ppc/pnv: Add chiptod and core timebase state machine models
cover ppc/pnv: Add chiptod and core timebase state machine models
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved to timebase_helper.c from misc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.
There is still a vast amount that is not modeled, but most of it related to error handling, injection, failover, etc that is very complicated and not required for normal operation.
Thanks,
Nick
Nicholas Piggin (4):
pnv/chiptod: Add POWER9/10 chiptod model
target/ppc: Tidy POWER book4 SPR registration target/ppc: add TFMR SPR implementation with read and write helpers target/ppc: Implement core timebase state machine and TFMR
hw/ppc/meson.build | 1 + hw/ppc/pnv.c | 38 +++ hw/ppc/pnv_chiptod.c | 488 +++++++++++++++++++++++++++++++++++ hw/ppc/pnv_xscom.c | 2 + hw/ppc/trace-events | 4 + include/hw/ppc/pnv_chip.h | 3 + include/hw/ppc/pnv_chiptod.h | 64 +++++ include/hw/ppc/pnv_core.h | 3 + include/hw/ppc/pnv_xscom.h | 9 + target/ppc/cpu.h | 40 +++ target/ppc/cpu_init.c | 92 ++++--- target/ppc/helper.h | 2 + target/ppc/spr_common.h | 2 + target/ppc/timebase_helper.c | 156 +++++++++++ target/ppc/translate.c | 10 + 15 files changed, 882 insertions(+), 32 deletions(-) create mode 100644 hw/ppc/pnv_chiptod.c create mode 100644 include/hw/ppc/pnv_chiptod.h
Signature · 1 line
2.40.1
PATCH 1/4 pnv/chiptod: Add POWER9/10 chiptod model
The chiptod is a pervasive facility which can keep TOD (time-of-day), synchronise it across multiple chips, and can move that TOD to or from the core timebase units.
This driver implements basic emulation of chiptod registers sufficient to successfully run the skiboot chiptod synchronisation procedure (with the following TFMR and timebase state-machine implementation).
The main way chiptod affects the rest of the system (relevant to the powernv model) is to interact with the timebase facility in the cores, influencing the timebase state machine and registers.
The way this chiptod driver implements that interaction is with two new flags in the CPUPPCState env, one is use for the core timebase to indicate it is ready to receive a TOD from chiptod, the other used by chiptod to indicate that it has sent TOD to the core timebase. The core timebase will be implemented in later changes.
--- hw/ppc/meson.build | 1 + hw/ppc/pnv.c | 38 +++ hw/ppc/pnv_chiptod.c | 488 +++++++++++++++++++++++++++++++++++ hw/ppc/pnv_xscom.c | 2 + hw/ppc/trace-events | 4 + include/hw/ppc/pnv_chip.h | 3 + include/hw/ppc/pnv_chiptod.h | 64 +++++ include/hw/ppc/pnv_core.h | 3 + include/hw/ppc/pnv_xscom.h | 9 + target/ppc/cpu.h | 6 + 10 files changed, 618 insertions(+) create mode 100644 hw/ppc/pnv_chiptod.c create mode 100644 include/hw/ppc/pnv_chiptod.h
hw/ppc/meson.build +1 -0
@@ -45,6 +45,7 @@ ppc_ss.add(when: 'CONFIG_POWERNV', if_true: files( 'pnv_core.c', 'pnv_lpc.c', 'pnv_psi.c', + 'pnv_chiptod.c', 'pnv_occ.c', 'pnv_sbe.c', 'pnv_bmc.c',
hw/ppc/pnv.c +38 -0
@@ -1414,6 +1414,8 @@ static void pnv_chip_power9_instance_init(Object *obj)object_initialize_child (obj,"lpc" , &chip9->lpc,TYPE_PNV9_LPC ); +object_initialize_child (obj,"chiptod" , &chip9->chiptod,TYPE_PNV9_CHIPTOD ); +object_initialize_child (obj,"occ" , &chip9->occ,TYPE_PNV9_OCC );object_initialize_child (obj,"sbe" , &chip9->sbe,TYPE_PNV9_SBE );
@@ -1558,6 +1560,15 @@ static void pnv_chip_power9_realize(DeviceState *dev, Error **errp) chip->dt_isa_nodename =g_strdup_printf ("/lpcm-opb@%" PRIx64"/lpc@0" , (uint64_t )PNV9_LPCM_BASE (chip)); +/* ChipTOD */ +object_property_set_link (OBJECT (&chip9->chiptod),"chip" ,OBJECT (chip), + &error_abort); +if (!qdev_realize (DEVICE (&chip9->chiptod),NULL , errp)) { +return ; + } +pnv_xscom_add_subregion (chip,PNV9_XSCOM_CHIPTOD_BASE , + &chip9->chiptod.xscom_regs); +/* Create the simplified OCC model */ if (!qdev_realize (DEVICE (&chip9->occ),NULL , errp)) {return ;
@@ -1644,6 +1655,7 @@ static void pnv_chip_power10_instance_init(Object *obj) "xive-fabric");object_initialize_child (obj,"psi" , &chip10->psi,TYPE_PNV10_PSI );object_initialize_child (obj,"lpc" , &chip10->lpc,TYPE_PNV10_LPC ); +object_initialize_child (obj,"chiptod" , &chip10->chiptod,TYPE_PNV10_CHIPTOD );object_initialize_child (obj,"occ" , &chip10->occ,TYPE_PNV10_OCC );object_initialize_child (obj,"sbe" , &chip10->sbe,TYPE_PNV10_SBE );object_initialize_child (obj,"homer" , &chip10->homer,TYPE_PNV10_HOMER );
@@ -1773,6 +1785,15 @@ static void pnv_chip_power10_realize(DeviceState *dev, Error **errp) chip->dt_isa_nodename =g_strdup_printf ("/lpcm-opb@%" PRIx64"/lpc@0" , (uint64_t )PNV10_LPCM_BASE (chip)); +/* ChipTOD */ +object_property_set_link (OBJECT (&chip10->chiptod),"chip" ,OBJECT (chip), + &error_abort); +if (!qdev_realize (DEVICE (&chip10->chiptod),NULL , errp)) { +return ; + } +pnv_xscom_add_subregion (chip,PNV10_XSCOM_CHIPTOD_BASE , + &chip10->chiptod.xscom_regs); +/* Create the simplified OCC model */ if (!qdev_realize (DEVICE (&chip10->occ),NULL , errp)) {return ;
@@ -1938,6 +1959,23 @@ static void pnv_chip_core_realize(PnvChip *chip, Error **errp) } } +PnvCore *pnv_get_vcpu_by_xscom_base (PnvChip *chip,uint32_t xscom_base) +{ +PnvChipClass *pcc =PNV_CHIP_GET_CLASS (chip); +int i; + +for (i =0 ; i < chip->nr_cores; i++) { +PnvCore *pc = chip->cores[i]; +CPUCore *cc =CPU_CORE (pc); +int core_hwid = cc->core_id; + +if (pcc->xscom_core_base (chip, core_hwid) == xscom_base) { +return pc; + } + } +return NULL ; +} +static void pnv_chip_realize (DeviceState *dev,Error **errp) {PnvChip *chip =PNV_CHIP (dev);
hw/ppc/pnv_chiptod.c new file +488 -0
488 lines not on this page · raw message
hw/ppc/pnv_xscom.c +2 -0
@@ -298,6 +298,8 @@ int pnv_dt_xscom(PnvChip *chip, void *fdt, int root_offset,_FDT ((fdt_setprop (fdt, xscom_offset,"scom-controller" ,NULL ,0 )));if (chip->chip_id ==0 ) {_FDT ((fdt_setprop (fdt, xscom_offset,"primary" ,NULL ,0 ))); + }else if (chip->chip_id ==1 ) { +_FDT ((fdt_setprop (fdt, xscom_offset,"secondary" ,NULL ,0 ))); } args.fdt = fdt;
hw/ppc/trace-events +4 -0
@@ -95,6 +95,10 @@ vof_write(uint32_t ih, unsigned cb, const char *msg) "ih=0x%x [%u] \"%s\"" vof_avail(uint64_t start, uint64_t end, uint64_t size) "0x%"PRIx64"..0x%"PRIx64" size=0x%"PRIx64 vof_claimed(uint64_t start, uint64_t end, uint64_t size) "0x%"PRIx64"..0x%"PRIx64" size=0x%"PRIx64 +# pnv_chiptod.c +pnv_chiptod_xscom_read(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64 +pnv_chiptod_xscom_write(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64 + # pnv_sbe.c pnv_sbe_xscom_ctrl_read(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64 pnv_sbe_xscom_ctrl_write(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64
include/hw/ppc/pnv_chip.h +3 -0
@@ -2,6 +2,7 @@#define PPC_PNV_CHIP_H #include "hw/pci-host/pnv_phb4.h" +#include "hw/ppc/pnv_chiptod.h" #include "hw/ppc/pnv_core.h" #include "hw/ppc/pnv_homer.h" #include "hw/ppc/pnv_lpc.h"
@@ -77,6 +78,7 @@ struct Pnv9Chip {PnvXive xive;Pnv9Psi psi;PnvLpcController lpc; +PnvChipTOD chiptod;PnvOCC occ;PnvSBE sbe;PnvHomer homer;
@@ -106,6 +108,7 @@ struct Pnv10Chip {PnvXive2 xive;Pnv9Psi psi;PnvLpcController lpc; +PnvChipTOD chiptod;PnvOCC occ;PnvSBE sbe;PnvHomer homer;
include/hw/ppc/pnv_chiptod.h new file +64 -0
64 lines not on this page · raw message
include/hw/ppc/pnv_core.h +3 -0
@@ -69,4 +69,7 @@ struct PnvQuad {uint32_t quad_id; MemoryRegion xscom_regs; }; + +PnvCore *pnv_get_vcpu_by_xscom_base (PnvChip *chip,uint32_t xscom_base); +#endif /* PPC_PNV_CORE_H */
include/hw/ppc/pnv_xscom.h +9 -0
@@ -64,6 +64,9 @@ struct PnvXScomInterfaceClass {#define PNV_XSCOM_PSIHB_BASE 0x2010900#define PNV_XSCOM_PSIHB_SIZE 0x20 +#define PNV_XSCOM_CHIPTOD_BASE 0x0040000 +#define PNV_XSCOM_CHIPTOD_SIZE 0x31 +#define PNV_XSCOM_OCC_BASE 0x0066000#define PNV_XSCOM_OCC_SIZE 0x6000
@@ -90,6 +93,9 @@ struct PnvXScomInterfaceClass { ((uint64_t)(((core) & 0x1C) + 0x40) << 22)#define PNV9_XSCOM_EQ_SIZE 0x100000 +#define PNV9_XSCOM_CHIPTOD_BASE PNV_XSCOM_CHIPTOD_BASE +#define PNV9_XSCOM_CHIPTOD_SIZE PNV_XSCOM_CHIPTOD_SIZE +#define PNV9_XSCOM_OCC_BASE PNV_XSCOM_OCC_BASE#define PNV9_XSCOM_OCC_SIZE 0x8000
@@ -138,6 +144,9 @@ struct PnvXScomInterfaceClass {#define PNV10_XSCOM_PSIHB_BASE 0x3011D00#define PNV10_XSCOM_PSIHB_SIZE 0x100 +#define PNV10_XSCOM_CHIPTOD_BASE PNV9_XSCOM_CHIPTOD_BASE +#define PNV10_XSCOM_CHIPTOD_SIZE PNV9_XSCOM_CHIPTOD_SIZE +#define PNV10_XSCOM_OCC_BASE PNV9_XSCOM_OCC_BASE#define PNV10_XSCOM_OCC_SIZE PNV9_XSCOM_OCC_SIZE
target/ppc/cpu.h +6 -0
@@ -1172,6 +1172,12 @@ struct CPUArchState {uint32_t tlb_need_flush;/* Delayed flush needed */ #define TLB_NEED_LOCAL_FLUSH 0x1#define TLB_NEED_GLOBAL_FLUSH 0x2 + +#if defined(TARGET_PPC64 ) +/* PowerNV chiptod / timebase facility state. */ +int tb_ready_for_tod;/* core TB ready to receive TOD from chiptod */ +int tod_sent_to_tb;/* chiptod sent TOD to the core TB */ +#endif #endif /* Other registers */
Signature · 1 line
2.40.1
PATCH 2/4 target/ppc: Tidy POWER book4 SPR registration
In reply to Nicholas PigginPOWER book4 (implementation-specific) SPRs are sometimes in their own functions, but in other cases are mixed with architected SPRs. Do some spring cleaning on these.
--- target/ppc/cpu_init.c | 92 ++++++++++++++++++++++++++++--------------- 1 file changed, 60 insertions(+), 32 deletions(-)
target/ppc/cpu_init.c +60 -32
@@ -5374,31 +5374,6 @@ static void register_book3s_ids_sprs(CPUPPCState *env) &spr_read_generic,SPR_NOACCESS , &spr_read_generic,NULL , 0x00000000); -spr_register_hv (env,SPR_HID0 ,"HID0" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_core_write_generic, -0x00000000 ); -spr_register_hv (env,SPR_TSCR ,"TSCR" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic32, -0x00000000 ); -spr_register_hv (env,SPR_HMER ,"HMER" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_hmer, -0x00000000 ); -spr_register_hv (env,SPR_HMEER ,"HMEER" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, -0x00000000 ); -spr_register_hv (env,SPR_TFMR ,"TFMR" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, -0x00000000 ); spr_register_hv(env, SPR_LPIDR, "LPIDR", SPR_NOACCESS, SPR_NOACCESS, SPR_NOACCESS, SPR_NOACCESS,
@@ -5454,11 +5429,6 @@ static void register_book3s_ids_sprs(CPUPPCState *env)SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic, 0x00000000); -spr_register_hv (env,SPR_LDBAR ,"LDBAR" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, -0x00000000 ); }static void register_rmor_sprs (CPUPPCState *env)
@@ -5665,14 +5635,65 @@ static void register_power8_ic_sprs(CPUPPCState *env)#endif } +/* SPRs specific to IBM POWER CPUs */ +static void register_power_common_book4_sprs (CPUPPCState *env) +{ +#if !defined(CONFIG_USER_ONLY ) +spr_register_hv (env,SPR_HID0 ,"HID0" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_core_write_generic, +0x00000000 ); +spr_register_hv (env,SPR_TSCR ,"TSCR" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic32, +0x00000000 ); +spr_register_hv (env,SPR_HMER ,"HMER" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_hmer, +0x00000000 ); +spr_register_hv (env,SPR_HMEER ,"HMEER" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +0x00000000 ); +spr_register_hv (env,SPR_TFMR ,"TFMR" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +0x00000000 ); +spr_register_hv (env,SPR_LDBAR ,"LDBAR" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +0x00000000 ); +#endif +} + +static void register_power9_book4_sprs (CPUPPCState *env) +{ +/* Add a number of P9 book4 registers */ +register_power_common_book4_sprs (env); +#if !defined(CONFIG_USER_ONLY ) +spr_register_kvm (env,SPR_WORT ,"WORT" , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +KVM_REG_PPC_WORT ,0 ); +#endif +} +static void register_power8_book4_sprs (CPUPPCState *env) {/* Add a number of P8 book4 registers */ +register_power_common_book4_sprs (env);#if !defined(CONFIG_USER_ONLY )spr_register_kvm (env,SPR_ACOP ,"ACOP" ,SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic,KVM_REG_PPC_ACOP ,0 ); +/* PID is only in BookE in ISA v2.07 */ spr_register_kvm(env, SPR_BOOKS_PID, "PID", SPR_NOACCESS, SPR_NOACCESS, &spr_read_generic, &spr_write_pidr,
@@ -5688,10 +5709,12 @@ static void register_power7_book4_sprs(CPUPPCState *env) {/* Add a number of P7 book4 registers */ #if !defined(CONFIG_USER_ONLY ) +register_power_common_book4_sprs (env);spr_register_kvm (env,SPR_ACOP ,"ACOP" ,SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic,KVM_REG_PPC_ACOP ,0 ); +/* PID is only in BookE in ISA v2.06 */ spr_register_kvm(env, SPR_BOOKS_PID, "PID", SPR_NOACCESS, SPR_NOACCESS, &spr_read_generic, &spr_write_generic32,
@@ -5725,6 +5748,11 @@ static void register_power9_mmu_sprs(CPUPPCState *env)SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic, 0x0000000000000000); +/* PID is part of the BookS ISA from v3.0 */ +spr_register_kvm (env,SPR_BOOKS_PID ,"PID" , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_pidr, +KVM_REG_PPC_PID ,0 );#endif }
@@ -6278,7 +6306,7 @@ static void init_proc_POWER9(CPUPPCState *env)register_power8_dpdes_sprs (env);register_vtb_sprs (env);register_power8_ic_sprs (env); -register_power8_book4_sprs (env); +register_power9_book4_sprs (env);register_power8_rpr_sprs (env);register_power9_mmu_sprs (env);
@@ -6471,7 +6499,7 @@ static void init_proc_POWER10(CPUPPCState *env)register_power8_dpdes_sprs (env);register_vtb_sprs (env);register_power8_ic_sprs (env); -register_power8_book4_sprs (env); +register_power9_book4_sprs (env);register_power8_rpr_sprs (env);register_power9_mmu_sprs (env);register_power10_hash_sprs (env);
Signature · 1 line
2.40.1
PATCH 4/4 target/ppc: Implement core timebase state machine and TFMR
In reply to Nicholas PigginThis implements the core timebase state machine, which is the core side of the time-of-day system in POWER processors. This facility is operated by control fields in the TFMR register, which also contains status fields.
The core timebase interacts with the chiptod hardware, primarily to receive TOD updates, to synchronise timebase with other cores. This model does not actually update TB values with TOD or updates received from the chiptod, as timebases are always synchronised. It does step through the states required to perform the update.
There are several asynchronous state transitions. These are modelled using using mfTFMR to drive state changes, because it is expected that firmware poll the register to wait for those states. This is good enough to test basic firmware behaviour without adding real timers. The values chosen are arbitrary.
--- target/ppc/cpu.h | 34 ++++++++ target/ppc/timebase_helper.c | 147 ++++++++++++++++++++++++++++++++++- 2 files changed, 179 insertions(+), 2 deletions(-)
target/ppc/cpu.h +34 -0
@@ -1177,6 +1177,13 @@ struct CPUArchState {/* PowerNV chiptod / timebase facility state. */ int tb_ready_for_tod;/* core TB ready to receive TOD from chiptod */ int tod_sent_to_tb;/* chiptod sent TOD to the core TB */ + +/* +* Timers for async events are simulated by mfTFAC because TFAC is to be +* polled for event. +*/ +int tb_state_timer; +int tb_sync_pulse_timer;#endif #endif
@@ -2527,6 +2534,33 @@ enum { HMER_XSCOM_STATUS_MASK = PPC_BITMASK(21, 23), }; +/* TFMR */ +enum { +TFMR_CONTROL_MASK =PPC_BITMASK (0 ,24 ), +TFMR_MASK_HMI =PPC_BIT (10 ), +TFMR_TB_ECLIPZ =PPC_BIT (14 ), +TFMR_LOAD_TOD_MOD =PPC_BIT (16 ), +TFMR_MOVE_CHIP_TOD_TO_TB =PPC_BIT (18 ), +TFMR_CLEAR_TB_ERRORS =PPC_BIT (24 ), +TFMR_STATUS_MASK =PPC_BITMASK (25 ,63 ), +TFMR_TBST_ENCODED =PPC_BITMASK (28 ,31 ),/* TBST = TB State */ +TFMR_TBST_LAST =PPC_BITMASK (32 ,35 ),/* Previous TBST */ +TFMR_TB_ENABLED =PPC_BIT (40 ), +TFMR_TB_VALID =PPC_BIT (41 ), +TFMR_TB_SYNC_OCCURED =PPC_BIT (42 ), +}; + +/* TFMR TBST */ +enum { +TBST_RESET =0x0 , +TBST_SEND_TOD_MOD =0x1 , +TBST_NOT_SET =0x2 , +TBST_SYNC_WAIT =0x6 , +TBST_GET_TOD =0x7 , +TBST_TB_RUNNING =0x8 , +TBST_TB_ERROR =0x9 , +}; +/*****************************************************************************/ #define is_isa300 (ctx) (!!(ctx->insns_flags2 & PPC2_ISA300))
target/ppc/timebase_helper.c +145 -2
159 lines not on this page · raw message
Signature · 1 line
2.40.1
PATCH 3/4 target/ppc: add TFMR SPR implementation with read and write helpers
In reply to Nicholas PigginTFMR is the Time Facility Management Register which is specific to POWER CPUs, and used for the purpose of timebase management (generally by firmware, not the OS).
This adds an initial simple TFMR register, which will form part of the core timebase facility model in the next patch.
--- target/ppc/cpu_init.c | 2 +- target/ppc/helper.h | 2 ++ target/ppc/spr_common.h | 2 ++ target/ppc/timebase_helper.c | 13 +++++++++++++ target/ppc/translate.c | 10 ++++++++++ 5 files changed, 28 insertions(+), 1 deletion(-)
target/ppc/cpu_init.c +1 -1
@@ -5662,7 +5662,7 @@ static void register_power_common_book4_sprs(CPUPPCState *env)spr_register_hv (env,SPR_TFMR ,"TFMR" ,SPR_NOACCESS ,SPR_NOACCESS ,SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, + &spr_read_tfmr, &spr_write_tfmr,0x00000000 ); spr_register_hv(env, SPR_LDBAR, "LDBAR", SPR_NOACCESS, SPR_NOACCESS,
target/ppc/helper.h +2 -0
@@ -746,6 +746,8 @@ DEF_HELPER_2(store_40x_dbcr0, void, env, tl) DEF_HELPER_2(store_40x_sler, void, env, tl) DEF_HELPER_FLAGS_2(store_booke_tcr, TCG_CALL_NO_RWG, void, env, tl) DEF_HELPER_FLAGS_2(store_booke_tsr, TCG_CALL_NO_RWG, void, env, tl) +DEF_HELPER_1(load_tfmr, tl, env) +DEF_HELPER_2(store_tfmr, void, env, tl) DEF_HELPER_3(store_ibatl, void, env, i32, tl) DEF_HELPER_3(store_ibatu, void, env, i32, tl)DEF_HELPER_3 (store_dbatl, void, env, i32, tl)
target/ppc/spr_common.h +2 -0
@@ -196,6 +196,8 @@ void spr_write_ebb(DisasContext *ctx, int sprn, int gprn);void spr_read_ebb_upper32 (DisasContext *ctx,int gprn,int sprn);void spr_write_ebb_upper32 (DisasContext *ctx,int sprn,int gprn);void spr_write_hmer (DisasContext *ctx,int sprn,int gprn); +void spr_read_tfmr (DisasContext *ctx,int gprn,int sprn); +void spr_write_tfmr (DisasContext *ctx,int sprn,int gprn);void spr_write_lpcr (DisasContext *ctx,int sprn,int gprn);void spr_read_dexcr_ureg (DisasContext *ctx,int gprn,int sprn);#endif
target/ppc/timebase_helper.c +13 -0
@@ -270,6 +270,19 @@ void helper_store_booke_tsr(CPUPPCState *env, target_ulong val)store_booke_tsr (env, val); } +#if defined(TARGET_PPC64 ) +/* POWER processor Timebase Facility */ +target_ulong helper_load_tfmr (CPUPPCState *env) +{ +return env->spr[SPR_TFMR ]; +} + +void helper_store_tfmr (CPUPPCState *env,target_ulong val) +{ + env->spr[SPR_TFMR ] = val; +} +#endif +/*****************************************************************************/ /* Embedded PowerPC specific helpers */
target/ppc/translate.c +10 -0
@@ -1255,6 +1255,16 @@ void spr_write_hmer(DisasContext *ctx, int sprn, int gprn)spr_store_dump_spr (sprn); } +void spr_read_tfmr (DisasContext *ctx,int gprn,int sprn) +{ +gen_helper_load_tfmr (cpu_gpr[gprn], cpu_env); +} + +void spr_write_tfmr (DisasContext *ctx,int sprn,int gprn) +{ +gen_helper_store_tfmr (cpu_env, cpu_gpr[gprn]); +} +void spr_write_lpcr (DisasContext *ctx,int sprn,int gprn) {gen_helper_store_lpcr (cpu_env, cpu_gpr[gprn]);
Signature · 1 line
2.40.1
Re: [PATCH 1/4] pnv/chiptod: Add POWER9/10 chiptod model
In reply to Nicholas PigginQuoted 135 lines · Nicholas Piggin
The chiptod is a pervasive facility which can keep TOD (time-of-day), synchronise it across multiple chips, and can move that TOD to or from the core timebase units.
This driver implements basic emulation of chiptod registers sufficient to successfully run the skiboot chiptod synchronisation procedure (with the following TFMR and timebase state-machine implementation).
The main way chiptod affects the rest of the system (relevant to the powernv model) is to interact with the timebase facility in the cores, influencing the timebase state machine and registers.
The way this chiptod driver implements that interaction is with two new flags in the CPUPPCState env, one is use for the core timebase to indicate it is ready to receive a TOD from chiptod, the other used by chiptod to indicate that it has sent TOD to the core timebase. The core timebase will be implemented in later changes.
Signed-off-by Nicholas Piggin <[email protected]>--- hw/ppc/meson.build | 1 + hw/ppc/pnv.c | 38 +++ hw/ppc/pnv_chiptod.c | 488 +++++++++++++++++++++++++++++++++++ hw/ppc/pnv_xscom.c | 2 + hw/ppc/trace-events | 4 + include/hw/ppc/pnv_chip.h | 3 + include/hw/ppc/pnv_chiptod.h | 64 +++++ include/hw/ppc/pnv_core.h | 3 + include/hw/ppc/pnv_xscom.h | 9 + target/ppc/cpu.h | 6 + 10 files changed, 618 insertions(+)create mode 100644
hw/ppc/pnv_chiptod.c
create mode 100644include/hw/ppc/pnv_chiptod.hhw/ppc/meson.build +1
-0@@ -45,6 +45,7 @@ ppc_ss.add(when: 'CONFIG_POWERNV', if_true: files( 'pnv_core.c', 'pnv_lpc.c', 'pnv_psi.c', + 'pnv_chiptod.c', 'pnv_occ.c', 'pnv_sbe.c', 'pnv_bmc.c',hw/ppc/pnv.c +38
-0@@ -1414,6 +1414,8 @@ static void pnv_chip_power9_instance_init(Object *obj)object_initialize_child (obj,"lpc" , &chip9->lpc,TYPE_PNV9_LPC ); +object_initialize_child (obj,"chiptod" , &chip9->chiptod,TYPE_PNV9_CHIPTOD ); +object_initialize_child (obj,"occ" , &chip9->occ,TYPE_PNV9_OCC );object_initialize_child (obj,"sbe" , &chip9->sbe,TYPE_PNV9_SBE );@@ -1558,6 +1560,15 @@ static void pnv_chip_power9_realize(DeviceState *dev, Error **errp) chip->dt_isa_nodename =g_strdup_printf ("/lpcm-opb@%" PRIx64"/lpc@0" , (uint64_t )PNV9_LPCM_BASE (chip)); +/* ChipTOD */ +object_property_set_link (OBJECT (&chip9->chiptod),"chip" ,OBJECT (chip), + &error_abort); +if (!qdev_realize (DEVICE (&chip9->chiptod),NULL , errp)) { +return ; + } +pnv_xscom_add_subregion (chip,PNV9_XSCOM_CHIPTOD_BASE , + &chip9->chiptod.xscom_regs); +/* Create the simplified OCC model */ if (!qdev_realize (DEVICE (&chip9->occ),NULL , errp)) {return ;@@ -1644,6 +1655,7 @@ static void pnv_chip_power10_instance_init(Object *obj) "xive-fabric");object_initialize_child (obj,"psi" , &chip10->psi,TYPE_PNV10_PSI );object_initialize_child (obj,"lpc" , &chip10->lpc,TYPE_PNV10_LPC ); +object_initialize_child (obj,"chiptod" , &chip10->chiptod,TYPE_PNV10_CHIPTOD );object_initialize_child (obj,"occ" , &chip10->occ,TYPE_PNV10_OCC );object_initialize_child (obj,"sbe" , &chip10->sbe,TYPE_PNV10_SBE );object_initialize_child (obj,"homer" , &chip10->homer,TYPE_PNV10_HOMER );@@ -1773,6 +1785,15 @@ static void pnv_chip_power10_realize(DeviceState *dev, Error **errp) chip->dt_isa_nodename =g_strdup_printf ("/lpcm-opb@%" PRIx64"/lpc@0" , (uint64_t )PNV10_LPCM_BASE (chip)); +/* ChipTOD */ +object_property_set_link (OBJECT (&chip10->chiptod),"chip" ,OBJECT (chip), + &error_abort); +if (!qdev_realize (DEVICE (&chip10->chiptod),NULL , errp)) { +return ; + } +pnv_xscom_add_subregion (chip,PNV10_XSCOM_CHIPTOD_BASE , + &chip10->chiptod.xscom_regs); +/* Create the simplified OCC model */ if (!qdev_realize (DEVICE (&chip10->occ),NULL , errp)) {return ;@@ -1938,6 +1959,23 @@ static void pnv_chip_core_realize(PnvChip *chip, Error **errp) } } +PnvCore *pnv_get_vcpu_by_xscom_base (PnvChip *chip,uint32_t xscom_base) +{ +PnvChipClass *pcc =PNV_CHIP_GET_CLASS (chip); +int i; + +for (i =0 ; i < chip->nr_cores; i++) { +PnvCore *pc = chip->cores[i]; +CPUCore *cc =CPU_CORE (pc); +int core_hwid = cc->core_id; + +if (pcc->xscom_core_base (chip, core_hwid) == xscom_base) { +return pc; + } + } +return NULL ; +} +static void pnv_chip_realize (DeviceState *dev,Error **errp) {PnvChip *chip =PNV_CHIP (dev);hw/ppc/pnv_chiptod.c new file +6
-0@@ -0,0 +1,488 @@ +/* +* QEMU PowerPC PowerNV Emulation of some CHIPTOD behaviour +* +* Copyright (c) 2022-2023, IBM Corporation. +* +* This program is free software; you can redistribute it and/or modify
hw/ppc/pnv_chiptod.c line 16
Quoted 10 lines
hw/ppc/pnv_chiptod.c new file +10
-0+* it under the terms of the GNU General Public License, version 2, as +* published by the Free Software Foundation. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>.
You can simplify the header with
* SPDX-License-Identifier: GPL-2.0-or-later
Quoted 99 lines
hw/ppc/pnv_chiptod.c new file +99
-0+*/ + +#include "qemu/osdep.h" +#include "target/ppc/cpu.h" +#include "qapi/error.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "hw/irq.h" +#include "hw/qdev-properties.h" +#include "hw/ppc/fdt.h" +#include "hw/ppc/ppc.h" +#include "hw/ppc/pnv.h" +#include "hw/ppc/pnv_chip.h" +#include "hw/ppc/pnv_core.h" +#include "hw/ppc/pnv_xscom.h" +#include "hw/ppc/pnv_chiptod.h" +#include "trace.h" + +#include <libfdt.h> + +/* TOD chip XSCOM addresses */ +#define TOD_M_PATH_CTRL_REG 0x00000000/* Master Path ctrl reg */ +#define TOD_PRI_PORT_0_CTRL_REG 0x00000001/* Primary port0 ctrl reg */ +#define TOD_PRI_PORT_1_CTRL_REG 0x00000002/* Primary port1 ctrl reg */ +#define TOD_SEC_PORT_0_CTRL_REG 0x00000003/* Secondary p0 ctrl reg */ +#define TOD_SEC_PORT_1_CTRL_REG 0x00000004/* Secondary p1 ctrl reg */ +#define TOD_S_PATH_CTRL_REG 0x00000005/* Slave Path ctrl reg */ +#define TOD_I_PATH_CTRL_REG 0x00000006/* Internal Path ctrl reg */ + +/* -- TOD primary/secondary master/slave control register -- */ +#define TOD_PSS_MSS_CTRL_REG 0x00000007 + +/* -- TOD primary/secondary master/slave status register -- */ +#define TOD_PSS_MSS_STATUS_REG 0x00000008 + +/* TOD chip XSCOM addresses */ +#define TOD_CHIP_CTRL_REG 0x00000010/* Chip control reg */ + +#define TOD_TX_TTYPE_0_REG 0x00000011 +#define TOD_TX_TTYPE_1_REG 0x00000012/* PSS switch reg */ +#define TOD_TX_TTYPE_2_REG 0x00000013/* Enable step checkers */ +#define TOD_TX_TTYPE_3_REG 0x00000014/* Request TOD reg */ +#define TOD_TX_TTYPE_4_REG 0x00000015/* Send TOD reg */ +#define TOD_TX_TTYPE_5_REG 0x00000016/* Invalidate TOD reg */ + +#define TOD_MOVE_TOD_TO_TB_REG 0x00000017 +#define TOD_LOAD_TOD_MOD_REG 0x00000018 +#define TOD_LOAD_TOD_REG 0x00000021 +#define TOD_FSM_REG 0x00000024 + +#define TOD_TX_TTYPE_CTRL_REG 0x00000027/* TX TTYPE Control reg */ +#define TOD_TX_TTYPE_PIB_SLAVE_ADDR PPC_BITMASK(26, 31) + +/* -- TOD Error interrupt register -- */ +#define TOD_ERROR_REG 0x00000030 + +/* PC unit PIB address which recieves the timebase transfer from TOD */ +#define PC_TOD 0x4A3 + +static uint64_t pnv_chiptod_xscom_read (void *opaque,hwaddr addr, +unsigned size) +{ +PnvChipTOD *chiptod =PNV_CHIPTOD (opaque); +uint32_t offset = addr >>3 ; +uint64_t val =0 ; + +switch (offset) { +case TOD_PSS_MSS_STATUS_REG : +/* +* ChipTOD does not support configurations other than primary +* master, does not support errors, etc. +*/ + val |=PPC_BITMASK (6 ,10 );/* STEP checker validity */ + val |=PPC_BIT (12 );/* Primary config master path select */ + val |=PPC_BIT (20 );/* Is running */ + val |=PPC_BIT (21 );/* Is using primary config */ + val |=PPC_BIT (26 );/* Is using master path select */ + +if (chiptod->primary) { + val |=PPC_BIT (23 );/* Is active master */ + }else if (chiptod->secondary) { + val |=PPC_BIT (24 );/* Is backup master */ + } +break ; +case TOD_PSS_MSS_CTRL_REG : + val = chiptod->pss_mss_ctrl_reg; +break ; +case TOD_TX_TTYPE_CTRL_REG : + val =0 ; +break ; +case TOD_ERROR_REG : + val = chiptod->tod_error; +break ; +case TOD_FSM_REG : +if (chiptod->tod_state == tod_running) { + val |=PPC_BIT (4 ); + } +break ; +default :
hw/ppc/pnv_chiptod.c line 125
Quoted 10 lines
hw/ppc/pnv_chiptod.c new file +10
-0+qemu_log_mask (LOG_UNIMP ,"pnv_chiptod: unimplemented register: Ox%" + HWADDR_PRIx"\n" , addr >>3 ); + } + +trace_pnv_chiptod_xscom_read (addr >>3 , val); + +return val; +} + +static void chiptod_power9_send_remotes (PnvChipTOD *chiptod)
Adding a class handler could be an alternative implementation.
hw/ppc/pnv_chiptod.c line 130
Quoted 5 lines
hw/ppc/pnv_chiptod.c new file +5
-0+{ +PnvMachineState *pnv =PNV_MACHINE (qdev_get_machine ()); +int i; + +for (i =0 ; i < pnv->num_chips; i++) {
There are a few other models (XIVE, XIVE2) which loop on the chips, is it time to introduce a pnv_foreach_chip(fn, data) routine ?
Quoted 76 lines
hw/ppc/pnv_chiptod.c new file +76
-0+Pnv9Chip *chip9 =PNV9_CHIP (pnv->chips[i]); +if (&chip9->chiptod != chiptod) { + chip9->chiptod.tod_state = tod_running; + } + } +} + +static void chiptod_power10_send_remotes (PnvChipTOD *chiptod) +{ +PnvMachineState *pnv =PNV_MACHINE (qdev_get_machine ()); +int i; + +for (i =0 ; i < pnv->num_chips; i++) { +Pnv10Chip *chip10 =PNV10_CHIP (pnv->chips[i]); +if (&chip10->chiptod != chiptod) { + chip10->chiptod.tod_state = tod_running; + } + } +} + +static void chiptod_power9_invalidate_remotes (PnvChipTOD *chiptod) +{ +PnvMachineState *pnv =PNV_MACHINE (qdev_get_machine ()); +int i; + +for (i =0 ; i < pnv->num_chips; i++) { +Pnv9Chip *chip9 =PNV9_CHIP (pnv->chips[i]); +if (&chip9->chiptod != chiptod) { + chip9->chiptod.tod_state = tod_not_set; + } + } +} + +static void chiptod_power10_invalidate_remotes (PnvChipTOD *chiptod) +{ +PnvMachineState *pnv =PNV_MACHINE (qdev_get_machine ()); +int i; + +for (i =0 ; i < pnv->num_chips; i++) { +Pnv10Chip *chip10 =PNV10_CHIP (pnv->chips[i]); +if (&chip10->chiptod != chiptod) { + chip10->chiptod.tod_state = tod_not_set; + } + } +} + +static void pnv_chiptod_xscom_write (void *opaque,hwaddr addr, +uint64_t val,unsigned size, +bool is_power9) +{ +PnvChipTOD *chiptod =PNV_CHIPTOD (opaque); +uint32_t offset = addr >>3 ; + +trace_pnv_chiptod_xscom_write (addr >>3 , val); + +switch (offset) { +case TOD_PSS_MSS_CTRL_REG : +/* Is this correct? */ +if (chiptod->primary) { + val |=PPC_BIT (1 );/* TOD is master */ + }else { + val &= ~PPC_BIT (1 ); + } + val |=PPC_BIT (2 );/* Drawer is master (don't simulate multi-drawer) */ + chiptod->pss_mss_ctrl_reg = val &PPC_BITMASK (0 ,31 ); +break ; + +case TOD_TX_TTYPE_CTRL_REG : +if (val &PPC_BIT (35 )) {/* SCOM addressing */ +uint32_t addr = val >>32 ; +uint32_t reg = addr &0xfff ; +PnvCore *pc; + +if (reg !=PC_TOD ) { + qemu_log_mask(LOG_GUEST_ERROR, "pnv_chiptod: SCOM addressing: " + "unimplemented slave register 0x%" PRIx32 "\n",
hw/ppc/pnv_chiptod.c line 216
Quoted 10 lines
hw/ppc/pnv_chiptod.c new file +10
-0+ reg); +return ; + } + +/* +* This may not deal with P10 big-core addressing at the moment. +* The big-core code in skiboot syncs small cores, but it targets +* the even PIR (first small-core) when syncing second small-core. +*/ + pc =pnv_get_vcpu_by_xscom_base (chiptod->chip, addr & ~0xfff );
hmm, couldn't we map xscom subregions, one for each thread instead ?
hw/ppc/pnv_chiptod.c line 223
Quoted 7 lines
hw/ppc/pnv_chiptod.c new file +7
-0+if (pc) { +/* +* If TCG implements SMT, TFMR is a per-core SPR and should +* be updated such that it is reflected in all threads. +* Same for TB if the chiptod model ever actually adjusted it. +*/ + chiptod->slave_cpu_target = pc->threads[0 ];
ah ! SMT is implemented :) The xscom subregions would help to get the CPU pointer.
Quoted 94 lines
hw/ppc/pnv_chiptod.c new file +94
-0+ }else { + chiptod->slave_cpu_target =NULL ; +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: xscom write reg" +" TOD_TX_TTYPE_CTRL_REG val 0x%" PRIx64 +" invalid slave PIR\n" , val); + } + + }else {/* PIR addressing */ +uint32_t pir; + +if (!is_power9) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: PIR addressing" +" is only implemented for POWER9\n" ); +return ; + } + + pir = (GETFIELD (TOD_TX_TTYPE_PIB_SLAVE_ADDR , val) &0x1f ) <<2 ; + chiptod->slave_cpu_target =ppc_get_vcpu_by_pir (pir); +if (chiptod->slave_cpu_target ==NULL ) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: xscom write reg" +" TOD_TX_TTYPE_CTRL_REG val 0x%" PRIx64 +" invalid slave PIR 0x%" PRIx32"\n" , val, pir); + } + } +break ; +case TOD_ERROR_REG : + chiptod->tod_error &= ~val; +break ; +case TOD_LOAD_TOD_MOD_REG : +if (!(val &PPC_BIT (0 ))) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: xscom write reg" +" TOD_LOAD_TOD_MOD_REG with bad val 0x%016lx\n" , val); + }else { + chiptod->tod_state = tod_not_set; + } +break ; +case TOD_LOAD_TOD_REG : + chiptod->tod_state = tod_running; +break ; +case TOD_MOVE_TOD_TO_TB_REG : +if (!(val &PPC_BIT (0 ))) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: xscom write reg" +" TOD_MOVE_TOD_TO_TB_REG with bad val 0x%016lx\n" , + val); + }else if (chiptod->slave_cpu_target ==NULL ) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: xscom write reg" +" TOD_MOVE_TOD_TO_TB_REG with no slave target\n" ); + }else { +PowerPCCPU *cpu = chiptod->slave_cpu_target; +CPUPPCState *env = &cpu->env; + +if (env->tb_ready_for_tod) { + env->tod_sent_to_tb =1 ; + }else { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: xscom write reg" +" TOD_MOVE_TOD_TO_TB_REG with TB not ready to" +" receive TOD\n" ); + } + } +break ; +case TOD_TX_TTYPE_4_REG : +if (is_power9) { +chiptod_power9_send_remotes (chiptod); + }else { +chiptod_power10_send_remotes (chiptod); + } +break ; +case TOD_TX_TTYPE_5_REG : +if (is_power9) { +chiptod_power9_invalidate_remotes (chiptod); + }else { +chiptod_power10_invalidate_remotes (chiptod); + } +break ; +default : +qemu_log_mask (LOG_UNIMP ,"pnv_chiptod: unimplemented register: Ox%" + HWADDR_PRIx"\n" , addr >>3 ); + } +} + +static void pnv_chiptod_power9_xscom_write (void *opaque,hwaddr addr, +uint64_t val,unsigned size) +{ +pnv_chiptod_xscom_write (opaque, addr, val, size, true); +} + +static const MemoryRegionOps pnv_chiptod_power9_xscom_ops = { + .read = pnv_chiptod_xscom_read, + .write = pnv_chiptod_power9_xscom_write, + .valid.min_access_size =8 , + .valid.max_access_size =8 , + .impl.min_access_size =8 , + .impl.max_access_size =8 , + .endianness =DEVICE_BIG_ENDIAN ,
hw/ppc/pnv_chiptod.c line 327
Quoted 10 lines
hw/ppc/pnv_chiptod.c new file +10
-0+}; + +static int pnv_chiptod_dt_xscom (PnvXScomInterface *dev,void *fdt, +int xscom_offset, +const char compat[],size_t compat_size) +{ +PnvChipTOD *chiptod =PNV_CHIPTOD (dev); + g_autofree char *name = NULL; +int offset; +uint32_t lpc_pcba =PNV9_XSCOM_CHIPTOD_BASE ;
lpc ?
Quoted 115 lines
hw/ppc/pnv_chiptod.c new file +115
-0+uint32_t reg[] = { +cpu_to_be32 (lpc_pcba), +cpu_to_be32 (PNV9_XSCOM_CHIPTOD_SIZE ) + }; + + name =g_strdup_printf ("chiptod@%x" , lpc_pcba); + offset =fdt_add_subnode (fdt, xscom_offset, name); +_FDT (offset); + +if (chiptod->primary) { +_FDT ((fdt_setprop (fdt, offset,"primary" ,NULL ,0 ))); + }else if (chiptod->secondary) { +_FDT ((fdt_setprop (fdt, offset,"secondary" ,NULL ,0 ))); + } + +_FDT ((fdt_setprop (fdt, offset,"reg" , reg,sizeof (reg)))); +_FDT ((fdt_setprop (fdt, offset,"compatible" , compat, compat_size))); +return 0 ; +} + +static int pnv_chiptod_power9_dt_xscom (PnvXScomInterface *dev,void *fdt, +int xscom_offset) +{ +const char compat[] ="ibm,power-chiptod\0ibm,power9-chiptod" ; + +return pnv_chiptod_dt_xscom (dev, fdt, xscom_offset, compat,sizeof (compat)); +} + +static Property pnv_chiptod_properties[] = { + DEFINE_PROP_LINK("chip", PnvChipTOD , chip, TYPE_PNV_CHIP, PnvChip *), +DEFINE_PROP_END_OF_LIST (), +}; + +static void pnv_chiptod_power9_class_init (ObjectClass *klass,void *data) +{ +PnvChipTODClass *pctc =PNV_CHIPTOD_CLASS (klass); +DeviceClass *dc =DEVICE_CLASS (klass); +PnvXScomInterfaceClass *xdc =PNV_XSCOM_INTERFACE_CLASS (klass); + + dc->desc ="PowerNV ChipTOD Controller (POWER9)" ; +device_class_set_props (dc, pnv_chiptod_properties); + + xdc->dt_xscom = pnv_chiptod_power9_dt_xscom; + + pctc->xscom_size =PNV_XSCOM_CHIPTOD_SIZE ; + pctc->xscom_ops = &pnv_chiptod_power9_xscom_ops; +} + +static const TypeInfo pnv_chiptod_power9_type_info = { + .name =TYPE_PNV9_CHIPTOD , + .parent =TYPE_PNV_CHIPTOD , + .instance_size =sizeof (PnvChipTOD), + .class_init = pnv_chiptod_power9_class_init, + .interfaces = (InterfaceInfo []) { + {TYPE_PNV_XSCOM_INTERFACE }, + { } + } +}; + +static void pnv_chiptod_power10_xscom_write (void *opaque,hwaddr addr, +uint64_t val,unsigned size) +{ +pnv_chiptod_xscom_write (opaque, addr, val, size, false); +} + +static const MemoryRegionOps pnv_chiptod_power10_xscom_ops = { + .read = pnv_chiptod_xscom_read, + .write = pnv_chiptod_power10_xscom_write, + .valid.min_access_size =8 , + .valid.max_access_size =8 , + .impl.min_access_size =8 , + .impl.max_access_size =8 , + .endianness =DEVICE_BIG_ENDIAN , +}; + +static int pnv_chiptod_power10_dt_xscom (PnvXScomInterface *dev,void *fdt, +int xscom_offset) +{ +const char compat[] ="ibm,power-chiptod\0ibm,power10-chiptod" ; + +return pnv_chiptod_dt_xscom (dev, fdt, xscom_offset, compat,sizeof (compat)); +} + +static void pnv_chiptod_power10_class_init (ObjectClass *klass,void *data) +{ +PnvChipTODClass *pctc =PNV_CHIPTOD_CLASS (klass); +DeviceClass *dc =DEVICE_CLASS (klass); +PnvXScomInterfaceClass *xdc =PNV_XSCOM_INTERFACE_CLASS (klass); + + dc->desc ="PowerNV ChipTOD Controller (POWER10)" ; +device_class_set_props (dc, pnv_chiptod_properties); + + xdc->dt_xscom = pnv_chiptod_power10_dt_xscom; + + pctc->xscom_size =PNV_XSCOM_CHIPTOD_SIZE ; + pctc->xscom_ops = &pnv_chiptod_power10_xscom_ops; +} + +static const TypeInfo pnv_chiptod_power10_type_info = { + .name =TYPE_PNV10_CHIPTOD , + .parent =TYPE_PNV_CHIPTOD , + .instance_size =sizeof (PnvChipTOD), + .class_init = pnv_chiptod_power10_class_init, + .interfaces = (InterfaceInfo []) { + {TYPE_PNV_XSCOM_INTERFACE }, + { } + } +}; + +static void pnv_chiptod_realize (DeviceState *dev,Error **errp) +{ +static bool got_primary = false; +static bool got_secondary = false; + +PnvChipTOD *chiptod =PNV_CHIPTOD (dev);
hw/ppc/pnv_chiptod.c line 452
Quoted 10 lines
hw/ppc/pnv_chiptod.c new file +10
-0+PnvChipTODClass *pctc =PNV_CHIPTOD_GET_CLASS (chiptod); + +if (!got_primary) { + got_primary = true; + chiptod->primary = true; + chiptod->pss_mss_ctrl_reg |=PPC_BIT (1 );/* TOD is master */ + }else if (!got_secondary) { + got_secondary = true; + chiptod->secondary = true; + }
It would be cleaner to introduce "primary" and "secondary" properties defined by the model realizing the PnvChipTOD objects.
Quoted 41 lines
hw/ppc/pnv_chiptod.c new file +36
-0+/* Drawer is master (we do not simulate multi-drawer) */ + chiptod->pss_mss_ctrl_reg |=PPC_BIT (2 ); + chiptod->tod_state = tod_running; + +/* XScom regions for ChipTOD registers */ +pnv_xscom_region_init (&chiptod->xscom_regs,OBJECT (dev), + pctc->xscom_ops, chiptod,"xscom-chiptod" , + pctc->xscom_size); +} + +static void pnv_chiptod_class_init (ObjectClass *klass,void *data) +{ +DeviceClass *dc =DEVICE_CLASS (klass); + + dc->realize = pnv_chiptod_realize; + dc->desc ="PowerNV ChipTOD Controller" ; + dc->user_creatable = false; +} + +static const TypeInfo pnv_chiptod_type_info = { + .name =TYPE_PNV_CHIPTOD , + .parent =TYPE_DEVICE , + .instance_size =sizeof (PnvChipTOD), + .class_init = pnv_chiptod_class_init, + .class_size =sizeof (PnvChipTODClass), + .abstract = true, +}; + +static void pnv_chiptod_register_types (void ) +{ +type_register_static (&pnv_chiptod_type_info); +type_register_static (&pnv_chiptod_power9_type_info); +type_register_static (&pnv_chiptod_power10_type_info); +} + +type_init (pnv_chiptod_register_types);hw/ppc/pnv_xscom.c +0
-0
hw/ppc/pnv_xscom.c int pnv_dt_xscom(PnvChip *chip, void *fdt, int root_offset, · line 302
Quoted 5 lines
hw/ppc/pnv_xscom.c +2
-0@@ -298,6 +298,8 @@ int pnv_dt_xscom(PnvChip *chip, void *fdt, int root_offset,_FDT ((fdt_setprop (fdt, xscom_offset,"scom-controller" ,NULL ,0 )));if (chip->chip_id ==0 ) {_FDT ((fdt_setprop (fdt, xscom_offset,"primary" ,NULL ,0 ))); + }else if (chip->chip_id ==1 ) { +_FDT ((fdt_setprop (fdt, xscom_offset,"secondary" ,NULL ,0 )));
This is a more-or-less related change. May be we should have "primary" and "secondary" property in PnvChip also.
Quoted 74 lines
hw/ppc/pnv_xscom.c +0
-0} args.fdt = fdt;hw/ppc/trace-events +4
-0@@ -95,6 +95,10 @@ vof_write(uint32_t ih, unsigned cb, const char *msg) "ih=0x%x [%u] \"%s\"" vof_avail(uint64_t start, uint64_t end, uint64_t size) "0x%"PRIx64"..0x%"PRIx64" size=0x%"PRIx64 vof_claimed(uint64_t start, uint64_t end, uint64_t size) "0x%"PRIx64"..0x%"PRIx64" size=0x%"PRIx64 +# pnv_chiptod.c +pnv_chiptod_xscom_read(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64 +pnv_chiptod_xscom_write(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64 + # pnv_sbe.c pnv_sbe_xscom_ctrl_read(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64 pnv_sbe_xscom_ctrl_write(uint64_t addr, uint64_t val) "addr 0x%" PRIx64 " val 0x%" PRIx64include/hw/ppc/pnv_chip.h +3
-0@@ -2,6 +2,7 @@#define PPC_PNV_CHIP_H #include "hw/pci-host/pnv_phb4.h" +#include "hw/ppc/pnv_chiptod.h" #include "hw/ppc/pnv_core.h" #include "hw/ppc/pnv_homer.h" #include "hw/ppc/pnv_lpc.h" @@ -77,6 +78,7 @@ struct Pnv9Chip {PnvXive xive;Pnv9Psi psi;PnvLpcController lpc; +PnvChipTOD chiptod;PnvOCC occ;PnvSBE sbe;PnvHomer homer;@@ -106,6 +108,7 @@ struct Pnv10Chip {PnvXive2 xive;Pnv9Psi psi;PnvLpcController lpc; +PnvChipTOD chiptod;PnvOCC occ;PnvSBE sbe;PnvHomer homer;include/hw/ppc/pnv_chiptod.h new file +22
-0@@ -0,0 +1,64 @@ +/* +* QEMU PowerPC PowerNV Emulation of some CHIPTOD behaviour +* +* Copyright (c) 2022-2023, IBM Corporation. +* +* This library is free software; you can redistribute it and/or +* modify it under the terms of the GNU Lesser General Public +* License as published by the Free Software Foundation; either +* version 2.1 of the License, or (at your option) any later version. +* +* This library is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +* Lesser General Public License for more details. +* +* You should have received a copy of the GNU Lesser General Public +* License along with this library; if not, see <http://www.gnu.org/licenses/>. +*/ + +#ifndef PPC_PNV_CHIPTOD_H +#define PPC_PNV_CHIPTOD_H +
include/hw/ppc/pnv_chiptod.h line 32
Quoted 10 lines
include/hw/ppc/pnv_chiptod.h new file +10
-0+#include "qom/object.h" + +#define TYPE_PNV_CHIPTOD "pnv-chiptod" +OBJECT_DECLARE_TYPE(PnvChipTOD, PnvChipTODClass, PNV_CHIPTOD) +#define TYPE_PNV9_CHIPTOD TYPE_PNV_CHIPTOD "-POWER9" +DECLARE_INSTANCE_CHECKER(PnvChipTOD, PNV9_CHIPTOD, TYPE_PNV9_CHIPTOD) +#define TYPE_PNV10_CHIPTOD TYPE_PNV_CHIPTOD "-POWER10" +DECLARE_INSTANCE_CHECKER(PnvChipTOD, PNV10_CHIPTOD, TYPE_PNV10_CHIPTOD) + +enum tod_state {
PnvChipTODState would be better naming.
Thanks,
C.
Quoted 95 lines
include/hw/ppc/pnv_chiptod.h new file +31
-0+ tod_error =0 , + tod_not_set =7 , + tod_not_set_step =11 , + tod_running =2 , + tod_running_step =10 , + tod_running_sync =14 , + tod_wait_for_sync =13 , + tod_stopped = 1, +}; +struct PnvChipTOD { +DeviceState xd; + +PnvChip *chip; +MemoryRegion xscom_regs; + +bool primary; +bool secondary; +enum tod_state tod_state; +uint64_t tod_error; +uint64_t pss_mss_ctrl_reg; +PowerPCCPU *slave_cpu_target; +}; + +struct PnvChipTODClass { +DeviceClass parent_class; + +int xscom_size; +const MemoryRegionOps *xscom_ops; +}; + +#endif /* PPC_PNV_CHIPTOD_H */ include/hw/ppc/pnv_core.h +3
-0@@ -69,4 +69,7 @@ struct PnvQuad {uint32_t quad_id; MemoryRegion xscom_regs; }; + +PnvCore *pnv_get_vcpu_by_xscom_base (PnvChip *chip,uint32_t xscom_base); +#endif /* PPC_PNV_CORE_H */ include/hw/ppc/pnv_xscom.h +9
-0@@ -64,6 +64,9 @@ struct PnvXScomInterfaceClass {#define PNV_XSCOM_PSIHB_BASE 0x2010900#define PNV_XSCOM_PSIHB_SIZE 0x20 +#define PNV_XSCOM_CHIPTOD_BASE 0x0040000 +#define PNV_XSCOM_CHIPTOD_SIZE 0x31 +#define PNV_XSCOM_OCC_BASE 0x0066000#define PNV_XSCOM_OCC_SIZE 0x6000@@ -90,6 +93,9 @@ struct PnvXScomInterfaceClass { ((uint64_t)(((core) & 0x1C) + 0x40) << 22)#define PNV9_XSCOM_EQ_SIZE 0x100000 +#define PNV9_XSCOM_CHIPTOD_BASE PNV_XSCOM_CHIPTOD_BASE +#define PNV9_XSCOM_CHIPTOD_SIZE PNV_XSCOM_CHIPTOD_SIZE +#define PNV9_XSCOM_OCC_BASE PNV_XSCOM_OCC_BASE#define PNV9_XSCOM_OCC_SIZE 0x8000@@ -138,6 +144,9 @@ struct PnvXScomInterfaceClass {#define PNV10_XSCOM_PSIHB_BASE 0x3011D00#define PNV10_XSCOM_PSIHB_SIZE 0x100 +#define PNV10_XSCOM_CHIPTOD_BASE PNV9_XSCOM_CHIPTOD_BASE +#define PNV10_XSCOM_CHIPTOD_SIZE PNV9_XSCOM_CHIPTOD_SIZE +#define PNV10_XSCOM_OCC_BASE PNV9_XSCOM_OCC_BASE#define PNV10_XSCOM_OCC_SIZE PNV9_XSCOM_OCC_SIZEtarget/ppc/cpu.h +6
-0@@ -1172,6 +1172,12 @@ struct CPUArchState {uint32_t tlb_need_flush;/* Delayed flush needed */ #define TLB_NEED_LOCAL_FLUSH 0x1#define TLB_NEED_GLOBAL_FLUSH 0x2 + +#if defined(TARGET_PPC64 ) +/* PowerNV chiptod / timebase facility state. */ +int tb_ready_for_tod;/* core TB ready to receive TOD from chiptod */ +int tod_sent_to_tb;/* chiptod sent TOD to the core TB */ +#endif #endif /* Other registers */
Re: [PATCH 2/4] target/ppc: Tidy POWER book4 SPR registration
In reply to Nicholas PigginQuoted 5 lines · Nicholas Piggin
POWER book4 (implementation-specific) SPRs are sometimes in their own functions, but in other cases are mixed with architected SPRs. Do some spring cleaning on these.
Signed-off-by Nicholas Piggin <[email protected]>
Thanks,
C.
Quoted 161 lines
--- target/ppc/cpu_init.c | 92 ++++++++++++++++++++++++++++--------------- 1 file changed, 60 insertions(+), 32 deletions(-)target/ppc/cpu_init.c +60
-32@@ -5374,31 +5374,6 @@ static void register_book3s_ids_sprs(CPUPPCState *env) &spr_read_generic,SPR_NOACCESS , &spr_read_generic,NULL , 0x00000000); -spr_register_hv (env,SPR_HID0 ,"HID0" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_core_write_generic, -0x00000000 ); -spr_register_hv (env,SPR_TSCR ,"TSCR" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic32, -0x00000000 ); -spr_register_hv (env,SPR_HMER ,"HMER" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_hmer, -0x00000000 ); -spr_register_hv (env,SPR_HMEER ,"HMEER" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, -0x00000000 ); -spr_register_hv (env,SPR_TFMR ,"TFMR" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, -0x00000000 ); spr_register_hv(env, SPR_LPIDR, "LPIDR", SPR_NOACCESS, SPR_NOACCESS, SPR_NOACCESS, SPR_NOACCESS,@@ -5454,11 +5429,6 @@ static void register_book3s_ids_sprs(CPUPPCState *env)SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic, 0x00000000); -spr_register_hv (env,SPR_LDBAR ,"LDBAR" , -SPR_NOACCESS ,SPR_NOACCESS , -SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, -0x00000000 ); }static void register_rmor_sprs (CPUPPCState *env)@@ -5665,14 +5635,65 @@ static void register_power8_ic_sprs(CPUPPCState *env)#endif } +/* SPRs specific to IBM POWER CPUs */ +static void register_power_common_book4_sprs (CPUPPCState *env) +{ +#if !defined(CONFIG_USER_ONLY ) +spr_register_hv (env,SPR_HID0 ,"HID0" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_core_write_generic, +0x00000000 ); +spr_register_hv (env,SPR_TSCR ,"TSCR" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic32, +0x00000000 ); +spr_register_hv (env,SPR_HMER ,"HMER" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_hmer, +0x00000000 ); +spr_register_hv (env,SPR_HMEER ,"HMEER" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +0x00000000 ); +spr_register_hv (env,SPR_TFMR ,"TFMR" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +0x00000000 ); +spr_register_hv (env,SPR_LDBAR ,"LDBAR" , +SPR_NOACCESS ,SPR_NOACCESS , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +0x00000000 ); +#endif +} + +static void register_power9_book4_sprs (CPUPPCState *env) +{ +/* Add a number of P9 book4 registers */ +register_power_common_book4_sprs (env); +#if !defined(CONFIG_USER_ONLY ) +spr_register_kvm (env,SPR_WORT ,"WORT" , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_generic, +KVM_REG_PPC_WORT ,0 ); +#endif +} +static void register_power8_book4_sprs (CPUPPCState *env) {/* Add a number of P8 book4 registers */ +register_power_common_book4_sprs (env);#if !defined(CONFIG_USER_ONLY )spr_register_kvm (env,SPR_ACOP ,"ACOP" ,SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic,KVM_REG_PPC_ACOP ,0 ); +/* PID is only in BookE in ISA v2.07 */ spr_register_kvm(env, SPR_BOOKS_PID, "PID", SPR_NOACCESS, SPR_NOACCESS, &spr_read_generic, &spr_write_pidr,@@ -5688,10 +5709,12 @@ static void register_power7_book4_sprs(CPUPPCState *env) {/* Add a number of P7 book4 registers */ #if !defined(CONFIG_USER_ONLY ) +register_power_common_book4_sprs (env);spr_register_kvm (env,SPR_ACOP ,"ACOP" ,SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic,KVM_REG_PPC_ACOP ,0 ); +/* PID is only in BookE in ISA v2.06 */ spr_register_kvm(env, SPR_BOOKS_PID, "PID", SPR_NOACCESS, SPR_NOACCESS, &spr_read_generic, &spr_write_generic32,@@ -5725,6 +5748,11 @@ static void register_power9_mmu_sprs(CPUPPCState *env)SPR_NOACCESS ,SPR_NOACCESS , &spr_read_generic, &spr_write_generic, 0x0000000000000000); +/* PID is part of the BookS ISA from v3.0 */ +spr_register_kvm (env,SPR_BOOKS_PID ,"PID" , +SPR_NOACCESS ,SPR_NOACCESS , + &spr_read_generic, &spr_write_pidr, +KVM_REG_PPC_PID ,0 );#endif }@@ -6278,7 +6306,7 @@ static void init_proc_POWER9(CPUPPCState *env)register_power8_dpdes_sprs (env);register_vtb_sprs (env);register_power8_ic_sprs (env); -register_power8_book4_sprs (env); +register_power9_book4_sprs (env);register_power8_rpr_sprs (env);register_power9_mmu_sprs (env);@@ -6471,7 +6499,7 @@ static void init_proc_POWER10(CPUPPCState *env)register_power8_dpdes_sprs (env);register_vtb_sprs (env);register_power8_ic_sprs (env); -register_power8_book4_sprs (env); +register_power9_book4_sprs (env);register_power8_rpr_sprs (env);register_power9_mmu_sprs (env);register_power10_hash_sprs (env);
Quoted 20 lines · Nicholas Piggin
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Quoted 6 lines
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved totimebase_helper.cfrommisc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.
The chiptod units are not exposed to the OS, it is all handled at FW level AFAIK. Could the OPAL people provide some feedback on the low level models ?
Thanks,
C.
Quoted 32 lines
There is still a vast amount that is not modeled, but most of it related to error handling, injection, failover, etc that is very complicated and not required for normal operation.
Thanks,
NickNicholas Piggin (4):
pnv/chiptod: Add POWER9/10 chiptod model
target/ppc: Tidy POWER book4 SPR registration target/ppc: add TFMR SPR implementation with read and write helpers target/ppc: Implement core timebase state machine and TFMRhw/ppc/meson.build | 1 + hw/ppc/pnv.c | 38 +++ hw/ppc/pnv_chiptod.c | 488 +++++++++++++++++++++++++++++++++++ hw/ppc/pnv_xscom.c | 2 + hw/ppc/trace-events | 4 + include/hw/ppc/pnv_chip.h | 3 + include/hw/ppc/pnv_chiptod.h | 64 +++++ include/hw/ppc/pnv_core.h | 3 + include/hw/ppc/pnv_xscom.h | 9 + target/ppc/cpu.h | 40 +++ target/ppc/cpu_init.c | 92 ++++--- target/ppc/helper.h | 2 + target/ppc/spr_common.h | 2 + target/ppc/timebase_helper.c | 156 +++++++++++ target/ppc/translate.c | 10 + 15 files changed, 882 insertions(+), 32 deletions(-)create mode 100644
hw/ppc/pnv_chiptod.c
create mode 100644include/hw/ppc/pnv_chiptod.h
Quoted 29 lines · Cédric Le Goater
Quoted 20 lines · Nicholas Piggin
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Quoted 6 lines
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved totimebase_helper.cfrommisc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.
Quoted 3 lines
The chiptod units are not exposed to the OS, it is all handled at FW level AFAIK. Could the OPAL people provide some feedback on the low level models ?
Well, no takers so far. I guess I'm a OPAL people :)
I did some of the P10 chiptod addressing in skiboot, at least. This patch does work with the skiboot chiptod driver at least.
I would eventually like to add in the ability to actually change the TB with it, and inject errors and generate HMIs because that's an area that seem to be a bit lacking (most of such testing seemed to be done on real hardware using special time facility corruption injection).
But yes for now it is a bit difficult to verify it does much useful aside from booting skiboot (+ patch to enable chiptod on QEMU I posted recently).
Thanks,
Nick
Re: [PATCH 1/4] pnv/chiptod: Add POWER9/10 chiptod model
In reply to Cédric Le GoaterQuoted 27 lines · Cédric Le Goater
Quoted 12 lines · Nicholas Piggin
hw/ppc/pnv_chiptod.c new file +6
-0@@ -0,0 +1,488 @@ +/* +* QEMU PowerPC PowerNV Emulation of some CHIPTOD behaviour +* +* Copyright (c) 2022-2023, IBM Corporation. +* +* This program is free software; you can redistribute it and/or modify hw/ppc/pnv_chiptod.c line 16
Quoted 10 lines
hw/ppc/pnv_chiptod.c new file +10
-0+* it under the terms of the GNU General Public License, version 2, as +* published by the Free Software Foundation. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. You can simplify the header with
Quoted 1 line
* SPDX-License-Identifier: GPL-2.0-or-later
Sure.
Quoted 4 lines
hw/ppc/pnv_chiptod.c line 18
Quoted 2 lines
hw/ppc/pnv_chiptod.c new file +2
-0+ +static void chiptod_power9_send_remotes (PnvChipTOD *chiptod)Adding a class handler could be an alternative implementation.
Might be a good idea, I'll see how it looks.
Quoted 16 lines
hw/ppc/pnv_chiptod.c line 130
Quoted 5 lines
hw/ppc/pnv_chiptod.c new file +5
-0+{ +PnvMachineState *pnv =PNV_MACHINE (qdev_get_machine ()); +int i; + +for (i =0 ; i < pnv->num_chips; i++) {There are a few other models (XIVE, XIVE2) which loop on the chips, is it time to introduce a
pnv_foreach_chip(fn, data)routine ?Quoted 6 lines
hw/ppc/pnv_chiptod.c new file +6
-0+Pnv9Chip *chip9 =PNV9_CHIP (pnv->chips[i]); +if (&chip9->chiptod != chiptod) { + chip9->chiptod.tod_state = tod_running; + } + } +}
[snip]
Quoted 13 lines
Quoted 13 lines
hw/ppc/pnv_chiptod.c new file +13
-0+case TOD_TX_TTYPE_CTRL_REG : +if (val &PPC_BIT (35 )) {/* SCOM addressing */ +uint32_t addr = val >>32 ; +uint32_t reg = addr &0xfff ; +PnvCore *pc; + +if (reg !=PC_TOD ) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: SCOM addressing: " +"unimplemented slave register 0x%" PRIx32"\n" , + reg); +return ; + } +
Quoted 8 lines
hw/ppc/pnv_chiptod.c line 216
Quoted 6 lines
hw/ppc/pnv_chiptod.c new file +6
-0+/* +* This may not deal with P10 big-core addressing at the moment. +* The big-core code in skiboot syncs small cores, but it targets +* the even PIR (first small-core) when syncing second small-core. +*/ + pc =pnv_get_vcpu_by_xscom_base (chiptod->chip, addr & ~0xfff );hmm, couldn't we map xscom subregions, one for each thread instead ?
I'm not sure what you mean. This scom-addressing uses the xscom address of the core's PC unit (where its time facilities are) to point nest chiptod transfers to cores.
Quoted 11 lines
hw/ppc/pnv_chiptod.c line 223
Quoted 8 lines
hw/ppc/pnv_chiptod.c new file +7
-0+if (pc) { +/* +* If TCG implements SMT, TFMR is a per-core SPR and should +* be updated such that it is reflected in all threads. +* Same for TB if the chiptod model ever actually adjusted it. +*/ + chiptod->slave_cpu_target = pc->threads[0 ];ah ! SMT is implemented :) The xscom subregions would help to get the CPU pointer.
I think I may be mistaken at least in the "get_vcpu" part. I think it should be get core, I don't know if chiptod is capable of addressing threads individually.
I could test it with SMT patches and see what skiboot does.
Quoted 10 lines
hw/ppc/pnv_chiptod.c line 327
Quoted 8 lines
hw/ppc/pnv_chiptod.c new file +8
-0+static int pnv_chiptod_dt_xscom (PnvXScomInterface *dev,void *fdt, +int xscom_offset, +const char compat[],size_t compat_size) +{ +PnvChipTOD *chiptod =PNV_CHIPTOD (dev); + g_autofree char *name = NULL; +int offset; +uint32_t lpc_pcba =PNV9_XSCOM_CHIPTOD_BASE ;lpc ?
Shh don't tell anybody I mostly code via copy and paste :P
[snip]
Quoted 11 lines
Quoted 11 lines
hw/ppc/pnv_chiptod.c new file +11
-0+static void pnv_chiptod_realize (DeviceState *dev,Error **errp) +{ +static bool got_primary = false; +static bool got_secondary = false; + +PnvChipTOD *chiptod =PNV_CHIPTOD (dev); +PnvChipTODClass *pctc =PNV_CHIPTOD_GET_CLASS (chiptod); + +if (!got_primary) { + got_primary = true; + chiptod->primary = true;
Quoted 8 lines
hw/ppc/pnv_chiptod.c line 452
Quoted 5 lines
hw/ppc/pnv_chiptod.c new file +5
-0+ chiptod->pss_mss_ctrl_reg |=PPC_BIT (1 );/* TOD is master */ + }else if (!got_secondary) { + got_secondary = true; + chiptod->secondary = true; + }It would be cleaner to introduce "primary" and "secondary" properties defined by the model realizing the PnvChipTOD objects.
Hum. There's a few hard-coded primaries on chip_id == 0 already in the tree. I don't know how closely related they are, chiptod can set other chips as primary AFAIK but maybe it just makes sense to make primary a chip property.
I might dig a bit more into what we (and other IBM firmware) want to test or expect with these primaries. Maybe another pass to tidy all that up can be done.
[...]
Quoted 7 lines
Quoted 7 lines
hw/ppc/pnv_chiptod.c new file +7
-0+#ifndef PPC_PNV_CHIPTOD_H +#define PPC_PNV_CHIPTOD_H + +#include "qom/object.h" + +#define TYPE_PNV_CHIPTOD "pnv-chiptod" +OBJECT_DECLARE_TYPE (PnvChipTOD, PnvChipTODClass,PNV_CHIPTOD )
Quoted 8 lines
hw/ppc/pnv_chiptod.c
Quoted 6 lines
hw/ppc/pnv_chiptod.c new file +6
-0+#define TYPE_PNV9_CHIPTOD TYPE_PNV_CHIPTOD "-POWER9" +DECLARE_INSTANCE_CHECKER(PnvChipTOD, PNV9_CHIPTOD, TYPE_PNV9_CHIPTOD) +#define TYPE_PNV10_CHIPTOD TYPE_PNV_CHIPTOD "-POWER10" +DECLARE_INSTANCE_CHECKER(PnvChipTOD, PNV10_CHIPTOD, TYPE_PNV10_CHIPTOD) + +enum tod_state {PnvChipTODState would be better naming.
Agree.
Thanks,
Nick
Re: [PATCH 1/4] pnv/chiptod: Add POWER9/10 chiptod model
Hello Nick,
[ ... ]
Quoted 22 lines
Quoted 13 lines
Quoted 13 lines
hw/ppc/pnv_chiptod.c new file +13
-0+case TOD_TX_TTYPE_CTRL_REG : +if (val &PPC_BIT (35 )) {/* SCOM addressing */ +uint32_t addr = val >>32 ; +uint32_t reg = addr &0xfff ; +PnvCore *pc; + +if (reg !=PC_TOD ) { +qemu_log_mask (LOG_GUEST_ERROR ,"pnv_chiptod: SCOM addressing: " +"unimplemented slave register 0x%" PRIx32"\n" , + reg); +return ; + } +Quoted 8 lines
hw/ppc/pnv_chiptod.c line 216
Quoted 6 lines
hw/ppc/pnv_chiptod.c new file +6
-0+/* +* This may not deal with P10 big-core addressing at the moment. +* The big-core code in skiboot syncs small cores, but it targets +* the even PIR (first small-core) when syncing second small-core. +*/ + pc =pnv_get_vcpu_by_xscom_base (chiptod->chip, addr & ~0xfff );hmm, couldn't we map xscom subregions, one for each thread instead ?
Quoted 3 lines
I'm not sure what you mean. This scom-addressing uses the xscom address of the core's PC unit (where its time facilities are) to point nest chiptod transfers to cores.
What I meant is that if you could map one xscom subregion per thread in an overall container region, each region could have its own 'opaque' data (the thread object you are interested in) and you wouldn't need to do the CPU lookup. A bit like for the ICP on POWER8 :
0003ffff80000000-0003ffff800fffff (prio 0, i/o): icp-0
0003ffff80008000-0003ffff80008fff (prio 0, i/o): icp-thread
0003ffff80010000-0003ffff80010fff (prio 0, i/o): icp-thread
0003ffff80018000-0003ffff80018fff (prio 0, i/o): icp-thread
0003ffff80020000-0003ffff80020fff (prio 0, i/o): icp-thread
0003ffff80028000-0003ffff80028fff (prio 0, i/o): icp-thread
0003ffff80030000-0003ffff80030fff (prio 0, i/o): icp-thread
0003ffff80048000-0003ffff80048fff (prio 0, i/o): icp-thread
0003ffff80050000-0003ffff80050fff (prio 0, i/o): icp-threadBut, I missed that part :
+ if (val & PPC_BIT(35)) { /* SCOM addressing */
+ uint32_t addr = val >> 32;
+ uint32_t reg = addr & 0xfff;
and pnv_get_vcpu_by_xscom_base() is a bit of a hack. That's why you need a backpointer to the chip which is always a bit suspicious for a sub unit. An address space would be cleaner since writing to this register generates another transaction it seems.
Do you plan to support other registers than PC_TOD ?
The part handling "SCOM addressing" deserves its own patch IMO. It would clarifies how the logic works and the modeling.
[ ... ]
Quoted 20 lines
Quoted 11 lines
Quoted 11 lines
hw/ppc/pnv_chiptod.c new file +11
-0+static void pnv_chiptod_realize (DeviceState *dev,Error **errp) +{ +static bool got_primary = false; +static bool got_secondary = false; + +PnvChipTOD *chiptod =PNV_CHIPTOD (dev); +PnvChipTODClass *pctc =PNV_CHIPTOD_GET_CLASS (chiptod); + +if (!got_primary) { + got_primary = true; + chiptod->primary = true;Quoted 8 lines
hw/ppc/pnv_chiptod.c line 452
Quoted 5 lines
hw/ppc/pnv_chiptod.c new file +5
-0+ chiptod->pss_mss_ctrl_reg |=PPC_BIT (1 );/* TOD is master */ + }else if (!got_secondary) { + got_secondary = true; + chiptod->secondary = true; + }It would be cleaner to introduce "primary" and "secondary" properties defined by the model realizing the PnvChipTOD objects.
Quoted 1 line
Hum. There's a few hard-coded primaries on
chip_id== 0 already
XCSOM doesn't have a QOM model but it should be done that way. XIVE is bit special because the TIMA is an overlapping mapping on all chips.
Quoted 3 lines
in the tree. I don't know how closely related they are, chiptod can set other chips as primary AFAIK but maybe it just makes sense to make primary a chip property.
It really shouldn't be too much work, mostly setting properties in the chip realize routine :
object_property_set_bool(OBJECT(&chip10->chiptod), "primary",
chip->chip_id == 0, &error_abort);
object_property_set_bool(OBJECT(&chip10->chiptod), "secondary",
chip->chip_id == 1, &error_abort);
if (!qdev_realize(DEVICE(&chip10->chiptod), NULL, errp)) {
return;
}Quoted 3 lines
I might dig a bit more into what we (and other IBM firmware) want to test or expect with these primaries. Maybe another pass to tidy all that up can be done.
Thanks,
C.
Re: [PATCH 3/4] target/ppc: add TFMR SPR implementation with read and write helpers
In reply to Nicholas PigginQuoted 8 lines · Nicholas Piggin
TFMR is the Time Facility Management Register which is specific to POWER CPUs, and used for the purpose of timebase management (generally by firmware, not the OS).
This adds an initial simple TFMR register, which will form part of the core timebase facility model in the next patch.
Signed-off-by Nicholas Piggin <[email protected]>
Thanks,
C.
Quoted 92 lines
--- target/ppc/cpu_init.c | 2 +- target/ppc/helper.h | 2 ++ target/ppc/spr_common.h | 2 ++ target/ppc/timebase_helper.c | 13 +++++++++++++ target/ppc/translate.c | 10 ++++++++++ 5 files changed, 28 insertions(+), 1 deletion(-)target/ppc/cpu_init.c +1
-1@@ -5662,7 +5662,7 @@ static void register_power_common_book4_sprs(CPUPPCState *env)spr_register_hv (env,SPR_TFMR ,"TFMR" ,SPR_NOACCESS ,SPR_NOACCESS ,SPR_NOACCESS ,SPR_NOACCESS , - &spr_read_generic, &spr_write_generic, + &spr_read_tfmr, &spr_write_tfmr,0x00000000 ); spr_register_hv(env, SPR_LDBAR, "LDBAR", SPR_NOACCESS, SPR_NOACCESS,target/ppc/helper.h +2
-0@@ -746,6 +746,8 @@ DEF_HELPER_2(store_40x_dbcr0, void, env, tl) DEF_HELPER_2(store_40x_sler, void, env, tl) DEF_HELPER_FLAGS_2(store_booke_tcr, TCG_CALL_NO_RWG, void, env, tl) DEF_HELPER_FLAGS_2(store_booke_tsr, TCG_CALL_NO_RWG, void, env, tl) +DEF_HELPER_1(load_tfmr, tl, env) +DEF_HELPER_2(store_tfmr, void, env, tl) DEF_HELPER_3(store_ibatl, void, env, i32, tl) DEF_HELPER_3(store_ibatu, void, env, i32, tl)DEF_HELPER_3 (store_dbatl, void, env, i32, tl)target/ppc/spr_common.h +2
-0@@ -196,6 +196,8 @@ void spr_write_ebb(DisasContext *ctx, int sprn, int gprn);void spr_read_ebb_upper32 (DisasContext *ctx,int gprn,int sprn);void spr_write_ebb_upper32 (DisasContext *ctx,int sprn,int gprn);void spr_write_hmer (DisasContext *ctx,int sprn,int gprn); +void spr_read_tfmr (DisasContext *ctx,int gprn,int sprn); +void spr_write_tfmr (DisasContext *ctx,int sprn,int gprn);void spr_write_lpcr (DisasContext *ctx,int sprn,int gprn);void spr_read_dexcr_ureg (DisasContext *ctx,int gprn,int sprn);#endif target/ppc/timebase_helper.c +13
-0@@ -270,6 +270,19 @@ void helper_store_booke_tsr(CPUPPCState *env, target_ulong val)store_booke_tsr (env, val); } +#if defined(TARGET_PPC64 ) +/* POWER processor Timebase Facility */ +target_ulong helper_load_tfmr (CPUPPCState *env) +{ +return env->spr[SPR_TFMR ]; +} + +void helper_store_tfmr (CPUPPCState *env,target_ulong val) +{ + env->spr[SPR_TFMR ] = val; +} +#endif +/*****************************************************************************/ /* Embedded PowerPC specific helpers */ target/ppc/translate.c +10
-0@@ -1255,6 +1255,16 @@ void spr_write_hmer(DisasContext *ctx, int sprn, int gprn)spr_store_dump_spr (sprn); } +void spr_read_tfmr (DisasContext *ctx,int gprn,int sprn) +{ +gen_helper_load_tfmr (cpu_gpr[gprn], cpu_env); +} + +void spr_write_tfmr (DisasContext *ctx,int sprn,int gprn) +{ +gen_helper_store_tfmr (cpu_env, cpu_gpr[gprn]); +} +void spr_write_lpcr (DisasContext *ctx,int sprn,int gprn) {gen_helper_store_lpcr (cpu_env, cpu_gpr[gprn]);
Quoted 30 lines · Nicholas Piggin
Quoted 29 lines · Cédric Le Goater
Quoted 26 lines · Nicholas Piggin
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved totimebase_helper.cfrommisc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.
Quoted 7 lines
Quoted 3 lines
The chiptod units are not exposed to the OS, it is all handled at FW level AFAIK. Could the OPAL people provide some feedback on the low level models ?
Well, no takers so far. I guess I'm a OPAL people :)
Quoted 1 line
I did some of the P10 chiptod addressing in skiboot, at least. This
patch does work with the skiboot chiptod driver at least.
cool, with 2 chips ?
Quoted 4 lines
I would eventually like to add in the ability to actually change the TB with it, and inject errors and generate HMIs because that's an area that seem to be a bit lacking (most of such testing seemed to be done on real hardware using special time facility corruption injection).
The MCE injection was a nice addon
https://lore.kernel.org/qemu-devel/[email protected]/
https://lore.kernel.org/qemu-devel/[email protected]/
I hope it will get more attention if you resend.
Quoted 3 lines
But yes for now it is a bit difficult to verify it does much useful aside from booting skiboot (+ patch to enable chiptod on QEMU I posted recently).
It's difficult to review PATCH 4 without some good knowledge of HW. I know you do but you can not review your own patches ! That said, the impact is limited to PowerNV machines, I guess we are fine.
Thanks,
C.
Re: [PATCH 4/4] target/ppc: Implement core timebase state machine and TFMR
In reply to Nicholas PigginQuoted 11 lines · Nicholas Piggin
This implements the core timebase state machine, which is the core side of the time-of-day system in POWER processors. This facility is operated by control fields in the TFMR register, which also contains status fields.
The core timebase interacts with the chiptod hardware, primarily to receive TOD updates, to synchronise timebase with other cores. This model does not actually update TB values with TOD or updates received from the chiptod, as timebases are always synchronised. It does step through the states required to perform the update.
Quoted 7 lines
There are several asynchronous state transitions. These are modelled using using mfTFMR to drive state changes, because it is expected that firmware poll the register to wait for those states. This is good enough to test basic firmware behaviour without adding real timers. The values chosen are arbitrary.
Signed-off-by Nicholas Piggin <[email protected]>
Looks correct,
Thanks,
C.
Quoted 221 lines
--- target/ppc/cpu.h | 34 ++++++++ target/ppc/timebase_helper.c | 147 ++++++++++++++++++++++++++++++++++- 2 files changed, 179 insertions(+), 2 deletions(-)target/ppc/cpu.h +34
-0@@ -1177,6 +1177,13 @@ struct CPUArchState {/* PowerNV chiptod / timebase facility state. */ int tb_ready_for_tod;/* core TB ready to receive TOD from chiptod */ int tod_sent_to_tb;/* chiptod sent TOD to the core TB */ + +/* +* Timers for async events are simulated by mfTFAC because TFAC is to be +* polled for event. +*/ +int tb_state_timer; +int tb_sync_pulse_timer;#endif #endif @@ -2527,6 +2534,33 @@ enum { HMER_XSCOM_STATUS_MASK = PPC_BITMASK(21, 23), }; +/* TFMR */ +enum { +TFMR_CONTROL_MASK =PPC_BITMASK (0 ,24 ), +TFMR_MASK_HMI =PPC_BIT (10 ), +TFMR_TB_ECLIPZ =PPC_BIT (14 ), +TFMR_LOAD_TOD_MOD =PPC_BIT (16 ), +TFMR_MOVE_CHIP_TOD_TO_TB =PPC_BIT (18 ), +TFMR_CLEAR_TB_ERRORS =PPC_BIT (24 ), +TFMR_STATUS_MASK =PPC_BITMASK (25 ,63 ), +TFMR_TBST_ENCODED =PPC_BITMASK (28 ,31 ),/* TBST = TB State */ +TFMR_TBST_LAST =PPC_BITMASK (32 ,35 ),/* Previous TBST */ +TFMR_TB_ENABLED =PPC_BIT (40 ), +TFMR_TB_VALID =PPC_BIT (41 ), +TFMR_TB_SYNC_OCCURED =PPC_BIT (42 ), +}; + +/* TFMR TBST */ +enum { +TBST_RESET =0x0 , +TBST_SEND_TOD_MOD =0x1 , +TBST_NOT_SET =0x2 , +TBST_SYNC_WAIT =0x6 , +TBST_GET_TOD =0x7 , +TBST_TB_RUNNING =0x8 , +TBST_TB_ERROR =0x9 , +}; +/*****************************************************************************/ #define is_isa300 (ctx) (!!(ctx->insns_flags2 & PPC2_ISA300))target/ppc/timebase_helper.c +145
-2@@ -272,14 +272,157 @@ void helper_store_booke_tsr(CPUPPCState *env, target_ulong val)#if defined(TARGET_PPC64 )/* POWER processor Timebase Facility */ +static unsigned int tfmr_get_tb_state (uint64_t tfmr) +{ +return (tfmr &TFMR_TBST_ENCODED ) >> (63 -31 ); +} + +static uint64_t tfmr_new_tb_state (uint64_t tfmr,unsigned int tbst) +{ + tfmr &= ~TFMR_TBST_LAST ; + tfmr |= (tfmr &TFMR_TBST_ENCODED ) >>4 ;/* move state to last state */ + tfmr &= ~TFMR_TBST_ENCODED ; + tfmr |= (uint64_t )tbst << (63 -31 );/* move new state to state */ + +if (tbst ==TBST_TB_RUNNING ) { + tfmr |=TFMR_TB_VALID ; + }else { + tfmr &= ~TFMR_TB_VALID ; + } + +return tfmr; +} + +static void tb_state_machine_step (CPUPPCState *env) +{ +uint64_t tfmr = env->spr[SPR_TFMR ]; +unsigned int tbst =tfmr_get_tb_state (tfmr); + +if (!(tfmr &TFMR_TB_ECLIPZ ) || tbst ==TBST_TB_ERROR ) { +return ; + } + +if (env->tb_sync_pulse_timer) { + env->tb_sync_pulse_timer--; + }else { + tfmr |=TFMR_TB_SYNC_OCCURED ; + env->spr[SPR_TFMR ] = tfmr; + } + +if (env->tb_state_timer) { + env->tb_state_timer--; +return ; + } + +if (tfmr &TFMR_LOAD_TOD_MOD ) { + tfmr &= ~TFMR_LOAD_TOD_MOD ; +if (tbst ==TBST_GET_TOD ) { + tfmr =tfmr_new_tb_state (tfmr,TBST_TB_ERROR ); + }else { + tfmr =tfmr_new_tb_state (tfmr,TBST_SEND_TOD_MOD ); +/* State seems to transition immediately */ + tfmr =tfmr_new_tb_state (tfmr,TBST_NOT_SET ); + } + }else if (tfmr &TFMR_MOVE_CHIP_TOD_TO_TB ) { +if (tbst ==TBST_SYNC_WAIT ) { + tfmr =tfmr_new_tb_state (tfmr,TBST_GET_TOD ); + env->tb_state_timer =3 ; + }else if (tbst ==TBST_GET_TOD ) { +if (env->tod_sent_to_tb) { + tfmr =tfmr_new_tb_state (tfmr,TBST_TB_RUNNING ); + tfmr &= ~TFMR_MOVE_CHIP_TOD_TO_TB ; + env->tb_ready_for_tod =0 ; + env->tod_sent_to_tb =0 ; + } + }else { +qemu_log_mask (LOG_GUEST_ERROR ,"TFMR error: MOVE_CHIP_TOD_TO_TB " +"state machine in invalid state 0x%x\n" , tbst); + tfmr =tfmr_new_tb_state (tfmr,TBST_TB_ERROR ); + env->tb_ready_for_tod =0 ; + } + } + + env->spr[SPR_TFMR ] = tfmr; +} +target_ulong helper_load_tfmr (CPUPPCState *env) { -return env->spr[SPR_TFMR ]; +tb_state_machine_step (env); + +return env->spr[SPR_TFMR ] |TFMR_TB_ECLIPZ ; }void helper_store_tfmr (CPUPPCState *env,target_ulong val) { - env->spr[SPR_TFMR ] = val; +uint64_t tfmr = env->spr[SPR_TFMR ]; +unsigned int tbst =tfmr_get_tb_state (tfmr); + +if (!(val &TFMR_TB_ECLIPZ )) { +qemu_log_mask (LOG_UNIMP ,"TFMR non-ECLIPZ mode not implemented\n" ); + tfmr &= ~TFMR_TBST_ENCODED ; + tfmr &= ~TFMR_TBST_LAST ; + goto out; + } + +/* Update control bits */ + tfmr = (tfmr & ~TFMR_CONTROL_MASK ) | (val &TFMR_CONTROL_MASK ); + +/* mtspr always clears this */ + tfmr &= ~TFMR_TB_SYNC_OCCURED ; + env->tb_sync_pulse_timer =1 ; + +/* +* We don't implement any of the error status bits that can be +* cleared by writing 1 to them. TB error injection / simulation +* would have to implement some. +* +* Also don't implement mfTB failing when the TB state machine is +* not running. +*/ + +if (((tfmr | val) & (TFMR_LOAD_TOD_MOD |TFMR_MOVE_CHIP_TOD_TO_TB )) == + (TFMR_LOAD_TOD_MOD |TFMR_MOVE_CHIP_TOD_TO_TB )) { +qemu_log_mask (LOG_GUEST_ERROR ,"TFMR error: LOAD_TOD_MOD and " +"MOVE_CHIP_TOD_TO_TB both set\n" ); + tfmr =tfmr_new_tb_state (tfmr,TBST_TB_ERROR ); + env->tb_ready_for_tod =0 ; + goto out; + } + +if (tfmr &TFMR_CLEAR_TB_ERRORS ) { + tfmr =tfmr_new_tb_state (tfmr,TBST_RESET ); + tfmr &= ~TFMR_CLEAR_TB_ERRORS ; + tfmr &= ~TFMR_LOAD_TOD_MOD ; + tfmr &= ~TFMR_MOVE_CHIP_TOD_TO_TB ; + env->tb_ready_for_tod =0 ; + env->tod_sent_to_tb =0 ; + goto out; + } + +if (tbst ==TBST_TB_ERROR ) { +qemu_log_mask (LOG_GUEST_ERROR ,"TFMR error: mtspr TFMR in TB_ERROR" +" state\n" ); +return ; + } + +if (tfmr &TFMR_LOAD_TOD_MOD ) { + env->tb_state_timer =3 ; + }else if (tfmr &TFMR_MOVE_CHIP_TOD_TO_TB ) { +if (tbst ==TBST_NOT_SET ) { + tfmr =tfmr_new_tb_state (tfmr,TBST_SYNC_WAIT ); + env->tb_ready_for_tod =1 ; + env->tb_state_timer =3 ; + }else { +qemu_log_mask (LOG_GUEST_ERROR ,"TFMR error: MOVE_CHIP_TOD_TO_TB " +"not in TB not set state 0x%x\n" , + tbst); + tfmr =tfmr_new_tb_state (tfmr,TBST_TB_ERROR ); + env->tb_ready_for_tod =0 ; + } + } + +out: + env->spr[SPR_TFMR ] = tfmr; }#endif
Quoted 35 lines · Cédric Le Goater
Quoted 34 lines · Nicholas Piggin
Quoted 32 lines · Cédric Le Goater
Quoted 20 lines · Nicholas Piggin
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Quoted 6 lines
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved totimebase_helper.cfrommisc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.The chiptod units are not exposed to the OS, it is all handled at FW level AFAIK. Could the OPAL people provide some feedback on the low level models ?
Quoted 5 lines
Quoted 3 lines
Well, no takers so far. I guess I'm a OPAL people :)
Quoted 1 line
I did some of the P10 chiptod addressing in skiboot, at least. This
patch does work with the skiboot chiptod driver at least.
cool, with 2 chips ?
Yes it worked with 2 chips.
Quoted 11 lines
Quoted 4 lines
I would eventually like to add in the ability to actually change the TB with it, and inject errors and generate HMIs because that's an area that seem to be a bit lacking (most of such testing seemed to be done on real hardware using special time facility corruption injection).
The MCE injection was a nice addon
https://lore.kernel.org/qemu-devel/[email protected]/ https://lore.kernel.org/qemu-devel/[email protected]/I hope it will get more attention if you resend.
Will take another look.
Quoted 7 lines
Quoted 3 lines
But yes for now it is a bit difficult to verify it does much useful aside from booting skiboot (+ patch to enable chiptod on QEMU I posted recently).
It's difficult to review PATCH 4 without some good knowledge of HW. I know you do but you can not review your own patches ! That said, the impact is limited to PowerNV machines, I guess we are fine.
Yeah. I appreciate all the review so far. It's pretty complicated even with the workbook. I might be able to add a simpler and higher-level description of the basic init sequence and states. You would still have to trust me, but it might make it easier to see what's going on.
Thanks,
Nick
[ ... ]
Quoted 1 line
Yes it worked with 2 chips.
I will give the next series a try.
[ ... ]
Quoted 8 lines
Quoted 3 lines
It's difficult to review PATCH 4 without some good knowledge of HW. I know you do but you can not review your own patches ! That said, the impact is limited to PowerNV machines, I guess we are fine.
Yeah. I appreciate all the review so far. It's pretty complicated even with the workbook. I might be able to add a simpler and higher-level description of the basic init sequence and states. You would still have to trust me, but it might make it easier to see what's going on.
Sure. Patches 2-4 are acked. I am only looking at patch 1 now.
Thanks,
C.
Quoted 31 lines · Nicholas Piggin
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved totimebase_helper.cfrommisc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.There is still a vast amount that is not modeled, but most of it related to error handling, injection, failover, etc that is very complicated and not required for normal operation.
Quoted 8 lines
Thanks,
NickNicholas Piggin (4):
pnv/chiptod: Add POWER9/10 chiptod model
target/ppc: Tidy POWER book4 SPR registration target/ppc: add TFMR SPR implementation with read and write helpers target/ppc: Implement core timebase state machine and TFMR
patch 2-4 could be merged in the next PR. Could you please rebase on ppc-next and resend ?
Then we still have 2+ weeks to polish pnv/chiptod which would be a nice addition to QEMU 8.1.
Thanks,
C.
Quoted 20 lines
hw/ppc/meson.build | 1 + hw/ppc/pnv.c | 38 +++ hw/ppc/pnv_chiptod.c | 488 +++++++++++++++++++++++++++++++++++ hw/ppc/pnv_xscom.c | 2 + hw/ppc/trace-events | 4 + include/hw/ppc/pnv_chip.h | 3 + include/hw/ppc/pnv_chiptod.h | 64 +++++ include/hw/ppc/pnv_core.h | 3 + include/hw/ppc/pnv_xscom.h | 9 + target/ppc/cpu.h | 40 +++ target/ppc/cpu_init.c | 92 ++++--- target/ppc/helper.h | 2 + target/ppc/spr_common.h | 2 + target/ppc/timebase_helper.c | 156 +++++++++++ target/ppc/translate.c | 10 + 15 files changed, 882 insertions(+), 32 deletions(-)create mode 100644
hw/ppc/pnv_chiptod.c
create mode 100644include/hw/ppc/pnv_chiptod.h
Quoted 35 lines · Cédric Le Goater
Quoted 34 lines · Nicholas Piggin
This adds support for chiptod and core timebase state machine models in the powernv POWER9 and POWER10 models.
This does not actually change the time or the value in TB registers (because they are alrady synced in QEMU), but it does go through the motions. It is enough to be able to run skiboot's chiptod initialisation code that synchronises core timebases (after a patch to prevent skiboot skipping chiptod for QEMU, posted to skiboot mailing list).
Sorry there was some delay since the last posting. There is a bit more interest in this recently but feedback and comments from RFC was not forgotten and is much appreciated.
https://lists.gnu.org/archive/html/qemu-ppc/2022-08/msg00324.html
I think I accounted for everything except moving register defines to the .h file. I'm on the fence about that but if they are only used in the .c file I think it's okay to keep them there for now. I cut out a lot of unused ones so it's not so cluttered now.
Lots of other changes and fixes since that RFC. Notably:
- Register names changed to match the workbook names instead of skiboot.
- TFMR moved totimebase_helper.cfrommisc_helper.c
- More comprehensive model and error checking, particularly of TFMR.
- POWER10 with multi-chip support.
- chiptod and core timebase linked via specific state instead of TFMR.There is still a vast amount that is not modeled, but most of it related to error handling, injection, failover, etc that is very complicated and not required for normal operation.
Thanks,
Nick
Quoted 8 lines
Quoted 5 lines
Nicholas Piggin (4):
pnv/chiptod: Add POWER9/10 chiptod model
target/ppc: Tidy POWER book4 SPR registration target/ppc: add TFMR SPR implementation with read and write helpers target/ppc: Implement core timebase state machine and TFMRpatch 2-4 could be merged in the next PR. Could you please rebase on ppc-next and resend ?
Good idea, I have a couple of other minor register additions that depend on patch 1 too.
Quoted 2 lines
Then we still have 2+ weeks to polish pnv/chiptod which would be a nice addition to QEMU 8.1.
Yeah. Been trying to get to it... Hopefully pseries SMT is close now so I will have some more time.
Thanks,
Nick