mirror of
https://github.com/torvalds/linux.git
synced 2024-11-18 01:51:53 +00:00
9600f261ac
This allows more code to be moved out of unrelocated regions. The system call KVMTEST is changed to be open-coded and remain in the tramp area to avoid having to move it to entry_64.S. The custom nature of the system call entry code means the hcall case can be made more streamlined than regular interrupt handlers. mpe: Incorporate fix from Nick: Moving KVM test to the common entry code missed the case of HMI and MCE, which do not do __GEN_COMMON_ENTRY (because they don't want to switch to virt mode). This means a MCE or HMI exception that is taken while KVM is running a guest context will not be switched out of that context, and KVM won't be notified. Found by running sigfuz in guest with patched host on POWER9 DD2.3, which causes some TM related HMI interrupts (which are expected and supposed to be handled by KVM). This fix adds a __GEN_REALMODE_COMMON_ENTRY for those handlers to add the KVM test. This makes them look a little more like other handlers that all use __GEN_COMMON_ENTRY. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200225173541.1549955-13-npiggin@gmail.com
410 lines
9.9 KiB
ArmAsm
410 lines
9.9 KiB
ArmAsm
/* SPDX-License-Identifier: GPL-2.0-only */
|
|
/*
|
|
*
|
|
* Copyright SUSE Linux Products GmbH 2010
|
|
*
|
|
* Authors: Alexander Graf <agraf@suse.de>
|
|
*/
|
|
|
|
/* Real mode helpers */
|
|
|
|
#include <asm/asm-compat.h>
|
|
#include <asm/feature-fixups.h>
|
|
|
|
#if defined(CONFIG_PPC_BOOK3S_64)
|
|
|
|
#define GET_SHADOW_VCPU(reg) \
|
|
mr reg, r13
|
|
|
|
#elif defined(CONFIG_PPC_BOOK3S_32)
|
|
|
|
#define GET_SHADOW_VCPU(reg) \
|
|
tophys(reg, r2); \
|
|
lwz reg, (THREAD + THREAD_KVM_SVCPU)(reg); \
|
|
tophys(reg, reg)
|
|
|
|
#endif
|
|
|
|
/* Disable for nested KVM */
|
|
#define USE_QUICK_LAST_INST
|
|
|
|
|
|
/* Get helper functions for subarch specific functionality */
|
|
|
|
#if defined(CONFIG_PPC_BOOK3S_64)
|
|
#include "book3s_64_slb.S"
|
|
#elif defined(CONFIG_PPC_BOOK3S_32)
|
|
#include "book3s_32_sr.S"
|
|
#endif
|
|
|
|
/******************************************************************************
|
|
* *
|
|
* Entry code *
|
|
* *
|
|
*****************************************************************************/
|
|
|
|
.global kvmppc_handler_trampoline_enter
|
|
kvmppc_handler_trampoline_enter:
|
|
|
|
/* Required state:
|
|
*
|
|
* MSR = ~IR|DR
|
|
* R1 = host R1
|
|
* R2 = host R2
|
|
* R4 = guest shadow MSR
|
|
* R5 = normal host MSR
|
|
* R6 = current host MSR (EE, IR, DR off)
|
|
* LR = highmem guest exit code
|
|
* all other volatile GPRS = free
|
|
* SVCPU[CR] = guest CR
|
|
* SVCPU[XER] = guest XER
|
|
* SVCPU[CTR] = guest CTR
|
|
* SVCPU[LR] = guest LR
|
|
*/
|
|
|
|
/* r3 = shadow vcpu */
|
|
GET_SHADOW_VCPU(r3)
|
|
|
|
/* Save guest exit handler address and MSR */
|
|
mflr r0
|
|
PPC_STL r0, HSTATE_VMHANDLER(r3)
|
|
PPC_STL r5, HSTATE_HOST_MSR(r3)
|
|
|
|
/* Save R1/R2 in the PACA (64-bit) or shadow_vcpu (32-bit) */
|
|
PPC_STL r1, HSTATE_HOST_R1(r3)
|
|
PPC_STL r2, HSTATE_HOST_R2(r3)
|
|
|
|
/* Activate guest mode, so faults get handled by KVM */
|
|
li r11, KVM_GUEST_MODE_GUEST
|
|
stb r11, HSTATE_IN_GUEST(r3)
|
|
|
|
/* Switch to guest segment. This is subarch specific. */
|
|
LOAD_GUEST_SEGMENTS
|
|
|
|
#ifdef CONFIG_PPC_BOOK3S_64
|
|
BEGIN_FTR_SECTION
|
|
/* Save host FSCR */
|
|
mfspr r8, SPRN_FSCR
|
|
std r8, HSTATE_HOST_FSCR(r13)
|
|
/* Set FSCR during guest execution */
|
|
ld r9, SVCPU_SHADOW_FSCR(r13)
|
|
mtspr SPRN_FSCR, r9
|
|
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
|
|
|
|
/* Some guests may need to have dcbz set to 32 byte length.
|
|
*
|
|
* Usually we ensure that by patching the guest's instructions
|
|
* to trap on dcbz and emulate it in the hypervisor.
|
|
*
|
|
* If we can, we should tell the CPU to use 32 byte dcbz though,
|
|
* because that's a lot faster.
|
|
*/
|
|
lbz r0, HSTATE_RESTORE_HID5(r3)
|
|
cmpwi r0, 0
|
|
beq no_dcbz32_on
|
|
|
|
mfspr r0,SPRN_HID5
|
|
ori r0, r0, 0x80 /* XXX HID5_dcbz32 = 0x80 */
|
|
mtspr SPRN_HID5,r0
|
|
no_dcbz32_on:
|
|
|
|
#endif /* CONFIG_PPC_BOOK3S_64 */
|
|
|
|
/* Enter guest */
|
|
|
|
PPC_LL r8, SVCPU_CTR(r3)
|
|
PPC_LL r9, SVCPU_LR(r3)
|
|
lwz r10, SVCPU_CR(r3)
|
|
PPC_LL r11, SVCPU_XER(r3)
|
|
|
|
mtctr r8
|
|
mtlr r9
|
|
mtcr r10
|
|
mtxer r11
|
|
|
|
/* Move SRR0 and SRR1 into the respective regs */
|
|
PPC_LL r9, SVCPU_PC(r3)
|
|
/* First clear RI in our current MSR value */
|
|
li r0, MSR_RI
|
|
andc r6, r6, r0
|
|
|
|
PPC_LL r0, SVCPU_R0(r3)
|
|
PPC_LL r1, SVCPU_R1(r3)
|
|
PPC_LL r2, SVCPU_R2(r3)
|
|
PPC_LL r5, SVCPU_R5(r3)
|
|
PPC_LL r7, SVCPU_R7(r3)
|
|
PPC_LL r8, SVCPU_R8(r3)
|
|
PPC_LL r10, SVCPU_R10(r3)
|
|
PPC_LL r11, SVCPU_R11(r3)
|
|
PPC_LL r12, SVCPU_R12(r3)
|
|
PPC_LL r13, SVCPU_R13(r3)
|
|
|
|
MTMSR_EERI(r6)
|
|
mtsrr0 r9
|
|
mtsrr1 r4
|
|
|
|
PPC_LL r4, SVCPU_R4(r3)
|
|
PPC_LL r6, SVCPU_R6(r3)
|
|
PPC_LL r9, SVCPU_R9(r3)
|
|
PPC_LL r3, (SVCPU_R3)(r3)
|
|
|
|
RFI_TO_GUEST
|
|
kvmppc_handler_trampoline_enter_end:
|
|
|
|
|
|
|
|
/******************************************************************************
|
|
* *
|
|
* Exit code *
|
|
* *
|
|
*****************************************************************************/
|
|
|
|
.global kvmppc_interrupt_pr
|
|
kvmppc_interrupt_pr:
|
|
/* 64-bit entry. Register usage at this point:
|
|
*
|
|
* SPRG_SCRATCH0 = guest R13
|
|
* R12 = (guest CR << 32) | exit handler id
|
|
* R13 = PACA
|
|
* HSTATE.SCRATCH0 = guest R12
|
|
*/
|
|
#ifdef CONFIG_PPC64
|
|
/* Match 32-bit entry */
|
|
rotldi r12, r12, 32 /* Flip R12 halves for stw */
|
|
stw r12, HSTATE_SCRATCH1(r13) /* CR is now in the low half */
|
|
srdi r12, r12, 32 /* shift trap into low half */
|
|
#endif
|
|
|
|
.global kvmppc_handler_trampoline_exit
|
|
kvmppc_handler_trampoline_exit:
|
|
/* Register usage at this point:
|
|
*
|
|
* SPRG_SCRATCH0 = guest R13
|
|
* R12 = exit handler id
|
|
* R13 = shadow vcpu (32-bit) or PACA (64-bit)
|
|
* HSTATE.SCRATCH0 = guest R12
|
|
* HSTATE.SCRATCH1 = guest CR
|
|
*/
|
|
|
|
/* Save registers */
|
|
|
|
PPC_STL r0, SVCPU_R0(r13)
|
|
PPC_STL r1, SVCPU_R1(r13)
|
|
PPC_STL r2, SVCPU_R2(r13)
|
|
PPC_STL r3, SVCPU_R3(r13)
|
|
PPC_STL r4, SVCPU_R4(r13)
|
|
PPC_STL r5, SVCPU_R5(r13)
|
|
PPC_STL r6, SVCPU_R6(r13)
|
|
PPC_STL r7, SVCPU_R7(r13)
|
|
PPC_STL r8, SVCPU_R8(r13)
|
|
PPC_STL r9, SVCPU_R9(r13)
|
|
PPC_STL r10, SVCPU_R10(r13)
|
|
PPC_STL r11, SVCPU_R11(r13)
|
|
|
|
/* Restore R1/R2 so we can handle faults */
|
|
PPC_LL r1, HSTATE_HOST_R1(r13)
|
|
PPC_LL r2, HSTATE_HOST_R2(r13)
|
|
|
|
/* Save guest PC and MSR */
|
|
#ifdef CONFIG_PPC64
|
|
BEGIN_FTR_SECTION
|
|
andi. r0, r12, 0x2
|
|
cmpwi cr1, r0, 0
|
|
beq 1f
|
|
mfspr r3,SPRN_HSRR0
|
|
mfspr r4,SPRN_HSRR1
|
|
andi. r12,r12,0x3ffd
|
|
b 2f
|
|
END_FTR_SECTION_IFSET(CPU_FTR_HVMODE)
|
|
#endif
|
|
1: mfsrr0 r3
|
|
mfsrr1 r4
|
|
2:
|
|
PPC_STL r3, SVCPU_PC(r13)
|
|
PPC_STL r4, SVCPU_SHADOW_SRR1(r13)
|
|
|
|
/* Get scratch'ed off registers */
|
|
GET_SCRATCH0(r9)
|
|
PPC_LL r8, HSTATE_SCRATCH0(r13)
|
|
lwz r7, HSTATE_SCRATCH1(r13)
|
|
|
|
PPC_STL r9, SVCPU_R13(r13)
|
|
PPC_STL r8, SVCPU_R12(r13)
|
|
stw r7, SVCPU_CR(r13)
|
|
|
|
/* Save more register state */
|
|
|
|
mfxer r5
|
|
mfdar r6
|
|
mfdsisr r7
|
|
mfctr r8
|
|
mflr r9
|
|
|
|
PPC_STL r5, SVCPU_XER(r13)
|
|
PPC_STL r6, SVCPU_FAULT_DAR(r13)
|
|
stw r7, SVCPU_FAULT_DSISR(r13)
|
|
PPC_STL r8, SVCPU_CTR(r13)
|
|
PPC_STL r9, SVCPU_LR(r13)
|
|
|
|
/*
|
|
* In order for us to easily get the last instruction,
|
|
* we got the #vmexit at, we exploit the fact that the
|
|
* virtual layout is still the same here, so we can just
|
|
* ld from the guest's PC address
|
|
*/
|
|
|
|
/* We only load the last instruction when it's safe */
|
|
cmpwi r12, BOOK3S_INTERRUPT_DATA_STORAGE
|
|
beq ld_last_inst
|
|
cmpwi r12, BOOK3S_INTERRUPT_PROGRAM
|
|
beq ld_last_inst
|
|
cmpwi r12, BOOK3S_INTERRUPT_SYSCALL
|
|
beq ld_last_prev_inst
|
|
cmpwi r12, BOOK3S_INTERRUPT_ALIGNMENT
|
|
beq- ld_last_inst
|
|
#ifdef CONFIG_PPC64
|
|
BEGIN_FTR_SECTION
|
|
cmpwi r12, BOOK3S_INTERRUPT_H_EMUL_ASSIST
|
|
beq- ld_last_inst
|
|
END_FTR_SECTION_IFSET(CPU_FTR_HVMODE)
|
|
BEGIN_FTR_SECTION
|
|
cmpwi r12, BOOK3S_INTERRUPT_FAC_UNAVAIL
|
|
beq- ld_last_inst
|
|
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
|
|
#endif
|
|
|
|
b no_ld_last_inst
|
|
|
|
ld_last_prev_inst:
|
|
addi r3, r3, -4
|
|
|
|
ld_last_inst:
|
|
/* Save off the guest instruction we're at */
|
|
|
|
/* In case lwz faults */
|
|
li r0, KVM_INST_FETCH_FAILED
|
|
|
|
#ifdef USE_QUICK_LAST_INST
|
|
|
|
/* Set guest mode to 'jump over instruction' so if lwz faults
|
|
* we'll just continue at the next IP. */
|
|
li r9, KVM_GUEST_MODE_SKIP
|
|
stb r9, HSTATE_IN_GUEST(r13)
|
|
|
|
/* 1) enable paging for data */
|
|
mfmsr r9
|
|
ori r11, r9, MSR_DR /* Enable paging for data */
|
|
mtmsr r11
|
|
sync
|
|
/* 2) fetch the instruction */
|
|
lwz r0, 0(r3)
|
|
/* 3) disable paging again */
|
|
mtmsr r9
|
|
sync
|
|
|
|
#endif
|
|
stw r0, SVCPU_LAST_INST(r13)
|
|
|
|
no_ld_last_inst:
|
|
|
|
/* Unset guest mode */
|
|
li r9, KVM_GUEST_MODE_NONE
|
|
stb r9, HSTATE_IN_GUEST(r13)
|
|
|
|
/* Switch back to host MMU */
|
|
LOAD_HOST_SEGMENTS
|
|
|
|
#ifdef CONFIG_PPC_BOOK3S_64
|
|
|
|
lbz r5, HSTATE_RESTORE_HID5(r13)
|
|
cmpwi r5, 0
|
|
beq no_dcbz32_off
|
|
|
|
li r4, 0
|
|
mfspr r5,SPRN_HID5
|
|
rldimi r5,r4,6,56
|
|
mtspr SPRN_HID5,r5
|
|
|
|
no_dcbz32_off:
|
|
|
|
BEGIN_FTR_SECTION
|
|
/* Save guest FSCR on a FAC_UNAVAIL interrupt */
|
|
cmpwi r12, BOOK3S_INTERRUPT_FAC_UNAVAIL
|
|
bne+ no_fscr_save
|
|
mfspr r7, SPRN_FSCR
|
|
std r7, SVCPU_SHADOW_FSCR(r13)
|
|
no_fscr_save:
|
|
/* Restore host FSCR */
|
|
ld r8, HSTATE_HOST_FSCR(r13)
|
|
mtspr SPRN_FSCR, r8
|
|
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
|
|
|
|
#endif /* CONFIG_PPC_BOOK3S_64 */
|
|
|
|
/*
|
|
* For some interrupts, we need to call the real Linux
|
|
* handler, so it can do work for us. This has to happen
|
|
* as if the interrupt arrived from the kernel though,
|
|
* so let's fake it here where most state is restored.
|
|
*
|
|
* Having set up SRR0/1 with the address where we want
|
|
* to continue with relocation on (potentially in module
|
|
* space), we either just go straight there with rfi[d],
|
|
* or we jump to an interrupt handler if there is an
|
|
* interrupt to be handled first. In the latter case,
|
|
* the rfi[d] at the end of the interrupt handler will
|
|
* get us back to where we want to continue.
|
|
*/
|
|
|
|
/* Register usage at this point:
|
|
*
|
|
* R1 = host R1
|
|
* R2 = host R2
|
|
* R10 = raw exit handler id
|
|
* R12 = exit handler id
|
|
* R13 = shadow vcpu (32-bit) or PACA (64-bit)
|
|
* SVCPU.* = guest *
|
|
*
|
|
*/
|
|
|
|
PPC_LL r6, HSTATE_HOST_MSR(r13)
|
|
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
|
|
/*
|
|
* We don't want to change MSR[TS] bits via rfi here.
|
|
* The actual TM handling logic will be in host with
|
|
* recovered DR/IR bits after HSTATE_VMHANDLER.
|
|
* And MSR_TM can be enabled in HOST_MSR so rfid may
|
|
* not suppress this change and can lead to exception.
|
|
* Manually set MSR to prevent TS state change here.
|
|
*/
|
|
mfmsr r7
|
|
rldicl r7, r7, 64 - MSR_TS_S_LG, 62
|
|
rldimi r6, r7, MSR_TS_S_LG, 63 - MSR_TS_T_LG
|
|
#endif
|
|
PPC_LL r8, HSTATE_VMHANDLER(r13)
|
|
|
|
#ifdef CONFIG_PPC64
|
|
BEGIN_FTR_SECTION
|
|
beq cr1, 1f
|
|
mtspr SPRN_HSRR1, r6
|
|
mtspr SPRN_HSRR0, r8
|
|
END_FTR_SECTION_IFSET(CPU_FTR_HVMODE)
|
|
#endif
|
|
1: /* Restore host msr -> SRR1 */
|
|
mtsrr1 r6
|
|
/* Load highmem handler address */
|
|
mtsrr0 r8
|
|
|
|
/* RFI into the highmem handler, or jump to interrupt handler */
|
|
cmpwi r12, BOOK3S_INTERRUPT_EXTERNAL
|
|
beqa BOOK3S_INTERRUPT_EXTERNAL
|
|
cmpwi r12, BOOK3S_INTERRUPT_DECREMENTER
|
|
beqa BOOK3S_INTERRUPT_DECREMENTER
|
|
cmpwi r12, BOOK3S_INTERRUPT_PERFMON
|
|
beqa BOOK3S_INTERRUPT_PERFMON
|
|
cmpwi r12, BOOK3S_INTERRUPT_DOORBELL
|
|
beqa BOOK3S_INTERRUPT_DOORBELL
|
|
|
|
RFI_TO_KERNEL
|
|
kvmppc_handler_trampoline_exit_end:
|