Changes

Jump to navigation Jump to search

14.0.0 (view source)

Revision as of 09:34, 22 March 2022

11,341 bytes added ,  09:34, 22 March 2022
→‎BootImagePackage: secmon/kernel
Line 76: Line 76:  
=== BootImagePackage ===
 
=== BootImagePackage ===
 
All files in RomFs were updated.
 
All files in RomFs were updated.
 +
 +
====Secure Monitor====
 +
* Compiler upgrade to latest llvm (now using same compiler revision as kernel).
 +
** Secure Monitor is now compiled with -fomit-frame-pointer.
 +
*** >:(
 +
* GenerateSeTestVectorImpl now uses a helper to mix each key into the vector.
 +
* ExceptionHandler is now linked in (@ .text + 0x3E04).
 +
** Previously, this was garbage collected/only present in debug secure monitors.
 +
** NOTE: This is unreachable, and stripped (as e.g. logging isn't emitted, likely because the macros are empty on release builds).
 +
 +
====Kernel====
 +
* Kernel is now compiled with -O3 again instead of -Os
 +
** >:(
 +
* crt0 no longer supports booting in EL2.
 +
** Infinite Loop/Panic is performed instead.
 +
* Initialize0 changes:
 +
** KernelStack setup now uses same helper to determine aslr as other random aligned regions.
 +
** KernelTemp setup now uses same helper to determine aslr as other random aligned regions.
 +
* Slab changes:
 +
** When assigned extra resource, the slab heap is now 0x148000 larger instead of 0x68000 larger.
 +
** Correspondingly, instead of increasing the thread resource limit by 160, the thread resource limit is now increased by 736.
 +
*** This corresponds to changes in userland for pm management of resource limits.
 +
*** Old Intended Resource Limits:
 +
**** System (96 + 512) -> (256 + 512)
 +
**** Applet 96 -> 96
 +
**** Application 96 -> 96
 +
*** New Intended Resource Limits:
 +
**** System (96 + 512) -> 1024
 +
**** Applet 96 -> 256
 +
**** Application 96 -> 256
 +
* SetupPoolPartitionMemoryRegions now panics if the end of the pool partition region is not coincidence with the end of dram.
 +
* KThreadContext was completely revised.
 +
** Most of KThreadContext is now stored inline in kernel stack.
 +
*** Kernel stack layout is now u8 stack[0xDB0]; KThreadContext thread_context; KThreadStackParameters stack_parameters;
 +
** KThreadContext now only stores the 8 callee-save FPU registers.
 +
*** The remaining 24 caller-save FPU registers are stored inside KThread, where KThreadContext used to be.
 +
*** NOTE that 32-bit fpu has 4 callee-save FPU registers and 12 caller-save registers, which use the start of the relevant 64-bit storages as usual.
 +
** KThreadStackParameters was revised to facilitate this.
 +
*** The pointer to KThreadContext previously stored in stack parameters now points to the external FPU register array.
 +
*** The members at end of params are now: u16 disable_count; u8 current_svc_id; u8 unused_2c; u8 exception_flags; u8 is_pinned; u8 unused_2f;
 +
**** The "exception_flags" field is a new set of bitflags (encoding old state was were previously separate bools + new state).
 +
***** Bit 0x1  = is_calling_svc
 +
***** Bit 0x2  = is_in_exception_handler
 +
***** Bit 0x4  = is_fpu_state_restore_needed
 +
***** Bit 0x8  = is_64_bit_fpu
 +
***** Bit 0x10 = has_exception_svc_permissions
 +
***** Bit 0x20 = is_in_cache_operation
 +
***** Bit 0x40 = is_in_tlb_operation
 +
** Exception exits now check is_fpu_state_restore_needed, and restore FPU registers only if needed (and clear is_fpu_state_restore_needed on restore).
 +
*** is_fpu_state_restore_needed is set to true *only* on thread switch with FPU enabled.
 +
**** Caller-save FPU registers are saved *only* if a thread is in an SVC and does not have exception svc permissions.
 +
**** All other thread switches save only the 8 (or 4) callee-save FPU registers.
 +
**** All thread switches now guarantee as post-condition that the fpu is disabled leaving the switch (it will be re-enabled on exception exit if needed).
 +
*** On SVC exception return, all caller-save FPU registers are set to zero unless the thread has exception svc permissions.
 +
** KThread::CloneFpuStatus now uses KScopedDisableInterrupt
 +
* Various hw maintenance changes:
 +
** KernelLdr no longer does cache maintenance by set/way when setting up initial identity mapping, no longer invalidates instruction cache/tlb, no longer does dsb after setting sctlr_el1.
 +
** FlushEntireDataCacheLocal/Shared in init now perform dsb sy, FlushEntireDataCacheAndInvalidateTlbForInit no longer does after calling them.
 +
** dsb sy/isb is now performed after setting sctlr_el1, when disabling mmu/icache.
 +
** KInitialPageTable::Map no longer does dsb ish after all attribute writes.
 +
*** Instead does it before writing table entries, and at the end of the function.
 +
** KInitialPageTable::PhysicallyRandomize no longer does StoreEntireCacheForInit.
 +
*** Now does dc cvac on randomized virtual address range, dsb ish, ic iallu, dsb ish, isb. (see weaker-barriers section of diff)
 +
** KInitialPageTable::SwapBlocks now does dsb ish after memcpy to swap blocks.
 +
** KInitialPageTable::Reprotect no longer does dsb ish before performing reprotection.
 +
** KInitialProcessReader::Load no longer calls cpu::FlushEntireDataCache/cpu::InvalidateInstructionCache.
 +
** Set/way cache operations now perform dsb sy before configuring csselr.
 +
*** This affects InvalidateDataCacheForResumeEntry, FlushEntireDataCache, KCacheHelperInterruptHandler, and the initial cache maintenance when disabling the mmu.
 +
** FlushEntireDataCache now does dsb sy after doing full set/way cache flush, instead of after each set/way op.
 +
*** NOTE: This is still only a local flush without coherence guarantees, set/way aren't supposed to be used after multiple cores are online.
 +
** KSystemControl::CpuSleepHandler no longer embeds unreachable cache maintenance assembly after CpuSuspend.
 +
** Kernel now performs different hw maintenance if a thread is in a hw maintenance operation when interrupted:
 +
*** If a thread is interrupted while performing cache maintenance in EL1 (tracked via new exception flags bit 0x20), KInterruptManager::OnHandleInterrupt performs dsb sy.
 +
**** Set and cleared for scope of cpu::InvalidateDataCache instead of disabling core migration.
 +
**** Set and cleared for scope of cpu::StoreDataCache instead of disabling core migration.
 +
**** Set and cleared for scope of cpu::FlushDataCache instead of disabling core migration.
 +
*** If a thread is interrupted while performing tlb maintenance in EL1 (tracked via new exception flags bit 0x40), KInterruptManager::OnHandleInterrupt performs dsb ish.
 +
**** Set and cleared for scope of KPageTable::NoteUpdated
 +
*** If a thread is interrupted while performing cache maintenance in EL0 (tracked via new bool @ TLS + 0x104), KInterruptManager::OnHandleInterrupt performs dsb sy.
 +
**** This is equivalent to the EL1 cache maintenance tracking above, providing an opt-in way for userland to ensure its cache maintenance is coherent even when interrupted.
 +
**** Note that official userland code now sets this bit before performing cache maintenance.
 +
** Memory barriers were revised in many places -- barriers were weakened in many places, and some functions which previously lacked barriers had them added, including:
 +
*** cpu::InvalidateEntireInstructionCache: dsb sy -> dsb ish
 +
*** cpu::EnsureInstructionConsistency: dsb sy; isb; -> dsb ish; isb;
 +
**** NOTE: Functions written in assembly still use the old pattern for ensuring instruction consistency.
 +
*** KCacheInterruptHandler::RequestOperation: dsb sy -> dsb ish
 +
*** KScheduler::EnableScheduling: dsb sy -> dsb ish
 +
*** KScheduler::SwitchThread no longer does dsb sy before setting ttbr0/contextidr_el1.
 +
*** KPageTable::NoteUpdated: dsb sy; if (m_kernel) { ... dsb sy; } else { ... dsb sy; isb; } -- dsb ishst; if (m_kernel) { ... dsb ish; } else { ... dsb ish; isb; }
 +
**** KPageTable::NoteSingleKernelPageUpdated now similarly does dsb ishst for outer and dsb ish for inner barriers.
 +
*** KPageTable::ClearPageTable: now does dsb ish after clearing page to zero via dc zva
 +
*** KPageTable::MapContiguous: now does dsb ishst after merging pages.
 +
*** KPageTable::MapPageGroup: now does dsb ishst after merging pages.
 +
*** KPageTable::PteDataSynchronizationBarrier: now dmb ishst instead of dsb ish (probably KPageTable::PteDataMemoryBarrier, now?)
 +
*** KPageTable::MapL2Blocks/MapL3Blocks: pattern for setting entry for new table went from Barrier(); WriteEntry(); Barrier(); -> Barrier(); WriteEntry();
 +
**** This was PteDataSynchronizationBarrier(), and correspondingly asm is dsb ish; str; dsb ish; -> dmb ishst; str;
 +
*** KSupervisorPageTable::SetTtbr0 no longer does dsb sy before setting ttbr0/contextidr_el1.
 +
** UserspaceAccess::InvalidateInstructionCache was removed (previously unused).
 +
* Various changes to KInterruptName/interrupt management:
 +
** Enum values for IPIs were revised:
 +
*** KInterruptName_ThreadTerminate    4 -> 0
 +
*** KInterruptName_CacheOperation    5 -> 1
 +
*** KInterruptName_Scheduler          6 -> 2
 +
** New KInterruptName (KInterruptName_CoreBarrier) = 3
 +
*** Interrupt handler for this is registered with KInterruptControllerPriority_Scheduler after ThreadTerminate handler is registered.
 +
** Interrupt handler for the user cycle counter interrupt is no longer registered.
 +
*** This is presumably now under the same ifdef that enables svc::InfoType_PerformanceCounter.
 +
* KCapability now has a new member "physical_core_mask", which tracks what physical cores are allowable.
 +
** KThread::FinishTermination now calls a new function (cpu::ForceSynchronizeAllCores) after waiting for the thread to not be current on any scheduler.
 +
*** This function sends an IPI (KInterruptName_CoreBarrier) to all cores, and waits for all cores to acknowledge the interrupt.
 +
* Changes to KMemoryManager allocation:
 +
** KPageHeap now has an additional KPageHeapBitmapRng @ 0x328 to facilitate additional allocation randomization.
 +
** KMemoryManager::AllocateAndOpenContinuous now uses a new KPageHeap method "AllocateRandomBlock"
 +
*** KPhysicalAddress KPageHeap::AllocateRandomBlock(s32 index, size_t num_pages, size_t align_pages);
 +
*** This method allocates `num_pages` pages (aligned to at least `align_pages`) at random.
 +
**** First, the kernel chooses a random block index to allocate from.
 +
***** This is done by increasing the block index until there are at least 4 possible random choices for the desired alignment, then selecting the block that corresponds to a random pick from those choices.
 +
**** Next, the kernel allocates a random block from within that index.
 +
**** Finally, the kernel selects a random (align_pages)-aligned offset within that block, frees the memory before/after the allocated chunk, and returns the memory.
 +
** Allocation of KPageGroups still uses a `random` argument, however:
 +
*** KPageHeap::PopBlock no longer takes a random argument.
 +
*** KPageHeap::AllocateBlock now calls new new KPageHeap method "AllocateRandomBlock".
 +
**** KPageHeap::AllocateRandomBlock(s32 index, size_t num_pages);
 +
**** This is effectively the same logic as above, but with align_pages == # of pages for the argument block index.
 +
* CreateProcess now calls a new function to validate the user-capabilities before creating the KProcess.
 +
** This checks that the capabilities are user-readable and that the map region capabilities correspond to actually-present regions.
 +
** This corresponds to changes in Loader allowing for map region capabilities (previously, these were only allowed via KIP, and Loader always rejected them).
 +
* New InfoType 0x1A ("InfoType_IsSvcPermitted").
 +
** Returns whether the current process can access a given SVC.
 +
** Nintendo returns InvalidCombination when checking SVCs other than SynchronizePreemptionState.
 +
*** Official userland code now aborts if the process does not have permission to use SynchronizePreemptionState before incrementing ThreadLocalRegion->disable_count for the first time.
    
==See Also==
 
==See Also==
Retrieved from "https://switchbrew.org/wiki/Special:MobileDiff/11554"

Navigation menu