}
}
-#define arch_make_huge_pte arch_make_huge_pte
-static inline pte_t arch_make_huge_pte(pte_t entry, struct vm_area_struct *vma,
- struct page *page, int writable)
-{
- unsigned long page_shift;
-
- if (!cpu_has_feature(CPU_FTR_POWER9_DD1))
- return entry;
-
- page_shift = huge_page_shift(hstate_vma(vma));
- /*
- * We don't support 1G hugetlb pages yet.
- */
- VM_WARN_ON(page_shift == mmu_psize_defs[MMU_PAGE_1G].shift);
- if (page_shift == mmu_psize_defs[MMU_PAGE_2M].shift)
- return __pte(pte_val(entry) | R_PAGE_LARGE);
- else
- return entry;
-}
-
#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
static inline bool gigantic_page_supported(void)
{
{
if (full && radix_enabled()) {
/*
- * Let's skip the DD1 style pte update here. We know that
- * this is a full mm pte clear and hence can be sure there is
- * no parallel set_pte.
+ * We know that this is a full mm pte clear and
+ * hence can be sure there is no parallel set_pte.
*/
return radix__ptep_get_and_clear_full(mm, addr, ptep, full);
}
#include <asm/book3s/64/radix-4k.h>
#endif
-/*
- * For P9 DD1 only, we need to track whether the pte's huge.
- */
-#define R_PAGE_LARGE _RPAGE_RSV1
-
-
#ifndef __ASSEMBLY__
#include <asm/book3s/64/tlbflush-radix.h>
#include <asm/cpu_has_feature.h>
{
unsigned long old_pte;
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
-
- unsigned long new_pte;
-
- old_pte = __radix_pte_update(ptep, ~0ul, 0);
- /*
- * new value of pte
- */
- new_pte = (old_pte | set) & ~clr;
- radix__flush_tlb_pte_p9_dd1(old_pte, mm, addr);
- if (new_pte)
- __radix_pte_update(ptep, 0, new_pte);
- } else
- old_pte = __radix_pte_update(ptep, clr, set);
+ old_pte = __radix_pte_update(ptep, clr, set);
if (!huge)
assert_pte_locked(mm, addr);
static inline pmd_t radix__pmd_mkhuge(pmd_t pmd)
{
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- return __pmd(pmd_val(pmd) | _PAGE_PTE | R_PAGE_LARGE);
return __pmd(pmd_val(pmd) | _PAGE_PTE);
}
unsigned long rts_field;
/*
* We support 52 bits, hence:
- * DD1 52-28 = 24, 0b11000
- * Others 52-31 = 21, 0b10101
+ * bits 52 - 31 = 21, 0b10101
* RTS encoding details
* bits 0 - 3 of rts -> bits 6 - 8 unsigned long
* bits 4 - 5 of rts -> bits 62 - 63 of unsigned long
*/
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- rts_field = (0x3UL << 61);
- else {
- rts_field = (0x5UL << 5); /* 6 - 8 bits */
- rts_field |= (0x2UL << 61);
- }
+ rts_field = (0x5UL << 5); /* 6 - 8 bits */
+ rts_field |= (0x2UL << 61);
+
return rts_field;
}
extern void radix__flush_tlb_pwc(struct mmu_gather *tlb, unsigned long addr);
extern void radix__flush_tlb_collapsed_pmd(struct mm_struct *mm, unsigned long addr);
extern void radix__flush_tlb_all(void);
-extern void radix__flush_tlb_pte_p9_dd1(unsigned long old_pte, struct mm_struct *mm,
- unsigned long address);
extern void radix__flush_tlb_lpid_page(unsigned int lpid,
unsigned long addr,
#define CPU_FTR_DAWR LONG_ASM_CONST(0x0000008000000000)
#define CPU_FTR_DABRX LONG_ASM_CONST(0x0000010000000000)
#define CPU_FTR_PMAO_BUG LONG_ASM_CONST(0x0000020000000000)
-#define CPU_FTR_POWER9_DD1 LONG_ASM_CONST(0x0000040000000000)
#define CPU_FTR_POWER9_DD2_1 LONG_ASM_CONST(0x0000080000000000)
#define CPU_FTR_P9_TM_HV_ASSIST LONG_ASM_CONST(0x0000100000000000)
#define CPU_FTR_P9_TM_XER_SO_BUG LONG_ASM_CONST(0x0000200000000000)
CPU_FTR_DBELL | CPU_FTR_HAS_PPR | CPU_FTR_ARCH_207S | \
CPU_FTR_TM_COMP | CPU_FTR_ARCH_300 | CPU_FTR_PKEY | \
CPU_FTR_P9_TLBIE_BUG | CPU_FTR_P9_TIDR)
-#define CPU_FTRS_POWER9_DD1 ((CPU_FTRS_POWER9 | CPU_FTR_POWER9_DD1) & \
- (~CPU_FTR_SAO))
#define CPU_FTRS_POWER9_DD2_0 CPU_FTRS_POWER9
#define CPU_FTRS_POWER9_DD2_1 (CPU_FTRS_POWER9 | CPU_FTR_POWER9_DD2_1)
#define CPU_FTRS_POWER9_DD2_2 (CPU_FTRS_POWER9 | CPU_FTR_POWER9_DD2_1 | \
#define CPU_FTRS_POSSIBLE \
(CPU_FTRS_POWER7 | CPU_FTRS_POWER8E | CPU_FTRS_POWER8 | \
CPU_FTRS_POWER8_DD1 | CPU_FTR_ALTIVEC_COMP | CPU_FTR_VSX_COMP | \
- CPU_FTRS_POWER9 | CPU_FTRS_POWER9_DD1 | CPU_FTRS_POWER9_DD2_1 | \
- CPU_FTRS_POWER9_DD2_2)
+ CPU_FTRS_POWER9 | CPU_FTRS_POWER9_DD2_1 | CPU_FTRS_POWER9_DD2_2)
#else
#define CPU_FTRS_POSSIBLE \
(CPU_FTRS_PPC970 | CPU_FTRS_POWER5 | \
CPU_FTRS_POWER6 | CPU_FTRS_POWER7 | CPU_FTRS_POWER8E | \
CPU_FTRS_POWER8 | CPU_FTRS_POWER8_DD1 | CPU_FTRS_CELL | \
CPU_FTRS_PA6T | CPU_FTR_VSX_COMP | CPU_FTR_ALTIVEC_COMP | \
- CPU_FTRS_POWER9 | CPU_FTRS_POWER9_DD1 | CPU_FTRS_POWER9_DD2_1 | \
- CPU_FTRS_POWER9_DD2_2)
+ CPU_FTRS_POWER9 | CPU_FTRS_POWER9_DD2_1 | CPU_FTRS_POWER9_DD2_2)
#endif /* CONFIG_CPU_LITTLE_ENDIAN */
#endif
#else
#define CPU_FTRS_ALWAYS \
(CPU_FTRS_POSSIBLE & ~CPU_FTR_HVMODE & CPU_FTRS_POWER7 & \
CPU_FTRS_POWER8E & CPU_FTRS_POWER8 & CPU_FTRS_POWER8_DD1 & \
- CPU_FTRS_POWER9 & CPU_FTRS_POWER9_DD1 & CPU_FTRS_POWER9_DD2_1 & \
+ CPU_FTRS_POWER9 & CPU_FTRS_POWER9_DD2_1 & \
CPU_FTRS_DT_CPU_BASE)
#else
#define CPU_FTRS_ALWAYS \
CPU_FTRS_POWER6 & CPU_FTRS_POWER7 & CPU_FTRS_CELL & \
CPU_FTRS_PA6T & CPU_FTRS_POWER8 & CPU_FTRS_POWER8E & \
CPU_FTRS_POWER8_DD1 & ~CPU_FTR_HVMODE & CPU_FTRS_POSSIBLE & \
- CPU_FTRS_POWER9 & CPU_FTRS_POWER9_DD1 & CPU_FTRS_POWER9_DD2_1 & \
+ CPU_FTRS_POWER9 & CPU_FTRS_POWER9_DD2_1 & \
CPU_FTRS_DT_CPU_BASE)
#endif /* CONFIG_CPU_LITTLE_ENDIAN */
#endif
u8 subcore_sibling_mask;
/* Flag to request this thread not to stop */
atomic_t dont_stop;
- /*
- * Pointer to an array which contains pointer
- * to the sibling threads' paca.
- */
- struct paca_struct **thread_sibling_pacas;
/* The PSSCR value that the kernel requested before going to stop */
u64 requested_psscr;
OFFSET(PACA_THREAD_IDLE_STATE, paca_struct, thread_idle_state);
OFFSET(PACA_THREAD_MASK, paca_struct, thread_mask);
OFFSET(PACA_SUBCORE_SIBLING_MASK, paca_struct, subcore_sibling_mask);
- OFFSET(PACA_SIBLING_PACA_PTRS, paca_struct, thread_sibling_pacas);
OFFSET(PACA_REQ_PSSCR, paca_struct, requested_psscr);
OFFSET(PACA_DONT_STOP, paca_struct, dont_stop);
#define STOP_SPR(x, f) OFFSET(x, paca_struct, stop_sprs.f)
.machine_check_early = __machine_check_early_realmode_p8,
.platform = "power8",
},
- { /* Power9 DD1*/
- .pvr_mask = 0xffffff00,
- .pvr_value = 0x004e0100,
- .cpu_name = "POWER9 (raw)",
- .cpu_features = CPU_FTRS_POWER9_DD1,
- .cpu_user_features = COMMON_USER_POWER9,
- .cpu_user_features2 = COMMON_USER2_POWER9,
- .mmu_features = MMU_FTRS_POWER9,
- .icache_bsize = 128,
- .dcache_bsize = 128,
- .num_pmcs = 6,
- .pmc_type = PPC_PMC_IBM,
- .oprofile_cpu_type = "ppc64/power9",
- .oprofile_type = PPC_OPROFILE_INVALID,
- .cpu_setup = __setup_cpu_power9,
- .cpu_restore = __restore_cpu_power9,
- .machine_check_early = __machine_check_early_realmode_p9,
- .platform = "power9",
- },
{ /* Power9 DD2.0 */
.pvr_mask = 0xffffefff,
.pvr_value = 0x004e0200,
/*
* Not all quirks can be derived from the cpufeatures device tree.
*/
- if ((version & 0xffffff00) == 0x004e0100)
- cur_cpu_spec->cpu_features |= CPU_FTR_POWER9_DD1;
- else if ((version & 0xffffefff) == 0x004e0200)
+ if ((version & 0xffffefff) == 0x004e0200)
; /* DD2.0 has no feature flag */
else if ((version & 0xffffefff) == 0x004e0201)
cur_cpu_spec->cpu_features |= CPU_FTR_POWER9_DD2_1;
*
* This interrupt can wake directly from idle. If that is the case,
* the machine check is handled then the idle wakeup code is called
- * to restore state. In that case, the POWER9 DD1 idle PACA workaround
- * is not applied in the early machine check code, which will cause
- * bugs.
+ * to restore state.
*/
mr r11,r1 /* Save r1 */
lhz r10,PACA_IN_MCE(r13)
blr /* return 0 for wakeup cause / SRR1 value */
#endif
-/*
- * On waking up from stop 0,1,2 with ESL=1 on POWER9 DD1,
- * HSPRG0 will be set to the HSPRG0 value of one of the
- * threads in this core. Thus the value we have in r13
- * may not be this thread's paca pointer.
- *
- * Fortunately, the TIR remains invariant. Since this thread's
- * paca pointer is recorded in all its sibling's paca, we can
- * correctly recover this thread's paca pointer if we
- * know the index of this thread in the core.
- *
- * This index can be obtained from the TIR.
- *
- * i.e, thread's position in the core = TIR.
- * If this value is i, then this thread's paca is
- * paca->thread_sibling_pacas[i].
- */
-power9_dd1_recover_paca:
- mfspr r4, SPRN_TIR
- /*
- * Since each entry in thread_sibling_pacas is 8 bytes
- * we need to left-shift by 3 bits. Thus r4 = i * 8
- */
- sldi r4, r4, 3
- /* Get &paca->thread_sibling_pacas[0] in r5 */
- ld r5, PACA_SIBLING_PACA_PTRS(r13)
- /* Load paca->thread_sibling_pacas[i] into r13 */
- ldx r13, r4, r5
- SET_PACA(r13)
- /*
- * Indicate that we have lost NVGPR state
- * which needs to be restored from the stack.
- */
- li r3, 1
- stb r3,PACA_NAPSTATELOST(r13)
- blr
-
/*
* Called from machine check handler for powersave wakeups.
* Low level machine check processing has already been done. Now just
ld r2, PACATOC(r13)
BEGIN_FTR_SECTION
-BEGIN_FTR_SECTION_NESTED(70)
- bl power9_dd1_recover_paca
-END_FTR_SECTION_NESTED_IFSET(CPU_FTR_POWER9_DD1, 70)
bl pnv_restore_hyp_resource_arch300
FTR_SECTION_ELSE
bl pnv_restore_hyp_resource_arch207
LOAD_REG_ADDRBASE(r5,pnv_first_deep_stop_state)
ld r4,ADDROFF(pnv_first_deep_stop_state)(r5)
-BEGIN_FTR_SECTION_NESTED(71)
- /*
- * Assume that we are waking up from the state
- * same as the Requested Level (RL) in the PSSCR
- * which are Bits 60-63
- */
- ld r5,PACA_REQ_PSSCR(r13)
- rldicl r5,r5,0,60
-FTR_SECTION_ELSE_NESTED(71)
/*
* 0-3 bits correspond to Power-Saving Level Status
* which indicates the idle state we are waking up from
*/
mfspr r5, SPRN_PSSCR
rldicl r5,r5,4,60
-ALT_FTR_SECTION_END_NESTED_IFSET(CPU_FTR_POWER9_DD1, 71)
li r0, 0 /* clear requested_psscr to say we're awake */
std r0, PACA_REQ_PSSCR(r13)
cmpd cr4,r5,r4
* mappings. If the new process has the foreign real address
* mappings, we must issue a cp_abort to clear any state and
* prevent snooping, corruption or a covert channel.
- *
- * DD1 allows paste into normal system memory so we do an
- * unpaired copy, rather than cp_abort, to clear the buffer,
- * since cp_abort is quite expensive.
*/
- if (current_thread_info()->task->thread.used_vas) {
+ if (current_thread_info()->task->thread.used_vas)
asm volatile(PPC_CP_ABORT);
- } else if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- asm volatile(PPC_COPY(%0, %1)
- : : "r"(dummy_copy_buffer), "r"(0));
- }
}
#endif /* CONFIG_PPC_BOOK3S_64 */
bits = root & RPDS_MASK;
root = root & RPDB_MASK;
- /* P9 DD1 interprets RTS (radix tree size) differently */
offset = rts + 31;
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- offset -= 3;
/* current implementations only support 52-bit space */
if (offset != 52)
unsigned long clr, unsigned long set,
unsigned long addr, unsigned int shift)
{
- unsigned long old = 0;
-
- if (!(clr & _PAGE_PRESENT) && cpu_has_feature(CPU_FTR_POWER9_DD1) &&
- pte_present(*ptep)) {
- /* have to invalidate it first */
- old = __radix_pte_update(ptep, _PAGE_PRESENT, 0);
- kvmppc_radix_tlbie_page(kvm, addr, shift);
- set |= _PAGE_PRESENT;
- old &= _PAGE_PRESENT;
- }
- return __radix_pte_update(ptep, clr, set) | old;
+ return __radix_pte_update(ptep, clr, set);
}
void kvmppc_radix_set_pte_at(struct kvm *kvm, unsigned long addr,
r = set_vpa(vcpu, &vcpu->arch.dtl, addr, len);
break;
case KVM_REG_PPC_TB_OFFSET:
- /*
- * POWER9 DD1 has an erratum where writing TBU40 causes
- * the timebase to lose ticks. So we don't let the
- * timebase offset be changed on P9 DD1. (It is
- * initialized to zero.)
- */
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- break;
/* round up to multiple of 2^24 */
vcpu->arch.vcore->tb_offset =
ALIGN(set_reg_val(id, *val), 1UL << 24);
/*
* Set the default HFSCR for the guest from the host value.
* This value is only used on POWER9.
- * On POWER9 DD1, TM doesn't work, so we make sure to
- * prevent the guest from using it.
* On POWER9, we want to virtualize the doorbell facility, so we
* turn off the HFSCR bit, which causes those instructions to trap.
*/
mtspr SPRN_BESCR, r6
mtspr SPRN_PID, r7
mtspr SPRN_WORT, r8
-BEGIN_FTR_SECTION
- PPC_INVALIDATE_ERAT
-END_FTR_SECTION_IFSET(CPU_FTR_POWER9_DD1)
BEGIN_FTR_SECTION
/* POWER8-only registers */
ld r5, VCPU_TCSCR(r4)
ld r5, VCPU_KVM(r9)
lbz r0, KVM_RADIX(r5)
cmpwi cr2, r0, 0
- beq cr2, 4f
+ beq cr2, 2f
/*
* Radix: do eieio; tlbsync; ptesync sequence in case we
bdnz 1b
ptesync
-2: /* Flush the ERAT on radix P9 DD1 guest exit */
-BEGIN_FTR_SECTION
- PPC_INVALIDATE_ERAT
-END_FTR_SECTION_IFSET(CPU_FTR_POWER9_DD1)
-4:
+2:
#endif /* CONFIG_PPC_RADIX_MMU */
/*
mtspr SPRN_CIABR, r0
mtspr SPRN_DAWRX, r0
- /* Flush the ERAT on radix P9 DD1 guest exit */
-BEGIN_FTR_SECTION
- PPC_INVALIDATE_ERAT
-END_FTR_SECTION_IFSET(CPU_FTR_POWER9_DD1)
-
BEGIN_MMU_FTR_SECTION
b 4f
END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
*/
eieio();
- /*
- * DD1 bug workaround: If PIPR is less favored than CPPR
- * ignore the interrupt or we might incorrectly lose an IPB
- * bit.
- */
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- __be64 qw1 = __x_readq(__x_tima + TM_QW1_OS);
- u8 pipr = be64_to_cpu(qw1) & 0xff;
- if (pipr >= xc->hw_cppr)
- return;
- }
-
/* Perform the acknowledge OS to register cycle. */
ack = be16_to_cpu(__x_readw(__x_tima + TM_SPC_ACK_OS_REG));
/* If the XIVE supports the new "store EOI facility, use it */
if (xd->flags & XIVE_IRQ_FLAG_STORE_EOI)
__x_writeq(0, __x_eoi_page(xd) + XIVE_ESB_STORE_EOI);
- else if (hw_irq && xd->flags & XIVE_IRQ_FLAG_EOI_FW) {
+ else if (hw_irq && xd->flags & XIVE_IRQ_FLAG_EOI_FW)
opal_int_eoi(hw_irq);
+ else if (xd->flags & XIVE_IRQ_FLAG_LSI) {
+ /*
+ * For LSIs the HW EOI cycle is used rather than PQ bits,
+ * as they are automatically re-triggred in HW when still
+ * pending.
+ */
+ __x_readq(__x_eoi_page(xd) + XIVE_ESB_LOAD_EOI);
} else {
uint64_t eoi_val;
*
* This allows us to then do a re-trigger if Q was set
* rather than synthetizing an interrupt in software
- *
- * For LSIs, using the HW EOI cycle works around a problem
- * on P9 DD1 PHBs where the other ESB accesses don't work
- * properly.
*/
- if (xd->flags & XIVE_IRQ_FLAG_LSI)
- __x_readq(__x_eoi_page(xd) + XIVE_ESB_LOAD_EOI);
- else {
- eoi_val = GLUE(X_PFX,esb_load)(xd, XIVE_ESB_SET_PQ_00);
-
- /* Re-trigger if needed */
- if ((eoi_val & 1) && __x_trig_page(xd))
- __x_writeq(0, __x_trig_page(xd));
- }
+ eoi_val = GLUE(X_PFX,esb_load)(xd, XIVE_ESB_SET_PQ_00);
+
+ /* Re-trigger if needed */
+ if ((eoi_val & 1) && __x_trig_page(xd))
+ __x_writeq(0, __x_trig_page(xd));
}
}
}
#endif /* CONFIG_MEMORY_HOTPLUG */
-static void update_hid_for_hash(void)
-{
- unsigned long hid0;
- unsigned long rb = 3UL << PPC_BITLSHIFT(53); /* IS = 3 */
-
- asm volatile("ptesync": : :"memory");
- /* prs = 0, ric = 2, rs = 0, r = 1 is = 3 */
- asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
- : : "r"(rb), "i"(0), "i"(0), "i"(2), "r"(0) : "memory");
- asm volatile("eieio; tlbsync; ptesync; isync; slbia": : :"memory");
- trace_tlbie(0, 0, rb, 0, 2, 0, 0);
-
- /*
- * now switch the HID
- */
- hid0 = mfspr(SPRN_HID0);
- hid0 &= ~HID0_POWER9_RADIX;
- mtspr(SPRN_HID0, hid0);
- asm volatile("isync": : :"memory");
-
- /* Wait for it to happen */
- while ((mfspr(SPRN_HID0) & HID0_POWER9_RADIX))
- cpu_relax();
-}
-
static void __init hash_init_partition_table(phys_addr_t hash_table,
unsigned long htab_size)
{
htab_size = __ilog2(htab_size) - 18;
mmu_partition_table_set_entry(0, hash_table | htab_size, 0);
pr_info("Partition table %p\n", partition_tb);
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- update_hid_for_hash();
}
static void __init htab_initialize(void)
/* Initialize hash table for that CPU */
if (!firmware_has_feature(FW_FEATURE_LPAR)) {
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- update_hid_for_hash();
-
if (!cpu_has_feature(CPU_FTR_ARCH_300))
mtspr(SPRN_SDR1, _SDR1);
else
* firmware we only add hugetlb support for page sizes that can be
* supported by linux page table layout.
* For now we have
- * Radix: 2M
+ * Radix: 2M and 1G
* Hash: 16M and 16G
*/
if (radix_enabled()) {
- if (mmu_psize != MMU_PAGE_2M) {
- if (cpu_has_feature(CPU_FTR_POWER9_DD1) ||
- (mmu_psize != MMU_PAGE_1G))
- return -EINVAL;
- }
+ if (mmu_psize != MMU_PAGE_2M && mmu_psize != MMU_PAGE_1G)
+ return -EINVAL;
} else {
if (mmu_psize != MMU_PAGE_16M && mmu_psize != MMU_PAGE_16G)
return -EINVAL;
#ifdef CONFIG_PPC_RADIX_MMU
void radix__switch_mmu_context(struct mm_struct *prev, struct mm_struct *next)
{
-
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- isync();
- mtspr(SPRN_PID, next->context.id);
- isync();
- asm volatile(PPC_INVALIDATE_ERAT : : :"memory");
- } else {
- mtspr(SPRN_PID, next->context.id);
- isync();
- }
+ mtspr(SPRN_PID, next->context.id);
+ isync();
}
#endif
{
unsigned long start, end;
- /*
- * mark_rodata_ro() will mark itself as !writable at some point.
- * Due to DD1 workaround in radix__pte_update(), we'll end up with
- * an invalid pte and the system will crash quite severly.
- */
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- pr_warn("Warning: Unable to mark rodata read only on P9 DD1\n");
- return;
- }
-
start = (unsigned long)_stext;
end = (unsigned long)__init_begin;
return;
}
-static void update_hid_for_radix(void)
-{
- unsigned long hid0;
- unsigned long rb = 3UL << PPC_BITLSHIFT(53); /* IS = 3 */
-
- asm volatile("ptesync": : :"memory");
- /* prs = 0, ric = 2, rs = 0, r = 1 is = 3 */
- asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
- : : "r"(rb), "i"(1), "i"(0), "i"(2), "r"(0) : "memory");
- /* prs = 1, ric = 2, rs = 0, r = 1 is = 3 */
- asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
- : : "r"(rb), "i"(1), "i"(1), "i"(2), "r"(0) : "memory");
- asm volatile("eieio; tlbsync; ptesync; isync; slbia": : :"memory");
- trace_tlbie(0, 0, rb, 0, 2, 0, 1);
- trace_tlbie(0, 0, rb, 0, 2, 1, 1);
-
- /*
- * now switch the HID
- */
- hid0 = mfspr(SPRN_HID0);
- hid0 |= HID0_POWER9_RADIX;
- mtspr(SPRN_HID0, hid0);
- asm volatile("isync": : :"memory");
-
- /* Wait for it to happen */
- while (!(mfspr(SPRN_HID0) & HID0_POWER9_RADIX))
- cpu_relax();
-}
-
static void radix_init_amor(void)
{
/*
static void radix_init_iamr(void)
{
- unsigned long iamr;
-
- /*
- * The IAMR should set to 0 on DD1.
- */
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- iamr = 0;
- else
- iamr = (1ul << 62);
-
/*
* Radix always uses key0 of the IAMR to determine if an access is
* allowed. We set bit 0 (IBM bit 1) of key0, to prevent instruction
* fetch.
*/
- mtspr(SPRN_IAMR, iamr);
+ mtspr(SPRN_IAMR, (1ul << 62));
}
void __init radix__early_init_mmu(void)
if (!firmware_has_feature(FW_FEATURE_LPAR)) {
radix_init_native();
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- update_hid_for_radix();
lpcr = mfspr(SPRN_LPCR);
mtspr(SPRN_LPCR, lpcr | LPCR_UPRT | LPCR_HR);
radix_init_partition_table();
* update partition table control register and UPRT
*/
if (!firmware_has_feature(FW_FEATURE_LPAR)) {
-
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- update_hid_for_radix();
-
lpcr = mfspr(SPRN_LPCR);
mtspr(SPRN_LPCR, lpcr | LPCR_UPRT | LPCR_HR);
* To avoid NMMU hang while relaxing access, we need mark
* the pte invalid in between.
*/
- if (cpu_has_feature(CPU_FTR_POWER9_DD1) ||
- atomic_read(&mm->context.copros) > 0) {
+ if (atomic_read(&mm->context.copros) > 0) {
unsigned long old_pte, new_pte;
old_pte = __radix_pte_update(ptep, ~0, 0);
asm volatile("eieio; tlbsync; ptesync": : :"memory");
}
-void radix__flush_tlb_pte_p9_dd1(unsigned long old_pte, struct mm_struct *mm,
- unsigned long address)
-{
- /*
- * We track page size in pte only for DD1, So we can
- * call this only on DD1.
- */
- if (!cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- VM_WARN_ON(1);
- return;
- }
-
- if (old_pte & R_PAGE_LARGE)
- radix__flush_tlb_page_psize(mm, address, MMU_PAGE_2M);
- else
- radix__flush_tlb_page_psize(mm, address, mmu_virtual_psize);
-}
-
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
extern void radix_kvm_prefetch_workaround(struct mm_struct *mm)
{
static void power_pmu_sched_task(struct perf_event_context *ctx, bool sched_in) {}
static inline void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) {}
static void pmao_restore_workaround(bool ebb) { }
-static bool use_ic(u64 event)
-{
- return false;
-}
#endif /* CONFIG_PPC32 */
static bool regs_use_siar(struct pt_regs *regs)
mtspr(SPRN_PMC6, pmcs[5]);
}
-static bool use_ic(u64 event)
-{
- if (cpu_has_feature(CPU_FTR_POWER9_DD1) &&
- (event == 0x200f2 || event == 0x300f2))
- return true;
-
- return false;
-}
#endif /* CONFIG_PPC64 */
static void perf_event_interrupt(struct pt_regs *regs);
static void power_pmu_read(struct perf_event *event)
{
s64 val, delta, prev;
- struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
if (event->hw.state & PERF_HES_STOPPED)
return;
if (is_ebb_event(event)) {
val = read_pmc(event->hw.idx);
- if (use_ic(event->attr.config)) {
- val = mfspr(SPRN_IC);
- if (val > cpuhw->ic_init)
- val = val - cpuhw->ic_init;
- else
- val = val + (0 - cpuhw->ic_init);
- }
local64_set(&event->hw.prev_count, val);
return;
}
prev = local64_read(&event->hw.prev_count);
barrier();
val = read_pmc(event->hw.idx);
- if (use_ic(event->attr.config)) {
- val = mfspr(SPRN_IC);
- if (val > cpuhw->ic_init)
- val = val - cpuhw->ic_init;
- else
- val = val + (0 - cpuhw->ic_init);
- }
delta = check_and_compute_delta(prev, val);
if (!delta)
return;
event->attr.branch_sample_type);
}
- /*
- * Workaround for POWER9 DD1 to use the Instruction Counter
- * register value for instruction counting
- */
- if (use_ic(event->attr.config))
- cpuhw->ic_init = mfspr(SPRN_IC);
-
perf_pmu_enable(event->pmu);
local_irq_restore(flags);
return ret;
{
u64 valid_mask = EVENT_VALID_MASK;
- if (cpu_has_feature(CPU_FTR_ARCH_300) && !cpu_has_feature(CPU_FTR_POWER9_DD1))
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
valid_mask = p9_EVENT_VALID_MASK;
return !(event & ~valid_mask);
* Incase of Power9:
* Marked event: MMCRA[SDAR_MODE] will be set to 0b00 ('No Updates'),
* or if group already have any marked events.
- * Non-Marked events (for DD1):
- * MMCRA[SDAR_MODE] will be set to 0b01
* For rest
* MMCRA[SDAR_MODE] will be set from event code.
* If sdar_mode from event is zero, default to 0b01. Hardware
if (cpu_has_feature(CPU_FTR_ARCH_300)) {
if (is_event_marked(event) || (*mmcra & MMCRA_SAMPLE_ENABLE))
*mmcra &= MMCRA_SDAR_MODE_NO_UPDATES;
- else if (!cpu_has_feature(CPU_FTR_POWER9_DD1) && p9_SDAR_MODE(event))
+ else if (p9_SDAR_MODE(event))
*mmcra |= p9_SDAR_MODE(event) << MMCRA_SDAR_MODE_SHIFT;
else
*mmcra |= MMCRA_SDAR_MODE_DCACHE;
static u64 thresh_cmp_val(u64 value)
{
- if (cpu_has_feature(CPU_FTR_ARCH_300) && !cpu_has_feature(CPU_FTR_POWER9_DD1))
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
return value << p9_MMCRA_THR_CMP_SHIFT;
return value << MMCRA_THR_CMP_SHIFT;
static unsigned long combine_from_event(u64 event)
{
- if (cpu_has_feature(CPU_FTR_ARCH_300) && !cpu_has_feature(CPU_FTR_POWER9_DD1))
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
return p9_EVENT_COMBINE(event);
return EVENT_COMBINE(event);
static unsigned long combine_shift(unsigned long pmc)
{
- if (cpu_has_feature(CPU_FTR_ARCH_300) && !cpu_has_feature(CPU_FTR_POWER9_DD1))
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
return p9_MMCR1_COMBINE_SHIFT(pmc);
return MMCR1_COMBINE_SHIFT(pmc);
CNST_PMC_VAL(1) | CNST_PMC_VAL(2) | CNST_PMC_VAL(3) | \
CNST_PMC_VAL(4) | CNST_PMC_VAL(5) | CNST_PMC_VAL(6) | CNST_NC_VAL
-/*
- * Lets restrict use of PMC5 for instruction counting.
- */
-#define P9_DD1_TEST_ADDER (ISA207_TEST_ADDER | CNST_PMC_VAL(5))
-
/* Bits in MMCR1 for PowerISA v2.07 */
#define MMCR1_UNIT_SHIFT(pmc) (60 - (4 * ((pmc) - 1)))
#define MMCR1_COMBINE_SHIFT(pmc) (35 - ((pmc) - 1))
.attrs = power9_events_attr,
};
-static const struct attribute_group *power9_isa207_pmu_attr_groups[] = {
- &isa207_pmu_format_group,
- &power9_pmu_events_group,
- NULL,
-};
-
PMU_FORMAT_ATTR(event, "config:0-51");
PMU_FORMAT_ATTR(pmcxsel, "config:0-7");
PMU_FORMAT_ATTR(mark, "config:8");
NULL,
};
-static int power9_generic_events_dd1[] = {
- [PERF_COUNT_HW_CPU_CYCLES] = PM_CYC,
- [PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = PM_ICT_NOSLOT_CYC,
- [PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = PM_CMPLU_STALL,
- [PERF_COUNT_HW_INSTRUCTIONS] = PM_INST_DISP,
- [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = PM_BR_CMPL_ALT,
- [PERF_COUNT_HW_BRANCH_MISSES] = PM_BR_MPRED_CMPL,
- [PERF_COUNT_HW_CACHE_REFERENCES] = PM_LD_REF_L1,
- [PERF_COUNT_HW_CACHE_MISSES] = PM_LD_MISS_L1_FIN,
-};
-
static int power9_generic_events[] = {
[PERF_COUNT_HW_CPU_CYCLES] = PM_CYC,
[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = PM_ICT_NOSLOT_CYC,
#undef C
-static struct power_pmu power9_isa207_pmu = {
- .name = "POWER9",
- .n_counter = MAX_PMU_COUNTERS,
- .add_fields = ISA207_ADD_FIELDS,
- .test_adder = P9_DD1_TEST_ADDER,
- .compute_mmcr = isa207_compute_mmcr,
- .config_bhrb = power9_config_bhrb,
- .bhrb_filter_map = power9_bhrb_filter_map,
- .get_constraint = isa207_get_constraint,
- .get_alternatives = power9_get_alternatives,
- .disable_pmc = isa207_disable_pmc,
- .flags = PPMU_NO_SIAR | PPMU_ARCH_207S,
- .n_generic = ARRAY_SIZE(power9_generic_events_dd1),
- .generic_events = power9_generic_events_dd1,
- .cache_events = &power9_cache_events,
- .attr_groups = power9_isa207_pmu_attr_groups,
- .bhrb_nr = 32,
-};
-
static struct power_pmu power9_pmu = {
.name = "POWER9",
.n_counter = MAX_PMU_COUNTERS,
}
}
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- /*
- * Since PM_INST_CMPL may not provide right counts in all
- * sampling scenarios in power9 DD1, instead use PM_INST_DISP.
- */
- EVENT_VAR(PM_INST_CMPL, _g).id = PM_INST_DISP;
- /*
- * Power9 DD1 should use PM_BR_CMPL_ALT event code for
- * "branches" to provide correct counter value.
- */
- EVENT_VAR(PM_BR_CMPL, _g).id = PM_BR_CMPL_ALT;
- EVENT_VAR(PM_BR_CMPL, _c).id = PM_BR_CMPL_ALT;
- rc = register_power_pmu(&power9_isa207_pmu);
- } else {
- rc = register_power_pmu(&power9_pmu);
- }
-
+ rc = register_power_pmu(&power9_pmu);
if (rc)
return rc;
paca_ptrs[cpu]->core_idle_state_ptr = core_idle_state;
paca_ptrs[cpu]->thread_idle_state = PNV_THREAD_RUNNING;
paca_ptrs[cpu]->thread_mask = 1 << j;
- if (!cpu_has_feature(CPU_FTR_POWER9_DD1))
- continue;
- paca_ptrs[cpu]->thread_sibling_pacas =
- kmalloc_node(paca_ptr_array_size,
- GFP_KERNEL, node);
}
}
pnv_alloc_idle_core_states();
- /*
- * For each CPU, record its PACA address in each of it's
- * sibling thread's PACA at the slot corresponding to this
- * CPU's index in the core.
- */
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- int cpu;
-
- pr_info("powernv: idle: Saving PACA pointers of all CPUs in their thread sibling PACA\n");
- for_each_present_cpu(cpu) {
- int base_cpu = cpu_first_thread_sibling(cpu);
- int idx = cpu_thread_in_core(cpu);
- int i;
-
- for (i = 0; i < threads_per_core; i++) {
- int j = base_cpu + i;
-
- paca_ptrs[j]->thread_sibling_pacas[idx] =
- paca_ptrs[cpu];
- }
- }
- }
-
if (supported_cpuidle_states & OPAL_PM_NAP_ENABLED)
ppc_md.power_save = power7_idle;
ic_cause_ipi(cpu);
}
-static void pnv_p9_dd1_cause_ipi(int cpu)
-{
- int this_cpu = get_cpu();
-
- /*
- * POWER9 DD1 has a global addressed msgsnd, but for now we restrict
- * IPIs to same core, because it requires additional synchronization
- * for inter-core doorbells which we do not implement.
- */
- if (cpumask_test_cpu(cpu, cpu_sibling_mask(this_cpu)))
- doorbell_global_ipi(cpu);
- else
- ic_cause_ipi(cpu);
-
- put_cpu();
-}
-
static void __init pnv_smp_probe(void)
{
if (xive_enabled())
ic_cause_ipi = smp_ops->cause_ipi;
WARN_ON(!ic_cause_ipi);
- if (cpu_has_feature(CPU_FTR_ARCH_300)) {
- if (cpu_has_feature(CPU_FTR_POWER9_DD1))
- smp_ops->cause_ipi = pnv_p9_dd1_cause_ipi;
- else
- smp_ops->cause_ipi = doorbell_global_ipi;
- } else {
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
+ smp_ops->cause_ipi = doorbell_global_ipi;
+ else
smp_ops->cause_ipi = pnv_cause_ipi;
- }
}
}
* The FW told us to call it. This happens for some
* interrupt sources that need additional HW whacking
* beyond the ESB manipulation. For example LPC interrupts
- * on P9 DD1.0 need a latch to be clared in the LPC bridge
+ * on P9 DD1.0 needed a latch to be clared in the LPC bridge
* itself. The Firmware will take care of it.
*/
if (WARN_ON_ONCE(!xive_ops->eoi))
* This allows us to then do a re-trigger if Q was set
* rather than synthesizing an interrupt in software
*
- * For LSIs, using the HW EOI cycle works around a problem
- * on P9 DD1 PHBs where the other ESB accesses don't work
- * properly.
+ * For LSIs the HW EOI cycle is used rather than PQ bits,
+ * as they are automatically re-triggred in HW when still
+ * pending.
*/
if (xd->flags & XIVE_IRQ_FLAG_LSI)
xive_esb_read(xd, XIVE_ESB_LOAD_EOI);
DUMP(p, thread_idle_state, "%#-*x");
DUMP(p, thread_mask, "%#-*x");
DUMP(p, subcore_sibling_mask, "%#-*x");
- DUMP(p, thread_sibling_pacas, "%-*px");
DUMP(p, requested_psscr, "%#-*llx");
DUMP(p, stop_sprs.pid, "%#-*llx");
DUMP(p, stop_sprs.ldbar, "%#-*llx");
return false;
}
-static inline bool cxl_is_power9_dd1(void)
-{
- if ((pvr_version_is(PVR_POWER9)) &&
- cpu_has_feature(CPU_FTR_POWER9_DD1))
- return true;
- return false;
-}
-
ssize_t cxl_pci_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
loff_t off, size_t count);
rc = cxl_get_xsl9_dsnctl(dev, capp_unit_id, &cfg->dsnctl);
if (rc)
return rc;
- if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
- /* workaround for DD1 - nbwind = capiind */
- cfg->dsnctl |= ((u64)0x02 << (63-47));
- }
cfg->version = CXL_XSL_CONFIG_CURRENT_VERSION;
cfg->log_bar_size = CXL_CAPI_WINDOW_LOG_SIZE;
/* nMMU_ID Defaults to: b’000001001’*/
xsl_dsnctl |= ((u64)0x09 << (63-28));
- if (!(cxl_is_power9_dd1())) {
- /*
- * Used to identify CAPI packets which should be sorted into
- * the Non-Blocking queues by the PHB. This field should match
- * the PHB PBL_NBW_CMPM register
- * nbwind=0x03, bits [57:58], must include capi indicator.
- * Not supported on P9 DD1.
- */
- xsl_dsnctl |= (nbwind << (63-55));
+ /*
+ * Used to identify CAPI packets which should be sorted into
+ * the Non-Blocking queues by the PHB. This field should match
+ * the PHB PBL_NBW_CMPM register
+ * nbwind=0x03, bits [57:58], must include capi indicator.
+ * Not supported on P9 DD1.
+ */
+ xsl_dsnctl |= (nbwind << (63-55));
- /*
- * Upper 16b address bits of ASB_Notify messages sent to the
- * system. Need to match the PHB’s ASN Compare/Mask Register.
- * Not supported on P9 DD1.
- */
- xsl_dsnctl |= asnind;
- }
+ /*
+ * Upper 16b address bits of ASB_Notify messages sent to the
+ * system. Need to match the PHB’s ASN Compare/Mask Register.
+ * Not supported on P9 DD1.
+ */
+ xsl_dsnctl |= asnind;
*reg = xsl_dsnctl;
return 0;
/* Snoop machines */
cxl_p1_write(adapter, CXL_PSL9_APCDEDALLOC, 0x800F000200000000ULL);
- if (cxl_is_power9_dd1()) {
- /* Disabling deadlock counter CAR */
- cxl_p1_write(adapter, CXL_PSL9_GP_CT, 0x0020000000000001ULL);
- /* Enable NORST */
- cxl_p1_write(adapter, CXL_PSL9_DEBUG, 0x8000000000000000ULL);
- } else {
- /* Enable NORST and DD2 features */
- cxl_p1_write(adapter, CXL_PSL9_DEBUG, 0xC000000000000000ULL);
- }
+ /* Enable NORST and DD2 features */
+ cxl_p1_write(adapter, CXL_PSL9_DEBUG, 0xC000000000000000ULL);
/*
* Check if PSL has data-cache. We need to flush adapter datacache