code if a DPC lasted more than a second.
- Do the appropriate bugcheck if DPCs are active during a context switch.
- Go over the FIXMEs in the assembly files too, and create a new FIXME-PERF category.
svn path=/trunk/; revision=34584
- Separate kernel FIXMEs into clearer groupings:
- TODO FIXMEs are highest priority, as they are missing required functionality.
- USER FIXMEs are the next highest priority, since they will be required for user mode support.
- V6 FIXMEs and future groupings are lowest priority, since they relate to hardware support we don't care about for now.
svn path=/trunk/; revision=34583
We now define _disable and _enable for ARM.
We shouldn't define KeRaiseIrqlToSynchLevel for each architecture, since the prototype is portable itself.
It was a mistake to guard against x86 only system calls -- the system calls should be the same on all archs, just return STATUS_NOT_IMPLEMENTED if they don't make sense. Undo the guards.
We now define KeGetPcr() as portable -- it's PCR itself that is a per-arch define.
We now support ARM in RtlWalkFrameChain.
We now support ARM in PspCreateThread.
We now define KeArchHaltProcessor for ARM by using Wait-For-Interrupt Mode.
We now define KeArmInitThreadWithContext for ARM.
KiRestore/SaveProcessorControlState are portable prototypes, we now define them as such.
Bochs KD code should use the portable WRITE/READ_PORT_UCHAR defines, we now do so.
We now support ARM in SharedUserData->ImageNumberLow/High during ExpInitializeExecutive.
NtQuerySytemInformation for SystemProcessorInformation has now been fixed to use the portable KeProcesssorXxx variables instead of reading from the non-portable PRCB values.
We now support NtFlushInstructionCache for ARM by flushing the I-Cache.
svn path=/trunk/; revision=32197
Added x86-only guards around certain structures and functions which are non-portable. Also guarded certain inline functions.
Added some more ARM structures, and filled out the KPRCB.
Created ARM-specific headers when required.
We can now have a buildable ARM kernel, minus a couple of compile errors.
svn path=/trunk/; revision=32178
For example, when you handle a page fault in a section, then page fault while handling that page fault (which is perfectly okay),
you shouldn't be trying to re-acquire the address space lock that you're already holding. After this fix, this scenario works
and countless others. Apps like QTInfo now work and load, and PictureViewer doesn't BSOD the system anymore. I've fixed this by changing
the lock to a pushlock. It not only increases speed inside the memory manager significantly (such as during page fault handling), but
does allow recursive acquisition without any problems.
- Now if that wasn't bad enough, here's a couple more tips. Fast Mutexes actually require APC_LEVEL to be effective. If you're going
to be using a Fast Mutex and calling it with the "Unsafe" version, then don't expect anything to work. Also, using functions like
"CcTryToAcquireBrokenMutex" where correct code is duplicated then hacked to work isn't a big help either. And that's not all. Fast Mutex
disables kernel APCs by setting the KernelApcDisable flag on, and it's expected that the count inside the fast mutex will match the count
inside the thread. In other words, LOCK ACQUISITION AND RELEASE MUST BE ORDERED. You can't acquire LOCK A and B, and then release lock A
and B, because that leads to deadlocks and other issues. So of course, the Cache Manager acquired a view lock, then acquired a segment lock,
then released the view lock, then released the segment lock, then re-acquired the view lock. Uh, no, that won't work. You know what else
doesn't work so well? Disabling APCs about 6-9 times to acquire a single lock, and using spinlocks in the same code path as well. Just how
paranoid are you about thread safety, but still manage to get it wrong? Okay, so we've got recursion, out-of-order lock acquision and
release, made-up "broken" acquire functions, and using a lock that depends on APC_LEVEL at PASSIVE_LEVEL. The best part is when Cc builds
an array of cache segments, and locks each of them... then during release, the list gets parsed head-first, so the first acquired locks
get released first. So locks a, b, c, d get acquired, then a, b, c, d get released. Great! Sounds about right for ReactOS's Cache Manager
design. I've changed the view lock to a guarded mutex -- which actually properly disables APCs and works at PASSIVE_LEVEL, and changed the
segment locks to be push locks. First it'll be 10 times faster then acquiring a bazillion fast mutexes, especially since APCs have already
been disabled at this point, and it also allows you to do most of the stupid things the Cache Manager does. Out-of-order release is still
not going to work well, so eventually on a multi-processor machine the code will completely die -- but at least it'll work on UP for now.
In the end, this makes things like the Inkscape installer and Quicktime Installer to work, and probably countless other things that generated
ASSERTS in the fast mutex code.
-- Alex Ionescu
svn path=/trunk/; revision=30401
- Fix typo in MMWSL.
- Add RtlRandom to NDK.
- Add MEMORY_PRIORITY values to NDK.
- Add KeAcquireSpinLockRaiseToSynch to NDK.
- Make MmInitializeProcessAddressSpace take two more parameters: one to specify flags, such as large page support, and another one to define the process being cloned, when fork() support will be added.
- Add KeInvalidAccessAllowed to deal with page faults in the special S-List code. The assembly code currently handles simple faults, but our MmAccessFault handler needs to start verifying the fault too.
- Mark LoaderReserve pages as LoaderFree, it seems they end up this way in Windows.
- Use MmNumberOfPhysicalPages instead of MmStats.NrTotalPages.
All NDK changes are discussed with Alex.
svn path=/trunk/; revision=29254
- Fix MM_EXTEND_INFO definition.
- Fix MMWSLE definition.
- Fix EPROCESS definition.
- Add quota functions to NDK.
- Add one more parameter to PspMapSystemDll to support mapping large pages.
- Don't make the quota functions do anything for the system process.
- Add page file quota functions.
- Other misc small fixes.
svn path=/trunk/; revision=29214
- Add KeAcquireQueuedSpinLock / KeAcquireReleaseQueuedSpinLock, RtlCompareMemoryUlong definitions missing from IFS.
- Add actual section object definition to NDK
- More NDK fixes
- Fix incorrect definition of MMPTE_LIST
- Fix incorrect prototype of NtFlushVirtualMemory
- Fix incorrect prototypes of KeAcquire / ReleaseQueuedSpinLock
- Add a second parameter to MmCreateKernelStack to select the NUMA node on which the stack should be allocated
- Avoid usage of PROS_SECTION_OBJECT in Ps where possible
- Unimplement PsChargeProcessQuota. It was not working right
- Use PFN_NUMBER defined in DDK instead of PFN_TYPE internal ROS Type.
svn path=/trunk/; revision=29188
Write the few lines of C code that were part of the "final" code into boot.S as
well, to avoid yet another trampoline jump. and
Thanks again to Thomas for finding and fixing the bug.
svn path=/trunk/; revision=28205
- Implement KeDisableInterrupts to disable interrupts and return whether or not they were enabled.
- Implement KiCheckTimerTable, in DBG mode, to validate the timer tables.
- Implement DPC Timeout detection, in DBG mode.
- Fix a bug in KiQuantumEnd which would've affected real-time threads.
- Fix some bugs in KiRetireDpcList to avoid issues should the DPC Queue Depth drop below 0, and solve some possible races.
- Fix KeRemoveQueueDpc only to enable interrupts if it was called with interrupts enabled.
svn path=/trunk/; revision=26139
- DebugPrint/DebugPrompt should have an int3 after the int2d, and the int2d handler should ++ the trap frame's EIP to compensate (merge from kd-branch).
- Remove KDB symbol hooks (merge from kd-branch).
- Make PSEH compialble in MSVC again after Greatlord's break.
- Fix KiSaveProcessorControlState/KiRestoreProcessorControlState (merge from kd-branch).
- Disable GDB hook/hacks (merge from kd-branch).
- Add KD64 directory from kd-branch with SharedUserData access enabled (no other code changed). It's not currently compiled though, just putting it here.
svn path=/trunk/; revision=25965
- Rename one more duplicated file in the kernel which wasn't caught previously.
- pseh, rossym, cmlib, kdcom, bootvid, rtl compile & link, and ntoskrnl is now down to only 23 linker errors.
svn path=/trunk/; revision=25956
- Fixes and compatible merges from KD Branch:
- Add stubs for KdSave, KdRestore, KdDebuggerInitialize0, KdSendPacket, KdReceivePacket to kdcom.dll
- Implement and export KeTryToAcquireSpinLockAtDpcLevel.
- Add EXCEPTION_RECORD64 and LIST_ENTRY64, KeTryToAcquireSpinLockAtDpcLevel, BREAKPOINT_COMMAND_STRING, Ke386SetCr2, Ke386SetDr3, Ke386SetDr6.
- Remove non-kernel routines from kdfuncs.h and remove deprecated routines from ke.h.
- Implement KiRestoreProcessorControlState, KeFreezeExecution, KeThawExecution, ExAcquireTimeRefreshLock, ExReleaseTimeRefreshLock.
- Rename ModuleLoadList to PsLoadedModuleList. Add PsNtosImageBase and set value in it.
- Add skeleton wdbgexts.h with what's needed until now, this is a PSDK header.
- Add kddll.h for KDCOM/1394/USB2.DLL prototypes.
- Add windbgkd.h with KD protocol definitions. Used to be an NT5 DDK header, but was removed, so this goes into include\reactos.
- Fix KiDebugService to load EDX from KTRAP_FRAME_EDX, not KTRAP_FRAME_EAX!.
- Fix CommonDispatchException to check for the argument count in ECX, not EAX. Previously we were ignoring parameter counts and never filling out exception records!
- Add KdDebuggerInitialize1 and enable call to it.
- Fix KD_SYMBOLS_INFO definition and DbgLoadImageSymbols prototype.
- Implement DbgUnLoadImageSymbols.
- Fix some small bugs in KeBugCheckWithTf and add various debugger calls/checks where needed.
- Fix bugcheck recursion code which was incorrect.
- Only save/restore CR4 if KeFeatureBits indicates CR4 support exists.
- Export KdDebuggerNotPresent since KDCOM needs it.
- Add KCONTINUE_STATUS.
- Add DBGKD_ANY_CONTROL_SET and X86/IA64/AMD64 control sets.
- Add DBGKD_MANIPULATE_STATE64 and all sub-structures (READ_MEMORY, WRITE_MEMORY, etc).
- Create GCC_ULONG64 type to hack around a bug in GCC which is incapable of creating entries for externals at compile-time for 64-bit pointers.
- Rename NameSpaceRoot to ObpRootDirectoryObject, IopLogListHead to IopErrorLogListHead, BugcheckCallbackListHead to KeBugcheckCallbackListHead, BugcheckReasonCallbackListHead to KeBugcheckReasonCallbackListHead, ObTypeObjectType to ObpTypeObjectType.
- Create ntverp.h and common.ver files. These are the standard files used by the NT/DDK build systems and we should try to support them as well instead of re-defining everything our own way (especially if we want to build ddk-compatible drivers later on).
- Made init.c use version data from ntverp.h instead of hard-coding.
- Defined NT 5.2.3790.1830 as the version we report.
- Fixed up .rc file to be correct and match DDK-sytnax/style.
- For now only the kernel uses this new versionning scheme, but we should change the build system later to use this for every component.
- Fix KiSaveProcessorControlState and KiRestoreProcessorControlSate. The latter doesn't freeze the CPU anymore so it's enabled, and the former doesn't cause WinDBG to panic anymore and display weird data.
- KPROCESSOR_STATE is not 4-byte aligned.
- Use DR_MASK and DR7_OVERRIDE_V in KiUpdateDr7, KiRecordDr7 instead of DR_ACTIVE_MASK.
- Add ExceptionRecord32To64.
- Fix generation of driver name for symbol load.
svn path=/trunk/; revision=25937
- Fix bugs in KdInitSystem; some 64-bit pointers must be sign-extneded, not zero-extended (thanks Myria!)
- Properly read kernel base instead of kernel stack.
- Fix compile issue due to incorrect KiBugCheckData definition.
- WinDBG reports: "Connected to Windows Vista 16199 x86 compatible target, ptr64 FALSE. Kernel Debugger connection established. (Initial Breakpoint requested)"
svn path=/branches/alex-kd-branch/; revision=25846
- Rename NameSpaceRoot to ObpRootDirectoryObject, IopLogListHead to IopErrorLogListHead, BugcheckCallbackListHead to KeBugcheckCallbackListHead, BugcheckReasonCallbackListHead to KeBugcheckReasonCallbackListHead, ObTypeObjectType to ObpTypeObjectType.
- Fill out KdDebuggerDataBlock with the variables that ROS currently supports. Most Mm variables we don't have yet -- it's unknown how much this will hurt WinDBG compatibility/functionality.
- Add KdPrint circular buffer and buffer location/data variables.
svn path=/branches/alex-kd-branch/; revision=25845
- Remove non-kernel routines from kdfuncs.h and remove deprecated routines from ke.h.
- Implement KiRestoreProcessorControlState, KeFreezeExecution, KeThawExecution, ExAcquireTimeRefreshLock, ExReleaseTimeRefreshLock.
- Rename ModuleLoadList to PsLoadedModuleList. Add PsNtosImageBase and set value in it.
- Add skeleton wdbgexts.h with what's needed until now, this is a PSDK header.
- Add kddll.h for KDCOM/1394/USB2.DLL prototypes.
- Add windbgkd.h with KD protocol definitions. Used to be an NT5 DDK header, but was removed, so this goes into include\reactos.
svn path=/branches/alex-kd-branch/; revision=25833
- Implement KeSetDisableBoostThread.
- Temporarily disable SEH/validation in NtQuery/SetProcess/ThreadInformation routines since it was making all calls fail.
- Fix bugs when using ThreadPriority, ThreadBasePriority, ThreadAffinityMask. Parameters are now validated properly, and affinity masked with the process.
- Implement ThreadSetIdealProcessor and ThreadPriorityBoost case.
- Implement ThreadZeroTlsCell case. TLS should now work properly.
- Make kernel32_winetest for threads usable.
- Update Kernel Fun.
svn path=/trunk/; revision=25631
- Replace mm failure detection code by dprint + while loop to avoid bugchecks which might make the error worse to see.
- Remove some deprecated code/functions and cleanup clock.c entirely.
svn path=/trunk/; revision=25626
- Create Phase 1 initialization for the SRM (SeInitSystem). Right now it inserts the system boot token into object manager, which is something we forgot to do before.
- Renamed ExPhase2Init to Phase1Initialization, since it's not Phase 2.
- Updated Phase 1 PS Initialization to get the KeLoaderBlock pointer and use it as a context parameter when calling Phase1Initialization.
- Split off Phase1Initialization into Phase1InitializationDiscard, which is the bulk of the phase 1 code (99% of it) and can be put in an .INIT section to be freed after boot.
- Modify parts of the Inbv setup code. Also implement support for /SOS, and try to mimic its behaviour on NT (not fully achieved). You will need /SOS to see boot messages on the screen! FreeLDR now adds this by default to the "Debug "configuration.
- Temporarily disable ReactOS Banner during boot. We will get this data from the .mc/.res file in a later patch instead of hard-coding it.
- Optimize calling and usage of ExpLoadInitialProcess.
- Add support for Y2K bug fix documented for Windows NT (/YEAR).
- Add support to detect WinPE/MiniNT/ReactOS Live CD.
- Add temporary debugging code to MmInit2 and some Mm functions to detect if these functions are being used too early, which could result in catastrophic to subtle bugs.
- Add more bugchecks when failures occur, and enhance others. Also add more codes to ntoskrnl.mc.
- Disable calls to ObfDereferenceDeviceMap since it's not yet implemented.
svn path=/trunk/; revision=25624
- Implement a new timer expiration function which takes advantage of the fact that timers are now hashed with the system tickcount, and thus only parses the "hot" lists (dpc.c).
- Detect timer expiration during KeUpdateSystemTime and edit the KPRCB to notify of timer expiration. This will be picked up by the timer expiration DPC (clock.s).
- Add support for new timer code in dispatcher (wait.c, queue.c, ke_x.h).
- Change system startup to support the new timer DPC/list (krnlinit.c, clock.c).
- Remove corresponding entry from kernel fun, and update the guidance plan with recent successes and setbacks. This patch is likely the last big architectural change in the kernel except for enabling the new scheduler at a later time.
svn path=/trunk/; revision=25611
- Implement KeReadStateThread.
- Fix PspTerminateProcess to handle case where there's no threads in the process.
- Fix check in PspTerminateProcess. ObClearProcessHandleTable gets called if there's a debug port, not if there's an object table.
- Simplfy PspReapRoutine.
- Fix PspExitThread to wait for all other threads before continuing to kill the last thread. Should fix lots of race/wait conditions.
- PspExitThread should check for !DeadThread and not !Terminated before determining if it should free the TEB. Also, the DbgK handle should only be closed if the thread isn't already dead.
- Fixup formatting of some code to warn less on MSVC.
- Fail various APIs if acquiring rundown protection failed.
- Fix Process Quantum/Priority settings.
- Grant PROCESS_TERMINATE by default.
- Add PROCESS_SET_INFORMATION, STANDARD_RIGHTS_ALL and PROCESS_SET_QUOTA to the default process granted access mask.
- Initialize process/thread/image notification callbacks durin phase 0 Ps initialization.
- The Audit Name belongs to the system process, not the idle process.
- Detect more failures in phase 0 startup.
- Fix various race conditions/incorrect checks in ps/security.c related to impersonation information. Also allow PspAssignPrimaryToken to be called directly with the token pointer and not only the handle.
- Wrap system thread startup stub in SEH and write a SEH filter function to print out debug information when a system thread dies unexpectedly.
svn path=/trunk/; revision=25504
- Add NtInitialUserProcessBuffer settings and validate/user them when launching smss.
- Use SePublicDefaultUnrestrictedSd for the root system link.
- Make enough space for the smss process parameters, there's 6 strings, not 4.
- Fix incorrect length of NtSystemRoot string.
- Disable SYSENTER on CPUs that don't support it.
- Detect and enable NX feature bit.
- Detect and enable DTS support.
- Implement Ki386EnableXMMIExceptions (the XMMI handler still needs to be done, however).
- Implement KiI386PentiumLockErrataFixup to work around the F00F bug.
- Implement KiIsNpxErrataPresent to detect the FDIV bug.
- Implement CPU Speed detection.
- Only setup Initial FX State if the CPU supports FXSR.
- Setup some more PCR fields during init.
- Use NX_SUPPORT_POLICY flags documented in ntddk.h
- Don't boot if CMPXCHG8B is not supported by the CPU. This is the same requirement as Windows.
- Don't overwrite all the PRCB locks in KiInitSpinlocks due to a stupid bug in the timer spinlock initialization.
svn path=/trunk/; revision=25503
- Guard some more code with #ifdef NEW_SCHEDULER.
- Fix bugs in KiDispatchInterrupt.
- Use PRCB fields instead of PriorityListHead/PriorityListMask to reduces number of code changes between NEW_SCHEDULER and old.
- Fully implement KiDeferredReadyThread, but only for Uni-Processor systems. Supports unwait boosts as well as lock boosts. Not yet used.
- Implement NEW_SCHEDULER version of KiSwapThread and NtYieldExecution.
svn path=/trunk/; revision=25502
- Disable KiSetAffinityThread to do nothing else but change affinity instead of doing re-scheduling (temporary change).
- Cleanup KiQuantumEnd to prepare for new scheduler.
- Fix up a large bug in KeInsertQueueDpc which was making every DPC become a Threaded DPC due to a bad assignment-instead-of-compare.
- Copy KiSelectNextThread for new scheduler branch, without SMT support.
- Fix a bug in KiAdjustQuantumThread under new scheduler.
- Re-implement KiSetPriorityThread under new scheduler, leave hacks for current scheduler support.
- Remove old testing assert from KiExitDispatcher.
svn path=/trunk/; revision=25491
- Fix architectural bug in the entire TrapFrame<->Context conversion system and Ring Privilege Transitions (Inter-ring and intra-ring) which was lacking proper sanitation and validation of segments, flags and debug registers. Among other things, IOPL is now respected, CS is not KGDT_R0_CODE | RPL_MASK anymore, and the GPF code is now properly being called. This completely fixes exception handling being totally broken and crashing firefox installer, mirc, and other applications.
- Rewrite the page fault handler base code in assembly instead of relying on a broken C routine. Detect VDM, V8086, detecting expected/normal fault in ExpInterlockedPopEntrySList and faults in the system handler code. Rewrite MmAccessFault to be the main function that calls out to other sub-fault functions, and use the same prototype as NT.
- Fix the KGDT boot table to have proper granularity and big flags, and extend it to 256 entries.
- Create proper thread context in RtlInitializeContext and cleanup Rtl Thread routines.
- Remove all int3 and breakpoints from trap handlers, and replace them with a much better "UNHANDLED_PATH" macro which freezes the system, beeps, and displays a message with the line of code that's unhandled. This is to clearly tell the user that something is unhandled, instead of nesting infinite exceptions due to the int3.
- Fix a bug in INT_PROLOG.
- Sanitize EFLAGS and Code Segments in KeContextToTrapFrame and KeTrapFrameToContext.
- Implement KiUpdateDr7 and KiRecordDr7 as well as DR_MASK and other DR-validation macros and functions to protect against DR-vulnerabilites as well as to properly account for each active hardware breakpoint in a per-thread fashion by using the dispatcher header.
- Allow CR0_EM when running in a VDM.
- Fix FPU/NPX Register handling in KeContextToTrapFrame and KeTrapFrameToContext, and also speed it up by manual copying instead of a memory move.
- Properly give IOPL 3 to user-mode threads if they requested it.
- Detect GPF during GPF.
- Detect pagefault with a trap-frame spread over two or more pages and nested.
- Properly sanitize and set correct trap frame in KiInitailizeUserApc.
- Return STATUS_ACCESS_VIOLATION during page faults instead of STATUS_UNSUCESSFUL.
- Fix assert in VdmSwapContext, as well as Code Selector check which was broken.
- Fix delayed object deletion (ObDeferDeleteObject) and the Ob Repear Routine and list.
- Update Kernel Fun.
- BUGBUG: Temporaily hack VMWare to detection to always detect VMWare.
svn path=/trunk/; revision=25238
- Add PE routines for traversing the import table and fixing up the imports.
- Add routine for getting the LOADER_MODULE for a freeldr module by name.
- Add the ability to load a not-yet-loaded-driver that's in an import entry (ie: dependencies). Only used for HAL now.
- Map HAL in the same was as the kernel, instead of relocating it as a hack from within the kernel.
- Refactor some routines to better share code. Don't read freed memory anymore.
- Use LdrRelocateImageWithBias instead of quintuplicating relocation code.
- Parse kernel and HAL imports while PE-loading from within freeldr.
- Remove dozens of freeldr.c hacks from the kernel. Remove slow code which was making two copies of the parameter block and wasting over 30KB in static strings that were immediately copied to other strings...
- Don't use HalModuleObject and NtoskrnlModuleObject anymore. Remove LdrSafePEProcessModule and other similar boot-time hacks. Reimplement MmGetSystemRoutineAddress.
- NDK/DDK fixes.
- Add KDCOM to bootcd/installation.
svn path=/trunk/; revision=24778
- Fix a bug in KeFreezeAllThreads which was causing us never to actually parse the next flink.
- Fix a bug in KeFreezeAllThreads which was causing us never to leave the critical region we entered at the beginning.
svn path=/trunk/; revision=24613
- Fix two bugs in KeSetBasePriorityThread and KeSetPriorityThread which were using an undefined value for "Lock Released" if the priority wasn't changing, potentially causing a missing spinlock release when none was required (and thus eventually freezing the scheduler)
svn path=/trunk/; revision=24611
- Improve Init bugchecks to give the exact module that failed. Add Kd initilization in the same block as the other subsystems.
- Rename and re-arrange some initlization calls.
svn path=/trunk/; revision=24438
- Rename some LastKrnl... addresses into MmFreeLdr to easily identify them, and remove the need to use MmFreeLdrMemLower and MmFreeLdrPageDirectoryStart.
- Remove the static KERNEL_BASE definition based on a GCC-specific external. Everything now uses the actual module entry's saved base address.
- Fix a Ps Initialization hack and properly set the boot PDE.
- Add a hack for NTLDR's lack of MmFreeLdrMemHigher and MmFreeLdrPageDirectoryEnd.
svn path=/trunk/; revision=24435
- Make ExpInitNls responsible for NLS initialization in Phase 0 as well, to clean up the code in ExpInitailizeExecutive a bit.
- Initialize the system time/clock in Phase 1, not in Phase 0.
- Do HAL Phase1 initialization as the first step in Phase 1 initialization, then initialize the system clock (since the HAL's RTC is now configured).
- Do Ob Phase 1 init in Phase 1 initialization, not in phase 0.
- Do Ke Phase 1 init after HAL, Ob and Ex phase 1 inits.
- Initialize NLS for Phase 1 after Ke Phase 1, instead of much later.
svn path=/trunk/; revision=24414
- Get rid of the CACHED_MODULE array and associated structures and stop using them, get rid of ParseAndCacheLoadedModules. Stop using KeLoadedModules and KeLoadedModuleCount outside of freeldr.c.
- Except for MmInit1, the kernel should now be totally isolated from FreeLDR-specific data.
svn path=/trunk/; revision=24409
- Detect Hyper-Threading and set Local APIC ID (put back Thomas's code that he added in an old revision and that I had accidentally removed and forgotten to re-convert into the new KiGetFeatureBits function).
- Add code to set the kernel MX Csr Mask in the FX Save Area.
- Startup the sysenter handler on the DPC stack.
- Detect and notify users with broken Pentiums.
- Do some NPX checks to forcefully disable any FPU flags that might be set (SSE, MMX, 3DNOW) if the NPX is disabled.
svn path=/trunk/; revision=24386
- Do all CPU configuration through IPIs and synchronize with potential multiple CPUs.
- Add mtrr.c and patpge.c for future support of MTRR, AMD K6 MTRR, PGE and PAT support, with initilization calls in each of them (which only print out a "Sorry" message).
- Add some stubbed code for extra initizliation to do later once some detection things are fixed.
svn path=/trunk/; revision=24385
- Remove intrinsics in winddk.h since they're now properly done in intrin.h (thanks KJK!!!)
- Make freeldr.c setup the boot KTSS like NTLDR does, so that the GDT entry for it is valid (and remove the code that was doing this from Ki386InitializeTss)
- Refactor KiSystemStartup to use 100% dynamic pointers and machine data queried from the Loader Block or actual GDT/IDT/Selectors in memory, isntead of hard-coded ntoskrnl offsets. This makes it possible to be loaded by NTLDR, which sets these system structures up by itself. (we do it in freeldr.c, as hacks).
svn path=/trunk/; revision=24306
- Implemented KiRosFrldrLpbtoNtLpb to do a lightweight conversion and setup. Next patches will try to get rid of PLOADER_MODULE and use LDR_DATA_TABLE_ENTRY as well as increase bootstrap compatibility.
svn path=/trunk/; revision=24305
- Optimize PspLookupUserEntrypoints to use static ANSI_STRINGs instead of building them at runtime.
- Fix mapping/loading of the System DLL so that it's loaded as executable code, not read-only code. Also fix a handle leak of the section, and re-factor some code into smaller shared functions to reduce code duplication.
svn path=/trunk/; revision=24154
- Modified KeDelayExecutionThread and KeWaitForSingleObject to be much simpler and readable, reducing some of the loops and continues, and 4th-level indentation. Also packed up common wait initialization at the beginning of the function, and into two new inline functions: KxDelayThreadWait and KxSingleThreadWait. No actual semantic changes, just re-ordering.
- Rename KiUnwakeQueue to KiActivateWaiterQueue.
svn path=/trunk/; revision=24110
- Make sure all a process's threads are ready if they're on the ready queue, during attachment.
- Merge in KiReadyThread from scheduler code, with a small code change to keep working with ROS's scheduler lists instead.
svn path=/trunk/; revision=24108
- Don't play with priorities in KiAbortWaitThread anymore, since we'll soon support Win2003 delayed "adjust increment" functionality, so that the code is not repeated many times.
- Rename KiAbortWaitThread to KiUnwaitThread, make it use KiUnlinkThread and delayed adjustment.
- Implement KxUnwaitThread and KxUnwaitThreadForEvent, optimized versions of KiWaitTest that can be used in special circumstances (notification events, thread termination, process signalling).
- Optimize KeSetEvent by handling signaled notification events without acquiring the dispatcher lock, and by using new inlined routines described above.
- Reimplement KeSetEventBoostPriority properly to actually do boosting.
- Fixup KeRundownQueue with a more typical/proper LIST_ENTRY loop.
- Let me know if you see regressions...
svn path=/trunk/; revision=24102
