OpenCloudOS-Kernel/arch/powerpc/kernel/entry_64.S

480 lines
12 KiB
ArmAsm
Raw Normal View History

/* SPDX-License-Identifier: GPL-2.0-or-later */
/*
* PowerPC version
* Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
* Rewritten by Cort Dougan (cort@cs.nmt.edu) for PReP
* Copyright (C) 1996 Cort Dougan <cort@cs.nmt.edu>
* Adapted for Power Macintosh by Paul Mackerras.
* Low-level exception handlers and MMU support
* rewritten by Paul Mackerras.
* Copyright (C) 1996 Paul Mackerras.
* MPC8xx modifications Copyright (C) 1997 Dan Malek (dmalek@jlc.net).
*
* This file contains the system call entry code, context switch
* code, and exception/interrupt return code for PowerPC.
*/
#include <linux/errno.h>
#include <linux/err.h>
powerpc/64s: Implement interrupt exit logic in C Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store. The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform the store. The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic. This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd. mpe: Includes fixes from Nick for _TIF_EMULATE_STACK_STORE handling (including the fast_interrupt_return path), to remove trace_hardirqs_on(), and fixes the interrupt-return part of the MSR_VSX restore bug caught by tm-unavailable selftest. mpe: Incorporate fix from Nick: The return-to-kernel path has to replay any soft-pending interrupts if it is returning to a context that had interrupts soft-enabled. It has to do this carefully and avoid plain enabling interrupts if this is an irq context, which can cause multiple nesting of interrupts on the stack, and other unexpected issues. The code which avoided this case got the soft-mask state wrong, and marked interrupts as enabled before going around again to retry. This seems to be mostly harmless except when PREEMPT=y, this calls preempt_schedule_irq with irqs apparently enabled and runs into a BUG in kernel/sched/core.c Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michal Suchanek <msuchanek@suse.de> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200225173541.1549955-29-npiggin@gmail.com
2020-02-26 01:35:37 +08:00
#include <asm/cache.h>
#include <asm/unistd.h>
#include <asm/processor.h>
#include <asm/page.h>
#include <asm/mmu.h>
#include <asm/thread_info.h>
#include <asm/code-patching-asm.h>
#include <asm/ppc_asm.h>
#include <asm/asm-offsets.h>
#include <asm/cputable.h>
#include <asm/firmware.h>
#include <asm/bug.h>
#include <asm/ptrace.h>
#include <asm/irqflags.h>
powerpc: Rework lazy-interrupt handling The current implementation of lazy interrupts handling has some issues that this tries to address. We don't do the various workarounds we need to do when re-enabling interrupts in some cases such as when returning from an interrupt and thus we may still lose or get delayed decrementer or doorbell interrupts. The current scheme also makes it much harder to handle the external "edge" interrupts provided by some BookE processors when using the EPR facility (External Proxy) and the Freescale Hypervisor. Additionally, we tend to keep interrupts hard disabled in a number of cases, such as decrementer interrupts, external interrupts, or when a masked decrementer interrupt is pending. This is sub-optimal. This is an attempt at fixing it all in one go by reworking the way we do the lazy interrupt disabling from the ground up. The base idea is to replace the "hard_enabled" field with a "irq_happened" field in which we store a bit mask of what interrupt occurred while soft-disabled. When re-enabling, either via arch_local_irq_restore() or when returning from an interrupt, we can now decide what to do by testing bits in that field. We then implement replaying of the missed interrupts either by re-using the existing exception frame (in exception exit case) or via the creation of a new one from an assembly trampoline (in the arch_local_irq_enable case). This removes the need to play with the decrementer to try to create fake interrupts, among others. In addition, this adds a few refinements: - We no longer hard disable decrementer interrupts that occur while soft-disabled. We now simply bump the decrementer back to max (on BookS) or leave it stopped (on BookE) and continue with hard interrupts enabled, which means that we'll potentially get better sample quality from performance monitor interrupts. - Timer, decrementer and doorbell interrupts now hard-enable shortly after removing the source of the interrupt, which means they no longer run entirely hard disabled. Again, this will improve perf sample quality. - On Book3E 64-bit, we now make the performance monitor interrupt act as an NMI like Book3S (the necessary C code for that to work appear to already be present in the FSL perf code, notably calling nmi_enter instead of irq_enter). (This also fixes a bug where BookE perfmon interrupts could clobber r14 ... oops) - We could make "masked" decrementer interrupts act as NMIs when doing timer-based perf sampling to improve the sample quality. Signed-off-by-yet: Benjamin Herrenschmidt <benh@kernel.crashing.org> --- v2: - Add hard-enable to decrementer, timer and doorbells - Fix CR clobber in masked irq handling on BookE - Make embedded perf interrupt act as an NMI - Add a PACA_HAPPENED_EE_EDGE for use by FSL if they want to retrigger an interrupt without preventing hard-enable v3: - Fix or vs. ori bug on Book3E - Fix enabling of interrupts for some exceptions on Book3E v4: - Fix resend of doorbells on return from interrupt on Book3E v5: - Rebased on top of my latest series, which involves some significant rework of some aspects of the patch. v6: - 32-bit compile fix - more compile fixes with various .config combos - factor out the asm code to soft-disable interrupts - remove the C wrapper around preempt_schedule_irq v7: - Fix a bug with hard irq state tracking on native power7
2012-03-06 15:27:59 +08:00
#include <asm/hw_irq.h>
#include <asm/context_tracking.h>
#include <asm/ppc-opcode.h>
#include <asm/barrier.h>
#include <asm/export.h>
#include <asm/asm-compat.h>
#ifdef CONFIG_PPC_BOOK3S
#include <asm/exception-64s.h>
#else
#include <asm/exception-64e.h>
#endif
#include <asm/feature-fixups.h>
#include <asm/kup.h>
/*
* System calls.
*/
.section ".text"
#ifdef CONFIG_PPC_BOOK3S_64
#define FLUSH_COUNT_CACHE \
1: nop; \
patch_site 1b, patch__call_flush_branch_caches1; \
1: nop; \
patch_site 1b, patch__call_flush_branch_caches2; \
1: nop; \
patch_site 1b, patch__call_flush_branch_caches3
.macro nops number
.rept \number
nop
.endr
.endm
.balign 32
.global flush_branch_caches
flush_branch_caches:
/* Save LR into r9 */
mflr r9
powerpc/book3s64: Fix link stack flush on context switch In commit ee13cb249fab ("powerpc/64s: Add support for software count cache flush"), I added support for software to flush the count cache (indirect branch cache) on context switch if firmware told us that was the required mitigation for Spectre v2. As part of that code we also added a software flush of the link stack (return address stack), which protects against Spectre-RSB between user processes. That is all correct for CPUs that activate that mitigation, which is currently Power9 Nimbus DD2.3. What I got wrong is that on older CPUs, where firmware has disabled the count cache, we also need to flush the link stack on context switch. To fix it we create a new feature bit which is not set by firmware, which tells us we need to flush the link stack. We set that when firmware tells us that either of the existing Spectre v2 mitigations are enabled. Then we adjust the patching code so that if we see that feature bit we enable the link stack flush. If we're also told to flush the count cache in software then we fall through and do that also. On the older CPUs we don't need to do do the software count cache flush, firmware has disabled it, so in that case we patch in an early return after the link stack flush. The naming of some of the functions is awkward after this patch, because they're called "count cache" but they also do link stack. But we'll fix that up in a later commit to ease backporting. This is the fix for CVE-2019-18660. Reported-by: Anthony Steinhauser <asteinhauser@google.com> Fixes: ee13cb249fab ("powerpc/64s: Add support for software count cache flush") Cc: stable@vger.kernel.org # v4.4+ Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2019-11-13 18:05:41 +08:00
// Flush the link stack
.rept 64
bl .+4
.endr
b 1f
nops 6
.balign 32
/* Restore LR */
1: mtlr r9
powerpc/book3s64: Fix link stack flush on context switch In commit ee13cb249fab ("powerpc/64s: Add support for software count cache flush"), I added support for software to flush the count cache (indirect branch cache) on context switch if firmware told us that was the required mitigation for Spectre v2. As part of that code we also added a software flush of the link stack (return address stack), which protects against Spectre-RSB between user processes. That is all correct for CPUs that activate that mitigation, which is currently Power9 Nimbus DD2.3. What I got wrong is that on older CPUs, where firmware has disabled the count cache, we also need to flush the link stack on context switch. To fix it we create a new feature bit which is not set by firmware, which tells us we need to flush the link stack. We set that when firmware tells us that either of the existing Spectre v2 mitigations are enabled. Then we adjust the patching code so that if we see that feature bit we enable the link stack flush. If we're also told to flush the count cache in software then we fall through and do that also. On the older CPUs we don't need to do do the software count cache flush, firmware has disabled it, so in that case we patch in an early return after the link stack flush. The naming of some of the functions is awkward after this patch, because they're called "count cache" but they also do link stack. But we'll fix that up in a later commit to ease backporting. This is the fix for CVE-2019-18660. Reported-by: Anthony Steinhauser <asteinhauser@google.com> Fixes: ee13cb249fab ("powerpc/64s: Add support for software count cache flush") Cc: stable@vger.kernel.org # v4.4+ Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2019-11-13 18:05:41 +08:00
// If we're just flushing the link stack, return here
3: nop
patch_site 3b patch__flush_link_stack_return
li r9,0x7fff
mtctr r9
PPC_BCCTR_FLUSH
2: nop
patch_site 2b patch__flush_count_cache_return
nops 3
.rept 278
.balign 32
PPC_BCCTR_FLUSH
nops 7
.endr
blr
#else
#define FLUSH_COUNT_CACHE
#endif /* CONFIG_PPC_BOOK3S_64 */
/*
* This routine switches between two different tasks. The process
* state of one is saved on its kernel stack. Then the state
* of the other is restored from its kernel stack. The memory
* management hardware is updated to the second process's state.
powerpc/64s: Implement interrupt exit logic in C Implement the bulk of interrupt return logic in C. The asm return code must handle a few cases: restoring full GPRs, and emulating stack store. The stack store emulation is significantly simplfied, rather than creating a new return frame and switching to that before performing the store, it uses the PACA to keep a scratch register around to perform the store. The asm return code is moved into 64e for now. The new logic has made allowance for 64e, but I don't have a full environment that works well to test it, and even booting in emulated qemu is not great for stress testing. 64e shouldn't be too far off working with this, given a bit more testing and auditing of the logic. This is slightly faster on a POWER9 (page fault speed increases about 1.1%), probably due to reduced mtmsrd. mpe: Includes fixes from Nick for _TIF_EMULATE_STACK_STORE handling (including the fast_interrupt_return path), to remove trace_hardirqs_on(), and fixes the interrupt-return part of the MSR_VSX restore bug caught by tm-unavailable selftest. mpe: Incorporate fix from Nick: The return-to-kernel path has to replay any soft-pending interrupts if it is returning to a context that had interrupts soft-enabled. It has to do this carefully and avoid plain enabling interrupts if this is an irq context, which can cause multiple nesting of interrupts on the stack, and other unexpected issues. The code which avoided this case got the soft-mask state wrong, and marked interrupts as enabled before going around again to retry. This seems to be mostly harmless except when PREEMPT=y, this calls preempt_schedule_irq with irqs apparently enabled and runs into a BUG in kernel/sched/core.c Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michal Suchanek <msuchanek@suse.de> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200225173541.1549955-29-npiggin@gmail.com
2020-02-26 01:35:37 +08:00
* Finally, we can return to the second process, via interrupt_return.
* On entry, r3 points to the THREAD for the current task, r4
* points to the THREAD for the new task.
*
* Note: there are two ways to get to the "going out" portion
* of this code; either by coming in via the entry (_switch)
* or via "fork" which must set up an environment equivalent
* to the "_switch" path. If you change this you'll have to change
* the fork code also.
*
* The code which creates the new task context is in 'copy_thread'
* in arch/powerpc/kernel/process.c
*/
.align 7
_GLOBAL(_switch)
mflr r0
std r0,16(r1)
stdu r1,-SWITCH_FRAME_SIZE(r1)
/* r3-r13 are caller saved -- Cort */
SAVE_NVGPRS(r1)
std r0,_NIP(r1) /* Return to switch caller */
mfcr r23
std r23,_CCR(r1)
std r1,KSP(r3) /* Set old stack pointer */
kuap_check_amr r9, r10
FLUSH_COUNT_CACHE /* Clobbers r9, ctr */
/*
* On SMP kernels, care must be taken because a task may be
* scheduled off CPUx and on to CPUy. Memory ordering must be
* considered.
*
* Cacheable stores on CPUx will be visible when the task is
* scheduled on CPUy by virtue of the core scheduler barriers
* (see "Notes on Program-Order guarantees on SMP systems." in
* kernel/sched/core.c).
*
* Uncacheable stores in the case of involuntary preemption must
* be taken care of. The smp_mb__after_spinlock() in __schedule()
* is implemented as hwsync on powerpc, which orders MMIO too. So
* long as there is an hwsync in the context switch path, it will
* be executed on the source CPU after the task has performed
* all MMIO ops on that CPU, and on the destination CPU before the
* task performs any MMIO ops there.
*/
powerpc: Feature nop out reservation clear when stcx checks address The POWER architecture does not require stcx to check that it is operating on the same address as the larx. This means it is possible for an an exception handler to execute a larx, get a reservation, decide not to do the stcx and then return back with an active reservation. If the interrupted code was in the middle of a larx/stcx sequence the stcx could incorrectly succeed. All recent POWER CPUs check the address before letting the stcx succeed so we can create a CPU feature and nop it out. As Ben suggested, we can only do this in our syscall path because there is a remote possibility some kernel code gets interrupted by an exception that ends up operating on the same cacheline. Thanks to Paul Mackerras and Derek Williams for the idea. To test this I used a very simple null syscall (actually getppid) testcase at http://ozlabs.org/~anton/junkcode/null_syscall.c I tested against 2.6.35-git10 with the following changes against the pseries_defconfig: CONFIG_VIRT_CPU_ACCOUNTING=n CONFIG_AUDIT=n CONFIG_PPC_4K_PAGES=n CONFIG_PPC_64K_PAGES=y CONFIG_FORCE_MAX_ZONEORDER=9 CONFIG_PPC_SUBPAGE_PROT=n CONFIG_FUNCTION_TRACER=n CONFIG_FUNCTION_GRAPH_TRACER=n CONFIG_IRQSOFF_TRACER=n CONFIG_STACK_TRACER=n to remove the overhead of virtual CPU accounting, syscall auditing and the ftrace mcount tracers. 64kB pages were enabled to minimise TLB misses. POWER6: +8.2% POWER7: +7.0% Another suggestion was to use a larx to something in the L1 instead of a stcx. This was almost as fast as removing the larx on POWER6, but only 3.5% faster on POWER7. We can use this to speed up the reservation clear in our exception exit code. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2010-08-11 09:40:27 +08:00
/*
* The kernel context switch path must contain a spin_lock,
* which contains larx/stcx, which will clear any reservation
* of the task being switched.
powerpc: Feature nop out reservation clear when stcx checks address The POWER architecture does not require stcx to check that it is operating on the same address as the larx. This means it is possible for an an exception handler to execute a larx, get a reservation, decide not to do the stcx and then return back with an active reservation. If the interrupted code was in the middle of a larx/stcx sequence the stcx could incorrectly succeed. All recent POWER CPUs check the address before letting the stcx succeed so we can create a CPU feature and nop it out. As Ben suggested, we can only do this in our syscall path because there is a remote possibility some kernel code gets interrupted by an exception that ends up operating on the same cacheline. Thanks to Paul Mackerras and Derek Williams for the idea. To test this I used a very simple null syscall (actually getppid) testcase at http://ozlabs.org/~anton/junkcode/null_syscall.c I tested against 2.6.35-git10 with the following changes against the pseries_defconfig: CONFIG_VIRT_CPU_ACCOUNTING=n CONFIG_AUDIT=n CONFIG_PPC_4K_PAGES=n CONFIG_PPC_64K_PAGES=y CONFIG_FORCE_MAX_ZONEORDER=9 CONFIG_PPC_SUBPAGE_PROT=n CONFIG_FUNCTION_TRACER=n CONFIG_FUNCTION_GRAPH_TRACER=n CONFIG_IRQSOFF_TRACER=n CONFIG_STACK_TRACER=n to remove the overhead of virtual CPU accounting, syscall auditing and the ftrace mcount tracers. 64kB pages were enabled to minimise TLB misses. POWER6: +8.2% POWER7: +7.0% Another suggestion was to use a larx to something in the L1 instead of a stcx. This was almost as fast as removing the larx on POWER6, but only 3.5% faster on POWER7. We can use this to speed up the reservation clear in our exception exit code. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2010-08-11 09:40:27 +08:00
*/
#ifdef CONFIG_PPC_BOOK3S
/* Cancel all explict user streams as they will have no use after context
* switch and will stop the HW from creating streams itself
*/
DCBT_BOOK3S_STOP_ALL_STREAM_IDS(r6)
#endif
addi r6,r4,-THREAD /* Convert THREAD to 'current' */
std r6,PACACURRENT(r13) /* Set new 'current' */
#if defined(CONFIG_STACKPROTECTOR)
ld r6, TASK_CANARY(r6)
std r6, PACA_CANARY(r13)
#endif
ld r8,KSP(r4) /* new stack pointer */
#ifdef CONFIG_PPC_64S_HASH_MMU
BEGIN_MMU_FTR_SECTION
b 2f
END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
BEGIN_FTR_SECTION
clrrdi r6,r8,28 /* get its ESID */
clrrdi r9,r1,28 /* get current sp ESID */
FTR_SECTION_ELSE
clrrdi r6,r8,40 /* get its 1T ESID */
clrrdi r9,r1,40 /* get current sp 1T ESID */
ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_1T_SEGMENT)
clrldi. r0,r6,2 /* is new ESID c00000000? */
cmpd cr1,r6,r9 /* or is new ESID the same as current ESID? */
cror eq,4*cr1+eq,eq
beq 2f /* if yes, don't slbie it */
/* Bolt in the new stack SLB entry */
ld r7,KSP_VSID(r4) /* Get new stack's VSID */
oris r0,r6,(SLB_ESID_V)@h
ori r0,r0,(SLB_NUM_BOLTED-1)@l
BEGIN_FTR_SECTION
li r9,MMU_SEGSIZE_1T /* insert B field */
oris r6,r6,(MMU_SEGSIZE_1T << SLBIE_SSIZE_SHIFT)@h
rldimi r7,r9,SLB_VSID_SSIZE_SHIFT,0
END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT)
/* Update the last bolted SLB. No write barriers are needed
* here, provided we only update the current CPU's SLB shadow
* buffer.
*/
ld r9,PACA_SLBSHADOWPTR(r13)
li r12,0
std r12,SLBSHADOW_STACKESID(r9) /* Clear ESID */
li r12,SLBSHADOW_STACKVSID
STDX_BE r7,r12,r9 /* Save VSID */
li r12,SLBSHADOW_STACKESID
STDX_BE r0,r12,r9 /* Save ESID */
/* No need to check for MMU_FTR_NO_SLBIE_B here, since when
* we have 1TB segments, the only CPUs known to have the errata
* only support less than 1TB of system memory and we'll never
* actually hit this code path.
*/
powerpc/mm/hash: Add missing isync prior to kernel stack SLB switch Currently we do not have an isync, or any other context synchronizing instruction prior to the slbie/slbmte in _switch() that updates the SLB entry for the kernel stack. However that is not correct as outlined in the ISA. From Power ISA Version 3.0B, Book III, Chapter 11, page 1133: "Changing the contents of ... the contents of SLB entries ... can have the side effect of altering the context in which data addresses and instruction addresses are interpreted, and in which instructions are executed and data accesses are performed. ... These side effects need not occur in program order, and therefore may require explicit synchronization by software. ... The synchronizing instruction before the context-altering instruction ensures that all instructions up to and including that synchronizing instruction are fetched and executed in the context that existed before the alteration." And page 1136: "For data accesses, the context synchronizing instruction before the slbie, slbieg, slbia, slbmte, tlbie, or tlbiel instruction ensures that all preceding instructions that access data storage have completed to a point at which they have reported all exceptions they will cause." We're not aware of any bugs caused by this, but it should be fixed regardless. Add the missing isync when updating kernel stack SLB entry. Cc: stable@vger.kernel.org Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> [mpe: Flesh out change log with more ISA text & explanation] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-05-30 21:18:04 +08:00
isync
slbie r6
BEGIN_FTR_SECTION
slbie r6 /* Workaround POWER5 < DD2.1 issue */
END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
slbmte r7,r0
isync
2:
#endif /* CONFIG_PPC_64S_HASH_MMU */
clrrdi r7, r8, THREAD_SHIFT /* base of new stack */
/* Note: this uses SWITCH_FRAME_SIZE rather than INT_FRAME_SIZE
because we don't need to leave the 288-byte ABI gap at the
top of the kernel stack. */
addi r7,r7,THREAD_SIZE-SWITCH_FRAME_SIZE
/*
* PMU interrupts in radix may come in here. They will use r1, not
* PACAKSAVE, so this stack switch will not cause a problem. They
* will store to the process stack, which may then be migrated to
* another CPU. However the rq lock release on this CPU paired with
* the rq lock acquire on the new CPU before the stack becomes
* active on the new CPU, will order those stores.
*/
mr r1,r8 /* start using new stack pointer */
std r7,PACAKSAVE(r13)
ld r6,_CCR(r1)
mtcrf 0xFF,r6
/* r3-r13 are destroyed -- Cort */
REST_NVGPRS(r1)
/* convert old thread to its task_struct for return value */
addi r3,r3,-THREAD
ld r7,_NIP(r1) /* Return to _switch caller in new task */
mtlr r7
addi r1,r1,SWITCH_FRAME_SIZE
blr
#ifdef CONFIG_PPC_RTAS
/*
* On CHRP, the Run-Time Abstraction Services (RTAS) have to be
* called with the MMU off.
*
* In addition, we need to be in 32b mode, at least for now.
*
* Note: r3 is an input parameter to rtas, so don't trash it...
*/
_GLOBAL(enter_rtas)
mflr r0
std r0,16(r1)
stdu r1,-SWITCH_FRAME_SIZE(r1) /* Save SP and create stack space. */
/* Because RTAS is running in 32b mode, it clobbers the high order half
* of all registers that it saves. We therefore save those registers
* RTAS might touch to the stack. (r0, r3-r13 are caller saved)
*/
SAVE_GPR(2, r1) /* Save the TOC */
SAVE_GPR(13, r1) /* Save paca */
SAVE_NVGPRS(r1) /* Save the non-volatiles */
mfcr r4
std r4,_CCR(r1)
mfctr r5
std r5,_CTR(r1)
mfspr r6,SPRN_XER
std r6,_XER(r1)
mfdar r7
std r7,_DAR(r1)
mfdsisr r8
std r8,_DSISR(r1)
/* Temporary workaround to clear CR until RTAS can be modified to
* ignore all bits.
*/
li r0,0
mtcr r0
powerpc/64: Change soft_enabled from flag to bitmask "paca->soft_enabled" is used as a flag to mask some of interrupts. Currently supported flags values and their details: soft_enabled MSR[EE] 0 0 Disabled (PMI and HMI not masked) 1 1 Enabled "paca->soft_enabled" is initialized to 1 to make the interripts as enabled. arch_local_irq_disable() will toggle the value when interrupts needs to disbled. At this point, the interrupts are not actually disabled, instead, interrupt vector has code to check for the flag and mask it when it occurs. By "mask it", it update interrupt paca->irq_happened and return. arch_local_irq_restore() is called to re-enable interrupts, which checks and replays interrupts if any occured. Now, as mentioned, current logic doesnot mask "performance monitoring interrupts" and PMIs are implemented as NMI. But this patchset depends on local_irq_* for a successful local_* update. Meaning, mask all possible interrupts during local_* update and replay them after the update. So the idea here is to reserve the "paca->soft_enabled" logic. New values and details: soft_enabled MSR[EE] 1 0 Disabled (PMI and HMI not masked) 0 1 Enabled Reason for the this change is to create foundation for a third mask value "0x2" for "soft_enabled" to add support to mask PMIs. When ->soft_enabled is set to a value "3", PMI interrupts are mask and when set to a value of "1", PMI are not mask. With this patch also extends soft_enabled as interrupt disable mask. Current flags are renamed from IRQ_[EN?DIS}ABLED to IRQS_ENABLED and IRQS_DISABLED. Patch also fixes the ptrace call to force the user to see the softe value to be alway 1. Reason being, even though userspace has no business knowing about softe, it is part of pt_regs. Like-wise in signal context. Signed-off-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-12-20 11:55:49 +08:00
#ifdef CONFIG_BUG
/* There is no way it is acceptable to get here with interrupts enabled,
* check it with the asm equivalent of WARN_ON
*/
lbz r0,PACAIRQSOFTMASK(r13)
powerpc/64: Change soft_enabled from flag to bitmask "paca->soft_enabled" is used as a flag to mask some of interrupts. Currently supported flags values and their details: soft_enabled MSR[EE] 0 0 Disabled (PMI and HMI not masked) 1 1 Enabled "paca->soft_enabled" is initialized to 1 to make the interripts as enabled. arch_local_irq_disable() will toggle the value when interrupts needs to disbled. At this point, the interrupts are not actually disabled, instead, interrupt vector has code to check for the flag and mask it when it occurs. By "mask it", it update interrupt paca->irq_happened and return. arch_local_irq_restore() is called to re-enable interrupts, which checks and replays interrupts if any occured. Now, as mentioned, current logic doesnot mask "performance monitoring interrupts" and PMIs are implemented as NMI. But this patchset depends on local_irq_* for a successful local_* update. Meaning, mask all possible interrupts during local_* update and replay them after the update. So the idea here is to reserve the "paca->soft_enabled" logic. New values and details: soft_enabled MSR[EE] 1 0 Disabled (PMI and HMI not masked) 0 1 Enabled Reason for the this change is to create foundation for a third mask value "0x2" for "soft_enabled" to add support to mask PMIs. When ->soft_enabled is set to a value "3", PMI interrupts are mask and when set to a value of "1", PMI are not mask. With this patch also extends soft_enabled as interrupt disable mask. Current flags are renamed from IRQ_[EN?DIS}ABLED to IRQS_ENABLED and IRQS_DISABLED. Patch also fixes the ptrace call to force the user to see the softe value to be alway 1. Reason being, even though userspace has no business knowing about softe, it is part of pt_regs. Like-wise in signal context. Signed-off-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-12-20 11:55:49 +08:00
1: tdeqi r0,IRQS_ENABLED
powerpc/bug: Provide better flexibility to WARN_ON/__WARN_FLAGS() with asm goto Using asm goto in __WARN_FLAGS() and WARN_ON() allows more flexibility to GCC. For that add an entry to the exception table so that program_check_exception() knowns where to resume execution after a WARNING. Here are two exemples. The first one is done on PPC32 (which benefits from the previous patch), the second is on PPC64. unsigned long test(struct pt_regs *regs) { int ret; WARN_ON(regs->msr & MSR_PR); return regs->gpr[3]; } unsigned long test9w(unsigned long a, unsigned long b) { if (WARN_ON(!b)) return 0; return a / b; } Before the patch: 000003a8 <test>: 3a8: 81 23 00 84 lwz r9,132(r3) 3ac: 71 29 40 00 andi. r9,r9,16384 3b0: 40 82 00 0c bne 3bc <test+0x14> 3b4: 80 63 00 0c lwz r3,12(r3) 3b8: 4e 80 00 20 blr 3bc: 0f e0 00 00 twui r0,0 3c0: 80 63 00 0c lwz r3,12(r3) 3c4: 4e 80 00 20 blr 0000000000000bf0 <.test9w>: bf0: 7c 89 00 74 cntlzd r9,r4 bf4: 79 29 d1 82 rldicl r9,r9,58,6 bf8: 0b 09 00 00 tdnei r9,0 bfc: 2c 24 00 00 cmpdi r4,0 c00: 41 82 00 0c beq c0c <.test9w+0x1c> c04: 7c 63 23 92 divdu r3,r3,r4 c08: 4e 80 00 20 blr c0c: 38 60 00 00 li r3,0 c10: 4e 80 00 20 blr After the patch: 000003a8 <test>: 3a8: 81 23 00 84 lwz r9,132(r3) 3ac: 71 29 40 00 andi. r9,r9,16384 3b0: 40 82 00 0c bne 3bc <test+0x14> 3b4: 80 63 00 0c lwz r3,12(r3) 3b8: 4e 80 00 20 blr 3bc: 0f e0 00 00 twui r0,0 0000000000000c50 <.test9w>: c50: 7c 89 00 74 cntlzd r9,r4 c54: 79 29 d1 82 rldicl r9,r9,58,6 c58: 0b 09 00 00 tdnei r9,0 c5c: 7c 63 23 92 divdu r3,r3,r4 c60: 4e 80 00 20 blr c70: 38 60 00 00 li r3,0 c74: 4e 80 00 20 blr In the first exemple, we see GCC doesn't need to duplicate what happens after the trap. In the second exemple, we see that GCC doesn't need to emit a test and a branch in the likely path in addition to the trap. We've got some WARN_ON() in .softirqentry.text section so it needs to be added in the OTHER_TEXT_SECTIONS in modpost.c Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/389962b1b702e3c78d169e59bcfac56282889173.1618331882.git.christophe.leroy@csgroup.eu
2021-04-14 00:38:10 +08:00
EMIT_WARN_ENTRY 1b,__FILE__,__LINE__,BUGFLAG_WARNING
#endif
powerpc/64: Change soft_enabled from flag to bitmask "paca->soft_enabled" is used as a flag to mask some of interrupts. Currently supported flags values and their details: soft_enabled MSR[EE] 0 0 Disabled (PMI and HMI not masked) 1 1 Enabled "paca->soft_enabled" is initialized to 1 to make the interripts as enabled. arch_local_irq_disable() will toggle the value when interrupts needs to disbled. At this point, the interrupts are not actually disabled, instead, interrupt vector has code to check for the flag and mask it when it occurs. By "mask it", it update interrupt paca->irq_happened and return. arch_local_irq_restore() is called to re-enable interrupts, which checks and replays interrupts if any occured. Now, as mentioned, current logic doesnot mask "performance monitoring interrupts" and PMIs are implemented as NMI. But this patchset depends on local_irq_* for a successful local_* update. Meaning, mask all possible interrupts during local_* update and replay them after the update. So the idea here is to reserve the "paca->soft_enabled" logic. New values and details: soft_enabled MSR[EE] 1 0 Disabled (PMI and HMI not masked) 0 1 Enabled Reason for the this change is to create foundation for a third mask value "0x2" for "soft_enabled" to add support to mask PMIs. When ->soft_enabled is set to a value "3", PMI interrupts are mask and when set to a value of "1", PMI are not mask. With this patch also extends soft_enabled as interrupt disable mask. Current flags are renamed from IRQ_[EN?DIS}ABLED to IRQS_ENABLED and IRQS_DISABLED. Patch also fixes the ptrace call to force the user to see the softe value to be alway 1. Reason being, even though userspace has no business knowing about softe, it is part of pt_regs. Like-wise in signal context. Signed-off-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-12-20 11:55:49 +08:00
/* Hard-disable interrupts */
mfmsr r6
rldicl r7,r6,48,1
rotldi r7,r7,16
mtmsrd r7,1
/* Unfortunately, the stack pointer and the MSR are also clobbered,
* so they are saved in the PACA which allows us to restore
* our original state after RTAS returns.
*/
std r1,PACAR1(r13)
std r6,PACASAVEDMSR(r13)
/* Setup our real return addr */
LOAD_REG_ADDR(r4,rtas_return_loc)
clrldi r4,r4,2 /* convert to realmode address */
mtlr r4
li r0,0
ori r0,r0,MSR_EE|MSR_SE|MSR_BE|MSR_RI
andc r0,r6,r0
li r9,1
rldicr r9,r9,MSR_SF_LG,(63-MSR_SF_LG)
ori r9,r9,MSR_IR|MSR_DR|MSR_FE0|MSR_FE1|MSR_FP|MSR_RI|MSR_LE
andc r6,r0,r9
__enter_rtas:
sync /* disable interrupts so SRR0/1 */
mtmsrd r0 /* don't get trashed */
LOAD_REG_ADDR(r4, rtas)
ld r5,RTASENTRY(r4) /* get the rtas->entry value */
ld r4,RTASBASE(r4) /* get the rtas->base value */
mtspr SPRN_SRR0,r5
mtspr SPRN_SRR1,r6
RFI_TO_KERNEL
b . /* prevent speculative execution */
rtas_return_loc:
FIXUP_ENDIAN
/*
* Clear RI and set SF before anything.
*/
mfmsr r6
li r0,MSR_RI
andc r6,r6,r0
sldi r0,r0,(MSR_SF_LG - MSR_RI_LG)
or r6,r6,r0
sync
mtmsrd r6
/* relocation is off at this point */
GET_PACA(r4)
clrldi r4,r4,2 /* convert to realmode address */
bcl 20,31,$+4
0: mflr r3
ld r3,(1f-0b)(r3) /* get &rtas_restore_regs */
ld r1,PACAR1(r4) /* Restore our SP */
ld r4,PACASAVEDMSR(r4) /* Restore our MSR */
mtspr SPRN_SRR0,r3
mtspr SPRN_SRR1,r4
RFI_TO_KERNEL
b . /* prevent speculative execution */
_ASM_NOKPROBE_SYMBOL(__enter_rtas)
_ASM_NOKPROBE_SYMBOL(rtas_return_loc)
.align 3
1: .8byte rtas_restore_regs
rtas_restore_regs:
/* relocation is on at this point */
REST_GPR(2, r1) /* Restore the TOC */
REST_GPR(13, r1) /* Restore paca */
REST_NVGPRS(r1) /* Restore the non-volatiles */
GET_PACA(r13)
ld r4,_CCR(r1)
mtcr r4
ld r5,_CTR(r1)
mtctr r5
ld r6,_XER(r1)
mtspr SPRN_XER,r6
ld r7,_DAR(r1)
mtdar r7
ld r8,_DSISR(r1)
mtdsisr r8
addi r1,r1,SWITCH_FRAME_SIZE /* Unstack our frame */
ld r0,16(r1) /* get return address */
mtlr r0
blr /* return to caller */
#endif /* CONFIG_PPC_RTAS */
_GLOBAL(enter_prom)
mflr r0
std r0,16(r1)
stdu r1,-SWITCH_FRAME_SIZE(r1) /* Save SP and create stack space */
/* Because PROM is running in 32b mode, it clobbers the high order half
* of all registers that it saves. We therefore save those registers
* PROM might touch to the stack. (r0, r3-r13 are caller saved)
*/
SAVE_GPR(2, r1)
SAVE_GPR(13, r1)
SAVE_NVGPRS(r1)
mfcr r10
mfmsr r11
std r10,_CCR(r1)
std r11,_MSR(r1)
/* Put PROM address in SRR0 */
mtsrr0 r4
/* Setup our trampoline return addr in LR */
bcl 20,31,$+4
0: mflr r4
addi r4,r4,(1f - 0b)
mtlr r4
/* Prepare a 32-bit mode big endian MSR
*/
#ifdef CONFIG_PPC_BOOK3E
rlwinm r11,r11,0,1,31
mtsrr1 r11
rfi
#else /* CONFIG_PPC_BOOK3E */
LOAD_REG_IMMEDIATE(r12, MSR_SF | MSR_LE)
andc r11,r11,r12
mtsrr1 r11
RFI_TO_KERNEL
#endif /* CONFIG_PPC_BOOK3E */
1: /* Return from OF */
FIXUP_ENDIAN
/* Just make sure that r1 top 32 bits didn't get
* corrupt by OF
*/
rldicl r1,r1,0,32
/* Restore the MSR (back to 64 bits) */
ld r0,_MSR(r1)
MTMSRD(r0)
isync
/* Restore other registers */
REST_GPR(2, r1)
REST_GPR(13, r1)
REST_NVGPRS(r1)
ld r4,_CCR(r1)
mtcr r4
addi r1,r1,SWITCH_FRAME_SIZE
ld r0,16(r1)
mtlr r0
blr