Add support for Supervisor Software Events extension
v2 cover Add support for Supervisor Software Events extension
The SBI Supervisor Software Events (SSE) extensions provides a mechanism to inject software events from an SBI implementation to supervisor software such that it preempts all other supervisor level traps and interrupts [1]. This RFC implements the V2 version of the spec.
Various events are defined and can be send asynchronously to supervisor software (RAS, PMU, DEBUG, Asynchronous page fault) from SBI as well as platform specific events. Events can be either local (per-hart) or global. Events can be nested on top of each other based on priority and can interrupt the supervisor mode at any time.
This PR adds support for this extension. First commit modifies interrupts handling to pass the sbi traps regs through interrupt handlers in order to inject SSE events from this context. Second one add the SSE support itself and the final two commits add SSE support for PMU interrupts.
---
Changes from v1:
- Implemented SSE Spec v2
- Change event state checking (Xiang W)
v1: http://lists.infradead.org/pipermail/opensbi/2023-November/006015.html
Cl?ment L?ger (4):
lib: sbi: provides regs to sbi_ipi_process()
lib: sbi: add support for Supervisor Software Events extension
lib: sbi: trigger SSE PMU event on PMU overflow IRQ
lib: sbi: add SBI_EXT_PMU_IRQ_CLEAR
include/sbi/sbi_ecall_interface.h | 49 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_ipi.h | 6 +- include/sbi/sbi_pmu.h | 5 + include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_pmu.c | 3 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 14 +- lib/sbi/sbi_pmu.c | 49 ++ lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ lib/sbi/sbi_tlb.c | 2 +- lib/sbi/sbi_trap.c | 10 +- lib/utils/irqchip/imsic.c | 2 +- 16 files changed, 1380 insertions(+), 14 deletions(-) create mode 100644 include/sbi/sbi_sse.h create mode 100644 lib/sbi/sbi_ecall_sse.c create mode 100644 lib/sbi/sbi_sse.c
Signature · 1 line
2.43.0
RFC PATCH v2 1/4 lib: sbi: provides regs to sbi_ipi_process()
In order to implement SSE on IPI notifications, provides regs to sbi_ipi_process(). This will be used in two cases, global event triggered from one cpu to another and ipi injection to a specific hart.
--- include/sbi/sbi_ipi.h | 6 ++++-- lib/sbi/sbi_ipi.c | 12 +++++++----- lib/sbi/sbi_tlb.c | 2 +- lib/sbi/sbi_trap.c | 4 ++-- lib/utils/irqchip/imsic.c | 2 +- 5 files changed, 15 insertions(+), 11 deletions(-)
include/sbi/sbi_ipi.h +4 -2
@@ -11,6 +11,7 @@#define __SBI_IPI_H__#include <sbi/sbi_types.h> +#include <sbi/sbi_trap.h> /* clang-format off */
@@ -68,7 +69,8 @@ struct sbi_ipi_event_ops {* Note: This is a mandatory callback and it is called on the * remote HART after IPI is triggered. */ -void (* process)(struct sbi_scratch *scratch); +void (* process)(struct sbi_scratch *scratch, +struct sbi_trap_regs *regs); };int sbi_ipi_send_many (ulong hmask,ulong hbase,u32 event,void *data);
@@ -83,7 +85,7 @@ void sbi_ipi_clear_smode(void);int sbi_ipi_send_halt (ulong hmask,ulong hbase); -void sbi_ipi_process (void ); +void sbi_ipi_process (struct sbi_trap_regs *regs);int sbi_ipi_raw_send (u32 hartindex);
lib/sbi/sbi_ipi.c +7 -5
@@ -186,7 +186,8 @@ void sbi_ipi_event_destroy(u32 event) ipi_ops_array[event] =NULL ; } -static void sbi_ipi_process_smode (struct sbi_scratch *scratch) +static void sbi_ipi_process_smode (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) {csr_set (CSR_MIP ,MIP_SSIP ); }
@@ -208,7 +209,8 @@ void sbi_ipi_clear_smode(void)csr_clear (CSR_MIP ,MIP_SSIP ); } -static void sbi_ipi_process_halt (struct sbi_scratch *scratch) +static void sbi_ipi_process_halt (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) {sbi_hsm_hart_stop (scratch, true); }
@@ -225,7 +227,7 @@ int sbi_ipi_send_halt(ulong hmask, ulong hbase)return sbi_ipi_send_many (hmask, hbase, ipi_halt_event,NULL ); } -void sbi_ipi_process (void ) +void sbi_ipi_process (struct sbi_trap_regs *regs) {unsigned long ipi_type;unsigned int ipi_event;
@@ -244,7 +246,7 @@ void sbi_ipi_process(void)if (ipi_type &1UL ) { ipi_ops = ipi_ops_array[ipi_event];if (ipi_ops) - ipi_ops->process (scratch); + ipi_ops->process (scratch, regs); } ipi_type = ipi_type >>1 ; ipi_event++;
@@ -349,7 +351,7 @@ void sbi_ipi_exit(struct sbi_scratch *scratch)csr_clear (CSR_MIE ,MIP_MSIP );/* Process pending IPIs */ -sbi_ipi_process (); +sbi_ipi_process (NULL );/* Platform exit */ sbi_platform_ipi_exit (sbi_platform_ptr (scratch));
lib/sbi/sbi_tlb.c +1 -1
@@ -240,7 +240,7 @@ static bool tlb_process_once(struct sbi_scratch *scratch)return false; } -static void tlb_process (struct sbi_scratch *scratch) +static void tlb_process (struct sbi_scratch *scratch,struct sbi_trap_regs *regs) {while (tlb_process_once (scratch)); }
lib/sbi/sbi_trap.c +2 -2
@@ -206,7 +206,7 @@ static int sbi_trap_nonaia_irq(struct sbi_trap_regs *regs, ulong mcause)sbi_timer_process ();break ;case IRQ_M_SOFT : -sbi_ipi_process (); +sbi_ipi_process (regs);break ;case IRQ_M_EXT :return sbi_irqchip_process (regs);
@@ -229,7 +229,7 @@ static int sbi_trap_aia_irq(struct sbi_trap_regs *regs, ulong mcause)sbi_timer_process ();break ;case IRQ_M_SOFT : -sbi_ipi_process (); +sbi_ipi_process (regs);break ;case IRQ_M_EXT : rc =sbi_irqchip_process (regs);
lib/utils/irqchip/imsic.c +1 -1
@@ -149,7 +149,7 @@ static int imsic_external_irqfn(struct sbi_trap_regs *regs)switch (mirq) {case IMSIC_IPI_ID : -sbi_ipi_process (); +sbi_ipi_process (regs);break ;default : sbi_printf("%s: unhandled IRQ%d\n",
Signature · 1 line
2.43.0
RFC PATCH v2 1/4 lib: sbi: provides regs to sbi_ipi_process()
Quoted 6 lines · On Tue, Jan 9, 2024 at 4:15?PM Cl?ment L?ger <[email protected]> wrote:
In order to implement SSE on IPI notifications, provides regs to
sbi_ipi_process(). This will be used in two cases, global event triggered from one cpu to another and ipi injection to a specific hart.Signed-off-by Cl?ment L?ger <[email protected]>
LGTM.
Regards,
Anup
Quoted 146 lines
--- include/sbi/sbi_ipi.h | 6 ++++-- lib/sbi/sbi_ipi.c | 12 +++++++----- lib/sbi/sbi_tlb.c | 2 +- lib/sbi/sbi_trap.c | 4 ++-- lib/utils/irqchip/imsic.c | 2 +- 5 files changed, 15 insertions(+), 11 deletions(-)include/sbi/sbi_ipi.h +4
-2@@ -11,6 +11,7 @@#define __SBI_IPI_H__#include <sbi/sbi_types.h> +#include <sbi/sbi_trap.h> /* clang-format off */ @@ -68,7 +69,8 @@ struct sbi_ipi_event_ops {* Note: This is a mandatory callback and it is called on the * remote HART after IPI is triggered. */ -void (* process)(struct sbi_scratch *scratch); +void (* process)(struct sbi_scratch *scratch, +struct sbi_trap_regs *regs); };int sbi_ipi_send_many (ulong hmask,ulong hbase,u32 event,void *data);@@ -83,7 +85,7 @@ void sbi_ipi_clear_smode(void);int sbi_ipi_send_halt (ulong hmask,ulong hbase); -void sbi_ipi_process (void ); +void sbi_ipi_process (struct sbi_trap_regs *regs);int sbi_ipi_raw_send (u32 hartindex);lib/sbi/sbi_ipi.c +7
-5@@ -186,7 +186,8 @@ void sbi_ipi_event_destroy(u32 event) ipi_ops_array[event] =NULL ; } -static void sbi_ipi_process_smode (struct sbi_scratch *scratch) +static void sbi_ipi_process_smode (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) {csr_set (CSR_MIP ,MIP_SSIP ); }@@ -208,7 +209,8 @@ void sbi_ipi_clear_smode(void)csr_clear (CSR_MIP ,MIP_SSIP ); } -static void sbi_ipi_process_halt (struct sbi_scratch *scratch) +static void sbi_ipi_process_halt (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) {sbi_hsm_hart_stop (scratch, true); }@@ -225,7 +227,7 @@ int sbi_ipi_send_halt(ulong hmask, ulong hbase)return sbi_ipi_send_many (hmask, hbase, ipi_halt_event,NULL ); } -void sbi_ipi_process (void ) +void sbi_ipi_process (struct sbi_trap_regs *regs) {unsigned long ipi_type;unsigned int ipi_event;@@ -244,7 +246,7 @@ void sbi_ipi_process(void)if (ipi_type &1UL ) { ipi_ops = ipi_ops_array[ipi_event];if (ipi_ops) - ipi_ops->process (scratch); + ipi_ops->process (scratch, regs); } ipi_type = ipi_type >>1 ; ipi_event++;@@ -349,7 +351,7 @@ void sbi_ipi_exit(struct sbi_scratch *scratch)csr_clear (CSR_MIE ,MIP_MSIP );/* Process pending IPIs */ -sbi_ipi_process (); +sbi_ipi_process (NULL );/* Platform exit */ sbi_platform_ipi_exit (sbi_platform_ptr (scratch));lib/sbi/sbi_tlb.c +1
-1@@ -240,7 +240,7 @@ static bool tlb_process_once(struct sbi_scratch *scratch)return false; } -static void tlb_process (struct sbi_scratch *scratch) +static void tlb_process (struct sbi_scratch *scratch,struct sbi_trap_regs *regs) {while (tlb_process_once (scratch)); }lib/sbi/sbi_trap.c +2
-2@@ -206,7 +206,7 @@ static int sbi_trap_nonaia_irq(struct sbi_trap_regs *regs, ulong mcause)sbi_timer_process ();break ;case IRQ_M_SOFT : -sbi_ipi_process (); +sbi_ipi_process (regs);break ;case IRQ_M_EXT :return sbi_irqchip_process (regs);@@ -229,7 +229,7 @@ static int sbi_trap_aia_irq(struct sbi_trap_regs *regs, ulong mcause)sbi_timer_process ();break ;case IRQ_M_SOFT : -sbi_ipi_process (); +sbi_ipi_process (regs);break ;case IRQ_M_EXT : rc =sbi_irqchip_process (regs);lib/utils/irqchip/imsic.c +1
-2@@ -149,7 +149,7 @@ static int imsic_external_irqfn(struct sbi_trap_regs *regs)switch (mirq) {case IMSIC_IPI_ID : -sbi_ipi_process (); +sbi_ipi_process (regs);break ;default : sbi_printf("%s: unhandled IRQ%d\n", --2.43.0
--
opensbi mailing list
opensbi at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
In reply to Clément LégerThis extension [1] allows to deliver events from SBI to supervisor via a software mecanism. This extensions defines events (either local or global) which are signaled by the SBI on specific signal sources (IRQ, traps, etc) and are injected to be executed in supervisor mode.
[1] https://lists.riscv.org/g/tech-prs/message/744
--- include/sbi/sbi_ecall_interface.h | 48 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 2 +- lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ 9 files changed, 1301 insertions(+), 3 deletions(-) create mode 100644 include/sbi/sbi_sse.h create mode 100644 lib/sbi/sbi_ecall_sse.c create mode 100644 lib/sbi/sbi_sse.c
include/sbi/sbi_ecall_interface.h +47 -1
@@ -32,6 +32,7 @@#define SBI_EXT_DBCN 0x4442434E#define SBI_EXT_SUSP 0x53555350#define SBI_EXT_CPPC 0x43505043 +#define SBI_EXT_SSE 0x535345/* SBI function IDs for BASE extension*/ #define SBI_EXT_BASE_GET_SPEC_VERSION 0x0
@@ -293,6 +294,48 @@ enum sbi_cppc_reg_id { SBI_CPPC_NON_ACPI_LAST = SBI_CPPC_TRANSITION_LATENCY, }; +/* SBI Function IDs for SSE extension */ +#define SBI_EXT_SSE_READ_ATTR 0x00000000 +#define SBI_EXT_SSE_WRITE_ATTR 0x00000001 +#define SBI_EXT_SSE_REGISTER 0x00000002 +#define SBI_EXT_SSE_UNREGISTER 0x00000003 +#define SBI_EXT_SSE_ENABLE 0x00000004 +#define SBI_EXT_SSE_DISABLE 0x00000005 +#define SBI_EXT_SSE_COMPLETE 0x00000006 +#define SBI_EXT_SSE_INJECT 0x00000007 + +/* SBI SSE Event Attributes. */ +#define SBI_SSE_ATTR_STATUS 0x00000000 +#define SBI_SSE_ATTR_PRIO 0x00000001 +#define SBI_SSE_ATTR_CONFIG 0x00000002 +#define SBI_SSE_ATTR_PREFERRED_HART 0x00000003 +#define SBI_SSE_ATTR_ENTRY_PC 0x00000004 +#define SBI_SSE_ATTR_ENTRY_A0 0x00000005 +#define SBI_SSE_ATTR_ENTRY_A6 0x00000006 +#define SBI_SSE_ATTR_ENTRY_A7 0x00000007 +#define SBI_SSE_ATTR_INTERRUPTED_MODE 0x00000008 +#define SBI_SSE_ATTR_INTERRUPTED_PC 0x00000009 +#define SBI_SSE_ATTR_INTERRUPTED_A0 0x0000000A +#define SBI_SSE_ATTR_INTERRUPTED_A6 0x0000000B +#define SBI_SSE_ATTR_INTERRUPTED_A7 0x0000000C + +#define SBI_SSE_ATTR_STATUS_STATE_OFFSET 0 +#define SBI_SSE_ATTR_STATUS_STATE_MASK 0x3 +#define SBI_SSE_ATTR_STATUS_PENDING_OFFSET 2 +#define SBI_SSE_ATTR_STATUS_INJECT_OFFSET 3 + +#define SBI_SSE_ATTR_CONFIG_ONESHOT (1 << 0) + +/* SBI SSE Event IDs. */ +#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000 +#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000 +#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000 +#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000 +#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000 + +#define SBI_SSE_EVENT_GLOBAL (1 << 15) +#define SBI_SSE_EVENT_PLATFORM (1 << 14) +/* SBI base specification related macros */ #define SBI_SPEC_VERSION_MAJOR_OFFSET 24#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f
@@ -313,8 +356,11 @@ enum sbi_cppc_reg_id {#define SBI_ERR_ALREADY_STARTED -7#define SBI_ERR_ALREADY_STOPPED -8#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_BUSY -12 -#define SBI_LAST_ERR SBI_ERR_NO_SHMEM +#define SBI_LAST_ERR SBI_ERR_BUSY/* clang-format on */
include/sbi/sbi_error.h +3 -1
@@ -24,6 +24,9 @@#define SBI_EALREADY_STARTED SBI_ERR_ALREADY_STARTED#define SBI_EALREADY_STOPPED SBI_ERR_ALREADY_STOPPED#define SBI_ENO_SHMEM SBI_ERR_NO_SHMEM +#define SBI_EINVALID_STATE SBI_ERR_INVALID_STATE +#define SBI_EBAD_RANGE SBI_ERR_BAD_RANGE +#define SBI_EBUSY SBI_ERR_BUSY#define SBI_ENODEV -1000#define SBI_ENOSYS -1001
@@ -34,7 +37,6 @@#define SBI_ENOMEM -1006#define SBI_EUNKNOWN -1007#define SBI_ENOENT -1008 -/* clang-format on */ #endif
include/sbi/sbi_sse.h new file +93 -0
93 lines not on this page · raw message
lib/sbi/Kconfig +4 -0
@@ -46,4 +46,8 @@ config SBI_ECALL_VENDOR bool "Platform-defined vendor extensions" default y +config SBI_ECALL_SSE + bool "SSE extension" + default y + endmenu
lib/sbi/objects.mk +4 -0
@@ -52,6 +52,9 @@ libsbi-objs-$(CONFIG_SBI_ECALL_LEGACY) += sbi_ecall_legacy.o carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_VENDOR) += ecall_vendor libsbi-objs-$(CONFIG_SBI_ECALL_VENDOR) += sbi_ecall_vendor.o +carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_SSE) += ecall_sse +libsbi-objs-$(CONFIG_SBI_ECALL_SSE) += sbi_ecall_sse.o + libsbi-objs-y += sbi_bitmap.o libsbi-objs-y += sbi_bitops.o libsbi-objs-y += sbi_console.o
@@ -71,6 +74,7 @@ libsbi-objs-y += sbi_misaligned_ldst.o libsbi-objs-y += sbi_platform.o libsbi-objs-y += sbi_pmu.o libsbi-objs-y += sbi_scratch.o +libsbi-objs-y += sbi_sse.o libsbi-objs-y += sbi_string.o libsbi-objs-y += sbi_system.o libsbi-objs-y += sbi_timer.o
lib/sbi/sbi_ecall_sse.c new file +58 -0
58 lines not on this page · raw message
lib/sbi/sbi_init.c +13 -0
@@ -23,6 +23,7 @@#include <sbi/sbi_irqchip.h> #include <sbi/sbi_platform.h> #include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> #include <sbi/sbi_system.h> #include <sbi/sbi_string.h> #include <sbi/sbi_timer.h>
@@ -315,6 +316,12 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, true); +if (rc) { +sbi_printf ("%s: sse init failed (error %d)\n" , __func__, rc); +sbi_hart_hang (); + } + rc =sbi_pmu_init (scratch, true);if (rc) { sbi_printf("%s: pmu init failed (error %d)\n",
@@ -435,6 +442,10 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, false); +if (rc) +sbi_hart_hang (); + rc =sbi_pmu_init (scratch, false);if (rc)sbi_hart_hang ();
@@ -639,6 +650,8 @@ void __noreturn sbi_exit(struct sbi_scratch *scratch)sbi_platform_early_exit (plat); +sbi_sse_exit (scratch); +sbi_pmu_exit (scratch);sbi_timer_exit (scratch);
lib/sbi/sbi_ipi.c +1 -1
@@ -66,7 +66,7 @@ static int sbi_ipi_send(struct sbi_scratch *scratch, u32 remote_hartindex,* SBI_IPI_UPDATE_BREAK for self-IPIs. For other events, check * for self-IPI and execute the callback directly here. */ - ipi_ops->process (scratch); + ipi_ops->process (scratch,NULL );return 0 ; }
lib/sbi/sbi_sse.c new file +1078 -0
1,078 lines not on this page · raw message
Signature · 1 line
2.43.0
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
lib/sbi/sbi_sse.c line 965
Quoted 11 lines · Cl?ment L?ger
lib/sbi/sbi_sse.c new file +11
-0+static int sse_global_init () +{ +struct sbi_sse_event *e; +unsigned int i, ev =0 ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) + global_event_count++; +else + local_event_count++; + }
nit: through out patch series global is used for global events. Here local event count and global event count both are setup. Although only global events are initialized.
Perhaps its better to take out event counting from here and put in sbi_sse_init or function/macro of it's own.
Quoted 24 lines
lib/sbi/sbi_sse.c new file +19
-0+ + global_events =sbi_zalloc (sizeof (*global_events) * global_event_count); +if (!global_events) +return SBI_ENOMEM ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (!EVENT_IS_GLOBAL (supported_events[i])) +continue ; + + e = &global_events[ev]; +sse_event_init (e, supported_events[i]); +SPIN_LOCK_INIT (e->lock); + + ev++; + } + +return 0 ; +} +Signature · 2 lines
-- 2.43.0
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
Quoted 10 lines · Deepak Gupta
Quoted 9 lines · Cl?ment L?ger
(quoted diff) +9
-0+static int sse_global_init() +{ +??? struct sbi_sse_event *e; +??? unsigned int i, ev = 0; + +??? for (i = 0; i < EVENT_COUNT; i++) { +??????? if (EVENT_IS_GLOBAL(supported_events[i])) +??????????? global_event_count++; +??????? else
Quoted 8 lines
Quoted 2 lines
+??????????? local_event_count++;
+??? }nit: through out patch series
globalis used for global events. Here local event count and global event count both are setup. Although only global events are initialized.
Perhaps its better to take out event counting from here and put insbi_sse_initor function/macro of it's own.
Oh yes clearly that would be less misleading. I'll do that.
Thanks,
Cl?ment
Quoted 27 lines
Quoted 27 lines
+
+???global_events=sbi_zalloc(sizeof(*global_events) * global_event_count);
+??? if (!global_events)
+??????? returnSBI_ENOMEM;(quoted diff) +15
-1+ +??? for (i = 0; i < EVENT_COUNT; i++) { +??????? if (!EVENT_IS_GLOBAL(supported_events[i])) +??????????? continue; + +??????? e = &global_events[ev]; +??????? sse_event_init(e, supported_events[i]); +??????? SPIN_LOCK_INIT(e->lock); + +??????? ev++; +??? } + +??? return 0; +} + --?2.43.0
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
In reply to Clément LégerQuoted 3 lines · On Tue, Jan 9, 2024 at 4:15?PM Cl?ment L?ger <[email protected]> wrote:
This extension [1] allows to deliver events from SBI to supervisor via a software mecanism. This extensions defines events (either local or
s/mecanism/mechanism/
Quoted 6 lines
global) which are signaled by the SBI on specific signal sources (IRQ, traps, etc) and are injected to be executed in supervisor mode.
[1] https://lists.riscv.org/g/tech-prs/message/744
Signed-off-by Cl?ment L?ger <[email protected]>
Please split this patch into two parts:
1) Add generic SSE framework
2) Add SBI SSE implementation
Regards,
Anup
Quoted 1,459 lines
--- include/sbi/sbi_ecall_interface.h | 48 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 2 +- lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ 9 files changed, 1301 insertions(+), 3 deletions(-) create mode 100644 include/sbi/sbi_sse.h create mode 100644 lib/sbi/sbi_ecall_sse.c create mode 100644 lib/sbi/sbi_sse.cinclude/sbi/sbi_ecall_interface.h +47
-1@@ -32,6 +32,7 @@#define SBI_EXT_DBCN 0x4442434E#define SBI_EXT_SUSP 0x53555350#define SBI_EXT_CPPC 0x43505043 +#define SBI_EXT_SSE 0x535345/* SBI function IDs for BASE extension*/ #define SBI_EXT_BASE_GET_SPEC_VERSION 0x0@@ -293,6 +294,48 @@ enum sbi_cppc_reg_id { SBI_CPPC_NON_ACPI_LAST = SBI_CPPC_TRANSITION_LATENCY, }; +/* SBI Function IDs for SSE extension */ +#define SBI_EXT_SSE_READ_ATTR 0x00000000 +#define SBI_EXT_SSE_WRITE_ATTR 0x00000001 +#define SBI_EXT_SSE_REGISTER 0x00000002 +#define SBI_EXT_SSE_UNREGISTER 0x00000003 +#define SBI_EXT_SSE_ENABLE 0x00000004 +#define SBI_EXT_SSE_DISABLE 0x00000005 +#define SBI_EXT_SSE_COMPLETE 0x00000006 +#define SBI_EXT_SSE_INJECT 0x00000007 + +/* SBI SSE Event Attributes. */ +#define SBI_SSE_ATTR_STATUS 0x00000000 +#define SBI_SSE_ATTR_PRIO 0x00000001 +#define SBI_SSE_ATTR_CONFIG 0x00000002 +#define SBI_SSE_ATTR_PREFERRED_HART 0x00000003 +#define SBI_SSE_ATTR_ENTRY_PC 0x00000004 +#define SBI_SSE_ATTR_ENTRY_A0 0x00000005 +#define SBI_SSE_ATTR_ENTRY_A6 0x00000006 +#define SBI_SSE_ATTR_ENTRY_A7 0x00000007 +#define SBI_SSE_ATTR_INTERRUPTED_MODE 0x00000008 +#define SBI_SSE_ATTR_INTERRUPTED_PC 0x00000009 +#define SBI_SSE_ATTR_INTERRUPTED_A0 0x0000000A +#define SBI_SSE_ATTR_INTERRUPTED_A6 0x0000000B +#define SBI_SSE_ATTR_INTERRUPTED_A7 0x0000000C + +#define SBI_SSE_ATTR_STATUS_STATE_OFFSET 0 +#define SBI_SSE_ATTR_STATUS_STATE_MASK 0x3 +#define SBI_SSE_ATTR_STATUS_PENDING_OFFSET 2 +#define SBI_SSE_ATTR_STATUS_INJECT_OFFSET 3 + +#define SBI_SSE_ATTR_CONFIG_ONESHOT (1 << 0) + +/* SBI SSE Event IDs. */ +#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000 +#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000 +#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000 +#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000 +#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000 + +#define SBI_SSE_EVENT_GLOBAL (1 << 15) +#define SBI_SSE_EVENT_PLATFORM (1 << 14) +/* SBI base specification related macros */ #define SBI_SPEC_VERSION_MAJOR_OFFSET 24#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f@@ -313,8 +356,11 @@ enum sbi_cppc_reg_id {#define SBI_ERR_ALREADY_STARTED -7#define SBI_ERR_ALREADY_STOPPED -8#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_BUSY -12 -#define SBI_LAST_ERR SBI_ERR_NO_SHMEM +#define SBI_LAST_ERR SBI_ERR_BUSY/* clang-format on */ include/sbi/sbi_error.h +3
-1@@ -24,6 +24,9 @@#define SBI_EALREADY_STARTED SBI_ERR_ALREADY_STARTED#define SBI_EALREADY_STOPPED SBI_ERR_ALREADY_STOPPED#define SBI_ENO_SHMEM SBI_ERR_NO_SHMEM +#define SBI_EINVALID_STATE SBI_ERR_INVALID_STATE +#define SBI_EBAD_RANGE SBI_ERR_BAD_RANGE +#define SBI_EBUSY SBI_ERR_BUSY#define SBI_ENODEV -1000#define SBI_ENOSYS -1001@@ -34,7 +37,6 @@#define SBI_ENOMEM -1006#define SBI_EUNKNOWN -1007#define SBI_ENOENT -1008 -/* clang-format on */ #endif include/sbi/sbi_sse.h new file +93
-0@@ -0,0 +1,93 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems. +*/ + +#ifndef __SBI_SSE_H__ +#define __SBI_SSE_H__ + +#include <sbi/sbi_types.h> +#include <sbi/sbi_list.h> +#include <sbi/riscv_locks.h> + +struct sbi_scratch ; + +#define EXC_MODE_PP_SHIFT 0 +#define EXC_MODE_PP BIT(EXC_MODE_PP_SHIFT) +#define EXC_MODE_PV_SHIFT 1 +#define EXC_MODE_PV BIT(EXC_MODE_PV_SHIFT) +#define EXC_MODE_SSTATUS_SPIE_SHIFT 2 +#define EXC_MODE_SSTATUS_SPIE BIT(EXC_MODE_SSTATUS_SPIE_SHIFT) + + +struct sbi_sse_cb_ops { +/** +* Called when hart_id is changed on the event. +*/ +void (*set_hartid_cb)(uint32_t event_id,unsigned long hart_id); + +/** +* Called when the SBI_EXT_SSE_COMPLETE is invoked on the event. +*/ +void (*complete_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_REGISTER is invoked on the event. +*/ +void (*register_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_UNREGISTER is invoked on the event. +*/ +void (*unregister_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_ENABLE is invoked on the event. +*/ +void (*enable_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_DISABLE is invoked on the event. +*/ +void (*disable_cb)(uint32_t event_id); +}; + +/* Set the callback operations for an event +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param cb_ops Callback operations +* @return 0 on success, error otherwise +*/ +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops); + +/* Inject an event to the current hard +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param regs Registers that were used on SBI entry +* @return 0 on success, error otherwise +*/ +int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs); + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot); +void sbi_sse_exit (struct sbi_scratch *scratch); + +/* Interface called from sbi_ecall_sse.c */ +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7); +int sbi_sse_unregister (uint32_t event_id); +int sbi_sse_enable (uint32_t event_id); +int sbi_sse_disable (uint32_t event_id); +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hart_id, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi); +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi); + +#endif lib/sbi/Kconfig +4
-0@@ -46,4 +46,8 @@ config SBI_ECALL_VENDOR bool "Platform-defined vendor extensions" default y +config SBI_ECALL_SSE + bool "SSE extension" + default y + endmenulib/sbi/objects.mk +4
-0@@ -52,6 +52,9 @@ libsbi-objs-$(CONFIG_SBI_ECALL_LEGACY) += sbi_ecall_legacy.o carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_VENDOR) += ecall_vendor libsbi-objs-$(CONFIG_SBI_ECALL_VENDOR) += sbi_ecall_vendor.o +carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_SSE) += ecall_sse +libsbi-objs-$(CONFIG_SBI_ECALL_SSE) += sbi_ecall_sse.o + libsbi-objs-y += sbi_bitmap.o libsbi-objs-y += sbi_bitops.o libsbi-objs-y += sbi_console.o@@ -71,6 +74,7 @@ libsbi-objs-y += sbi_misaligned_ldst.o libsbi-objs-y += sbi_platform.o libsbi-objs-y += sbi_pmu.o libsbi-objs-y += sbi_scratch.o +libsbi-objs-y += sbi_sse.o libsbi-objs-y += sbi_string.o libsbi-objs-y += sbi_system.o libsbi-objs-y += sbi_timer.olib/sbi/sbi_ecall_sse.c new file +58
-0@@ -0,0 +1,58 @@ +#include <sbi/sbi_error.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_trap.h> +#include <sbi/sbi_sse.h> + +static int sbi_ecall_sse_handler (unsigned long extid,unsigned long funcid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; + +switch (funcid) { +case SBI_EXT_SSE_READ_ATTR : + ret =sbi_sse_read_attrs (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a4); +break ; +case SBI_EXT_SSE_WRITE_ATTR : + ret =sbi_sse_write_attrs (regs->a0, regs->a1, regs->a2, + regs->a3, regs->a4); +break ; +case SBI_EXT_SSE_REGISTER : + ret =sbi_sse_register (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a2); +break ; +case SBI_EXT_SSE_UNREGISTER : + ret =sbi_sse_unregister (regs->a0); +break ; +case SBI_EXT_SSE_ENABLE : + ret =sbi_sse_enable (regs->a0); +break ; +case SBI_EXT_SSE_DISABLE : + ret =sbi_sse_disable (regs->a0); +break ; +case SBI_EXT_SSE_COMPLETE : + ret =sbi_sse_complete (regs->a0, regs, out); +break ; +case SBI_EXT_SSE_INJECT : + ret =sbi_sse_inject_from_ecall (regs->a0, regs->a1, regs, out); +break ; +default : + ret =SBI_ENOTSUPP ; + } +return ret; +} + +struct sbi_ecall_extension ecall_sse; + +static int sbi_ecall_sse_register_extensions (void ) +{ +return sbi_ecall_register_extension (&ecall_sse); +} + +struct sbi_ecall_extension ecall_sse = { + .extid_start =SBI_EXT_SSE , + .extid_end =SBI_EXT_SSE , + .register_extensions = sbi_ecall_sse_register_extensions, + .handle = sbi_ecall_sse_handler, +};lib/sbi/sbi_init.c +13
-0@@ -23,6 +23,7 @@#include <sbi/sbi_irqchip.h> #include <sbi/sbi_platform.h> #include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> #include <sbi/sbi_system.h> #include <sbi/sbi_string.h> #include <sbi/sbi_timer.h> @@ -315,6 +316,12 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, true); +if (rc) { +sbi_printf ("%s: sse init failed (error %d)\n" , __func__, rc); +sbi_hart_hang (); + } + rc =sbi_pmu_init (scratch, true);if (rc) { sbi_printf("%s: pmu init failed (error %d)\n",@@ -435,6 +442,10 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, false); +if (rc) +sbi_hart_hang (); + rc =sbi_pmu_init (scratch, false);if (rc)sbi_hart_hang ();@@ -639,6 +650,8 @@ void __noreturn sbi_exit(struct sbi_scratch *scratch)sbi_platform_early_exit (plat); +sbi_sse_exit (scratch); +sbi_pmu_exit (scratch);sbi_timer_exit (scratch);lib/sbi/sbi_ipi.c +1
-1@@ -66,7 +66,7 @@ static int sbi_ipi_send(struct sbi_scratch *scratch, u32 remote_hartindex,* SBI_IPI_UPDATE_BREAK for self-IPIs. For other events, check * for self-IPI and execute the callback directly here. */ - ipi_ops->process (scratch); + ipi_ops->process (scratch,NULL );return 0 ; }lib/sbi/sbi_sse.c new file +1078
-1@@ -0,0 +1,1078 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems Inc. +* +*/ + +#include <sbi/riscv_asm.h> +#include <sbi/riscv_barrier.h> +#include <sbi/riscv_encoding.h> +#include <sbi/riscv_locks.h> +#include <sbi/sbi_domain.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_ecall_interface.h> +#include <sbi/sbi_error.h> +#include <sbi/sbi_fifo.h> +#include <sbi/sbi_hart.h> +#include <sbi/sbi_heap.h> +#include <sbi/sbi_hsm.h> +#include <sbi/sbi_ipi.h> +#include <sbi/sbi_list.h> +#include <sbi/sbi_platform.h> +#include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> +#include <sbi/sbi_scratch.h> +#include <sbi/sbi_string.h> +#include <sbi/sbi_trap.h> + +#include <sbi/sbi_console.h> + +#define sse_get_hart_state_ptr (__scratch) \ + sbi_scratch_read_type((__scratch), void *, shs_ptr_off) + +#define sse_thishart_state_ptr () \ + sse_get_hart_state_ptr(sbi_scratch_thishart_ptr()) + +#define sse_set_hart_state_ptr (__scratch, __sse_state) \ + sbi_scratch_write_type((__scratch), void *, shs_ptr_off, (__sse_state)) + +/* +* Rather than using memcpy to copy the context (which does it byte-per-byte), +* copy each field which generates ld/lw. +*/ +#define sse_regs_copy (dst, src) \ + (dst)->a0 = (src)->a0; \ + (dst)->a6 = (src)->a6; \ + (dst)->a7 = (src)->a7 + +#define EVENT_IS_GLOBAL (__event_id) ((__event_id) & SBI_SSE_EVENT_GLOBAL) + +static const uint32_t supported_events[] = +{ +SBI_SSE_EVENT_LOCAL_RAS , +SBI_SSE_EVENT_GLOBAL_RAS , +SBI_SSE_EVENT_LOCAL_PMU , +SBI_SSE_EVENT_LOCAL_SOFTWARE , +SBI_SSE_EVENT_GLOBAL_SOFTWARE , +}; + +#define EVENT_COUNT array_size(supported_events) + +#define sse_event_invoke_cb (_event, _cb, ...) \ +{ \ + if (_event->cb_ops && _event->cb_ops->_cb) \ + _event->cb_ops->_cb(_event->event_id, ##__VA_ARGS__); \ +} + +#define SSE_EVENT_STATE (__e) __e->attrs.status.state +#define SSE_EVENT_PENDING (__e) __e->attrs.status.pending +#define SSE_EVENT_CAN_INJECT (__e) __e->attrs.status.inject +#define SSE_EVENT_HARTID (__e) __e->attrs.hartid +#define SSE_EVENT_PRIO (__e) __e->attrs.prio +#define SSE_EVENT_CONFIG (__e) __e->attrs.config +#define SSE_EVENT_ENTRY (__e) __e->attrs.entry +#define SSE_EVENT_INTERRUPTED (__e) __e->attrs.interrupted + +struct sse_entry_state { +/** Entry program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +struct sse_interrupted_state { +/** Exception mode */ +unsigned long exc_mode; +/** Interrupted program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +enum sbi_sse_state { +SSE_STATE_UNUSED =0 , +SSE_STATE_REGISTERED =1 , +SSE_STATE_ENABLED =2 , +SSE_STATE_RUNNING =3 , +}; + +struct sse_ipi_inject_data { +uint32_t event_id; +}; + +struct sbi_sse_event_status { +unsigned long state:2 ; +unsigned long pending:1 ; +unsigned long inject:1 ; +} __packed; + +struct sbi_sse_event_attrs { +struct sbi_sse_event_status status; +unsigned long prio; +unsigned long config; +unsigned long hartid; +struct sse_entry_state entry; +struct sse_interrupted_state interrupted; +}; + +/* Make sure all attributes are packed for direct memcpy in ATTR_READ */ +#define assert_field_offset (field, attr_offset) \ + _Static_assert( \ + ((offsetof(struct sbi_sse_event_attrs, field)) / sizeof(unsigned long)) == attr_offset, \ + "field "#field " from struct sbi_sse_event_attrs invalid offset, expected "#attr_offset \ + ) + +assert_field_offset (status,SBI_SSE_ATTR_STATUS ); +assert_field_offset (prio,SBI_SSE_ATTR_PRIO ); +assert_field_offset (config,SBI_SSE_ATTR_CONFIG ); +assert_field_offset (hartid,SBI_SSE_ATTR_PREFERRED_HART ); +assert_field_offset (entry.pc,SBI_SSE_ATTR_ENTRY_PC ); +assert_field_offset (entry.a0,SBI_SSE_ATTR_ENTRY_A0 ); +assert_field_offset (entry.a6,SBI_SSE_ATTR_ENTRY_A6 ); +assert_field_offset (entry.a7,SBI_SSE_ATTR_ENTRY_A7 ); +assert_field_offset (interrupted.exc_mode,SBI_SSE_ATTR_INTERRUPTED_MODE ); +assert_field_offset (interrupted.pc,SBI_SSE_ATTR_INTERRUPTED_PC ); +assert_field_offset (interrupted.a0,SBI_SSE_ATTR_INTERRUPTED_A0 ); +assert_field_offset (interrupted.a6,SBI_SSE_ATTR_INTERRUPTED_A6 ); +assert_field_offset (interrupted.a7,SBI_SSE_ATTR_INTERRUPTED_A7 ); + +struct sbi_sse_event { +struct sbi_sse_event_attrs attrs; +uint32_t event_id; +const struct sbi_sse_cb_ops *cb_ops; +struct sbi_dlist node; +/* Only global events are using the lock, local ones don't need it */ +spinlock_t lock; +}; + +struct sse_hart_state { +struct sbi_dlist event_list; +spinlock_t list_lock; +struct sbi_sse_event *local_events; +}; + +static unsigned int local_event_count; +static unsigned int global_event_count; +static struct sbi_sse_event *global_events; + +static unsigned long sse_inject_fifo_off; +static unsigned long sse_inject_fifo_mem_off; +/* Offset of pointer to SSE HART state in scratch space */ +static unsigned long shs_ptr_off; + +static u32 sse_ipi_inject_event =SBI_IPI_EVENT_MAX ; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id); + +static bool sse_event_is_global (struct sbi_sse_event *e) +{ +return EVENT_IS_GLOBAL (e->event_id); +} + +static void sse_global_event_lock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_lock (&e->lock); +} + +static void sse_global_event_unlock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_unlock (&e->lock); +} + +static void sse_event_set_state (struct sbi_sse_event *e, +enum sbi_sse_state new_state) +{ +enum sbi_sse_state prev_state =SSE_EVENT_STATE (e); + +if ((new_state - prev_state ==1 ) || (prev_state - new_state ==1 )) { +SSE_EVENT_STATE (e) = new_state; +return ; + } + +sbi_panic ("Invalid SSE state transition: %d -> %d\n" , prev_state, + new_state); +} + +static struct sbi_sse_event *sse_event_get (uint32_t event) +{ +unsigned int i; +struct sbi_sse_event *events, *e; +unsigned int count; +struct sse_hart_state *shs; + +if (EVENT_IS_GLOBAL (event)) { + count = global_event_count; + events = global_events; + }else { + count = local_event_count; + shs =sse_thishart_state_ptr (); + events = shs->local_events; + } + +for (i =0 ; i < count; i++) { + e = &events[i]; +if (e->event_id == event) +return e; + } + +return NULL ; +} + +static struct sse_hart_state *sse_event_get_hart_state (struct sbi_sse_event *e) +{ +struct sbi_scratch *s =sbi_hartid_to_scratch (SSE_EVENT_HARTID (e)); + +return sse_get_hart_state_ptr (s); +} + +static void sse_event_remove_from_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); + +spin_lock (&state->list_lock); +sbi_list_del (&e->node); +spin_unlock (&state->list_lock); +} + +static void sse_event_add_to_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); +struct sbi_sse_event *tmp; + +spin_lock (&state->list_lock); +sbi_list_for_each_entry (tmp, &state->event_list, node) { +if (SSE_EVENT_PRIO (e) <SSE_EVENT_PRIO (tmp)) +break ; +if (SSE_EVENT_PRIO (e) ==SSE_EVENT_PRIO (tmp) && + e->event_id < tmp->event_id) +break ; + } +sbi_list_add_tail (&e->node, &tmp->node); + +spin_unlock (&state->list_lock); +} + +static int sse_event_disable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, disable_cb); + +sse_event_remove_from_list (e); +sse_event_set_state (e,SSE_STATE_REGISTERED ); + +return SBI_OK ; +} + +static int sse_event_set_hart_id_check (struct sbi_sse_event *e, +unsigned long new_hartid) +{ +int hstate; +unsigned int hartid = (uint32_t ) new_hartid; +struct sbi_domain * hd =sbi_domain_thishart_ptr (); + +if (!sse_event_is_global (e)) +return SBI_EDENIED ; + +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) +return SBI_EBUSY ; + +if (!sbi_domain_is_assigned_hart (hd, new_hartid)) +return SBI_EINVAL ; + + hstate =sbi_hsm_hart_get_state (hd, hartid); +if (hstate !=SBI_HSM_STATE_STARTED ) +return SBI_EINVAL ; + +return SBI_OK ; +} + +static int sse_event_set_attr_check (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +int ret =SBI_OK ; + +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +if (val & ~SBI_SSE_ATTR_CONFIG_ONESHOT ) + ret =SBI_ERR_INVALID_PARAM ; +break ; +case SBI_SSE_ATTR_PRIO : +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) + ret =SBI_EINVALID_STATE ; +break ; +case SBI_SSE_ATTR_PREFERRED_HART : + ret =sse_event_set_hart_id_check (e, val); +break ; +default : + ret =SBI_EBAD_RANGE ; +break ; + } + +return ret; +} + +static void sse_event_set_attr (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +SSE_EVENT_CONFIG (e) = val; +break ; +case SBI_SSE_ATTR_PRIO : +SSE_EVENT_PRIO (e) = (uint32_t )val; +break ; +case SBI_SSE_ATTR_PREFERRED_HART : +SSE_EVENT_HARTID (e) = val; +sse_event_invoke_cb (e, set_hartid_cb, val); +break ; + } +} + +static int sse_event_register (struct sbi_sse_event *e, +unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +if (sse_event_is_global (e) &&SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EINVAL ; + +if (SSE_EVENT_STATE (e) !=SSE_STATE_UNUSED ) +return SBI_EINVALID_STATE ; + +SSE_EVENT_ENTRY (e).a0 = handler_entry_a0; +SSE_EVENT_ENTRY (e).a6 = handler_entry_a6; +SSE_EVENT_ENTRY (e).a7 = handler_entry_a7; +SSE_EVENT_ENTRY (e).pc = handler_entry_pc; + +sse_event_set_state (e,SSE_STATE_REGISTERED ); + +sse_event_invoke_cb (e, register_cb); + +return 0 ; +} + +static int sse_event_unregister (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, unregister_cb); + +sse_event_set_state (e,SSE_STATE_UNUSED ); + +return 0 ; +} + +static void sse_event_inject (struct sbi_sse_event *e, +struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +ulong prev_smode, prev_virt; +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); +struct sse_interrupted_state *prev_i_ctx; +struct sse_entry_state *e_ctx = &SSE_EVENT_ENTRY (e); + +sse_event_set_state (e,SSE_STATE_RUNNING ); +SSE_EVENT_PENDING (e) = false; + +if (prev_e) { +/* back-to-back injection after another event, copy previous +* event context for correct restoration +*/ + prev_i_ctx = &SSE_EVENT_INTERRUPTED (prev_e); + +sse_regs_copy (i_ctx, prev_i_ctx); + i_ctx->exc_mode = prev_i_ctx->exc_mode; + i_ctx->pc = prev_i_ctx->pc; + }else { +sse_regs_copy (i_ctx, regs); + + prev_smode = (regs->mstatus &MSTATUS_MPP ) >>MSTATUS_MPP_SHIFT ; +#if __riscv_xlen ==32 + prev_virt = (regs->mstatusH &MSTATUSH_MPV ) ?1 :0 ; +#else + prev_virt = (regs->mstatus &MSTATUS_MPV ) ?1 :0 ; +#endif + + i_ctx->exc_mode = prev_smode <<EXC_MODE_PP_SHIFT ; + i_ctx->exc_mode |= prev_virt <<EXC_MODE_PV_SHIFT ; +if (regs->mstatus &MSTATUS_SPIE ) + i_ctx->exc_mode |=EXC_MODE_SSTATUS_SPIE ; + i_ctx->pc = regs->mepc; + +/* We only want to set SPIE for the first event injected after +* entering M-Mode. For the event injected right after another +* event (after calling sse_event_complete(), we will keep the +* saved SPIE). +*/ + regs->mstatus &= ~MSTATUS_SPIE ; +if (regs->mstatus &MSTATUS_SIE ) + regs->mstatus |=MSTATUS_SPIE ; + } + +sse_regs_copy (regs, e_ctx); + regs->mepc = e_ctx->pc; + + regs->mstatus &= ~MSTATUS_MPP ; + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +#endif + + regs->mstatus &= ~MSTATUS_SIE ; +} + +static void sse_event_resume (struct sbi_sse_event *e,struct sbi_trap_regs *regs) +{ +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); + +sse_regs_copy (regs, i_ctx); + +/* Restore previous virtualization state */ +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatusH |=MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatus |=MSTATUS_MPV ; +#endif + + regs->mstatus &= ~MSTATUS_MPP ; +if (i_ctx->exc_mode &EXC_MODE_PP ) + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + + regs->mstatus &= ~MSTATUS_SIE ; +if (regs->mstatus &MSTATUS_SPIE ) + regs->mstatus |=MSTATUS_SIE ; + + regs->mstatus &= ~MSTATUS_SPIE ; +if (i_ctx->exc_mode &EXC_MODE_SSTATUS_SPIE ) + regs->mstatus |=MSTATUS_SPIE ; + + regs->mepc = i_ctx->pc; +} + +static bool sse_event_is_ready (struct sbi_sse_event *e) +{ +if (!SSE_EVENT_PENDING (e) ||SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED || +SSE_EVENT_HARTID (e) !=current_hartid ()) { +return false; + } + +return true; +} + +/* Return true if an event has been injected, false otherwise */ +static bool sse_process_pending_events (struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +struct sbi_sse_event *e, *to_run =NULL ; +struct sse_hart_state *state =sse_thishart_state_ptr (); + +retry: +spin_lock (&state->list_lock); + +if (sbi_list_empty (&state->event_list)) { +spin_unlock (&state->list_lock); +return false; + } + +sbi_list_for_each_entry (e, &state->event_list, node) { +/* +* List of event is ordered by priority, stop at first running +* event since all other events after this one are of lower +* priority. +*/ +if (SSE_EVENT_STATE (e) ==SSE_STATE_RUNNING ) +break ; + +if (sse_event_is_ready (e)) { + to_run = e; +break ; + } + } + +spin_unlock (&state->list_lock); + +if (!to_run) +return false; + +sse_global_event_lock (e); +/* +* If the event is global, the event might have been moved to another +* hart or disabled, evaluate readiness again. +*/ +if (sse_event_is_global (e) && !sse_event_is_ready (e)) { +sse_global_event_unlock (e); + goto retry; + } + +sse_event_inject (e, prev_e, regs); +sse_global_event_unlock (e); + +return true; +} + +static int sse_event_set_pending (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_RUNNING && +SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_ERR_INVALID_STATE ; + +SSE_EVENT_PENDING (e) = true; + +return SBI_OK ; +} + +static void sse_ipi_inject_process (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) +{ +struct sbi_sse_event *e; +struct sse_ipi_inject_data evt; +struct sbi_fifo *sse_inject_fifo_r = +sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + +/* This can be the case when sbi_exit() is called */ +if (!regs) +return ; + +/* Mark all queued events as pending */ +while (!sbi_fifo_dequeue (sse_inject_fifo_r, &evt)) { + e =sse_event_get (evt.event_id); +if (!e) +continue ; + +sse_global_event_lock (e); +sse_event_set_pending (e); +sse_global_event_unlock (e); + } + +sse_process_pending_events (NULL , regs); +} + +static struct sbi_ipi_event_ops sse_ipi_inject_ops = { + .name ="IPI_SSE_INJECT" , + .process = sse_ipi_inject_process, +}; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id) +{ +int ret; +struct sbi_scratch *remote_scratch =NULL ; +struct sse_ipi_inject_data evt = {event_id}; +struct sbi_fifo *sse_inject_fifo_r; + + remote_scratch =sbi_hartid_to_scratch (hartid); +if (!remote_scratch) +return SBI_EINVAL ; + sse_inject_fifo_r =sbi_scratch_offset_ptr (remote_scratch, + sse_inject_fifo_off); + + ret =sbi_fifo_enqueue (sse_inject_fifo_r, &evt); +if (ret) +return SBI_EFAIL ; + + ret =sbi_ipi_send_many (1 , hartid, sse_ipi_inject_event,NULL ); +if (ret) +return SBI_EFAIL ; + +return SBI_OK ; +} + +static int sse_inject_event (uint32_t event_id,unsigned long hartid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out,bool from_ecall) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +/* In case of global event, provided hart_id is ignored */ +if (sse_event_is_global (e)) + hartid =SSE_EVENT_HARTID (e); + +/* Event is for another hart, send it through IPI */ +if (hartid !=current_hartid ()) { +sse_global_event_unlock (e); +return sse_ipi_inject_send (hartid, event_id); + } + + ret =sse_event_set_pending (e); +sse_global_event_unlock (e); +if (ret) +return ret; + +if (from_ecall) { +/* Due to skip_regs_update = true being set if injection +* succeed an the fact that we need to modify regs here before +* injecting the event (regs are copied by the event), we need +* to set out->skip_regs_update to true to avoid. Moreover +* injection might not be done right now if another event of +* higher priority is already running so always set +* out->skip_regs_update to true there. +*/ + regs->mepc +=4 ; + regs->a0 =SBI_OK ; + out->skip_regs_update = true; + } + +if (sse_process_pending_events (NULL , regs)) + out->skip_regs_update = true; + +return SBI_OK ; +} + +static int sse_event_enable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_set_state (e,SSE_STATE_ENABLED ); +sse_event_add_to_list (e); + +if (SSE_EVENT_PENDING (e)) +sbi_ipi_send_many (1 ,SSE_EVENT_HARTID (e), sse_ipi_inject_event, +NULL ); + +sse_event_invoke_cb (e, enable_cb); + +return SBI_OK ; +} + +static int sse_event_complete (struct sbi_sse_event *e, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +bool inject; + +if (SSE_EVENT_STATE (e) !=SSE_STATE_RUNNING ) +return SBI_EINVALID_STATE ; + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EDENIED ; + +if (SSE_EVENT_CONFIG (e) &SBI_SSE_ATTR_CONFIG_ONESHOT ) +sse_event_disable (e); +else +sse_event_set_state (e,SSE_STATE_ENABLED ); + +sse_event_invoke_cb (e, complete_cb); + + inject =sse_process_pending_events (e, regs); +if (!inject) +sse_event_resume (e, regs); + + out->skip_regs_update = true; + +return SBI_OK ; +} + +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_complete (e, regs, out); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_enable (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_enable (e); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_disable (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_disable (e); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hartid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +if (!sbi_domain_is_assigned_hart (sbi_domain_thishart_ptr (), hartid)) +return SBI_EINVAL ; + +return sse_inject_event (event_id, hartid, regs, out, true); +} + +int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs) +{ +/* We don't really care about return value here */ +struct sbi_ecall_return out; + +return sse_inject_event (event_id,current_hartid (), regs, &out, false); +} + +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops) +{ +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +if (cb_ops->set_hartid_cb && !sse_event_is_global (e)) +return SBI_EINVAL ; + +sse_global_event_lock (e); + e->cb_ops = cb_ops; +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_attr_check (uint32_t base_attr_id,uint32_t attr_count, +unsigned long phys_lo, +unsigned long phys_hi, +unsigned long access) +{ +const unsigned align = __riscv_xlen >>3 ; +/* Avoid 32 bits overflow */ +uint64_t end_id = (uint64_t )base_attr_id + attr_count; + +if (end_id >SBI_SSE_ATTR_INTERRUPTED_A7 ) +return SBI_EBAD_RANGE ; + +if (phys_lo & (align -1 )) +return SBI_EINVALID_ADDR ; + +/* +* On RV32, the M-mode can only access the first 4GB of +* the physical address space because M-mode does not have +* MMU to access full 34-bit physical address space. +* +* Based on above, we simply fail if the upper 32bits of +* the physical address (i.e. a2 register) is non-zero on +* RV32. +*/ +if (phys_hi) +return SBI_EINVALID_ADDR ; + +if (!sbi_domain_check_addr_range (sbi_domain_thishart_ptr (), phys_lo, +sizeof (unsigned long ) * attr_count,1 , + access)) +return SBI_EINVALID_ADDR ; + +return SBI_OK ; +} + + +static void copy_attrs (unsigned long *out,const unsigned long *in, +unsigned int long_count) +{ +int i =0 ; + +/* +* sbi_memcpy() does byte-per-byte copy, using this yields long-per-long +* copy +*/ +for (i =0 ; i < long_count; i++) + out[i] = in[i]; +} + +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi) +{ +int ret; +unsigned long *e_attrs; +struct sbi_sse_event *e; +unsigned long *attrs; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, output_phys_lo, + output_phys_hi,SBI_DOMAIN_WRITE ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (output_phys_lo,sizeof (unsigned long ) * attr_count); + +/* +* Copy all attributes at once since struct sse_event_attrs is matching +* the SBI_SSE_ATTR_* attributes. While WRITE_ATTR attribute is not used +* in s-mode sse handling path, READ_ATTR is used to retrieve the value +* of registers when interrupted. rather than doing multiple SBI calls, +* a single one is done allowing to retrieve them all at once. +*/ + e_attrs = (unsigned long *)&e->attrs; + attrs = (unsigned long *)output_phys_lo; +copy_attrs (attrs, &e_attrs[base_attr_id], attr_count); + +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi) +{ +int ret; +struct sbi_sse_event *e; +unsigned long attr =0 , val; +uint32_t id, end_id = base_attr_id + attr_count; +unsigned long *attrs = (unsigned long *)input_phys_lo; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, input_phys_lo, + input_phys_hi,SBI_DOMAIN_READ ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (input_phys_lo,sizeof (unsigned long ) * attr_count); + +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; + ret =sse_event_set_attr_check (e, id, val); +if (ret) + goto out; + } + + attr =0 ; +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; +sse_event_set_attr (e, id, val); + } +out: +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_register (e, handler_entry_pc, handler_entry_a0, + handler_entry_a6, handler_entry_a7); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_unregister (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_unregister (e); +sse_global_event_unlock (e); + +return ret; +} + +static void sse_event_init (struct sbi_sse_event *e,uint32_t event_id) +{ + e->event_id = event_id; +SSE_EVENT_HARTID (e) =current_hartid (); +/* Declare all events as injectable */ +SSE_EVENT_CAN_INJECT (e) =1 ; +} + +static int sse_global_init () +{ +struct sbi_sse_event *e; +unsigned int i, ev =0 ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) + global_event_count++; +else + local_event_count++; + } + + global_events =sbi_zalloc (sizeof (*global_events) * global_event_count); +if (!global_events) +return SBI_ENOMEM ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (!EVENT_IS_GLOBAL (supported_events[i])) +continue ; + + e = &global_events[ev]; +sse_event_init (e, supported_events[i]); +SPIN_LOCK_INIT (e->lock); + + ev++; + } + +return 0 ; +} + +static void sse_local_init (struct sse_hart_state *shs) +{ +unsigned int i, ev =0 ; + +SBI_INIT_LIST_HEAD (&shs->event_list); +SPIN_LOCK_INIT (shs->list_lock); + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) +continue ; + +sse_event_init (&shs->local_events[ev++], supported_events[i]); + } +} + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot) +{ +int ret; +void *sse_inject_mem; +struct sse_hart_state *shs; +struct sbi_fifo *sse_inject_q; + +if (cold_boot) { + ret =sse_global_init (); +if (ret) +return ret; + + shs_ptr_off =sbi_scratch_alloc_offset (sizeof (void *)); +if (!shs_ptr_off) +return SBI_ENOMEM ; + + sse_inject_fifo_off =sbi_scratch_alloc_offset ( +sizeof (*sse_inject_q)); +if (!sse_inject_fifo_off) { +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + sse_inject_fifo_mem_off =sbi_scratch_alloc_offset ( +EVENT_COUNT *sizeof (struct sse_ipi_inject_data )); +if (!sse_inject_fifo_mem_off) { +sbi_scratch_free_offset (sse_inject_fifo_off); +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + ret =sbi_ipi_event_create (&sse_ipi_inject_ops); +if (ret <0 ) { +sbi_scratch_free_offset (shs_ptr_off); +return ret; + } + sse_ipi_inject_event = ret; + } + + shs =sse_get_hart_state_ptr (scratch); +if (!shs) { +/* Allocate per hart state and local events at once */ + shs =sbi_zalloc (sizeof (*shs) + +sizeof (struct sbi_sse_event ) * local_event_count); +if (!shs) +return SBI_ENOMEM ; + + shs->local_events = (struct sbi_sse_event *)(shs +1 ); + +sse_set_hart_state_ptr (scratch, shs); + } + +sse_local_init (shs); + + sse_inject_q =sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + sse_inject_mem =sbi_scratch_offset_ptr (scratch, + sse_inject_fifo_mem_off); + +sbi_fifo_init (sse_inject_q, sse_inject_mem,EVENT_COUNT , +sizeof (struct sse_ipi_inject_data )); + +return 0 ; +} + +void sbi_sse_exit (struct sbi_scratch *scratch) +{ +int i; +struct sbi_sse_event *e; + +for (i =0 ; i <EVENT_COUNT ; i++) { + e =sse_event_get (supported_events[i]); + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +continue ; + +if (SSE_EVENT_STATE (e) >SSE_STATE_REGISTERED ) +sbi_printf ("Event %d in invalid state at exit" , i); + } +} --2.43.0
--
opensbi mailing list
opensbi at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
In reply to Clément LégerQuoted 22 lines
lib/sbi/sbi_sse.c new file +22
-0+int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi) +{ +int ret; +unsigned long *e_attrs; +struct sbi_sse_event *e; +unsigned long *attrs; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, output_phys_lo, + output_phys_hi,SBI_DOMAIN_WRITE ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (output_phys_lo,sizeof (unsigned long ) * attr_count); +
lib/sbi/sbi_sse.c line 859
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+/* +* Copy all attributes at once since struct sse_event_attrs is matching +* the SBI_SSE_ATTR_* attributes. While WRITE_ATTR attribute is not used +* in s-mode sse handling path, READ_ATTR is used to retrieve the value +* of registers when interrupted. rather than doing multiple SBI calls, +* a single one is done allowing to retrieve them all at once. +*/ + e_attrs = (unsigned long *)&e->attrs; + attrs = (unsigned long *)output_phys_lo; +copy_attrs (attrs, &e_attrs[base_attr_id], attr_count);
I don't know how sbi_domain memory regions work. It looks like it's a recent addition.
Skimming through sources it look's like SBI own management of which physical regions are readable and writable.
Question:
Exposing get/set on physical memory region doesn't allow kernel to read/write arbitrary firmware regions too?
Quoted 222 lines
lib/sbi/sbi_sse.c new file +219
-1+ +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi) +{ +int ret; +struct sbi_sse_event *e; +unsigned long attr =0 , val; +uint32_t id, end_id = base_attr_id + attr_count; +unsigned long *attrs = (unsigned long *)input_phys_lo; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, input_phys_lo, + input_phys_hi,SBI_DOMAIN_READ ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (input_phys_lo,sizeof (unsigned long ) * attr_count); + +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; + ret =sse_event_set_attr_check (e, id, val); +if (ret) + goto out; + } + + attr =0 ; +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; +sse_event_set_attr (e, id, val); + } +out: +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_register (e, handler_entry_pc, handler_entry_a0, + handler_entry_a6, handler_entry_a7); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_unregister (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_unregister (e); +sse_global_event_unlock (e); + +return ret; +} + +static void sse_event_init (struct sbi_sse_event *e,uint32_t event_id) +{ + e->event_id = event_id; +SSE_EVENT_HARTID (e) =current_hartid (); +/* Declare all events as injectable */ +SSE_EVENT_CAN_INJECT (e) =1 ; +} + +static int sse_global_init () +{ +struct sbi_sse_event *e; +unsigned int i, ev =0 ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) + global_event_count++; +else + local_event_count++; + } + + global_events =sbi_zalloc (sizeof (*global_events) * global_event_count); +if (!global_events) +return SBI_ENOMEM ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (!EVENT_IS_GLOBAL (supported_events[i])) +continue ; + + e = &global_events[ev]; +sse_event_init (e, supported_events[i]); +SPIN_LOCK_INIT (e->lock); + + ev++; + } + +return 0 ; +} + +static void sse_local_init (struct sse_hart_state *shs) +{ +unsigned int i, ev =0 ; + +SBI_INIT_LIST_HEAD (&shs->event_list); +SPIN_LOCK_INIT (shs->list_lock); + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) +continue ; + +sse_event_init (&shs->local_events[ev++], supported_events[i]); + } +} + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot) +{ +int ret; +void *sse_inject_mem; +struct sse_hart_state *shs; +struct sbi_fifo *sse_inject_q; + +if (cold_boot) { + ret =sse_global_init (); +if (ret) +return ret; + + shs_ptr_off =sbi_scratch_alloc_offset (sizeof (void *)); +if (!shs_ptr_off) +return SBI_ENOMEM ; + + sse_inject_fifo_off =sbi_scratch_alloc_offset ( +sizeof (*sse_inject_q)); +if (!sse_inject_fifo_off) { +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + sse_inject_fifo_mem_off =sbi_scratch_alloc_offset ( +EVENT_COUNT *sizeof (struct sse_ipi_inject_data )); +if (!sse_inject_fifo_mem_off) { +sbi_scratch_free_offset (sse_inject_fifo_off); +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + ret =sbi_ipi_event_create (&sse_ipi_inject_ops); +if (ret <0 ) { +sbi_scratch_free_offset (shs_ptr_off); +return ret; + } + sse_ipi_inject_event = ret; + } + + shs =sse_get_hart_state_ptr (scratch); +if (!shs) { +/* Allocate per hart state and local events at once */ + shs =sbi_zalloc (sizeof (*shs) + +sizeof (struct sbi_sse_event ) * local_event_count); +if (!shs) +return SBI_ENOMEM ; + + shs->local_events = (struct sbi_sse_event *)(shs +1 ); + +sse_set_hart_state_ptr (scratch, shs); + } + +sse_local_init (shs); + + sse_inject_q =sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + sse_inject_mem =sbi_scratch_offset_ptr (scratch, + sse_inject_fifo_mem_off); + +sbi_fifo_init (sse_inject_q, sse_inject_mem,EVENT_COUNT , +sizeof (struct sse_ipi_inject_data )); + +return 0 ; +} + +void sbi_sse_exit (struct sbi_scratch *scratch) +{ +int i; +struct sbi_sse_event *e; + +for (i =0 ; i <EVENT_COUNT ; i++) { + e =sse_event_get (supported_events[i]); + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +continue ; + +if (SSE_EVENT_STATE (e) >SSE_STATE_REGISTERED ) +sbi_printf ("Event %d in invalid state at exit" , i); + } +} --2.43.0
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
In reply to Clément LégerHi Clement,
Thanks for the patch! My comments inline.
Quoted 66 lines · Cl?ment L?ger
This extension [1] allows to deliver events from SBI to supervisor via a software mecanism. This extensions defines events (either local or global) which are signaled by the SBI on specific signal sources (IRQ, traps, etc) and are injected to be executed in supervisor mode.
[1] https://lists.riscv.org/g/tech-prs/message/744
Signed-off-by Cl?ment L?ger <[email protected]>---
include/sbi/sbi_ecall_interface.h | 48 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 2 +- lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ 9 files changed, 1301 insertions(+), 3 deletions(-)create mode 100644
include/sbi/sbi_sse.h
create mode 100644lib/sbi/sbi_ecall_sse.c
create mode 100644lib/sbi/sbi_sse.cinclude/sbi/sbi_ecall_interface.h +26
-0@@ -32,6 +32,7 @@#define SBI_EXT_DBCN 0x4442434E#define SBI_EXT_SUSP 0x53555350#define SBI_EXT_CPPC 0x43505043 +#define SBI_EXT_SSE 0x535345/* SBI function IDs for BASE extension*/ #define SBI_EXT_BASE_GET_SPEC_VERSION 0x0@@ -293,6 +294,48 @@ enum sbi_cppc_reg_id { SBI_CPPC_NON_ACPI_LAST = SBI_CPPC_TRANSITION_LATENCY, }; +/* SBI Function IDs for SSE extension */ +#define SBI_EXT_SSE_READ_ATTR 0x00000000 +#define SBI_EXT_SSE_WRITE_ATTR 0x00000001 +#define SBI_EXT_SSE_REGISTER 0x00000002 +#define SBI_EXT_SSE_UNREGISTER 0x00000003 +#define SBI_EXT_SSE_ENABLE 0x00000004 +#define SBI_EXT_SSE_DISABLE 0x00000005 +#define SBI_EXT_SSE_COMPLETE 0x00000006 +#define SBI_EXT_SSE_INJECT 0x00000007 + +/* SBI SSE Event Attributes. */ +#define SBI_SSE_ATTR_STATUS 0x00000000 +#define SBI_SSE_ATTR_PRIO 0x00000001 +#define SBI_SSE_ATTR_CONFIG 0x00000002 +#define SBI_SSE_ATTR_PREFERRED_HART 0x00000003 +#define SBI_SSE_ATTR_ENTRY_PC 0x00000004 +#define SBI_SSE_ATTR_ENTRY_A0 0x00000005 +#define SBI_SSE_ATTR_ENTRY_A6 0x00000006 +#define SBI_SSE_ATTR_ENTRY_A7 0x00000007 +#define SBI_SSE_ATTR_INTERRUPTED_MODE 0x00000008 +#define SBI_SSE_ATTR_INTERRUPTED_PC 0x00000009 +#define SBI_SSE_ATTR_INTERRUPTED_A0 0x0000000A +#define SBI_SSE_ATTR_INTERRUPTED_A6 0x0000000B +#define SBI_SSE_ATTR_INTERRUPTED_A7 0x0000000C +
include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 331
Quoted 10 lines
include/sbi/sbi_ecall_interface.h +10
-0+#define SBI_SSE_ATTR_STATUS_STATE_OFFSET 0 +#define SBI_SSE_ATTR_STATUS_STATE_MASK 0x3 +#define SBI_SSE_ATTR_STATUS_PENDING_OFFSET 2 +#define SBI_SSE_ATTR_STATUS_INJECT_OFFSET 3 + +#define SBI_SSE_ATTR_CONFIG_ONESHOT (1 << 0) + +/* SBI SSE Event IDs. */ +#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000 +#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000
Can we add a define for start of platform specific local events (0x4000)?
include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 335
Quoted 4 lines
include/sbi/sbi_ecall_interface.h +4
-0+#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000 +#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000 +#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000 +
IMHO, we should add all the ranges defined in spec, used or not.
Quoted 114 lines
include/sbi/sbi_ecall_interface.h +7
-1+#define SBI_SSE_EVENT_GLOBAL (1 << 15) +#define SBI_SSE_EVENT_PLATFORM (1 << 14) +/* SBI base specification related macros */ #define SBI_SPEC_VERSION_MAJOR_OFFSET 24#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f@@ -313,8 +356,11 @@ enum sbi_cppc_reg_id {#define SBI_ERR_ALREADY_STARTED -7#define SBI_ERR_ALREADY_STOPPED -8#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_BUSY -12 -#define SBI_LAST_ERR SBI_ERR_NO_SHMEM +#define SBI_LAST_ERR SBI_ERR_BUSY/* clang-format on */ include/sbi/sbi_error.h +3
-1@@ -24,6 +24,9 @@#define SBI_EALREADY_STARTED SBI_ERR_ALREADY_STARTED#define SBI_EALREADY_STOPPED SBI_ERR_ALREADY_STOPPED#define SBI_ENO_SHMEM SBI_ERR_NO_SHMEM +#define SBI_EINVALID_STATE SBI_ERR_INVALID_STATE +#define SBI_EBAD_RANGE SBI_ERR_BAD_RANGE +#define SBI_EBUSY SBI_ERR_BUSY#define SBI_ENODEV -1000#define SBI_ENOSYS -1001@@ -34,7 +37,6 @@#define SBI_ENOMEM -1006#define SBI_EUNKNOWN -1007#define SBI_ENOENT -1008 -/* clang-format on */ #endif include/sbi/sbi_sse.h new file +67
-0@@ -0,0 +1,93 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems. +*/ + +#ifndef __SBI_SSE_H__ +#define __SBI_SSE_H__ + +#include <sbi/sbi_types.h> +#include <sbi/sbi_list.h> +#include <sbi/riscv_locks.h> + +struct sbi_scratch ; + +#define EXC_MODE_PP_SHIFT 0 +#define EXC_MODE_PP BIT(EXC_MODE_PP_SHIFT) +#define EXC_MODE_PV_SHIFT 1 +#define EXC_MODE_PV BIT(EXC_MODE_PV_SHIFT) +#define EXC_MODE_SSTATUS_SPIE_SHIFT 2 +#define EXC_MODE_SSTATUS_SPIE BIT(EXC_MODE_SSTATUS_SPIE_SHIFT) + + +struct sbi_sse_cb_ops { +/** +* Called when hart_id is changed on the event. +*/ +void (*set_hartid_cb)(uint32_t event_id,unsigned long hart_id); + +/** +* Called when the SBI_EXT_SSE_COMPLETE is invoked on the event. +*/ +void (*complete_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_REGISTER is invoked on the event. +*/ +void (*register_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_UNREGISTER is invoked on the event. +*/ +void (*unregister_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_ENABLE is invoked on the event. +*/ +void (*enable_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_DISABLE is invoked on the event. +*/ +void (*disable_cb)(uint32_t event_id); +}; + +/* Set the callback operations for an event +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param cb_ops Callback operations +* @return 0 on success, error otherwise +*/ +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops); + +/* Inject an event to the current hard +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param regs Registers that were used on SBI entry +* @return 0 on success, error otherwise +*/
include/sbi/sbi_sse.h line 77
Quoted 10 lines
include/sbi/sbi_sse.h new file +10
-0+int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs); + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot); +void sbi_sse_exit (struct sbi_scratch *scratch); + +/* Interface called from sbi_ecall_sse.c */ +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7);
Wouldn?t it be better to only use defined length data types like uint32_t, uint64_t everywhere? Instead of mixing them up?
Quoted 285 lines
include/sbi/sbi_sse.h new file +16
-0+int sbi_sse_unregister (uint32_t event_id); +int sbi_sse_enable (uint32_t event_id); +int sbi_sse_disable (uint32_t event_id); +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hart_id, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi); +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi); + +#endif lib/sbi/Kconfig +0
-0@@ -46,4 +46,8 @@ config SBI_ECALL_VENDORbool "Platform-defined vendor extensions"
default ylib/sbi/Kconfig +4
-0+config SBI_ECALL_SSE + bool "SSE extension" + default y +endmenu
lib/sbi/objects.mk +0
-0@@ -52,6 +52,9 @@ libsbi-objs-$(CONFIG_SBI_ECALL_LEGACY) += sbi_ecall_legacy.ocarray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_VENDOR) +=
ecall_vendorlibsbi-objs-$(CONFIG_SBI_ECALL_VENDOR) += sbi_ecall_vendor.olib/sbi/objects.mk +3
-0+carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_SSE) += ecall_sse +libsbi-objs-$(CONFIG_SBI_ECALL_SSE) += sbi_ecall_sse.o +libsbi-objs-y += sbi_bitmap.o
libsbi-objs-y += sbi_bitops.o
libsbi-objs-y += sbi_console.olib/sbi/objects.mk +0
-0@@ -71,6 +74,7 @@ libsbi-objs-y += sbi_misaligned_ldst.olibsbi-objs-y += sbi_platform.o
libsbi-objs-y += sbi_pmu.o
libsbi-objs-y += sbi_scratch.o
+libsbi-objs-y += sbi_sse.o
libsbi-objs-y += sbi_string.o
libsbi-objs-y += sbi_system.o
libsbi-objs-y += sbi_timer.olib/sbi/sbi_ecall_sse.c new file +58
-0@@ -0,0 +1,58 @@ +#include <sbi/sbi_error.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_trap.h> +#include <sbi/sbi_sse.h> + +static int sbi_ecall_sse_handler (unsigned long extid,unsigned long funcid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; + +switch (funcid) { +case SBI_EXT_SSE_READ_ATTR : + ret =sbi_sse_read_attrs (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a4); +break ; +case SBI_EXT_SSE_WRITE_ATTR : + ret =sbi_sse_write_attrs (regs->a0, regs->a1, regs->a2, + regs->a3, regs->a4); +break ; +case SBI_EXT_SSE_REGISTER : + ret =sbi_sse_register (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a2); +break ; +case SBI_EXT_SSE_UNREGISTER : + ret =sbi_sse_unregister (regs->a0); +break ; +case SBI_EXT_SSE_ENABLE : + ret =sbi_sse_enable (regs->a0); +break ; +case SBI_EXT_SSE_DISABLE : + ret =sbi_sse_disable (regs->a0); +break ; +case SBI_EXT_SSE_COMPLETE : + ret =sbi_sse_complete (regs->a0, regs, out); +break ; +case SBI_EXT_SSE_INJECT : + ret =sbi_sse_inject_from_ecall (regs->a0, regs->a1, regs, out); +break ; +default : + ret =SBI_ENOTSUPP ; + } +return ret; +} + +struct sbi_ecall_extension ecall_sse; + +static int sbi_ecall_sse_register_extensions (void ) +{ +return sbi_ecall_register_extension (&ecall_sse); +} + +struct sbi_ecall_extension ecall_sse = { + .extid_start =SBI_EXT_SSE , + .extid_end =SBI_EXT_SSE , + .register_extensions = sbi_ecall_sse_register_extensions, + .handle = sbi_ecall_sse_handler, +};lib/sbi/sbi_init.c +13
-0@@ -23,6 +23,7 @@#include <sbi/sbi_irqchip.h> #include <sbi/sbi_platform.h> #include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> #include <sbi/sbi_system.h> #include <sbi/sbi_string.h> #include <sbi/sbi_timer.h> @@ -315,6 +316,12 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, true); +if (rc) { +sbi_printf ("%s: sse init failed (error %d)\n" , __func__, rc); +sbi_hart_hang (); + } + rc =sbi_pmu_init (scratch, true);if (rc) { sbi_printf("%s: pmu init failed (error %d)\n",@@ -435,6 +442,10 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, false); +if (rc) +sbi_hart_hang (); + rc =sbi_pmu_init (scratch, false);if (rc)sbi_hart_hang ();@@ -639,6 +650,8 @@ void __noreturn sbi_exit(struct sbi_scratch *scratch)sbi_platform_early_exit (plat); +sbi_sse_exit (scratch); +sbi_pmu_exit (scratch);sbi_timer_exit (scratch);lib/sbi/sbi_ipi.c +1
-1@@ -66,7 +66,7 @@ static int sbi_ipi_send(struct sbi_scratch *scratch, u32 remote_hartindex,* SBI_IPI_UPDATE_BREAK for self-IPIs. For other events, check * for self-IPI and execute the callback directly here. */ - ipi_ops->process (scratch); + ipi_ops->process (scratch,NULL );return 0 ; }lib/sbi/sbi_sse.c new file +106
-0@@ -0,0 +1,1078 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems Inc. +* +*/ + +#include <sbi/riscv_asm.h> +#include <sbi/riscv_barrier.h> +#include <sbi/riscv_encoding.h> +#include <sbi/riscv_locks.h> +#include <sbi/sbi_domain.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_ecall_interface.h> +#include <sbi/sbi_error.h> +#include <sbi/sbi_fifo.h> +#include <sbi/sbi_hart.h> +#include <sbi/sbi_heap.h> +#include <sbi/sbi_hsm.h> +#include <sbi/sbi_ipi.h> +#include <sbi/sbi_list.h> +#include <sbi/sbi_platform.h> +#include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> +#include <sbi/sbi_scratch.h> +#include <sbi/sbi_string.h> +#include <sbi/sbi_trap.h> + +#include <sbi/sbi_console.h> + +#define sse_get_hart_state_ptr (__scratch) \ + sbi_scratch_read_type((__scratch), void *, shs_ptr_off) + +#define sse_thishart_state_ptr () \ + sse_get_hart_state_ptr(sbi_scratch_thishart_ptr()) + +#define sse_set_hart_state_ptr (__scratch, __sse_state) \ + sbi_scratch_write_type((__scratch), void *, shs_ptr_off, (__sse_state)) + +/* +* Rather than using memcpy to copy the context (which does it byte-per-byte), +* copy each field which generates ld/lw. +*/ +#define sse_regs_copy (dst, src) \ + (dst)->a0 = (src)->a0; \ + (dst)->a6 = (src)->a6; \ + (dst)->a7 = (src)->a7 + +#define EVENT_IS_GLOBAL (__event_id) ((__event_id) & SBI_SSE_EVENT_GLOBAL) + +static const uint32_t supported_events[] = +{ +SBI_SSE_EVENT_LOCAL_RAS , +SBI_SSE_EVENT_GLOBAL_RAS , +SBI_SSE_EVENT_LOCAL_PMU , +SBI_SSE_EVENT_LOCAL_SOFTWARE , +SBI_SSE_EVENT_GLOBAL_SOFTWARE , +}; + +#define EVENT_COUNT array_size(supported_events) + +#define sse_event_invoke_cb (_event, _cb, ...) \ +{ \ + if (_event->cb_ops && _event->cb_ops->_cb) \ + _event->cb_ops->_cb(_event->event_id, ##__VA_ARGS__); \ +} + +#define SSE_EVENT_STATE (__e) __e->attrs.status.state +#define SSE_EVENT_PENDING (__e) __e->attrs.status.pending +#define SSE_EVENT_CAN_INJECT (__e) __e->attrs.status.inject +#define SSE_EVENT_HARTID (__e) __e->attrs.hartid +#define SSE_EVENT_PRIO (__e) __e->attrs.prio +#define SSE_EVENT_CONFIG (__e) __e->attrs.config +#define SSE_EVENT_ENTRY (__e) __e->attrs.entry +#define SSE_EVENT_INTERRUPTED (__e) __e->attrs.interrupted + +struct sse_entry_state { +/** Entry program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +struct sse_interrupted_state { +/** Exception mode */ +unsigned long exc_mode; +/** Interrupted program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +enum sbi_sse_state { +SSE_STATE_UNUSED =0 , +SSE_STATE_REGISTERED =1 , +SSE_STATE_ENABLED =2 , +SSE_STATE_RUNNING =3 , +};
lib/sbi/sbi_sse.c line 116
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +struct sse_ipi_inject_data { +uint32_t event_id; +}; + +struct sbi_sse_event_status { +unsigned long state:2 ; +unsigned long pending:1 ; +unsigned long inject:1 ; +} __packed;
Can we have do away with bit fields? Like everywhere else use positions and masks? Why packed? We are making sure sse_event_attrs is aligned to unsigned long so that We can copy with loops. Probably make it a long with reserved entry? Unless you had Something else in mind.
Quoted 78 lines
lib/sbi/sbi_sse.c new file +78
-0+ +struct sbi_sse_event_attrs { +struct sbi_sse_event_status status; +unsigned long prio; +unsigned long config; +unsigned long hartid; +struct sse_entry_state entry; +struct sse_interrupted_state interrupted; +}; + +/* Make sure all attributes are packed for direct memcpy in ATTR_READ */ +#define assert_field_offset (field, attr_offset) \ + _Static_assert( \ + ((offsetof(struct sbi_sse_event_attrs, field)) / sizeof(unsigned long)) == attr_offset, \ + "field "#field " from struct sbi_sse_event_attrs invalid offset, expected "#attr_offset \ + ) + +assert_field_offset (status,SBI_SSE_ATTR_STATUS ); +assert_field_offset (prio,SBI_SSE_ATTR_PRIO ); +assert_field_offset (config,SBI_SSE_ATTR_CONFIG ); +assert_field_offset (hartid,SBI_SSE_ATTR_PREFERRED_HART ); +assert_field_offset (entry.pc,SBI_SSE_ATTR_ENTRY_PC ); +assert_field_offset (entry.a0,SBI_SSE_ATTR_ENTRY_A0 ); +assert_field_offset (entry.a6,SBI_SSE_ATTR_ENTRY_A6 ); +assert_field_offset (entry.a7,SBI_SSE_ATTR_ENTRY_A7 ); +assert_field_offset (interrupted.exc_mode,SBI_SSE_ATTR_INTERRUPTED_MODE ); +assert_field_offset (interrupted.pc,SBI_SSE_ATTR_INTERRUPTED_PC ); +assert_field_offset (interrupted.a0,SBI_SSE_ATTR_INTERRUPTED_A0 ); +assert_field_offset (interrupted.a6,SBI_SSE_ATTR_INTERRUPTED_A6 ); +assert_field_offset (interrupted.a7,SBI_SSE_ATTR_INTERRUPTED_A7 ); + +struct sbi_sse_event { +struct sbi_sse_event_attrs attrs; +uint32_t event_id; +const struct sbi_sse_cb_ops *cb_ops; +struct sbi_dlist node; +/* Only global events are using the lock, local ones don't need it */ +spinlock_t lock; +}; + +struct sse_hart_state { +struct sbi_dlist event_list; +spinlock_t list_lock; +struct sbi_sse_event *local_events; +}; + +static unsigned int local_event_count; +static unsigned int global_event_count; +static struct sbi_sse_event *global_events; + +static unsigned long sse_inject_fifo_off; +static unsigned long sse_inject_fifo_mem_off; +/* Offset of pointer to SSE HART state in scratch space */ +static unsigned long shs_ptr_off; + +static u32 sse_ipi_inject_event =SBI_IPI_EVENT_MAX ; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id); + +static bool sse_event_is_global (struct sbi_sse_event *e) +{ +return EVENT_IS_GLOBAL (e->event_id); +} + +static void sse_global_event_lock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_lock (&e->lock); +} + +static void sse_global_event_unlock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_unlock (&e->lock); +} + +static void sse_event_set_state (struct sbi_sse_event *e, +enum sbi_sse_state new_state)
lib/sbi/sbi_sse.c line 204
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+{ +enum sbi_sse_state prev_state =SSE_EVENT_STATE (e); + +if ((new_state - prev_state ==1 ) || (prev_state - new_state ==1 )) { +SSE_EVENT_STATE (e) = new_state; +return ; + } + +sbi_panic ("Invalid SSE state transition: %d -> %d\n" , prev_state, + new_state);
Panic in a running system is not a good idea, IMHO. Thinking out loud, what if we let the state as is and return error (handled by caller and decide). In worst case, the event can be disabled but definitely not panic.
Quoted 73 lines
lib/sbi/sbi_sse.c new file +73
-0+} + +static struct sbi_sse_event *sse_event_get (uint32_t event) +{ +unsigned int i; +struct sbi_sse_event *events, *e; +unsigned int count; +struct sse_hart_state *shs; + +if (EVENT_IS_GLOBAL (event)) { + count = global_event_count; + events = global_events; + }else { + count = local_event_count; + shs =sse_thishart_state_ptr (); + events = shs->local_events; + } + +for (i =0 ; i < count; i++) { + e = &events[i]; +if (e->event_id == event) +return e; + } + +return NULL ; +} + +static struct sse_hart_state *sse_event_get_hart_state (struct sbi_sse_event *e) +{ +struct sbi_scratch *s =sbi_hartid_to_scratch (SSE_EVENT_HARTID (e)); + +return sse_get_hart_state_ptr (s); +} + +static void sse_event_remove_from_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); + +spin_lock (&state->list_lock); +sbi_list_del (&e->node); +spin_unlock (&state->list_lock); +} + +static void sse_event_add_to_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); +struct sbi_sse_event *tmp; + +spin_lock (&state->list_lock); +sbi_list_for_each_entry (tmp, &state->event_list, node) { +if (SSE_EVENT_PRIO (e) <SSE_EVENT_PRIO (tmp)) +break ; +if (SSE_EVENT_PRIO (e) ==SSE_EVENT_PRIO (tmp) && + e->event_id < tmp->event_id) +break ; + } +sbi_list_add_tail (&e->node, &tmp->node); + +spin_unlock (&state->list_lock); +} + +static int sse_event_disable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, disable_cb); + +sse_event_remove_from_list (e); +sse_event_set_state (e,SSE_STATE_REGISTERED ); + +return SBI_OK ; +}
lib/sbi/sbi_sse.c line 287
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +static int sse_event_set_hart_id_check (struct sbi_sse_event *e, +unsigned long new_hartid) +{ +int hstate; +unsigned int hartid = (uint32_t ) new_hartid; +struct sbi_domain * hd =sbi_domain_thishart_ptr (); + +if (!sse_event_is_global (e)) +return SBI_EDENIED ;
Why denied? Why not invalid param?
lib/sbi/sbi_sse.c line 290
Quoted 3 lines
lib/sbi/sbi_sse.c new file +3
-0+ +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) +return SBI_EBUSY ;
In the spec, whenever the state doesn?t match the required state, we say we return INVALID_STATE.
Can we do that? This is the only place where EBUSY is used. It might be good to have EBUSY but then it?s out of the scope of this patch set.
Quoted 35 lines
lib/sbi/sbi_sse.c new file +35
-0+ +if (!sbi_domain_is_assigned_hart (hd, new_hartid)) +return SBI_EINVAL ; + + hstate =sbi_hsm_hart_get_state (hd, hartid); +if (hstate !=SBI_HSM_STATE_STARTED ) +return SBI_EINVAL ; + +return SBI_OK ; +} + +static int sse_event_set_attr_check (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +int ret =SBI_OK ; + +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +if (val & ~SBI_SSE_ATTR_CONFIG_ONESHOT ) + ret =SBI_ERR_INVALID_PARAM ; +break ; +case SBI_SSE_ATTR_PRIO : +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) + ret =SBI_EINVALID_STATE ; +break ; +case SBI_SSE_ATTR_PREFERRED_HART : + ret =sse_event_set_hart_id_check (e, val); +break ; +default : + ret =SBI_EBAD_RANGE ; +break ; + } + +return ret; +}
lib/sbi/sbi_sse.c line 335
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +static void sse_event_set_attr (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +SSE_EVENT_CONFIG (e) = val; +break ; +case SBI_SSE_ATTR_PRIO : +SSE_EVENT_PRIO (e) = (uint32_t )val;
Another reason to used defined length data types. I think we can do away with typecasts like this.
Quoted 16 lines
lib/sbi/sbi_sse.c new file +16
-0+break ; +case SBI_SSE_ATTR_PREFERRED_HART : +SSE_EVENT_HARTID (e) = val; +sse_event_invoke_cb (e, set_hartid_cb, val); +break ; + } +} + +static int sse_event_register (struct sbi_sse_event *e, +unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +if (sse_event_is_global (e) &&SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EINVAL ;
lib/sbi/sbi_sse.c line 361
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +if (SSE_EVENT_STATE (e) !=SSE_STATE_UNUSED ) +return SBI_EINVALID_STATE ; + +SSE_EVENT_ENTRY (e).a0 = handler_entry_a0; +SSE_EVENT_ENTRY (e).a6 = handler_entry_a6; +SSE_EVENT_ENTRY (e).a7 = handler_entry_a7; +SSE_EVENT_ENTRY (e).pc = handler_entry_pc; + +sse_event_set_state (e,SSE_STATE_REGISTERED );
Better return an error from set state (my previous comment). Handle it here based on what transition is expected. Here we just have crossed fingers that system won?t panic.
Quoted 87 lines
lib/sbi/sbi_sse.c new file +87
-0+ +sse_event_invoke_cb (e, register_cb); + +return 0 ; +} + +static int sse_event_unregister (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, unregister_cb); + +sse_event_set_state (e,SSE_STATE_UNUSED ); + +return 0 ; +} + +static void sse_event_inject (struct sbi_sse_event *e, +struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +ulong prev_smode, prev_virt; +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); +struct sse_interrupted_state *prev_i_ctx; +struct sse_entry_state *e_ctx = &SSE_EVENT_ENTRY (e); + +sse_event_set_state (e,SSE_STATE_RUNNING ); +SSE_EVENT_PENDING (e) = false; + +if (prev_e) { +/* back-to-back injection after another event, copy previous +* event context for correct restoration +*/ + prev_i_ctx = &SSE_EVENT_INTERRUPTED (prev_e); + +sse_regs_copy (i_ctx, prev_i_ctx); + i_ctx->exc_mode = prev_i_ctx->exc_mode; + i_ctx->pc = prev_i_ctx->pc; + }else { +sse_regs_copy (i_ctx, regs); + + prev_smode = (regs->mstatus &MSTATUS_MPP ) >>MSTATUS_MPP_SHIFT ; +#if __riscv_xlen ==32 + prev_virt = (regs->mstatusH &MSTATUSH_MPV ) ?1 :0 ; +#else + prev_virt = (regs->mstatus &MSTATUS_MPV ) ?1 :0 ; +#endif + + i_ctx->exc_mode = prev_smode <<EXC_MODE_PP_SHIFT ; + i_ctx->exc_mode |= prev_virt <<EXC_MODE_PV_SHIFT ; +if (regs->mstatus &MSTATUS_SPIE ) + i_ctx->exc_mode |=EXC_MODE_SSTATUS_SPIE ; + i_ctx->pc = regs->mepc; + +/* We only want to set SPIE for the first event injected after +* entering M-Mode. For the event injected right after another +* event (after calling sse_event_complete(), we will keep the +* saved SPIE). +*/ + regs->mstatus &= ~MSTATUS_SPIE ; +if (regs->mstatus &MSTATUS_SIE ) + regs->mstatus |=MSTATUS_SPIE ; + } + +sse_regs_copy (regs, e_ctx); + regs->mepc = e_ctx->pc; + + regs->mstatus &= ~MSTATUS_MPP ; + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +#endif + + regs->mstatus &= ~MSTATUS_SIE ; +} + +static void sse_event_resume (struct sbi_sse_event *e,struct sbi_trap_regs *regs) +{ +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); + +sse_regs_copy (regs, i_ctx); + +/* Restore previous virtualization state */
lib/sbi/sbi_sse.c line 458
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatusH |=MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatus |=MSTATUS_MPV ; +#endif +
Probably have #error for undefined __riscv_xlen.
Quoted 577 lines
lib/sbi/sbi_sse.c new file +577
-0+ regs->mstatus &= ~MSTATUS_MPP ; +if (i_ctx->exc_mode &EXC_MODE_PP ) + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + + regs->mstatus &= ~MSTATUS_SIE ; +if (regs->mstatus &MSTATUS_SPIE ) + regs->mstatus |=MSTATUS_SIE ; + + regs->mstatus &= ~MSTATUS_SPIE ; +if (i_ctx->exc_mode &EXC_MODE_SSTATUS_SPIE ) + regs->mstatus |=MSTATUS_SPIE ; + + regs->mepc = i_ctx->pc; +} + +static bool sse_event_is_ready (struct sbi_sse_event *e) +{ +if (!SSE_EVENT_PENDING (e) ||SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED || +SSE_EVENT_HARTID (e) !=current_hartid ()) { +return false; + } + +return true; +} + +/* Return true if an event has been injected, false otherwise */ +static bool sse_process_pending_events (struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +struct sbi_sse_event *e, *to_run =NULL ; +struct sse_hart_state *state =sse_thishart_state_ptr (); + +retry: +spin_lock (&state->list_lock); + +if (sbi_list_empty (&state->event_list)) { +spin_unlock (&state->list_lock); +return false; + } + +sbi_list_for_each_entry (e, &state->event_list, node) { +/* +* List of event is ordered by priority, stop at first running +* event since all other events after this one are of lower +* priority. +*/ +if (SSE_EVENT_STATE (e) ==SSE_STATE_RUNNING ) +break ; + +if (sse_event_is_ready (e)) { + to_run = e; +break ; + } + } + +spin_unlock (&state->list_lock); + +if (!to_run) +return false; + +sse_global_event_lock (e); +/* +* If the event is global, the event might have been moved to another +* hart or disabled, evaluate readiness again. +*/ +if (sse_event_is_global (e) && !sse_event_is_ready (e)) { +sse_global_event_unlock (e); + goto retry; + } + +sse_event_inject (e, prev_e, regs); +sse_global_event_unlock (e); + +return true; +} + +static int sse_event_set_pending (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_RUNNING && +SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_ERR_INVALID_STATE ; + +SSE_EVENT_PENDING (e) = true; + +return SBI_OK ; +} + +static void sse_ipi_inject_process (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) +{ +struct sbi_sse_event *e; +struct sse_ipi_inject_data evt; +struct sbi_fifo *sse_inject_fifo_r = +sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + +/* This can be the case when sbi_exit() is called */ +if (!regs) +return ; + +/* Mark all queued events as pending */ +while (!sbi_fifo_dequeue (sse_inject_fifo_r, &evt)) { + e =sse_event_get (evt.event_id); +if (!e) +continue ; + +sse_global_event_lock (e); +sse_event_set_pending (e); +sse_global_event_unlock (e); + } + +sse_process_pending_events (NULL , regs); +} + +static struct sbi_ipi_event_ops sse_ipi_inject_ops = { + .name ="IPI_SSE_INJECT" , + .process = sse_ipi_inject_process, +}; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id) +{ +int ret; +struct sbi_scratch *remote_scratch =NULL ; +struct sse_ipi_inject_data evt = {event_id}; +struct sbi_fifo *sse_inject_fifo_r; + + remote_scratch =sbi_hartid_to_scratch (hartid); +if (!remote_scratch) +return SBI_EINVAL ; + sse_inject_fifo_r =sbi_scratch_offset_ptr (remote_scratch, + sse_inject_fifo_off); + + ret =sbi_fifo_enqueue (sse_inject_fifo_r, &evt); +if (ret) +return SBI_EFAIL ; + + ret =sbi_ipi_send_many (1 , hartid, sse_ipi_inject_event,NULL ); +if (ret) +return SBI_EFAIL ; + +return SBI_OK ; +} + +static int sse_inject_event (uint32_t event_id,unsigned long hartid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out,bool from_ecall) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +/* In case of global event, provided hart_id is ignored */ +if (sse_event_is_global (e)) + hartid =SSE_EVENT_HARTID (e); + +/* Event is for another hart, send it through IPI */ +if (hartid !=current_hartid ()) { +sse_global_event_unlock (e); +return sse_ipi_inject_send (hartid, event_id); + } + + ret =sse_event_set_pending (e); +sse_global_event_unlock (e); +if (ret) +return ret; + +if (from_ecall) { +/* Due to skip_regs_update = true being set if injection +* succeed an the fact that we need to modify regs here before +* injecting the event (regs are copied by the event), we need +* to set out->skip_regs_update to true to avoid. Moreover +* injection might not be done right now if another event of +* higher priority is already running so always set +* out->skip_regs_update to true there. +*/ + regs->mepc +=4 ; + regs->a0 =SBI_OK ; + out->skip_regs_update = true; + } + +if (sse_process_pending_events (NULL , regs)) + out->skip_regs_update = true; + +return SBI_OK ; +} + +static int sse_event_enable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_set_state (e,SSE_STATE_ENABLED ); +sse_event_add_to_list (e); + +if (SSE_EVENT_PENDING (e)) +sbi_ipi_send_many (1 ,SSE_EVENT_HARTID (e), sse_ipi_inject_event, +NULL ); + +sse_event_invoke_cb (e, enable_cb); + +return SBI_OK ; +} + +static int sse_event_complete (struct sbi_sse_event *e, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +bool inject; + +if (SSE_EVENT_STATE (e) !=SSE_STATE_RUNNING ) +return SBI_EINVALID_STATE ; + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EDENIED ; + +if (SSE_EVENT_CONFIG (e) &SBI_SSE_ATTR_CONFIG_ONESHOT ) +sse_event_disable (e); +else +sse_event_set_state (e,SSE_STATE_ENABLED ); + +sse_event_invoke_cb (e, complete_cb); + + inject =sse_process_pending_events (e, regs); +if (!inject) +sse_event_resume (e, regs); + + out->skip_regs_update = true; + +return SBI_OK ; +} + +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_complete (e, regs, out); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_enable (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_enable (e); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_disable (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_disable (e); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hartid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +if (!sbi_domain_is_assigned_hart (sbi_domain_thishart_ptr (), hartid)) +return SBI_EINVAL ; + +return sse_inject_event (event_id, hartid, regs, out, true); +} + +int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs) +{ +/* We don't really care about return value here */ +struct sbi_ecall_return out; + +return sse_inject_event (event_id,current_hartid (), regs, &out, false); +} + +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops) +{ +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +if (cb_ops->set_hartid_cb && !sse_event_is_global (e)) +return SBI_EINVAL ; + +sse_global_event_lock (e); + e->cb_ops = cb_ops; +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_attr_check (uint32_t base_attr_id,uint32_t attr_count, +unsigned long phys_lo, +unsigned long phys_hi, +unsigned long access) +{ +const unsigned align = __riscv_xlen >>3 ; +/* Avoid 32 bits overflow */ +uint64_t end_id = (uint64_t )base_attr_id + attr_count; + +if (end_id >SBI_SSE_ATTR_INTERRUPTED_A7 ) +return SBI_EBAD_RANGE ; + +if (phys_lo & (align -1 )) +return SBI_EINVALID_ADDR ; + +/* +* On RV32, the M-mode can only access the first 4GB of +* the physical address space because M-mode does not have +* MMU to access full 34-bit physical address space. +* +* Based on above, we simply fail if the upper 32bits of +* the physical address (i.e. a2 register) is non-zero on +* RV32. +*/ +if (phys_hi) +return SBI_EINVALID_ADDR ; + +if (!sbi_domain_check_addr_range (sbi_domain_thishart_ptr (), phys_lo, +sizeof (unsigned long ) * attr_count,1 , + access)) +return SBI_EINVALID_ADDR ; + +return SBI_OK ; +} + + +static void copy_attrs (unsigned long *out,const unsigned long *in, +unsigned int long_count) +{ +int i =0 ; + +/* +* sbi_memcpy() does byte-per-byte copy, using this yields long-per-long +* copy +*/ +for (i =0 ; i < long_count; i++) + out[i] = in[i]; +} + +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi) +{ +int ret; +unsigned long *e_attrs; +struct sbi_sse_event *e; +unsigned long *attrs; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, output_phys_lo, + output_phys_hi,SBI_DOMAIN_WRITE ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (output_phys_lo,sizeof (unsigned long ) * attr_count); + +/* +* Copy all attributes at once since struct sse_event_attrs is matching +* the SBI_SSE_ATTR_* attributes. While WRITE_ATTR attribute is not used +* in s-mode sse handling path, READ_ATTR is used to retrieve the value +* of registers when interrupted. rather than doing multiple SBI calls, +* a single one is done allowing to retrieve them all at once. +*/ + e_attrs = (unsigned long *)&e->attrs; + attrs = (unsigned long *)output_phys_lo; +copy_attrs (attrs, &e_attrs[base_attr_id], attr_count); + +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi) +{ +int ret; +struct sbi_sse_event *e; +unsigned long attr =0 , val; +uint32_t id, end_id = base_attr_id + attr_count; +unsigned long *attrs = (unsigned long *)input_phys_lo; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, input_phys_lo, + input_phys_hi,SBI_DOMAIN_READ ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (input_phys_lo,sizeof (unsigned long ) * attr_count); + +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; + ret =sse_event_set_attr_check (e, id, val); +if (ret) + goto out; + } + + attr =0 ; +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; +sse_event_set_attr (e, id, val); + } +out: +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_register (e, handler_entry_pc, handler_entry_a0, + handler_entry_a6, handler_entry_a7); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_unregister (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_unregister (e); +sse_global_event_unlock (e); + +return ret; +} + +static void sse_event_init (struct sbi_sse_event *e,uint32_t event_id) +{ + e->event_id = event_id; +SSE_EVENT_HARTID (e) =current_hartid (); +/* Declare all events as injectable */ +SSE_EVENT_CAN_INJECT (e) =1 ; +} + +static int sse_global_init () +{ +struct sbi_sse_event *e; +unsigned int i, ev =0 ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) + global_event_count++; +else + local_event_count++; + } + + global_events =sbi_zalloc (sizeof (*global_events) * global_event_count); +if (!global_events) +return SBI_ENOMEM ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (!EVENT_IS_GLOBAL (supported_events[i])) +continue ; + + e = &global_events[ev]; +sse_event_init (e, supported_events[i]); +SPIN_LOCK_INIT (e->lock); + + ev++; + } + +return 0 ; +} + +static void sse_local_init (struct sse_hart_state *shs) +{ +unsigned int i, ev =0 ; + +SBI_INIT_LIST_HEAD (&shs->event_list); +SPIN_LOCK_INIT (shs->list_lock); + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) +continue ; + +sse_event_init (&shs->local_events[ev++], supported_events[i]); + } +} + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot) +{ +int ret; +void *sse_inject_mem; +struct sse_hart_state *shs; +struct sbi_fifo *sse_inject_q; + +if (cold_boot) { + ret =sse_global_init (); +if (ret) +return ret; + + shs_ptr_off =sbi_scratch_alloc_offset (sizeof (void *)); +if (!shs_ptr_off) +return SBI_ENOMEM ; + + sse_inject_fifo_off =sbi_scratch_alloc_offset ( +sizeof (*sse_inject_q)); +if (!sse_inject_fifo_off) { +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + sse_inject_fifo_mem_off =sbi_scratch_alloc_offset ( +EVENT_COUNT *sizeof (struct sse_ipi_inject_data )); +if (!sse_inject_fifo_mem_off) { +sbi_scratch_free_offset (sse_inject_fifo_off); +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + ret =sbi_ipi_event_create (&sse_ipi_inject_ops); +if (ret <0 ) { +sbi_scratch_free_offset (shs_ptr_off); +return ret; + }
lib/sbi/sbi_sse.c line 1045
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ sse_ipi_inject_event = ret; + } + + shs =sse_get_hart_state_ptr (scratch); +if (!shs) { +/* Allocate per hart state and local events at once */ + shs =sbi_zalloc (sizeof (*shs) + +sizeof (struct sbi_sse_event ) * local_event_count); +if (!shs) +return SBI_ENOMEM ;
What about scratch offset allocated above. Shouldn?t that be freed? I suggest to have a goto to undo stuff in case of error.
Quoted 36 lines
lib/sbi/sbi_sse.c new file +33
-0+ + shs->local_events = (struct sbi_sse_event *)(shs +1 ); + +sse_set_hart_state_ptr (scratch, shs); + } + +sse_local_init (shs); + + sse_inject_q =sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + sse_inject_mem =sbi_scratch_offset_ptr (scratch, + sse_inject_fifo_mem_off); + +sbi_fifo_init (sse_inject_q, sse_inject_mem,EVENT_COUNT , +sizeof (struct sse_ipi_inject_data )); + +return 0 ; +} + +void sbi_sse_exit (struct sbi_scratch *scratch) +{ +int i; +struct sbi_sse_event *e; + +for (i =0 ; i <EVENT_COUNT ; i++) { + e =sse_event_get (supported_events[i]); + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +continue ; + +if (SSE_EVENT_STATE (e) >SSE_STATE_REGISTERED ) +sbi_printf ("Event %d in invalid state at exit" , i); + } +}Signature · 1 line
2.43.0
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
Quoted 212 lines · Himanshu Chauhan
Hi Clement,
Thanks for the patch! My comments inline.
Quoted 66 lines · Cl?ment L?ger
This extension [1] allows to deliver events from SBI to supervisor via a software mecanism. This extensions defines events (either local or global) which are signaled by the SBI on specific signal sources (IRQ, traps, etc) and are injected to be executed in supervisor mode.
[1] https://lists.riscv.org/g/tech-prs/message/744
Signed-off-by Cl?ment L?ger <[email protected]>---
include/sbi/sbi_ecall_interface.h | 48 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 2 +- lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ 9 files changed, 1301 insertions(+), 3 deletions(-)create mode 100644
include/sbi/sbi_sse.h
create mode 100644lib/sbi/sbi_ecall_sse.c
create mode 100644lib/sbi/sbi_sse.cinclude/sbi/sbi_ecall_interface.h +26
-0@@ -32,6 +32,7 @@#define SBI_EXT_DBCN 0x4442434E#define SBI_EXT_SUSP 0x53555350#define SBI_EXT_CPPC 0x43505043 +#define SBI_EXT_SSE 0x535345/* SBI function IDs for BASE extension*/ #define SBI_EXT_BASE_GET_SPEC_VERSION 0x0@@ -293,6 +294,48 @@ enum sbi_cppc_reg_id { SBI_CPPC_NON_ACPI_LAST = SBI_CPPC_TRANSITION_LATENCY, }; +/* SBI Function IDs for SSE extension */ +#define SBI_EXT_SSE_READ_ATTR 0x00000000 +#define SBI_EXT_SSE_WRITE_ATTR 0x00000001 +#define SBI_EXT_SSE_REGISTER 0x00000002 +#define SBI_EXT_SSE_UNREGISTER 0x00000003 +#define SBI_EXT_SSE_ENABLE 0x00000004 +#define SBI_EXT_SSE_DISABLE 0x00000005 +#define SBI_EXT_SSE_COMPLETE 0x00000006 +#define SBI_EXT_SSE_INJECT 0x00000007 + +/* SBI SSE Event Attributes. */ +#define SBI_SSE_ATTR_STATUS 0x00000000 +#define SBI_SSE_ATTR_PRIO 0x00000001 +#define SBI_SSE_ATTR_CONFIG 0x00000002 +#define SBI_SSE_ATTR_PREFERRED_HART 0x00000003 +#define SBI_SSE_ATTR_ENTRY_PC 0x00000004 +#define SBI_SSE_ATTR_ENTRY_A0 0x00000005 +#define SBI_SSE_ATTR_ENTRY_A6 0x00000006 +#define SBI_SSE_ATTR_ENTRY_A7 0x00000007 +#define SBI_SSE_ATTR_INTERRUPTED_MODE 0x00000008 +#define SBI_SSE_ATTR_INTERRUPTED_PC 0x00000009 +#define SBI_SSE_ATTR_INTERRUPTED_A0 0x0000000A +#define SBI_SSE_ATTR_INTERRUPTED_A6 0x0000000B +#define SBI_SSE_ATTR_INTERRUPTED_A7 0x0000000C +include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 331
Quoted 10 lines
include/sbi/sbi_ecall_interface.h +10
-0+#define SBI_SSE_ATTR_STATUS_STATE_OFFSET 0 +#define SBI_SSE_ATTR_STATUS_STATE_MASK 0x3 +#define SBI_SSE_ATTR_STATUS_PENDING_OFFSET 2 +#define SBI_SSE_ATTR_STATUS_INJECT_OFFSET 3 + +#define SBI_SSE_ATTR_CONFIG_ONESHOT (1 << 0) + +/* SBI SSE Event IDs. */ +#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000 +#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000Can we add a define for start of platform specific local events (
0x4000)?include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 335
Quoted 4 lines
include/sbi/sbi_ecall_interface.h +4
-0+#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000 +#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000 +#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000 +IMHO, we should add all the ranges defined in spec, used or not.
Quoted 118 lines
include/sbi/sbi_ecall_interface.h +7
-1+#define SBI_SSE_EVENT_GLOBAL (1 << 15) +#define SBI_SSE_EVENT_PLATFORM (1 << 14) +/* SBI base specification related macros */ #define SBI_SPEC_VERSION_MAJOR_OFFSET 24#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f@@ -313,8 +356,11 @@ enum sbi_cppc_reg_id {#define SBI_ERR_ALREADY_STARTED -7#define SBI_ERR_ALREADY_STOPPED -8#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_BUSY -12 -#define SBI_LAST_ERR SBI_ERR_NO_SHMEM +#define SBI_LAST_ERR SBI_ERR_BUSY/* clang-format on */ include/sbi/sbi_error.h +3
-1@@ -24,6 +24,9 @@#define SBI_EALREADY_STARTED SBI_ERR_ALREADY_STARTED#define SBI_EALREADY_STOPPED SBI_ERR_ALREADY_STOPPED#define SBI_ENO_SHMEM SBI_ERR_NO_SHMEM +#define SBI_EINVALID_STATE SBI_ERR_INVALID_STATE +#define SBI_EBAD_RANGE SBI_ERR_BAD_RANGE +#define SBI_EBUSY SBI_ERR_BUSY#define SBI_ENODEV -1000#define SBI_ENOSYS -1001@@ -34,7 +37,6 @@#define SBI_ENOMEM -1006#define SBI_EUNKNOWN -1007#define SBI_ENOENT -1008 -/* clang-format on */ #endif include/sbi/sbi_sse.h new file +71
-0@@ -0,0 +1,93 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems. +*/ + +#ifndef __SBI_SSE_H__ +#define __SBI_SSE_H__ + +#include <sbi/sbi_types.h> +#include <sbi/sbi_list.h> +#include <sbi/riscv_locks.h> + +struct sbi_scratch ; + +#define EXC_MODE_PP_SHIFT 0 +#define EXC_MODE_PP BIT(EXC_MODE_PP_SHIFT) +#define EXC_MODE_PV_SHIFT 1 +#define EXC_MODE_PV BIT(EXC_MODE_PV_SHIFT) +#define EXC_MODE_SSTATUS_SPIE_SHIFT 2 +#define EXC_MODE_SSTATUS_SPIE BIT(EXC_MODE_SSTATUS_SPIE_SHIFT) + + +struct sbi_sse_cb_ops { +/** +* Called when hart_id is changed on the event. +*/ +void (*set_hartid_cb)(uint32_t event_id,unsigned long hart_id); + +/** +* Called when the SBI_EXT_SSE_COMPLETE is invoked on the event. +*/ +void (*complete_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_REGISTER is invoked on the event. +*/ +void (*register_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_UNREGISTER is invoked on the event. +*/ +void (*unregister_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_ENABLE is invoked on the event. +*/ +void (*enable_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_DISABLE is invoked on the event. +*/ +void (*disable_cb)(uint32_t event_id); +}; + +/* Set the callback operations for an event +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param cb_ops Callback operations +* @return 0 on success, error otherwise +*/ +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops); + +/* Inject an event to the current hard +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param regs Registers that were used on SBI entry +* @return 0 on success, error otherwise +*/ +int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs); + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot); +void sbi_sse_exit (struct sbi_scratch *scratch);
Quoted 8 lines
include/sbi/sbi_sse.h line 77
Quoted 6 lines
include/sbi/sbi_sse.h new file +6
-0+ +/* Interface called from sbi_ecall_sse.c */ +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7);Wouldn?t it be better to only use defined length data types like
uint32_t,uint64_teverywhere? Instead of mixing them up?
Sorry. I stand corrected here. It has to be unsigned long since it has work for both RV32/64. Please disregard this comment.
Quoted 1,291 lines
Quoted 286 lines
include/sbi/sbi_sse.h new file +16
-0+int sbi_sse_unregister (uint32_t event_id); +int sbi_sse_enable (uint32_t event_id); +int sbi_sse_disable (uint32_t event_id); +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hart_id, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi); +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi); + +#endif lib/sbi/Kconfig +0
-0@@ -46,4 +46,8 @@ config SBI_ECALL_VENDORbool "Platform-defined vendor extensions"
default ylib/sbi/Kconfig +4
-0+config SBI_ECALL_SSE + bool "SSE extension" + default y +endmenu
lib/sbi/objects.mk +0
-0@@ -52,6 +52,9 @@ libsbi-objs-$(CONFIG_SBI_ECALL_LEGACY) += sbi_ecall_legacy.ocarray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_VENDOR) +=
ecall_vendorlibsbi-objs-$(CONFIG_SBI_ECALL_VENDOR) += sbi_ecall_vendor.olib/sbi/objects.mk +3
-0+carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_SSE) += ecall_sse +libsbi-objs-$(CONFIG_SBI_ECALL_SSE) += sbi_ecall_sse.o +libsbi-objs-y += sbi_bitmap.o
libsbi-objs-y += sbi_bitops.o
libsbi-objs-y += sbi_console.olib/sbi/objects.mk +0
-0@@ -71,6 +74,7 @@ libsbi-objs-y += sbi_misaligned_ldst.olibsbi-objs-y += sbi_platform.o
libsbi-objs-y += sbi_pmu.o
libsbi-objs-y += sbi_scratch.o
+libsbi-objs-y += sbi_sse.o
libsbi-objs-y += sbi_string.o
libsbi-objs-y += sbi_system.o
libsbi-objs-y += sbi_timer.olib/sbi/sbi_ecall_sse.c new file +58
-0@@ -0,0 +1,58 @@ +#include <sbi/sbi_error.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_trap.h> +#include <sbi/sbi_sse.h> + +static int sbi_ecall_sse_handler (unsigned long extid,unsigned long funcid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; + +switch (funcid) { +case SBI_EXT_SSE_READ_ATTR : + ret =sbi_sse_read_attrs (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a4); +break ; +case SBI_EXT_SSE_WRITE_ATTR : + ret =sbi_sse_write_attrs (regs->a0, regs->a1, regs->a2, + regs->a3, regs->a4); +break ; +case SBI_EXT_SSE_REGISTER : + ret =sbi_sse_register (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a2); +break ; +case SBI_EXT_SSE_UNREGISTER : + ret =sbi_sse_unregister (regs->a0); +break ; +case SBI_EXT_SSE_ENABLE : + ret =sbi_sse_enable (regs->a0); +break ; +case SBI_EXT_SSE_DISABLE : + ret =sbi_sse_disable (regs->a0); +break ; +case SBI_EXT_SSE_COMPLETE : + ret =sbi_sse_complete (regs->a0, regs, out); +break ; +case SBI_EXT_SSE_INJECT : + ret =sbi_sse_inject_from_ecall (regs->a0, regs->a1, regs, out); +break ; +default : + ret =SBI_ENOTSUPP ; + } +return ret; +} + +struct sbi_ecall_extension ecall_sse; + +static int sbi_ecall_sse_register_extensions (void ) +{ +return sbi_ecall_register_extension (&ecall_sse); +} + +struct sbi_ecall_extension ecall_sse = { + .extid_start =SBI_EXT_SSE , + .extid_end =SBI_EXT_SSE , + .register_extensions = sbi_ecall_sse_register_extensions, + .handle = sbi_ecall_sse_handler, +};lib/sbi/sbi_init.c +13
-0@@ -23,6 +23,7 @@#include <sbi/sbi_irqchip.h> #include <sbi/sbi_platform.h> #include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> #include <sbi/sbi_system.h> #include <sbi/sbi_string.h> #include <sbi/sbi_timer.h> @@ -315,6 +316,12 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, true); +if (rc) { +sbi_printf ("%s: sse init failed (error %d)\n" , __func__, rc); +sbi_hart_hang (); + } + rc =sbi_pmu_init (scratch, true);if (rc) { sbi_printf("%s: pmu init failed (error %d)\n",@@ -435,6 +442,10 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, false); +if (rc) +sbi_hart_hang (); + rc =sbi_pmu_init (scratch, false);if (rc)sbi_hart_hang ();@@ -639,6 +650,8 @@ void __noreturn sbi_exit(struct sbi_scratch *scratch)sbi_platform_early_exit (plat); +sbi_sse_exit (scratch); +sbi_pmu_exit (scratch);sbi_timer_exit (scratch);lib/sbi/sbi_ipi.c +1
-1@@ -66,7 +66,7 @@ static int sbi_ipi_send(struct sbi_scratch *scratch, u32 remote_hartindex,* SBI_IPI_UPDATE_BREAK for self-IPIs. For other events, check * for self-IPI and execute the callback directly here. */ - ipi_ops->process (scratch); + ipi_ops->process (scratch,NULL );return 0 ; }lib/sbi/sbi_sse.c new file +106
-0@@ -0,0 +1,1078 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems Inc. +* +*/ + +#include <sbi/riscv_asm.h> +#include <sbi/riscv_barrier.h> +#include <sbi/riscv_encoding.h> +#include <sbi/riscv_locks.h> +#include <sbi/sbi_domain.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_ecall_interface.h> +#include <sbi/sbi_error.h> +#include <sbi/sbi_fifo.h> +#include <sbi/sbi_hart.h> +#include <sbi/sbi_heap.h> +#include <sbi/sbi_hsm.h> +#include <sbi/sbi_ipi.h> +#include <sbi/sbi_list.h> +#include <sbi/sbi_platform.h> +#include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> +#include <sbi/sbi_scratch.h> +#include <sbi/sbi_string.h> +#include <sbi/sbi_trap.h> + +#include <sbi/sbi_console.h> + +#define sse_get_hart_state_ptr (__scratch) \ + sbi_scratch_read_type((__scratch), void *, shs_ptr_off) + +#define sse_thishart_state_ptr () \ + sse_get_hart_state_ptr(sbi_scratch_thishart_ptr()) + +#define sse_set_hart_state_ptr (__scratch, __sse_state) \ + sbi_scratch_write_type((__scratch), void *, shs_ptr_off, (__sse_state)) + +/* +* Rather than using memcpy to copy the context (which does it byte-per-byte), +* copy each field which generates ld/lw. +*/ +#define sse_regs_copy (dst, src) \ + (dst)->a0 = (src)->a0; \ + (dst)->a6 = (src)->a6; \ + (dst)->a7 = (src)->a7 + +#define EVENT_IS_GLOBAL (__event_id) ((__event_id) & SBI_SSE_EVENT_GLOBAL) + +static const uint32_t supported_events[] = +{ +SBI_SSE_EVENT_LOCAL_RAS , +SBI_SSE_EVENT_GLOBAL_RAS , +SBI_SSE_EVENT_LOCAL_PMU , +SBI_SSE_EVENT_LOCAL_SOFTWARE , +SBI_SSE_EVENT_GLOBAL_SOFTWARE , +}; + +#define EVENT_COUNT array_size(supported_events) + +#define sse_event_invoke_cb (_event, _cb, ...) \ +{ \ + if (_event->cb_ops && _event->cb_ops->_cb) \ + _event->cb_ops->_cb(_event->event_id, ##__VA_ARGS__); \ +} + +#define SSE_EVENT_STATE (__e) __e->attrs.status.state +#define SSE_EVENT_PENDING (__e) __e->attrs.status.pending +#define SSE_EVENT_CAN_INJECT (__e) __e->attrs.status.inject +#define SSE_EVENT_HARTID (__e) __e->attrs.hartid +#define SSE_EVENT_PRIO (__e) __e->attrs.prio +#define SSE_EVENT_CONFIG (__e) __e->attrs.config +#define SSE_EVENT_ENTRY (__e) __e->attrs.entry +#define SSE_EVENT_INTERRUPTED (__e) __e->attrs.interrupted + +struct sse_entry_state { +/** Entry program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +struct sse_interrupted_state { +/** Exception mode */ +unsigned long exc_mode; +/** Interrupted program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +enum sbi_sse_state { +SSE_STATE_UNUSED =0 , +SSE_STATE_REGISTERED =1 , +SSE_STATE_ENABLED =2 , +SSE_STATE_RUNNING =3 , +};lib/sbi/sbi_sse.c line 116
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +struct sse_ipi_inject_data { +uint32_t event_id; +}; + +struct sbi_sse_event_status { +unsigned long state:2 ; +unsigned long pending:1 ; +unsigned long inject:1 ; +} __packed;Can we have do away with bit fields? Like everywhere else use positions and masks? Why packed? We are making sure
sse_event_attrsis aligned to unsigned long so that We can copy with loops. Probably make it a long with reserved entry? Unless you had Something else in mind.Quoted 78 lines
lib/sbi/sbi_sse.c new file +78
-0+ +struct sbi_sse_event_attrs { +struct sbi_sse_event_status status; +unsigned long prio; +unsigned long config; +unsigned long hartid; +struct sse_entry_state entry; +struct sse_interrupted_state interrupted; +}; + +/* Make sure all attributes are packed for direct memcpy in ATTR_READ */ +#define assert_field_offset (field, attr_offset) \ + _Static_assert( \ + ((offsetof(struct sbi_sse_event_attrs, field)) / sizeof(unsigned long)) == attr_offset, \ + "field "#field " from struct sbi_sse_event_attrs invalid offset, expected "#attr_offset \ + ) + +assert_field_offset (status,SBI_SSE_ATTR_STATUS ); +assert_field_offset (prio,SBI_SSE_ATTR_PRIO ); +assert_field_offset (config,SBI_SSE_ATTR_CONFIG ); +assert_field_offset (hartid,SBI_SSE_ATTR_PREFERRED_HART ); +assert_field_offset (entry.pc,SBI_SSE_ATTR_ENTRY_PC ); +assert_field_offset (entry.a0,SBI_SSE_ATTR_ENTRY_A0 ); +assert_field_offset (entry.a6,SBI_SSE_ATTR_ENTRY_A6 ); +assert_field_offset (entry.a7,SBI_SSE_ATTR_ENTRY_A7 ); +assert_field_offset (interrupted.exc_mode,SBI_SSE_ATTR_INTERRUPTED_MODE ); +assert_field_offset (interrupted.pc,SBI_SSE_ATTR_INTERRUPTED_PC ); +assert_field_offset (interrupted.a0,SBI_SSE_ATTR_INTERRUPTED_A0 ); +assert_field_offset (interrupted.a6,SBI_SSE_ATTR_INTERRUPTED_A6 ); +assert_field_offset (interrupted.a7,SBI_SSE_ATTR_INTERRUPTED_A7 ); + +struct sbi_sse_event { +struct sbi_sse_event_attrs attrs; +uint32_t event_id; +const struct sbi_sse_cb_ops *cb_ops; +struct sbi_dlist node; +/* Only global events are using the lock, local ones don't need it */ +spinlock_t lock; +}; + +struct sse_hart_state { +struct sbi_dlist event_list; +spinlock_t list_lock; +struct sbi_sse_event *local_events; +}; + +static unsigned int local_event_count; +static unsigned int global_event_count; +static struct sbi_sse_event *global_events; + +static unsigned long sse_inject_fifo_off; +static unsigned long sse_inject_fifo_mem_off; +/* Offset of pointer to SSE HART state in scratch space */ +static unsigned long shs_ptr_off; + +static u32 sse_ipi_inject_event =SBI_IPI_EVENT_MAX ; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id); + +static bool sse_event_is_global (struct sbi_sse_event *e) +{ +return EVENT_IS_GLOBAL (e->event_id); +} + +static void sse_global_event_lock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_lock (&e->lock); +} + +static void sse_global_event_unlock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_unlock (&e->lock); +} + +static void sse_event_set_state (struct sbi_sse_event *e, +enum sbi_sse_state new_state)lib/sbi/sbi_sse.c line 204
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+{ +enum sbi_sse_state prev_state =SSE_EVENT_STATE (e); + +if ((new_state - prev_state ==1 ) || (prev_state - new_state ==1 )) { +SSE_EVENT_STATE (e) = new_state; +return ; + } + +sbi_panic ("Invalid SSE state transition: %d -> %d\n" , prev_state, + new_state);Panic in a running system is not a good idea, IMHO. Thinking out loud, what if we let the state as is and return error (handled by caller and decide). In worst case, the event can be disabled but definitely not panic.
Quoted 73 lines
lib/sbi/sbi_sse.c new file +73
-0+} + +static struct sbi_sse_event *sse_event_get (uint32_t event) +{ +unsigned int i; +struct sbi_sse_event *events, *e; +unsigned int count; +struct sse_hart_state *shs; + +if (EVENT_IS_GLOBAL (event)) { + count = global_event_count; + events = global_events; + }else { + count = local_event_count; + shs =sse_thishart_state_ptr (); + events = shs->local_events; + } + +for (i =0 ; i < count; i++) { + e = &events[i]; +if (e->event_id == event) +return e; + } + +return NULL ; +} + +static struct sse_hart_state *sse_event_get_hart_state (struct sbi_sse_event *e) +{ +struct sbi_scratch *s =sbi_hartid_to_scratch (SSE_EVENT_HARTID (e)); + +return sse_get_hart_state_ptr (s); +} + +static void sse_event_remove_from_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); + +spin_lock (&state->list_lock); +sbi_list_del (&e->node); +spin_unlock (&state->list_lock); +} + +static void sse_event_add_to_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); +struct sbi_sse_event *tmp; + +spin_lock (&state->list_lock); +sbi_list_for_each_entry (tmp, &state->event_list, node) { +if (SSE_EVENT_PRIO (e) <SSE_EVENT_PRIO (tmp)) +break ; +if (SSE_EVENT_PRIO (e) ==SSE_EVENT_PRIO (tmp) && + e->event_id < tmp->event_id) +break ; + } +sbi_list_add_tail (&e->node, &tmp->node); + +spin_unlock (&state->list_lock); +} + +static int sse_event_disable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, disable_cb); + +sse_event_remove_from_list (e); +sse_event_set_state (e,SSE_STATE_REGISTERED ); + +return SBI_OK ; +}lib/sbi/sbi_sse.c line 287
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +static int sse_event_set_hart_id_check (struct sbi_sse_event *e, +unsigned long new_hartid) +{ +int hstate; +unsigned int hartid = (uint32_t ) new_hartid; +struct sbi_domain * hd =sbi_domain_thishart_ptr (); + +if (!sse_event_is_global (e)) +return SBI_EDENIED ;Why denied? Why not invalid param?
lib/sbi/sbi_sse.c line 290
Quoted 3 lines
lib/sbi/sbi_sse.c new file +3
-0+ +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) +return SBI_EBUSY ;In the spec, whenever the state doesn?t match the required state, we say we return
INVALID_STATE.
Can we do that? This is the only place where EBUSY is used. It might be good to have EBUSY but then it?s out of the scope of this patch set.Quoted 35 lines
lib/sbi/sbi_sse.c new file +35
-0+ +if (!sbi_domain_is_assigned_hart (hd, new_hartid)) +return SBI_EINVAL ; + + hstate =sbi_hsm_hart_get_state (hd, hartid); +if (hstate !=SBI_HSM_STATE_STARTED ) +return SBI_EINVAL ; + +return SBI_OK ; +} + +static int sse_event_set_attr_check (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +int ret =SBI_OK ; + +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +if (val & ~SBI_SSE_ATTR_CONFIG_ONESHOT ) + ret =SBI_ERR_INVALID_PARAM ; +break ; +case SBI_SSE_ATTR_PRIO : +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) + ret =SBI_EINVALID_STATE ; +break ; +case SBI_SSE_ATTR_PREFERRED_HART : + ret =sse_event_set_hart_id_check (e, val); +break ; +default : + ret =SBI_EBAD_RANGE ; +break ; + } + +return ret; +}lib/sbi/sbi_sse.c line 335
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +static void sse_event_set_attr (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +SSE_EVENT_CONFIG (e) = val; +break ; +case SBI_SSE_ATTR_PRIO : +SSE_EVENT_PRIO (e) = (uint32_t )val;Another reason to used defined length data types. I think we can do away with typecasts like this.
Quoted 16 lines
lib/sbi/sbi_sse.c new file +16
-0+break ; +case SBI_SSE_ATTR_PREFERRED_HART : +SSE_EVENT_HARTID (e) = val; +sse_event_invoke_cb (e, set_hartid_cb, val); +break ; + } +} + +static int sse_event_register (struct sbi_sse_event *e, +unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +if (sse_event_is_global (e) &&SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EINVAL ;lib/sbi/sbi_sse.c line 361
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +if (SSE_EVENT_STATE (e) !=SSE_STATE_UNUSED ) +return SBI_EINVALID_STATE ; + +SSE_EVENT_ENTRY (e).a0 = handler_entry_a0; +SSE_EVENT_ENTRY (e).a6 = handler_entry_a6; +SSE_EVENT_ENTRY (e).a7 = handler_entry_a7; +SSE_EVENT_ENTRY (e).pc = handler_entry_pc; + +sse_event_set_state (e,SSE_STATE_REGISTERED );Better return an error from set state (my previous comment). Handle it here based on what transition is expected. Here we just have crossed fingers that system won?t panic.
Quoted 87 lines
lib/sbi/sbi_sse.c new file +87
-0+ +sse_event_invoke_cb (e, register_cb); + +return 0 ; +} + +static int sse_event_unregister (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, unregister_cb); + +sse_event_set_state (e,SSE_STATE_UNUSED ); + +return 0 ; +} + +static void sse_event_inject (struct sbi_sse_event *e, +struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +ulong prev_smode, prev_virt; +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); +struct sse_interrupted_state *prev_i_ctx; +struct sse_entry_state *e_ctx = &SSE_EVENT_ENTRY (e); + +sse_event_set_state (e,SSE_STATE_RUNNING ); +SSE_EVENT_PENDING (e) = false; + +if (prev_e) { +/* back-to-back injection after another event, copy previous +* event context for correct restoration +*/ + prev_i_ctx = &SSE_EVENT_INTERRUPTED (prev_e); + +sse_regs_copy (i_ctx, prev_i_ctx); + i_ctx->exc_mode = prev_i_ctx->exc_mode; + i_ctx->pc = prev_i_ctx->pc; + }else { +sse_regs_copy (i_ctx, regs); + + prev_smode = (regs->mstatus &MSTATUS_MPP ) >>MSTATUS_MPP_SHIFT ; +#if __riscv_xlen ==32 + prev_virt = (regs->mstatusH &MSTATUSH_MPV ) ?1 :0 ; +#else + prev_virt = (regs->mstatus &MSTATUS_MPV ) ?1 :0 ; +#endif + + i_ctx->exc_mode = prev_smode <<EXC_MODE_PP_SHIFT ; + i_ctx->exc_mode |= prev_virt <<EXC_MODE_PV_SHIFT ; +if (regs->mstatus &MSTATUS_SPIE ) + i_ctx->exc_mode |=EXC_MODE_SSTATUS_SPIE ; + i_ctx->pc = regs->mepc; + +/* We only want to set SPIE for the first event injected after +* entering M-Mode. For the event injected right after another +* event (after calling sse_event_complete(), we will keep the +* saved SPIE). +*/ + regs->mstatus &= ~MSTATUS_SPIE ; +if (regs->mstatus &MSTATUS_SIE ) + regs->mstatus |=MSTATUS_SPIE ; + } + +sse_regs_copy (regs, e_ctx); + regs->mepc = e_ctx->pc; + + regs->mstatus &= ~MSTATUS_MPP ; + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +#endif + + regs->mstatus &= ~MSTATUS_SIE ; +} + +static void sse_event_resume (struct sbi_sse_event *e,struct sbi_trap_regs *regs) +{ +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); + +sse_regs_copy (regs, i_ctx); + +/* Restore previous virtualization state */ lib/sbi/sbi_sse.c line 458
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatusH |=MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatus |=MSTATUS_MPV ; +#endif +Probably have #error for undefined
__riscv_xlen.Quoted 577 lines
lib/sbi/sbi_sse.c new file +577
-0+ regs->mstatus &= ~MSTATUS_MPP ; +if (i_ctx->exc_mode &EXC_MODE_PP ) + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + + regs->mstatus &= ~MSTATUS_SIE ; +if (regs->mstatus &MSTATUS_SPIE ) + regs->mstatus |=MSTATUS_SIE ; + + regs->mstatus &= ~MSTATUS_SPIE ; +if (i_ctx->exc_mode &EXC_MODE_SSTATUS_SPIE ) + regs->mstatus |=MSTATUS_SPIE ; + + regs->mepc = i_ctx->pc; +} + +static bool sse_event_is_ready (struct sbi_sse_event *e) +{ +if (!SSE_EVENT_PENDING (e) ||SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED || +SSE_EVENT_HARTID (e) !=current_hartid ()) { +return false; + } + +return true; +} + +/* Return true if an event has been injected, false otherwise */ +static bool sse_process_pending_events (struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +struct sbi_sse_event *e, *to_run =NULL ; +struct sse_hart_state *state =sse_thishart_state_ptr (); + +retry: +spin_lock (&state->list_lock); + +if (sbi_list_empty (&state->event_list)) { +spin_unlock (&state->list_lock); +return false; + } + +sbi_list_for_each_entry (e, &state->event_list, node) { +/* +* List of event is ordered by priority, stop at first running +* event since all other events after this one are of lower +* priority. +*/ +if (SSE_EVENT_STATE (e) ==SSE_STATE_RUNNING ) +break ; + +if (sse_event_is_ready (e)) { + to_run = e; +break ; + } + } + +spin_unlock (&state->list_lock); + +if (!to_run) +return false; + +sse_global_event_lock (e); +/* +* If the event is global, the event might have been moved to another +* hart or disabled, evaluate readiness again. +*/ +if (sse_event_is_global (e) && !sse_event_is_ready (e)) { +sse_global_event_unlock (e); + goto retry; + } + +sse_event_inject (e, prev_e, regs); +sse_global_event_unlock (e); + +return true; +} + +static int sse_event_set_pending (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_RUNNING && +SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_ERR_INVALID_STATE ; + +SSE_EVENT_PENDING (e) = true; + +return SBI_OK ; +} + +static void sse_ipi_inject_process (struct sbi_scratch *scratch, +struct sbi_trap_regs *regs) +{ +struct sbi_sse_event *e; +struct sse_ipi_inject_data evt; +struct sbi_fifo *sse_inject_fifo_r = +sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + +/* This can be the case when sbi_exit() is called */ +if (!regs) +return ; + +/* Mark all queued events as pending */ +while (!sbi_fifo_dequeue (sse_inject_fifo_r, &evt)) { + e =sse_event_get (evt.event_id); +if (!e) +continue ; + +sse_global_event_lock (e); +sse_event_set_pending (e); +sse_global_event_unlock (e); + } + +sse_process_pending_events (NULL , regs); +} + +static struct sbi_ipi_event_ops sse_ipi_inject_ops = { + .name ="IPI_SSE_INJECT" , + .process = sse_ipi_inject_process, +}; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id) +{ +int ret; +struct sbi_scratch *remote_scratch =NULL ; +struct sse_ipi_inject_data evt = {event_id}; +struct sbi_fifo *sse_inject_fifo_r; + + remote_scratch =sbi_hartid_to_scratch (hartid); +if (!remote_scratch) +return SBI_EINVAL ; + sse_inject_fifo_r =sbi_scratch_offset_ptr (remote_scratch, + sse_inject_fifo_off); + + ret =sbi_fifo_enqueue (sse_inject_fifo_r, &evt); +if (ret) +return SBI_EFAIL ; + + ret =sbi_ipi_send_many (1 , hartid, sse_ipi_inject_event,NULL ); +if (ret) +return SBI_EFAIL ; + +return SBI_OK ; +} + +static int sse_inject_event (uint32_t event_id,unsigned long hartid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out,bool from_ecall) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +/* In case of global event, provided hart_id is ignored */ +if (sse_event_is_global (e)) + hartid =SSE_EVENT_HARTID (e); + +/* Event is for another hart, send it through IPI */ +if (hartid !=current_hartid ()) { +sse_global_event_unlock (e); +return sse_ipi_inject_send (hartid, event_id); + } + + ret =sse_event_set_pending (e); +sse_global_event_unlock (e); +if (ret) +return ret; + +if (from_ecall) { +/* Due to skip_regs_update = true being set if injection +* succeed an the fact that we need to modify regs here before +* injecting the event (regs are copied by the event), we need +* to set out->skip_regs_update to true to avoid. Moreover +* injection might not be done right now if another event of +* higher priority is already running so always set +* out->skip_regs_update to true there. +*/ + regs->mepc +=4 ; + regs->a0 =SBI_OK ; + out->skip_regs_update = true; + } + +if (sse_process_pending_events (NULL , regs)) + out->skip_regs_update = true; + +return SBI_OK ; +} + +static int sse_event_enable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_set_state (e,SSE_STATE_ENABLED ); +sse_event_add_to_list (e); + +if (SSE_EVENT_PENDING (e)) +sbi_ipi_send_many (1 ,SSE_EVENT_HARTID (e), sse_ipi_inject_event, +NULL ); + +sse_event_invoke_cb (e, enable_cb); + +return SBI_OK ; +} + +static int sse_event_complete (struct sbi_sse_event *e, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +bool inject; + +if (SSE_EVENT_STATE (e) !=SSE_STATE_RUNNING ) +return SBI_EINVALID_STATE ; + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EDENIED ; + +if (SSE_EVENT_CONFIG (e) &SBI_SSE_ATTR_CONFIG_ONESHOT ) +sse_event_disable (e); +else +sse_event_set_state (e,SSE_STATE_ENABLED ); + +sse_event_invoke_cb (e, complete_cb); + + inject =sse_process_pending_events (e, regs); +if (!inject) +sse_event_resume (e, regs); + + out->skip_regs_update = true; + +return SBI_OK ; +} + +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_complete (e, regs, out); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_enable (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_enable (e); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_disable (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_disable (e); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hartid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +if (!sbi_domain_is_assigned_hart (sbi_domain_thishart_ptr (), hartid)) +return SBI_EINVAL ; + +return sse_inject_event (event_id, hartid, regs, out, true); +} + +int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs) +{ +/* We don't really care about return value here */ +struct sbi_ecall_return out; + +return sse_inject_event (event_id,current_hartid (), regs, &out, false); +} + +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops) +{ +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +if (cb_ops->set_hartid_cb && !sse_event_is_global (e)) +return SBI_EINVAL ; + +sse_global_event_lock (e); + e->cb_ops = cb_ops; +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_attr_check (uint32_t base_attr_id,uint32_t attr_count, +unsigned long phys_lo, +unsigned long phys_hi, +unsigned long access) +{ +const unsigned align = __riscv_xlen >>3 ; +/* Avoid 32 bits overflow */ +uint64_t end_id = (uint64_t )base_attr_id + attr_count; + +if (end_id >SBI_SSE_ATTR_INTERRUPTED_A7 ) +return SBI_EBAD_RANGE ; + +if (phys_lo & (align -1 )) +return SBI_EINVALID_ADDR ; + +/* +* On RV32, the M-mode can only access the first 4GB of +* the physical address space because M-mode does not have +* MMU to access full 34-bit physical address space. +* +* Based on above, we simply fail if the upper 32bits of +* the physical address (i.e. a2 register) is non-zero on +* RV32. +*/ +if (phys_hi) +return SBI_EINVALID_ADDR ; + +if (!sbi_domain_check_addr_range (sbi_domain_thishart_ptr (), phys_lo, +sizeof (unsigned long ) * attr_count,1 , + access)) +return SBI_EINVALID_ADDR ; + +return SBI_OK ; +} + + +static void copy_attrs (unsigned long *out,const unsigned long *in, +unsigned int long_count) +{ +int i =0 ; + +/* +* sbi_memcpy() does byte-per-byte copy, using this yields long-per-long +* copy +*/ +for (i =0 ; i < long_count; i++) + out[i] = in[i]; +} + +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi) +{ +int ret; +unsigned long *e_attrs; +struct sbi_sse_event *e; +unsigned long *attrs; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, output_phys_lo, + output_phys_hi,SBI_DOMAIN_WRITE ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (output_phys_lo,sizeof (unsigned long ) * attr_count); + +/* +* Copy all attributes at once since struct sse_event_attrs is matching +* the SBI_SSE_ATTR_* attributes. While WRITE_ATTR attribute is not used +* in s-mode sse handling path, READ_ATTR is used to retrieve the value +* of registers when interrupted. rather than doing multiple SBI calls, +* a single one is done allowing to retrieve them all at once. +*/ + e_attrs = (unsigned long *)&e->attrs; + attrs = (unsigned long *)output_phys_lo; +copy_attrs (attrs, &e_attrs[base_attr_id], attr_count); + +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi) +{ +int ret; +struct sbi_sse_event *e; +unsigned long attr =0 , val; +uint32_t id, end_id = base_attr_id + attr_count; +unsigned long *attrs = (unsigned long *)input_phys_lo; + + ret =sbi_sse_attr_check (base_attr_id, attr_count, input_phys_lo, + input_phys_hi,SBI_DOMAIN_READ ); +if (ret) +return ret; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + +sbi_hart_map_saddr (input_phys_lo,sizeof (unsigned long ) * attr_count); + +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; + ret =sse_event_set_attr_check (e, id, val); +if (ret) + goto out; + } + + attr =0 ; +for (id = base_attr_id; id < end_id; id++) { + val = attrs[attr++]; +sse_event_set_attr (e, id, val); + } +out: +sbi_hart_unmap_saddr (); + +sse_global_event_unlock (e); + +return SBI_OK ; +} + +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_register (e, handler_entry_pc, handler_entry_a0, + handler_entry_a6, handler_entry_a7); +sse_global_event_unlock (e); + +return ret; +} + +int sbi_sse_unregister (uint32_t event_id) +{ +int ret; +struct sbi_sse_event *e; + + e =sse_event_get (event_id); +if (!e) +return SBI_EINVAL ; + +sse_global_event_lock (e); + ret =sse_event_unregister (e); +sse_global_event_unlock (e); + +return ret; +} + +static void sse_event_init (struct sbi_sse_event *e,uint32_t event_id) +{ + e->event_id = event_id; +SSE_EVENT_HARTID (e) =current_hartid (); +/* Declare all events as injectable */ +SSE_EVENT_CAN_INJECT (e) =1 ; +} + +static int sse_global_init () +{ +struct sbi_sse_event *e; +unsigned int i, ev =0 ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) + global_event_count++; +else + local_event_count++; + } + + global_events =sbi_zalloc (sizeof (*global_events) * global_event_count); +if (!global_events) +return SBI_ENOMEM ; + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (!EVENT_IS_GLOBAL (supported_events[i])) +continue ; + + e = &global_events[ev]; +sse_event_init (e, supported_events[i]); +SPIN_LOCK_INIT (e->lock); + + ev++; + } + +return 0 ; +} + +static void sse_local_init (struct sse_hart_state *shs) +{ +unsigned int i, ev =0 ; + +SBI_INIT_LIST_HEAD (&shs->event_list); +SPIN_LOCK_INIT (shs->list_lock); + +for (i =0 ; i <EVENT_COUNT ; i++) { +if (EVENT_IS_GLOBAL (supported_events[i])) +continue ; + +sse_event_init (&shs->local_events[ev++], supported_events[i]); + } +} + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot) +{ +int ret; +void *sse_inject_mem; +struct sse_hart_state *shs; +struct sbi_fifo *sse_inject_q; + +if (cold_boot) { + ret =sse_global_init (); +if (ret) +return ret; + + shs_ptr_off =sbi_scratch_alloc_offset (sizeof (void *)); +if (!shs_ptr_off) +return SBI_ENOMEM ; + + sse_inject_fifo_off =sbi_scratch_alloc_offset ( +sizeof (*sse_inject_q)); +if (!sse_inject_fifo_off) { +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + sse_inject_fifo_mem_off =sbi_scratch_alloc_offset ( +EVENT_COUNT *sizeof (struct sse_ipi_inject_data )); +if (!sse_inject_fifo_mem_off) { +sbi_scratch_free_offset (sse_inject_fifo_off); +sbi_scratch_free_offset (shs_ptr_off); +return SBI_ENOMEM ; + } + + ret =sbi_ipi_event_create (&sse_ipi_inject_ops); +if (ret <0 ) { +sbi_scratch_free_offset (shs_ptr_off); +return ret; + }lib/sbi/sbi_sse.c line 1045
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ sse_ipi_inject_event = ret; + } + + shs =sse_get_hart_state_ptr (scratch); +if (!shs) { +/* Allocate per hart state and local events at once */ + shs =sbi_zalloc (sizeof (*shs) + +sizeof (struct sbi_sse_event ) * local_event_count); +if (!shs) +return SBI_ENOMEM ;What about scratch offset allocated above. Shouldn?t that be freed? I suggest to have a goto to undo stuff in case of error.
Quoted 35 lines
lib/sbi/sbi_sse.c new file +33
-0+ + shs->local_events = (struct sbi_sse_event *)(shs +1 ); + +sse_set_hart_state_ptr (scratch, shs); + } + +sse_local_init (shs); + + sse_inject_q =sbi_scratch_offset_ptr (scratch, sse_inject_fifo_off); + sse_inject_mem =sbi_scratch_offset_ptr (scratch, + sse_inject_fifo_mem_off); + +sbi_fifo_init (sse_inject_q, sse_inject_mem,EVENT_COUNT , +sizeof (struct sse_ipi_inject_data )); + +return 0 ; +} + +void sbi_sse_exit (struct sbi_scratch *scratch) +{ +int i; +struct sbi_sse_event *e; + +for (i =0 ; i <EVENT_COUNT ; i++) { + e =sse_event_get (supported_events[i]); + +if (SSE_EVENT_HARTID (e) !=current_hartid ()) +continue ; + +if (SSE_EVENT_STATE (e) >SSE_STATE_REGISTERED ) +sbi_printf ("Event %d in invalid state at exit" , i); + } +}Signature · 1 line
2.43.0
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
In reply to Himanshu ChauhanHi Himanshu and thanks for the review,
Quoted 83 lines · Himanshu Chauhan
Hi Clement,
Thanks for the patch! My comments inline.
Quoted 66 lines · Cl?ment L?ger
This extension [1] allows to deliver events from SBI to supervisor via a software mecanism. This extensions defines events (either local or global) which are signaled by the SBI on specific signal sources (IRQ, traps, etc) and are injected to be executed in supervisor mode.
[1] https://lists.riscv.org/g/tech-prs/message/744
Signed-off-by Cl?ment L?ger <[email protected]>---
include/sbi/sbi_ecall_interface.h | 48 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 2 +- lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ 9 files changed, 1301 insertions(+), 3 deletions(-)create mode 100644
include/sbi/sbi_sse.h
create mode 100644lib/sbi/sbi_ecall_sse.c
create mode 100644lib/sbi/sbi_sse.cinclude/sbi/sbi_ecall_interface.h +26
-0@@ -32,6 +32,7 @@#define SBI_EXT_DBCN 0x4442434E#define SBI_EXT_SUSP 0x53555350#define SBI_EXT_CPPC 0x43505043 +#define SBI_EXT_SSE 0x535345/* SBI function IDs for BASE extension*/ #define SBI_EXT_BASE_GET_SPEC_VERSION 0x0@@ -293,6 +294,48 @@ enum sbi_cppc_reg_id { SBI_CPPC_NON_ACPI_LAST = SBI_CPPC_TRANSITION_LATENCY, }; +/* SBI Function IDs for SSE extension */ +#define SBI_EXT_SSE_READ_ATTR 0x00000000 +#define SBI_EXT_SSE_WRITE_ATTR 0x00000001 +#define SBI_EXT_SSE_REGISTER 0x00000002 +#define SBI_EXT_SSE_UNREGISTER 0x00000003 +#define SBI_EXT_SSE_ENABLE 0x00000004 +#define SBI_EXT_SSE_DISABLE 0x00000005 +#define SBI_EXT_SSE_COMPLETE 0x00000006 +#define SBI_EXT_SSE_INJECT 0x00000007 + +/* SBI SSE Event Attributes. */ +#define SBI_SSE_ATTR_STATUS 0x00000000 +#define SBI_SSE_ATTR_PRIO 0x00000001 +#define SBI_SSE_ATTR_CONFIG 0x00000002 +#define SBI_SSE_ATTR_PREFERRED_HART 0x00000003 +#define SBI_SSE_ATTR_ENTRY_PC 0x00000004 +#define SBI_SSE_ATTR_ENTRY_A0 0x00000005 +#define SBI_SSE_ATTR_ENTRY_A6 0x00000006 +#define SBI_SSE_ATTR_ENTRY_A7 0x00000007 +#define SBI_SSE_ATTR_INTERRUPTED_MODE 0x00000008 +#define SBI_SSE_ATTR_INTERRUPTED_PC 0x00000009 +#define SBI_SSE_ATTR_INTERRUPTED_A0 0x0000000A +#define SBI_SSE_ATTR_INTERRUPTED_A6 0x0000000B +#define SBI_SSE_ATTR_INTERRUPTED_A7 0x0000000C +include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 331
Quoted 10 lines
include/sbi/sbi_ecall_interface.h +10
-0+#define SBI_SSE_ATTR_STATUS_STATE_OFFSET 0 +#define SBI_SSE_ATTR_STATUS_STATE_MASK 0x3 +#define SBI_SSE_ATTR_STATUS_PENDING_OFFSET 2 +#define SBI_SSE_ATTR_STATUS_INJECT_OFFSET 3 + +#define SBI_SSE_ATTR_CONFIG_ONESHOT (1 << 0) + +/* SBI SSE Event IDs. */ +#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000 +#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000Can we add a define for start of platform specific local events (
0x4000)?
Quoted 6 lines
include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 335
Quoted 4 lines
include/sbi/sbi_ecall_interface.h +4
-0+#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000 +#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000 +#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000 +IMHO, we should add all the ranges defined in spec, used or not.
Since there are multiple ranges reserved for the same kind of events, it will need a numbering of some sort, is something like this ok for you:
#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000
#define SBI_SSE_EVENT_LOCAL_PLAT_0_START 0x00004000
#define SBI_SSE_EVENT_LOCAL_PLAT_0_END 0x00007fff
#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000
#define SBI_SSE_EVENT_GLOBAL_PLAT_0_START 0x00004000
#define SBI_SSE_EVENT_GLOBAL_PLAT_0_END 0x00007fff
#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000
#define SBI_SSE_EVENT_LOCAL_PLAT_1_START 0x00014000
#define SBI_SSE_EVENT_LOCAL_PLAT_1_END 0x00017fff
#define SBI_SSE_EVENT_GLOBAL_PLAT_1_START 0x0001c000
#define SBI_SSE_EVENT_GLOBAL_PLAT_1_END 0x0001ffff
#define SBI_SSE_EVENT_LOCAL_PLAT_2_START 0x00024000
#define SBI_SSE_EVENT_LOCAL_PLAT_2_END 0x00027fff
#define SBI_SSE_EVENT_GLOBAL_PLAT_2_START 0x0002c000
#define SBI_SSE_EVENT_GLOBAL_PLAT_2_END 0x0002ffff
#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000
#define SBI_SSE_EVENT_LOCAL_PLAT_3_START 0xffff4000
#define SBI_SSE_EVENT_LOCAL_PLAT_3_END 0xffff7fff
#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000
#define SBI_SSE_EVENT_GLOBAL_PLAT_3_START 0xffffc000
#define SBI_SSE_EVENT_GLOBAL_PLAT_3_END 0xffffffff
Quoted 420 lines
Quoted 115 lines
include/sbi/sbi_ecall_interface.h +7
-1+#define SBI_SSE_EVENT_GLOBAL (1 << 15) +#define SBI_SSE_EVENT_PLATFORM (1 << 14) +/* SBI base specification related macros */ #define SBI_SPEC_VERSION_MAJOR_OFFSET 24#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f@@ -313,8 +356,11 @@ enum sbi_cppc_reg_id {#define SBI_ERR_ALREADY_STARTED -7#define SBI_ERR_ALREADY_STOPPED -8#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_BUSY -12 -#define SBI_LAST_ERR SBI_ERR_NO_SHMEM +#define SBI_LAST_ERR SBI_ERR_BUSY/* clang-format on */ include/sbi/sbi_error.h +3
-1@@ -24,6 +24,9 @@#define SBI_EALREADY_STARTED SBI_ERR_ALREADY_STARTED#define SBI_EALREADY_STOPPED SBI_ERR_ALREADY_STOPPED#define SBI_ENO_SHMEM SBI_ERR_NO_SHMEM +#define SBI_EINVALID_STATE SBI_ERR_INVALID_STATE +#define SBI_EBAD_RANGE SBI_ERR_BAD_RANGE +#define SBI_EBUSY SBI_ERR_BUSY#define SBI_ENODEV -1000#define SBI_ENOSYS -1001@@ -34,7 +37,6 @@#define SBI_ENOMEM -1006#define SBI_EUNKNOWN -1007#define SBI_ENOENT -1008 -/* clang-format on */ #endif include/sbi/sbi_sse.h new file +67
-0@@ -0,0 +1,93 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems. +*/ + +#ifndef __SBI_SSE_H__ +#define __SBI_SSE_H__ + +#include <sbi/sbi_types.h> +#include <sbi/sbi_list.h> +#include <sbi/riscv_locks.h> + +struct sbi_scratch ; + +#define EXC_MODE_PP_SHIFT 0 +#define EXC_MODE_PP BIT(EXC_MODE_PP_SHIFT) +#define EXC_MODE_PV_SHIFT 1 +#define EXC_MODE_PV BIT(EXC_MODE_PV_SHIFT) +#define EXC_MODE_SSTATUS_SPIE_SHIFT 2 +#define EXC_MODE_SSTATUS_SPIE BIT(EXC_MODE_SSTATUS_SPIE_SHIFT) + + +struct sbi_sse_cb_ops { +/** +* Called when hart_id is changed on the event. +*/ +void (*set_hartid_cb)(uint32_t event_id,unsigned long hart_id); + +/** +* Called when the SBI_EXT_SSE_COMPLETE is invoked on the event. +*/ +void (*complete_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_REGISTER is invoked on the event. +*/ +void (*register_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_UNREGISTER is invoked on the event. +*/ +void (*unregister_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_ENABLE is invoked on the event. +*/ +void (*enable_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_DISABLE is invoked on the event. +*/ +void (*disable_cb)(uint32_t event_id); +}; + +/* Set the callback operations for an event +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param cb_ops Callback operations +* @return 0 on success, error otherwise +*/ +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops); + +/* Inject an event to the current hard +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param regs Registers that were used on SBI entry +* @return 0 on success, error otherwise +*/ include/sbi/sbi_sse.h line 77
Quoted 10 lines
include/sbi/sbi_sse.h new file +10
-0+int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs); + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot); +void sbi_sse_exit (struct sbi_scratch *scratch); + +/* Interface called from sbi_ecall_sse.c */ +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7);Wouldn?t it be better to only use defined length data types like
uint32_t,uint64_teverywhere? Instead of mixing them up?Quoted 292 lines
include/sbi/sbi_sse.h new file +16
-0+int sbi_sse_unregister (uint32_t event_id); +int sbi_sse_enable (uint32_t event_id); +int sbi_sse_disable (uint32_t event_id); +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hart_id, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi); +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi); + +#endif lib/sbi/Kconfig +0
-0@@ -46,4 +46,8 @@ config SBI_ECALL_VENDORbool "Platform-defined vendor extensions"
default ylib/sbi/Kconfig +4
-0+config SBI_ECALL_SSE + bool "SSE extension" + default y +endmenu
lib/sbi/objects.mk +0
-0@@ -52,6 +52,9 @@ libsbi-objs-$(CONFIG_SBI_ECALL_LEGACY) += sbi_ecall_legacy.ocarray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_VENDOR) +=
ecall_vendorlibsbi-objs-$(CONFIG_SBI_ECALL_VENDOR) += sbi_ecall_vendor.olib/sbi/objects.mk +3
-0+carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_SSE) += ecall_sse +libsbi-objs-$(CONFIG_SBI_ECALL_SSE) += sbi_ecall_sse.o +libsbi-objs-y += sbi_bitmap.o
libsbi-objs-y += sbi_bitops.o
libsbi-objs-y += sbi_console.olib/sbi/objects.mk +0
-0@@ -71,6 +74,7 @@ libsbi-objs-y += sbi_misaligned_ldst.olibsbi-objs-y += sbi_platform.o
libsbi-objs-y += sbi_pmu.o
libsbi-objs-y += sbi_scratch.o
+libsbi-objs-y += sbi_sse.o
libsbi-objs-y += sbi_string.o
libsbi-objs-y += sbi_system.o
libsbi-objs-y += sbi_timer.olib/sbi/sbi_ecall_sse.c new file +58
-0@@ -0,0 +1,58 @@ +#include <sbi/sbi_error.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_trap.h> +#include <sbi/sbi_sse.h> + +static int sbi_ecall_sse_handler (unsigned long extid,unsigned long funcid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; + +switch (funcid) { +case SBI_EXT_SSE_READ_ATTR : + ret =sbi_sse_read_attrs (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a4); +break ; +case SBI_EXT_SSE_WRITE_ATTR : + ret =sbi_sse_write_attrs (regs->a0, regs->a1, regs->a2, + regs->a3, regs->a4); +break ; +case SBI_EXT_SSE_REGISTER : + ret =sbi_sse_register (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a2); +break ; +case SBI_EXT_SSE_UNREGISTER : + ret =sbi_sse_unregister (regs->a0); +break ; +case SBI_EXT_SSE_ENABLE : + ret =sbi_sse_enable (regs->a0); +break ; +case SBI_EXT_SSE_DISABLE : + ret =sbi_sse_disable (regs->a0); +break ; +case SBI_EXT_SSE_COMPLETE : + ret =sbi_sse_complete (regs->a0, regs, out); +break ; +case SBI_EXT_SSE_INJECT : + ret =sbi_sse_inject_from_ecall (regs->a0, regs->a1, regs, out); +break ; +default : + ret =SBI_ENOTSUPP ; + } +return ret; +} + +struct sbi_ecall_extension ecall_sse; + +static int sbi_ecall_sse_register_extensions (void ) +{ +return sbi_ecall_register_extension (&ecall_sse); +} + +struct sbi_ecall_extension ecall_sse = { + .extid_start =SBI_EXT_SSE , + .extid_end =SBI_EXT_SSE , + .register_extensions = sbi_ecall_sse_register_extensions, + .handle = sbi_ecall_sse_handler, +};lib/sbi/sbi_init.c +13
-0@@ -23,6 +23,7 @@#include <sbi/sbi_irqchip.h> #include <sbi/sbi_platform.h> #include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> #include <sbi/sbi_system.h> #include <sbi/sbi_string.h> #include <sbi/sbi_timer.h> @@ -315,6 +316,12 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, true); +if (rc) { +sbi_printf ("%s: sse init failed (error %d)\n" , __func__, rc); +sbi_hart_hang (); + } + rc =sbi_pmu_init (scratch, true);if (rc) { sbi_printf("%s: pmu init failed (error %d)\n",@@ -435,6 +442,10 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, false); +if (rc) +sbi_hart_hang (); + rc =sbi_pmu_init (scratch, false);if (rc)sbi_hart_hang ();@@ -639,6 +650,8 @@ void __noreturn sbi_exit(struct sbi_scratch *scratch)sbi_platform_early_exit (plat); +sbi_sse_exit (scratch); +sbi_pmu_exit (scratch);sbi_timer_exit (scratch);lib/sbi/sbi_ipi.c +1
-1@@ -66,7 +66,7 @@ static int sbi_ipi_send(struct sbi_scratch *scratch, u32 remote_hartindex,* SBI_IPI_UPDATE_BREAK for self-IPIs. For other events, check * for self-IPI and execute the callback directly here. */ - ipi_ops->process (scratch); + ipi_ops->process (scratch,NULL );return 0 ; }lib/sbi/sbi_sse.c new file +113
-0@@ -0,0 +1,1078 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems Inc. +* +*/ + +#include <sbi/riscv_asm.h> +#include <sbi/riscv_barrier.h> +#include <sbi/riscv_encoding.h> +#include <sbi/riscv_locks.h> +#include <sbi/sbi_domain.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_ecall_interface.h> +#include <sbi/sbi_error.h> +#include <sbi/sbi_fifo.h> +#include <sbi/sbi_hart.h> +#include <sbi/sbi_heap.h> +#include <sbi/sbi_hsm.h> +#include <sbi/sbi_ipi.h> +#include <sbi/sbi_list.h> +#include <sbi/sbi_platform.h> +#include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> +#include <sbi/sbi_scratch.h> +#include <sbi/sbi_string.h> +#include <sbi/sbi_trap.h> + +#include <sbi/sbi_console.h> + +#define sse_get_hart_state_ptr (__scratch) \ + sbi_scratch_read_type((__scratch), void *, shs_ptr_off) + +#define sse_thishart_state_ptr () \ + sse_get_hart_state_ptr(sbi_scratch_thishart_ptr()) + +#define sse_set_hart_state_ptr (__scratch, __sse_state) \ + sbi_scratch_write_type((__scratch), void *, shs_ptr_off, (__sse_state)) + +/* +* Rather than using memcpy to copy the context (which does it byte-per-byte), +* copy each field which generates ld/lw. +*/ +#define sse_regs_copy (dst, src) \ + (dst)->a0 = (src)->a0; \ + (dst)->a6 = (src)->a6; \ + (dst)->a7 = (src)->a7 + +#define EVENT_IS_GLOBAL (__event_id) ((__event_id) & SBI_SSE_EVENT_GLOBAL) + +static const uint32_t supported_events[] = +{ +SBI_SSE_EVENT_LOCAL_RAS , +SBI_SSE_EVENT_GLOBAL_RAS , +SBI_SSE_EVENT_LOCAL_PMU , +SBI_SSE_EVENT_LOCAL_SOFTWARE , +SBI_SSE_EVENT_GLOBAL_SOFTWARE , +}; + +#define EVENT_COUNT array_size(supported_events) + +#define sse_event_invoke_cb (_event, _cb, ...) \ +{ \ + if (_event->cb_ops && _event->cb_ops->_cb) \ + _event->cb_ops->_cb(_event->event_id, ##__VA_ARGS__); \ +} + +#define SSE_EVENT_STATE (__e) __e->attrs.status.state +#define SSE_EVENT_PENDING (__e) __e->attrs.status.pending +#define SSE_EVENT_CAN_INJECT (__e) __e->attrs.status.inject +#define SSE_EVENT_HARTID (__e) __e->attrs.hartid +#define SSE_EVENT_PRIO (__e) __e->attrs.prio +#define SSE_EVENT_CONFIG (__e) __e->attrs.config +#define SSE_EVENT_ENTRY (__e) __e->attrs.entry +#define SSE_EVENT_INTERRUPTED (__e) __e->attrs.interrupted + +struct sse_entry_state { +/** Entry program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +struct sse_interrupted_state { +/** Exception mode */ +unsigned long exc_mode; +/** Interrupted program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +enum sbi_sse_state { +SSE_STATE_UNUSED =0 , +SSE_STATE_REGISTERED =1 , +SSE_STATE_ENABLED =2 , +SSE_STATE_RUNNING =3 , +}; + +struct sse_ipi_inject_data { +uint32_t event_id; +}; + +struct sbi_sse_event_status { +unsigned long state:2 ;
Quoted 8 lines
lib/sbi/sbi_sse.c line 116
Quoted 3 lines
lib/sbi/sbi_sse.c new file +3
-0+ unsigned long pending:1; + unsigned long inject:1; +} __packed;Can we have do away with bit fields? Like everywhere else use positions and masks? Why packed? We are making sure
sse_event_attrsis aligned to unsigned long so that We can copy with loops. Probably make it a long with reserved entry? Unless you had Something else in mind.
Using bitfield was purely for the ease of use. I could of course use a long with some shift masks if you prefer that over bitfields.
Quoted 84 lines
Quoted 84 lines
lib/sbi/sbi_sse.c new file +83
-0+ +struct sbi_sse_event_attrs { +struct sbi_sse_event_status status; +unsigned long prio; +unsigned long config; +unsigned long hartid; +struct sse_entry_state entry; +struct sse_interrupted_state interrupted; +}; + +/* Make sure all attributes are packed for direct memcpy in ATTR_READ */ +#define assert_field_offset (field, attr_offset) \ + _Static_assert( \ + ((offsetof(struct sbi_sse_event_attrs, field)) / sizeof(unsigned long)) == attr_offset, \ + "field "#field " from struct sbi_sse_event_attrs invalid offset, expected "#attr_offset \ + ) + +assert_field_offset (status,SBI_SSE_ATTR_STATUS ); +assert_field_offset (prio,SBI_SSE_ATTR_PRIO ); +assert_field_offset (config,SBI_SSE_ATTR_CONFIG ); +assert_field_offset (hartid,SBI_SSE_ATTR_PREFERRED_HART ); +assert_field_offset (entry.pc,SBI_SSE_ATTR_ENTRY_PC ); +assert_field_offset (entry.a0,SBI_SSE_ATTR_ENTRY_A0 ); +assert_field_offset (entry.a6,SBI_SSE_ATTR_ENTRY_A6 ); +assert_field_offset (entry.a7,SBI_SSE_ATTR_ENTRY_A7 ); +assert_field_offset (interrupted.exc_mode,SBI_SSE_ATTR_INTERRUPTED_MODE ); +assert_field_offset (interrupted.pc,SBI_SSE_ATTR_INTERRUPTED_PC ); +assert_field_offset (interrupted.a0,SBI_SSE_ATTR_INTERRUPTED_A0 ); +assert_field_offset (interrupted.a6,SBI_SSE_ATTR_INTERRUPTED_A6 ); +assert_field_offset (interrupted.a7,SBI_SSE_ATTR_INTERRUPTED_A7 ); + +struct sbi_sse_event { +struct sbi_sse_event_attrs attrs; +uint32_t event_id; +const struct sbi_sse_cb_ops *cb_ops; +struct sbi_dlist node; +/* Only global events are using the lock, local ones don't need it */ +spinlock_t lock; +}; + +struct sse_hart_state { +struct sbi_dlist event_list; +spinlock_t list_lock; +struct sbi_sse_event *local_events; +}; + +static unsigned int local_event_count; +static unsigned int global_event_count; +static struct sbi_sse_event *global_events; + +static unsigned long sse_inject_fifo_off; +static unsigned long sse_inject_fifo_mem_off; +/* Offset of pointer to SSE HART state in scratch space */ +static unsigned long shs_ptr_off; + +static u32 sse_ipi_inject_event =SBI_IPI_EVENT_MAX ; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id); + +static bool sse_event_is_global (struct sbi_sse_event *e) +{ +return EVENT_IS_GLOBAL (e->event_id); +} + +static void sse_global_event_lock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_lock (&e->lock); +} + +static void sse_global_event_unlock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_unlock (&e->lock); +} + +static void sse_event_set_state (struct sbi_sse_event *e, +enum sbi_sse_state new_state) +{ +enum sbi_sse_state prev_state =SSE_EVENT_STATE (e); + +if ((new_state - prev_state ==1 ) || (prev_state - new_state ==1 )) { +SSE_EVENT_STATE (e) = new_state;
Quoted 8 lines
lib/sbi/sbi_sse.c line 204
Quoted 5 lines
lib/sbi/sbi_sse.c new file +5
-0+return ; + } + +sbi_panic ("Invalid SSE state transition: %d -> %d\n" , prev_state, + new_state);Panic in a running system is not a good idea, IMHO. Thinking out loud, what if we let the state as is and return error (handled by caller and decide). In worst case, the event can be disabled but definitely not panic.
Yeah makes sense, Even if this is mostly debugging code, it could help catch some nasty bug without crashing the system. I'll try to convert that to return a value.
Quoted 78 lines
Quoted 78 lines
lib/sbi/sbi_sse.c new file +77
-0+} + +static struct sbi_sse_event *sse_event_get (uint32_t event) +{ +unsigned int i; +struct sbi_sse_event *events, *e; +unsigned int count; +struct sse_hart_state *shs; + +if (EVENT_IS_GLOBAL (event)) { + count = global_event_count; + events = global_events; + }else { + count = local_event_count; + shs =sse_thishart_state_ptr (); + events = shs->local_events; + } + +for (i =0 ; i < count; i++) { + e = &events[i]; +if (e->event_id == event) +return e; + } + +return NULL ; +} + +static struct sse_hart_state *sse_event_get_hart_state (struct sbi_sse_event *e) +{ +struct sbi_scratch *s =sbi_hartid_to_scratch (SSE_EVENT_HARTID (e)); + +return sse_get_hart_state_ptr (s); +} + +static void sse_event_remove_from_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); + +spin_lock (&state->list_lock); +sbi_list_del (&e->node); +spin_unlock (&state->list_lock); +} + +static void sse_event_add_to_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); +struct sbi_sse_event *tmp; + +spin_lock (&state->list_lock); +sbi_list_for_each_entry (tmp, &state->event_list, node) { +if (SSE_EVENT_PRIO (e) <SSE_EVENT_PRIO (tmp)) +break ; +if (SSE_EVENT_PRIO (e) ==SSE_EVENT_PRIO (tmp) && + e->event_id < tmp->event_id) +break ; + } +sbi_list_add_tail (&e->node, &tmp->node); + +spin_unlock (&state->list_lock); +} + +static int sse_event_disable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, disable_cb); + +sse_event_remove_from_list (e); +sse_event_set_state (e,SSE_STATE_REGISTERED ); + +return SBI_OK ; +} + +static int sse_event_set_hart_id_check (struct sbi_sse_event *e, +unsigned long new_hartid) +{
Quoted 8 lines
lib/sbi/sbi_sse.c line 287
Quoted 6 lines
lib/sbi/sbi_sse.c new file +6
-0+int hstate; +unsigned int hartid = (uint32_t ) new_hartid; +struct sbi_domain * hd =sbi_domain_thishart_ptr (); + +if (!sse_event_is_global (e)) +return SBI_EDENIED ;Why denied? Why not invalid param?
Oops, probably a left over of some previous version.
Quoted 7 lines
lib/sbi/sbi_sse.c line 290
Quoted 4 lines
lib/sbi/sbi_sse.c new file +3
-0+ +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) +return SBI_EBUSY ;In the spec, whenever the state doesn?t match the required state, we say we return
INVALID_STATE.
Can we do that? This is the only place where EBUSY is used. It might be good to have EBUSY but then it?s out of the scope of this patch set.
Agreed, this was specified in a previous spec version also. I'll change that.
Quoted 170 lines
Quoted 36 lines
lib/sbi/sbi_sse.c new file +35
-0+ +if (!sbi_domain_is_assigned_hart (hd, new_hartid)) +return SBI_EINVAL ; + + hstate =sbi_hsm_hart_get_state (hd, hartid); +if (hstate !=SBI_HSM_STATE_STARTED ) +return SBI_EINVAL ; + +return SBI_OK ; +} + +static int sse_event_set_attr_check (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +int ret =SBI_OK ; + +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +if (val & ~SBI_SSE_ATTR_CONFIG_ONESHOT ) + ret =SBI_ERR_INVALID_PARAM ; +break ; +case SBI_SSE_ATTR_PRIO : +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) + ret =SBI_EINVALID_STATE ; +break ; +case SBI_SSE_ATTR_PREFERRED_HART : + ret =sse_event_set_hart_id_check (e, val); +break ; +default : + ret =SBI_EBAD_RANGE ; +break ; + } + +return ret; +}lib/sbi/sbi_sse.c line 335
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +static void sse_event_set_attr (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +SSE_EVENT_CONFIG (e) = val; +break ; +case SBI_SSE_ATTR_PRIO : +SSE_EVENT_PRIO (e) = (uint32_t )val;Another reason to used defined length data types. I think we can do away with typecasts like this.
Quoted 16 lines
lib/sbi/sbi_sse.c new file +16
-0+break ; +case SBI_SSE_ATTR_PREFERRED_HART : +SSE_EVENT_HARTID (e) = val; +sse_event_invoke_cb (e, set_hartid_cb, val); +break ; + } +} + +static int sse_event_register (struct sbi_sse_event *e, +unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +if (sse_event_is_global (e) &&SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EINVAL ;lib/sbi/sbi_sse.c line 361
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +if (SSE_EVENT_STATE (e) !=SSE_STATE_UNUSED ) +return SBI_EINVALID_STATE ; + +SSE_EVENT_ENTRY (e).a0 = handler_entry_a0; +SSE_EVENT_ENTRY (e).a6 = handler_entry_a6; +SSE_EVENT_ENTRY (e).a7 = handler_entry_a7; +SSE_EVENT_ENTRY (e).pc = handler_entry_pc; + +sse_event_set_state (e,SSE_STATE_REGISTERED );Better return an error from set state (my previous comment). Handle it here based on what transition is expected. Here we just have crossed fingers that system won?t panic.
Quoted 91 lines
lib/sbi/sbi_sse.c new file +91
-0+ +sse_event_invoke_cb (e, register_cb); + +return 0 ; +} + +static int sse_event_unregister (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, unregister_cb); + +sse_event_set_state (e,SSE_STATE_UNUSED ); + +return 0 ; +} + +static void sse_event_inject (struct sbi_sse_event *e, +struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +ulong prev_smode, prev_virt; +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); +struct sse_interrupted_state *prev_i_ctx; +struct sse_entry_state *e_ctx = &SSE_EVENT_ENTRY (e); + +sse_event_set_state (e,SSE_STATE_RUNNING ); +SSE_EVENT_PENDING (e) = false; + +if (prev_e) { +/* back-to-back injection after another event, copy previous +* event context for correct restoration +*/ + prev_i_ctx = &SSE_EVENT_INTERRUPTED (prev_e); + +sse_regs_copy (i_ctx, prev_i_ctx); + i_ctx->exc_mode = prev_i_ctx->exc_mode; + i_ctx->pc = prev_i_ctx->pc; + }else { +sse_regs_copy (i_ctx, regs); + + prev_smode = (regs->mstatus &MSTATUS_MPP ) >>MSTATUS_MPP_SHIFT ; +#if __riscv_xlen ==32 + prev_virt = (regs->mstatusH &MSTATUSH_MPV ) ?1 :0 ; +#else + prev_virt = (regs->mstatus &MSTATUS_MPV ) ?1 :0 ; +#endif + + i_ctx->exc_mode = prev_smode <<EXC_MODE_PP_SHIFT ; + i_ctx->exc_mode |= prev_virt <<EXC_MODE_PV_SHIFT ; +if (regs->mstatus &MSTATUS_SPIE ) + i_ctx->exc_mode |=EXC_MODE_SSTATUS_SPIE ; + i_ctx->pc = regs->mepc; + +/* We only want to set SPIE for the first event injected after +* entering M-Mode. For the event injected right after another +* event (after calling sse_event_complete(), we will keep the +* saved SPIE). +*/ + regs->mstatus &= ~MSTATUS_SPIE ; +if (regs->mstatus &MSTATUS_SIE ) + regs->mstatus |=MSTATUS_SPIE ; + } + +sse_regs_copy (regs, e_ctx); + regs->mepc = e_ctx->pc; + + regs->mstatus &= ~MSTATUS_MPP ; + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +#endif + + regs->mstatus &= ~MSTATUS_SIE ; +} + +static void sse_event_resume (struct sbi_sse_event *e,struct sbi_trap_regs *regs) +{ +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); + +sse_regs_copy (regs, i_ctx); + +/* Restore previous virtualization state */ +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatusH |=MSTATUSH_MPV ;
Quoted 8 lines
lib/sbi/sbi_sse.c line 458
Quoted 6 lines
lib/sbi/sbi_sse.c new file +6
-0+#else + regs->mstatus &= ~MSTATUS_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatus |=MSTATUS_MPV ; +#endif +Probably have #error for undefined
__riscv_xlen.
Acked
Thanks,
Cl?ment
RFC PATCH v2 2/4 lib: sbi: add support for Supervisor Software Events extension
Hi Clement,
Quoted 116 lines · Cl?ment L?ger
Hi Himanshu and thanks for the review,
Quoted 83 lines · Himanshu Chauhan
Hi Clement,
Thanks for the patch! My comments inline.
Quoted 66 lines · Cl?ment L?ger
This extension [1] allows to deliver events from SBI to supervisor via a software mecanism. This extensions defines events (either local or global) which are signaled by the SBI on specific signal sources (IRQ, traps, etc) and are injected to be executed in supervisor mode.
[1] https://lists.riscv.org/g/tech-prs/message/744
Signed-off-by Cl?ment L?ger <[email protected]>---
include/sbi/sbi_ecall_interface.h | 48 +- include/sbi/sbi_error.h | 4 +- include/sbi/sbi_sse.h | 93 +++ lib/sbi/Kconfig | 4 + lib/sbi/objects.mk | 4 + lib/sbi/sbi_ecall_sse.c | 58 ++ lib/sbi/sbi_init.c | 13 + lib/sbi/sbi_ipi.c | 2 +- lib/sbi/sbi_sse.c | 1078 +++++++++++++++++++++++++++++ 9 files changed, 1301 insertions(+), 3 deletions(-)create mode 100644
include/sbi/sbi_sse.h
create mode 100644lib/sbi/sbi_ecall_sse.c
create mode 100644lib/sbi/sbi_sse.cinclude/sbi/sbi_ecall_interface.h +26
-0@@ -32,6 +32,7 @@#define SBI_EXT_DBCN 0x4442434E#define SBI_EXT_SUSP 0x53555350#define SBI_EXT_CPPC 0x43505043 +#define SBI_EXT_SSE 0x535345/* SBI function IDs for BASE extension*/ #define SBI_EXT_BASE_GET_SPEC_VERSION 0x0@@ -293,6 +294,48 @@ enum sbi_cppc_reg_id { SBI_CPPC_NON_ACPI_LAST = SBI_CPPC_TRANSITION_LATENCY, }; +/* SBI Function IDs for SSE extension */ +#define SBI_EXT_SSE_READ_ATTR 0x00000000 +#define SBI_EXT_SSE_WRITE_ATTR 0x00000001 +#define SBI_EXT_SSE_REGISTER 0x00000002 +#define SBI_EXT_SSE_UNREGISTER 0x00000003 +#define SBI_EXT_SSE_ENABLE 0x00000004 +#define SBI_EXT_SSE_DISABLE 0x00000005 +#define SBI_EXT_SSE_COMPLETE 0x00000006 +#define SBI_EXT_SSE_INJECT 0x00000007 + +/* SBI SSE Event Attributes. */ +#define SBI_SSE_ATTR_STATUS 0x00000000 +#define SBI_SSE_ATTR_PRIO 0x00000001 +#define SBI_SSE_ATTR_CONFIG 0x00000002 +#define SBI_SSE_ATTR_PREFERRED_HART 0x00000003 +#define SBI_SSE_ATTR_ENTRY_PC 0x00000004 +#define SBI_SSE_ATTR_ENTRY_A0 0x00000005 +#define SBI_SSE_ATTR_ENTRY_A6 0x00000006 +#define SBI_SSE_ATTR_ENTRY_A7 0x00000007 +#define SBI_SSE_ATTR_INTERRUPTED_MODE 0x00000008 +#define SBI_SSE_ATTR_INTERRUPTED_PC 0x00000009 +#define SBI_SSE_ATTR_INTERRUPTED_A0 0x0000000A +#define SBI_SSE_ATTR_INTERRUPTED_A6 0x0000000B +#define SBI_SSE_ATTR_INTERRUPTED_A7 0x0000000C +include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 331
Quoted 10 lines
include/sbi/sbi_ecall_interface.h +10
-0+#define SBI_SSE_ATTR_STATUS_STATE_OFFSET 0 +#define SBI_SSE_ATTR_STATUS_STATE_MASK 0x3 +#define SBI_SSE_ATTR_STATUS_PENDING_OFFSET 2 +#define SBI_SSE_ATTR_STATUS_INJECT_OFFSET 3 + +#define SBI_SSE_ATTR_CONFIG_ONESHOT (1 << 0) + +/* SBI SSE Event IDs. */ +#define SBI_SSE_EVENT_LOCAL_RAS 0x00000000 +#define SBI_SSE_EVENT_GLOBAL_RAS 0x00008000Can we add a define for start of platform specific local events (
0x4000)?Quoted 6 lines
include/sbi/sbi_ecall_interface.h enum sbi_cppc_reg_id { · line 335
Quoted 4 lines
include/sbi/sbi_ecall_interface.h +4
-0+#define SBI_SSE_EVENT_LOCAL_PMU 0x00010000 +#define SBI_SSE_EVENT_LOCAL_SOFTWARE 0xffff0000 +#define SBI_SSE_EVENT_GLOBAL_SOFTWARE 0xffff8000 +IMHO, we should add all the ranges defined in spec, used or not.
Since there are multiple ranges reserved for the same kind of events, it will need a numbering of some sort, is something like this ok for you:
#define
SBI_SSE_EVENT_LOCAL_RAS0x00000000#defineSBI_SSE_EVENT_LOCAL_PLAT_0_START0x00004000#defineSBI_SSE_EVENT_LOCAL_PLAT_0_END0x00007fff#defineSBI_SSE_EVENT_GLOBAL_RAS0x00008000#defineSBI_SSE_EVENT_GLOBAL_PLAT_0_START0x00004000#defineSBI_SSE_EVENT_GLOBAL_PLAT_0_END0x00007fff#define
SBI_SSE_EVENT_LOCAL_PMU0x00010000#defineSBI_SSE_EVENT_LOCAL_PLAT_1_START0x00014000#defineSBI_SSE_EVENT_LOCAL_PLAT_1_END0x00017fff#defineSBI_SSE_EVENT_GLOBAL_PLAT_1_START0x0001c000#defineSBI_SSE_EVENT_GLOBAL_PLAT_1_END0x0001ffff#define
SBI_SSE_EVENT_LOCAL_PLAT_2_START0x00024000#defineSBI_SSE_EVENT_LOCAL_PLAT_2_END0x00027fff#defineSBI_SSE_EVENT_GLOBAL_PLAT_2_START0x0002c000#defineSBI_SSE_EVENT_GLOBAL_PLAT_2_END0x0002ffff
Quoted 6 lines
#define
SBI_SSE_EVENT_LOCAL_SOFTWARE0xffff0000#defineSBI_SSE_EVENT_LOCAL_PLAT_3_START0xffff4000#defineSBI_SSE_EVENT_LOCAL_PLAT_3_END0xffff7fff#defineSBI_SSE_EVENT_GLOBAL_SOFTWARE0xffff8000#defineSBI_SSE_EVENT_GLOBAL_PLAT_3_START0xffffc000#defineSBI_SSE_EVENT_GLOBAL_PLAT_3_END0xffffffff
This looks okay to me.
Quoted 430 lines
Quoted 421 lines
Quoted 116 lines
include/sbi/sbi_ecall_interface.h +7
-1+#define SBI_SSE_EVENT_GLOBAL (1 << 15) +#define SBI_SSE_EVENT_PLATFORM (1 << 14) +/* SBI base specification related macros */ #define SBI_SPEC_VERSION_MAJOR_OFFSET 24#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f@@ -313,8 +356,11 @@ enum sbi_cppc_reg_id {#define SBI_ERR_ALREADY_STARTED -7#define SBI_ERR_ALREADY_STOPPED -8#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_BUSY -12 -#define SBI_LAST_ERR SBI_ERR_NO_SHMEM +#define SBI_LAST_ERR SBI_ERR_BUSY/* clang-format on */ include/sbi/sbi_error.h +3
-1@@ -24,6 +24,9 @@#define SBI_EALREADY_STARTED SBI_ERR_ALREADY_STARTED#define SBI_EALREADY_STOPPED SBI_ERR_ALREADY_STOPPED#define SBI_ENO_SHMEM SBI_ERR_NO_SHMEM +#define SBI_EINVALID_STATE SBI_ERR_INVALID_STATE +#define SBI_EBAD_RANGE SBI_ERR_BAD_RANGE +#define SBI_EBUSY SBI_ERR_BUSY#define SBI_ENODEV -1000#define SBI_ENOSYS -1001@@ -34,7 +37,6 @@#define SBI_ENOMEM -1006#define SBI_EUNKNOWN -1007#define SBI_ENOENT -1008 -/* clang-format on */ #endif include/sbi/sbi_sse.h new file +67
-0@@ -0,0 +1,93 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems. +*/ + +#ifndef __SBI_SSE_H__ +#define __SBI_SSE_H__ + +#include <sbi/sbi_types.h> +#include <sbi/sbi_list.h> +#include <sbi/riscv_locks.h> + +struct sbi_scratch ; + +#define EXC_MODE_PP_SHIFT 0 +#define EXC_MODE_PP BIT(EXC_MODE_PP_SHIFT) +#define EXC_MODE_PV_SHIFT 1 +#define EXC_MODE_PV BIT(EXC_MODE_PV_SHIFT) +#define EXC_MODE_SSTATUS_SPIE_SHIFT 2 +#define EXC_MODE_SSTATUS_SPIE BIT(EXC_MODE_SSTATUS_SPIE_SHIFT) + + +struct sbi_sse_cb_ops { +/** +* Called when hart_id is changed on the event. +*/ +void (*set_hartid_cb)(uint32_t event_id,unsigned long hart_id); + +/** +* Called when the SBI_EXT_SSE_COMPLETE is invoked on the event. +*/ +void (*complete_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_REGISTER is invoked on the event. +*/ +void (*register_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_UNREGISTER is invoked on the event. +*/ +void (*unregister_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_ENABLE is invoked on the event. +*/ +void (*enable_cb)(uint32_t event_id); + +/** +* Called when the SBI_EXT_SSE_DISABLE is invoked on the event. +*/ +void (*disable_cb)(uint32_t event_id); +}; + +/* Set the callback operations for an event +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param cb_ops Callback operations +* @return 0 on success, error otherwise +*/ +int sbi_sse_set_cb_ops (uint32_t event_id,const struct sbi_sse_cb_ops *cb_ops); + +/* Inject an event to the current hard +* @param event_id Event identifier (SBI_SSE_EVENT_*) +* @param regs Registers that were used on SBI entry +* @return 0 on success, error otherwise +*/ include/sbi/sbi_sse.h line 77
Quoted 10 lines
include/sbi/sbi_sse.h new file +10
-0+int sbi_sse_inject_event (uint32_t event_id,struct sbi_trap_regs *regs); + +int sbi_sse_init (struct sbi_scratch *scratch,bool cold_boot); +void sbi_sse_exit (struct sbi_scratch *scratch); + +/* Interface called from sbi_ecall_sse.c */ +int sbi_sse_register (uint32_t event_id,unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7);Wouldn?t it be better to only use defined length data types like
uint32_t,uint64_teverywhere? Instead of mixing them up?Quoted 292 lines
include/sbi/sbi_sse.h new file +16
-0+int sbi_sse_unregister (uint32_t event_id); +int sbi_sse_enable (uint32_t event_id); +int sbi_sse_disable (uint32_t event_id); +int sbi_sse_complete (uint32_t event_id,struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_inject_from_ecall (uint32_t event_id,unsigned long hart_id, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out); +int sbi_sse_read_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long output_phys_lo, +unsigned long output_phys_hi); +int sbi_sse_write_attrs (uint32_t event_id,uint32_t base_attr_id, +uint32_t attr_count,unsigned long input_phys_lo, +unsigned long input_phys_hi); + +#endif lib/sbi/Kconfig +0
-0@@ -46,4 +46,8 @@ config SBI_ECALL_VENDORbool "Platform-defined vendor extensions"
default ylib/sbi/Kconfig +4
-0+config SBI_ECALL_SSE + bool "SSE extension" + default y +endmenu
lib/sbi/objects.mk +0
-0@@ -52,6 +52,9 @@ libsbi-objs-$(CONFIG_SBI_ECALL_LEGACY) += sbi_ecall_legacy.ocarray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_VENDOR) +=
ecall_vendorlibsbi-objs-$(CONFIG_SBI_ECALL_VENDOR) += sbi_ecall_vendor.olib/sbi/objects.mk +3
-0+carray-sbi_ecall_exts-$(CONFIG_SBI_ECALL_SSE) += ecall_sse +libsbi-objs-$(CONFIG_SBI_ECALL_SSE) += sbi_ecall_sse.o +libsbi-objs-y += sbi_bitmap.o
libsbi-objs-y += sbi_bitops.o
libsbi-objs-y += sbi_console.olib/sbi/objects.mk +0
-0@@ -71,6 +74,7 @@ libsbi-objs-y += sbi_misaligned_ldst.olibsbi-objs-y += sbi_platform.o
libsbi-objs-y += sbi_pmu.o
libsbi-objs-y += sbi_scratch.o
+libsbi-objs-y += sbi_sse.o
libsbi-objs-y += sbi_string.o
libsbi-objs-y += sbi_system.o
libsbi-objs-y += sbi_timer.olib/sbi/sbi_ecall_sse.c new file +58
-0@@ -0,0 +1,58 @@ +#include <sbi/sbi_error.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_trap.h> +#include <sbi/sbi_sse.h> + +static int sbi_ecall_sse_handler (unsigned long extid,unsigned long funcid, +struct sbi_trap_regs *regs, +struct sbi_ecall_return *out) +{ +int ret; + +switch (funcid) { +case SBI_EXT_SSE_READ_ATTR : + ret =sbi_sse_read_attrs (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a4); +break ; +case SBI_EXT_SSE_WRITE_ATTR : + ret =sbi_sse_write_attrs (regs->a0, regs->a1, regs->a2, + regs->a3, regs->a4); +break ; +case SBI_EXT_SSE_REGISTER : + ret =sbi_sse_register (regs->a0, regs->a1, regs->a2, regs->a3, + regs->a2); +break ; +case SBI_EXT_SSE_UNREGISTER : + ret =sbi_sse_unregister (regs->a0); +break ; +case SBI_EXT_SSE_ENABLE : + ret =sbi_sse_enable (regs->a0); +break ; +case SBI_EXT_SSE_DISABLE : + ret =sbi_sse_disable (regs->a0); +break ; +case SBI_EXT_SSE_COMPLETE : + ret =sbi_sse_complete (regs->a0, regs, out); +break ; +case SBI_EXT_SSE_INJECT : + ret =sbi_sse_inject_from_ecall (regs->a0, regs->a1, regs, out); +break ; +default : + ret =SBI_ENOTSUPP ; + } +return ret; +} + +struct sbi_ecall_extension ecall_sse; + +static int sbi_ecall_sse_register_extensions (void ) +{ +return sbi_ecall_register_extension (&ecall_sse); +} + +struct sbi_ecall_extension ecall_sse = { + .extid_start =SBI_EXT_SSE , + .extid_end =SBI_EXT_SSE , + .register_extensions = sbi_ecall_sse_register_extensions, + .handle = sbi_ecall_sse_handler, +};lib/sbi/sbi_init.c +13
-0@@ -23,6 +23,7 @@#include <sbi/sbi_irqchip.h> #include <sbi/sbi_platform.h> #include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> #include <sbi/sbi_system.h> #include <sbi/sbi_string.h> #include <sbi/sbi_timer.h> @@ -315,6 +316,12 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, true); +if (rc) { +sbi_printf ("%s: sse init failed (error %d)\n" , __func__, rc); +sbi_hart_hang (); + } + rc =sbi_pmu_init (scratch, true);if (rc) { sbi_printf("%s: pmu init failed (error %d)\n",@@ -435,6 +442,10 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,if (rc)sbi_hart_hang (); + rc =sbi_sse_init (scratch, false); +if (rc) +sbi_hart_hang (); + rc =sbi_pmu_init (scratch, false);if (rc)sbi_hart_hang ();@@ -639,6 +650,8 @@ void __noreturn sbi_exit(struct sbi_scratch *scratch)sbi_platform_early_exit (plat); +sbi_sse_exit (scratch); +sbi_pmu_exit (scratch);sbi_timer_exit (scratch);lib/sbi/sbi_ipi.c +1
-1@@ -66,7 +66,7 @@ static int sbi_ipi_send(struct sbi_scratch *scratch, u32 remote_hartindex,* SBI_IPI_UPDATE_BREAK for self-IPIs. For other events, check * for self-IPI and execute the callback directly here. */ - ipi_ops->process (scratch); + ipi_ops->process (scratch,NULL );return 0 ; }lib/sbi/sbi_sse.c new file +113
-0@@ -0,0 +1,1078 @@ +/* +* SPDX-License-Identifier: BSD-2-Clause +* +* Copyright (c) 2023 Rivos Systems Inc. +* +*/ + +#include <sbi/riscv_asm.h> +#include <sbi/riscv_barrier.h> +#include <sbi/riscv_encoding.h> +#include <sbi/riscv_locks.h> +#include <sbi/sbi_domain.h> +#include <sbi/sbi_ecall.h> +#include <sbi/sbi_ecall_interface.h> +#include <sbi/sbi_error.h> +#include <sbi/sbi_fifo.h> +#include <sbi/sbi_hart.h> +#include <sbi/sbi_heap.h> +#include <sbi/sbi_hsm.h> +#include <sbi/sbi_ipi.h> +#include <sbi/sbi_list.h> +#include <sbi/sbi_platform.h> +#include <sbi/sbi_pmu.h> +#include <sbi/sbi_sse.h> +#include <sbi/sbi_scratch.h> +#include <sbi/sbi_string.h> +#include <sbi/sbi_trap.h> + +#include <sbi/sbi_console.h> + +#define sse_get_hart_state_ptr (__scratch) \ + sbi_scratch_read_type((__scratch), void *, shs_ptr_off) + +#define sse_thishart_state_ptr () \ + sse_get_hart_state_ptr(sbi_scratch_thishart_ptr()) + +#define sse_set_hart_state_ptr (__scratch, __sse_state) \ + sbi_scratch_write_type((__scratch), void *, shs_ptr_off, (__sse_state)) + +/* +* Rather than using memcpy to copy the context (which does it byte-per-byte), +* copy each field which generates ld/lw. +*/ +#define sse_regs_copy (dst, src) \ + (dst)->a0 = (src)->a0; \ + (dst)->a6 = (src)->a6; \ + (dst)->a7 = (src)->a7 + +#define EVENT_IS_GLOBAL (__event_id) ((__event_id) & SBI_SSE_EVENT_GLOBAL) + +static const uint32_t supported_events[] = +{ +SBI_SSE_EVENT_LOCAL_RAS , +SBI_SSE_EVENT_GLOBAL_RAS , +SBI_SSE_EVENT_LOCAL_PMU , +SBI_SSE_EVENT_LOCAL_SOFTWARE , +SBI_SSE_EVENT_GLOBAL_SOFTWARE , +}; + +#define EVENT_COUNT array_size(supported_events) + +#define sse_event_invoke_cb (_event, _cb, ...) \ +{ \ + if (_event->cb_ops && _event->cb_ops->_cb) \ + _event->cb_ops->_cb(_event->event_id, ##__VA_ARGS__); \ +} + +#define SSE_EVENT_STATE (__e) __e->attrs.status.state +#define SSE_EVENT_PENDING (__e) __e->attrs.status.pending +#define SSE_EVENT_CAN_INJECT (__e) __e->attrs.status.inject +#define SSE_EVENT_HARTID (__e) __e->attrs.hartid +#define SSE_EVENT_PRIO (__e) __e->attrs.prio +#define SSE_EVENT_CONFIG (__e) __e->attrs.config +#define SSE_EVENT_ENTRY (__e) __e->attrs.entry +#define SSE_EVENT_INTERRUPTED (__e) __e->attrs.interrupted + +struct sse_entry_state { +/** Entry program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +struct sse_interrupted_state { +/** Exception mode */ +unsigned long exc_mode; +/** Interrupted program counter */ +unsigned long pc; +/** a0 register state */ +unsigned long a0; +/** a6 register state */ +unsigned long a6; +/** a7 register state */ +unsigned long a7; +}; + +enum sbi_sse_state { +SSE_STATE_UNUSED =0 , +SSE_STATE_REGISTERED =1 , +SSE_STATE_ENABLED =2 , +SSE_STATE_RUNNING =3 , +}; + +struct sse_ipi_inject_data { +uint32_t event_id; +}; + +struct sbi_sse_event_status { +unsigned long state:2 ;Quoted 8 lines
lib/sbi/sbi_sse.c line 116
Quoted 3 lines
lib/sbi/sbi_sse.c new file +3
-0+ unsigned long pending:1; + unsigned long inject:1; +} __packed;Can we have do away with bit fields? Like everywhere else use positions and masks? Why packed? We are making sure
sse_event_attrsis aligned to unsigned long so that We can copy with loops. Probably make it a long with reserved entry? Unless you had Something else in mind.
Quoted 2 lines
Using bitfield was purely for the ease of use. I could of course use a long with some shift masks if you prefer that over bitfields.
The general ask on the list is to do away with bit fields. I will let others pitch in.
Quoted 387 lines
Quoted 85 lines
Quoted 85 lines
lib/sbi/sbi_sse.c new file +83
-0+ +struct sbi_sse_event_attrs { +struct sbi_sse_event_status status; +unsigned long prio; +unsigned long config; +unsigned long hartid; +struct sse_entry_state entry; +struct sse_interrupted_state interrupted; +}; + +/* Make sure all attributes are packed for direct memcpy in ATTR_READ */ +#define assert_field_offset (field, attr_offset) \ + _Static_assert( \ + ((offsetof(struct sbi_sse_event_attrs, field)) / sizeof(unsigned long)) == attr_offset, \ + "field "#field " from struct sbi_sse_event_attrs invalid offset, expected "#attr_offset \ + ) + +assert_field_offset (status,SBI_SSE_ATTR_STATUS ); +assert_field_offset (prio,SBI_SSE_ATTR_PRIO ); +assert_field_offset (config,SBI_SSE_ATTR_CONFIG ); +assert_field_offset (hartid,SBI_SSE_ATTR_PREFERRED_HART ); +assert_field_offset (entry.pc,SBI_SSE_ATTR_ENTRY_PC ); +assert_field_offset (entry.a0,SBI_SSE_ATTR_ENTRY_A0 ); +assert_field_offset (entry.a6,SBI_SSE_ATTR_ENTRY_A6 ); +assert_field_offset (entry.a7,SBI_SSE_ATTR_ENTRY_A7 ); +assert_field_offset (interrupted.exc_mode,SBI_SSE_ATTR_INTERRUPTED_MODE ); +assert_field_offset (interrupted.pc,SBI_SSE_ATTR_INTERRUPTED_PC ); +assert_field_offset (interrupted.a0,SBI_SSE_ATTR_INTERRUPTED_A0 ); +assert_field_offset (interrupted.a6,SBI_SSE_ATTR_INTERRUPTED_A6 ); +assert_field_offset (interrupted.a7,SBI_SSE_ATTR_INTERRUPTED_A7 ); + +struct sbi_sse_event { +struct sbi_sse_event_attrs attrs; +uint32_t event_id; +const struct sbi_sse_cb_ops *cb_ops; +struct sbi_dlist node; +/* Only global events are using the lock, local ones don't need it */ +spinlock_t lock; +}; + +struct sse_hart_state { +struct sbi_dlist event_list; +spinlock_t list_lock; +struct sbi_sse_event *local_events; +}; + +static unsigned int local_event_count; +static unsigned int global_event_count; +static struct sbi_sse_event *global_events; + +static unsigned long sse_inject_fifo_off; +static unsigned long sse_inject_fifo_mem_off; +/* Offset of pointer to SSE HART state in scratch space */ +static unsigned long shs_ptr_off; + +static u32 sse_ipi_inject_event =SBI_IPI_EVENT_MAX ; + +static int sse_ipi_inject_send (unsigned long hartid,uint32_t event_id); + +static bool sse_event_is_global (struct sbi_sse_event *e) +{ +return EVENT_IS_GLOBAL (e->event_id); +} + +static void sse_global_event_lock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_lock (&e->lock); +} + +static void sse_global_event_unlock (struct sbi_sse_event *e) +{ +if (sse_event_is_global (e)) +spin_unlock (&e->lock); +} + +static void sse_event_set_state (struct sbi_sse_event *e, +enum sbi_sse_state new_state) +{ +enum sbi_sse_state prev_state =SSE_EVENT_STATE (e); + +if ((new_state - prev_state ==1 ) || (prev_state - new_state ==1 )) { +SSE_EVENT_STATE (e) = new_state;Quoted 8 lines
lib/sbi/sbi_sse.c line 204
Quoted 5 lines
lib/sbi/sbi_sse.c new file +5
-0+return ; + } + +sbi_panic ("Invalid SSE state transition: %d -> %d\n" , prev_state, + new_state);Panic in a running system is not a good idea, IMHO. Thinking out loud, what if we let the state as is and return error (handled by caller and decide). In worst case, the event can be disabled but definitely not panic.
Yeah makes sense, Even if this is mostly debugging code, it could help catch some nasty bug without crashing the system. I'll try to convert that to return a value.
Quoted 78 lines
Quoted 78 lines
lib/sbi/sbi_sse.c new file +77
-0+} + +static struct sbi_sse_event *sse_event_get (uint32_t event) +{ +unsigned int i; +struct sbi_sse_event *events, *e; +unsigned int count; +struct sse_hart_state *shs; + +if (EVENT_IS_GLOBAL (event)) { + count = global_event_count; + events = global_events; + }else { + count = local_event_count; + shs =sse_thishart_state_ptr (); + events = shs->local_events; + } + +for (i =0 ; i < count; i++) { + e = &events[i]; +if (e->event_id == event) +return e; + } + +return NULL ; +} + +static struct sse_hart_state *sse_event_get_hart_state (struct sbi_sse_event *e) +{ +struct sbi_scratch *s =sbi_hartid_to_scratch (SSE_EVENT_HARTID (e)); + +return sse_get_hart_state_ptr (s); +} + +static void sse_event_remove_from_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); + +spin_lock (&state->list_lock); +sbi_list_del (&e->node); +spin_unlock (&state->list_lock); +} + +static void sse_event_add_to_list (struct sbi_sse_event *e) +{ +struct sse_hart_state *state =sse_event_get_hart_state (e); +struct sbi_sse_event *tmp; + +spin_lock (&state->list_lock); +sbi_list_for_each_entry (tmp, &state->event_list, node) { +if (SSE_EVENT_PRIO (e) <SSE_EVENT_PRIO (tmp)) +break ; +if (SSE_EVENT_PRIO (e) ==SSE_EVENT_PRIO (tmp) && + e->event_id < tmp->event_id) +break ; + } +sbi_list_add_tail (&e->node, &tmp->node); + +spin_unlock (&state->list_lock); +} + +static int sse_event_disable (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_ENABLED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, disable_cb); + +sse_event_remove_from_list (e); +sse_event_set_state (e,SSE_STATE_REGISTERED ); + +return SBI_OK ; +} + +static int sse_event_set_hart_id_check (struct sbi_sse_event *e, +unsigned long new_hartid) +{Quoted 8 lines
lib/sbi/sbi_sse.c line 287
Quoted 6 lines
lib/sbi/sbi_sse.c new file +6
-0+int hstate; +unsigned int hartid = (uint32_t ) new_hartid; +struct sbi_domain * hd =sbi_domain_thishart_ptr (); + +if (!sse_event_is_global (e)) +return SBI_EDENIED ;Why denied? Why not invalid param?
Oops, probably a left over of some previous version.
Quoted 7 lines
lib/sbi/sbi_sse.c line 290
Quoted 4 lines
lib/sbi/sbi_sse.c new file +3
-0+ +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) +return SBI_EBUSY ;In the spec, whenever the state doesn?t match the required state, we say we return
INVALID_STATE.
Can we do that? This is the only place where EBUSY is used. It might be good to have EBUSY but then it?s out of the scope of this patch set.Agreed, this was specified in a previous spec version also. I'll change that.
Quoted 170 lines
Quoted 36 lines
lib/sbi/sbi_sse.c new file +35
-0+ +if (!sbi_domain_is_assigned_hart (hd, new_hartid)) +return SBI_EINVAL ; + + hstate =sbi_hsm_hart_get_state (hd, hartid); +if (hstate !=SBI_HSM_STATE_STARTED ) +return SBI_EINVAL ; + +return SBI_OK ; +} + +static int sse_event_set_attr_check (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +int ret =SBI_OK ; + +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +if (val & ~SBI_SSE_ATTR_CONFIG_ONESHOT ) + ret =SBI_ERR_INVALID_PARAM ; +break ; +case SBI_SSE_ATTR_PRIO : +if (SSE_EVENT_STATE (e) >=SSE_STATE_ENABLED ) + ret =SBI_EINVALID_STATE ; +break ; +case SBI_SSE_ATTR_PREFERRED_HART : + ret =sse_event_set_hart_id_check (e, val); +break ; +default : + ret =SBI_EBAD_RANGE ; +break ; + } + +return ret; +}lib/sbi/sbi_sse.c line 335
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +static void sse_event_set_attr (struct sbi_sse_event *e,uint32_t attr_id, +unsigned long val) +{ +switch (attr_id) { +case SBI_SSE_ATTR_CONFIG : +SSE_EVENT_CONFIG (e) = val; +break ; +case SBI_SSE_ATTR_PRIO : +SSE_EVENT_PRIO (e) = (uint32_t )val;Another reason to used defined length data types. I think we can do away with typecasts like this.
Quoted 16 lines
lib/sbi/sbi_sse.c new file +16
-0+break ; +case SBI_SSE_ATTR_PREFERRED_HART : +SSE_EVENT_HARTID (e) = val; +sse_event_invoke_cb (e, set_hartid_cb, val); +break ; + } +} + +static int sse_event_register (struct sbi_sse_event *e, +unsigned long handler_entry_pc, +unsigned long handler_entry_a0, +unsigned long handler_entry_a6, +unsigned long handler_entry_a7) +{ +if (sse_event_is_global (e) &&SSE_EVENT_HARTID (e) !=current_hartid ()) +return SBI_EINVAL ;lib/sbi/sbi_sse.c line 361
Quoted 10 lines
lib/sbi/sbi_sse.c new file +10
-0+ +if (SSE_EVENT_STATE (e) !=SSE_STATE_UNUSED ) +return SBI_EINVALID_STATE ; + +SSE_EVENT_ENTRY (e).a0 = handler_entry_a0; +SSE_EVENT_ENTRY (e).a6 = handler_entry_a6; +SSE_EVENT_ENTRY (e).a7 = handler_entry_a7; +SSE_EVENT_ENTRY (e).pc = handler_entry_pc; + +sse_event_set_state (e,SSE_STATE_REGISTERED );Better return an error from set state (my previous comment). Handle it here based on what transition is expected. Here we just have crossed fingers that system won?t panic.
Quoted 91 lines
lib/sbi/sbi_sse.c new file +91
-0+ +sse_event_invoke_cb (e, register_cb); + +return 0 ; +} + +static int sse_event_unregister (struct sbi_sse_event *e) +{ +if (SSE_EVENT_STATE (e) !=SSE_STATE_REGISTERED ) +return SBI_EINVALID_STATE ; + +sse_event_invoke_cb (e, unregister_cb); + +sse_event_set_state (e,SSE_STATE_UNUSED ); + +return 0 ; +} + +static void sse_event_inject (struct sbi_sse_event *e, +struct sbi_sse_event *prev_e, +struct sbi_trap_regs *regs) +{ +ulong prev_smode, prev_virt; +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); +struct sse_interrupted_state *prev_i_ctx; +struct sse_entry_state *e_ctx = &SSE_EVENT_ENTRY (e); + +sse_event_set_state (e,SSE_STATE_RUNNING ); +SSE_EVENT_PENDING (e) = false; + +if (prev_e) { +/* back-to-back injection after another event, copy previous +* event context for correct restoration +*/ + prev_i_ctx = &SSE_EVENT_INTERRUPTED (prev_e); + +sse_regs_copy (i_ctx, prev_i_ctx); + i_ctx->exc_mode = prev_i_ctx->exc_mode; + i_ctx->pc = prev_i_ctx->pc; + }else { +sse_regs_copy (i_ctx, regs); + + prev_smode = (regs->mstatus &MSTATUS_MPP ) >>MSTATUS_MPP_SHIFT ; +#if __riscv_xlen ==32 + prev_virt = (regs->mstatusH &MSTATUSH_MPV ) ?1 :0 ; +#else + prev_virt = (regs->mstatus &MSTATUS_MPV ) ?1 :0 ; +#endif + + i_ctx->exc_mode = prev_smode <<EXC_MODE_PP_SHIFT ; + i_ctx->exc_mode |= prev_virt <<EXC_MODE_PV_SHIFT ; +if (regs->mstatus &MSTATUS_SPIE ) + i_ctx->exc_mode |=EXC_MODE_SSTATUS_SPIE ; + i_ctx->pc = regs->mepc; + +/* We only want to set SPIE for the first event injected after +* entering M-Mode. For the event injected right after another +* event (after calling sse_event_complete(), we will keep the +* saved SPIE). +*/ + regs->mstatus &= ~MSTATUS_SPIE ; +if (regs->mstatus &MSTATUS_SIE ) + regs->mstatus |=MSTATUS_SPIE ; + } + +sse_regs_copy (regs, e_ctx); + regs->mepc = e_ctx->pc; + + regs->mstatus &= ~MSTATUS_MPP ; + regs->mstatus |= (PRV_S <<MSTATUS_MPP_SHIFT ); + +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +#else + regs->mstatus &= ~MSTATUS_MPV ; +#endif + + regs->mstatus &= ~MSTATUS_SIE ; +} + +static void sse_event_resume (struct sbi_sse_event *e,struct sbi_trap_regs *regs) +{ +struct sse_interrupted_state *i_ctx = &SSE_EVENT_INTERRUPTED (e); + +sse_regs_copy (regs, i_ctx); + +/* Restore previous virtualization state */ +#if __riscv_xlen ==32 + regs->mstatusH &= ~MSTATUSH_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatusH |=MSTATUSH_MPV ;Quoted 8 lines
lib/sbi/sbi_sse.c line 458
Quoted 6 lines
lib/sbi/sbi_sse.c new file +6
-0+#else + regs->mstatus &= ~MSTATUS_MPV ; +if (i_ctx->exc_mode &EXC_MODE_PV ) + regs->mstatus |=MSTATUS_MPV ; +#endif +Probably have #error for undefined
__riscv_xlen.Acked
Thanks,
Cl?ment
Signature · 3 lines
opensbi mailing list opensbi at lists.infradead.org http://lists.infradead.org/mailman/listinfo/opensbi
RFC PATCH v2 3/4 lib: sbi: trigger SSE PMU event on PMU overflow IRQ
In reply to Clément LégerIn order to send SSE local PMU event and receive PMU interrupt in openSBI, we need to disable interrupt delegation for PMU interrupts and send the SSE event upon LCOFIP IRQ.
--- include/sbi/sbi_pmu.h | 3 +++ lib/sbi/sbi_pmu.c | 49 +++++++++++++++++++++++++++++++++++++++++++ lib/sbi/sbi_trap.c | 6 ++++++ 3 files changed, 58 insertions(+)
include/sbi/sbi_pmu.h +3 -0
@@ -11,6 +11,7 @@#define __SBI_PMU_H__#include <sbi/sbi_types.h> +#include <sbi/sbi_trap.h> struct sbi_scratch ;
@@ -150,4 +151,6 @@ int sbi_pmu_ctr_cfg_match(unsigned long cidx_base, unsigned long cidx_mask,int sbi_pmu_ctr_incr_fw (enum sbi_pmu_fw_event_code_id fw_id); +void sbi_pmu_ovf_irq (struct sbi_trap_regs *regs); +#endif
lib/sbi/sbi_pmu.c +49 -0
91 lines not on this page · raw message
lib/sbi/sbi_trap.c +6 -0
@@ -208,6 +208,9 @@ static int sbi_trap_nonaia_irq(struct sbi_trap_regs *regs, ulong mcause)case IRQ_M_SOFT :sbi_ipi_process (regs);break ; +case IRQ_PMU_OVF : +sbi_pmu_ovf_irq (regs); +break ;case IRQ_M_EXT :return sbi_irqchip_process (regs);default :
@@ -231,6 +234,9 @@ static int sbi_trap_aia_irq(struct sbi_trap_regs *regs, ulong mcause)case IRQ_M_SOFT :sbi_ipi_process (regs);break ; +case IRQ_PMU_OVF : +sbi_pmu_ovf_irq (regs); +break ;case IRQ_M_EXT : rc =sbi_irqchip_process (regs);if (rc)
Signature · 1 line
2.43.0
RFC PATCH v2 4/4 lib: sbi: add SBI_EXT_PMU_IRQ_CLEAR
In reply to Clément LégerNow that we have sse support for such interrupt, the IRQ can only be cleared by the SBI. In some cases in S-mode OS, it might be needed to clear the pending IRQ. Add SBI_EXT_PMU_IRQ_CLEAR ecall to clear such IRQ.
NOTE: this is temporary until the counter delegation support from Atish lands in Linux/OpenSBI.
--- include/sbi/sbi_ecall_interface.h | 1 + include/sbi/sbi_pmu.h | 2 ++ lib/sbi/sbi_ecall_pmu.c | 3 +++ 3 files changed, 6 insertions(+)
include/sbi/sbi_ecall_interface.h +1 -0
@@ -105,6 +105,7 @@#define SBI_EXT_PMU_COUNTER_FW_READ 0x5#define SBI_EXT_PMU_COUNTER_FW_READ_HI 0x6#define SBI_EXT_PMU_SNAPSHOT_SET_SHMEM 0x7 +#define SBI_EXT_PMU_IRQ_CLEAR 0x8/** General pmu event codes specified in SBI PMU extension */ enum sbi_pmu_hw_generic_events_t {
include/sbi/sbi_pmu.h +2 -0
@@ -153,4 +153,6 @@ int sbi_pmu_ctr_incr_fw(enum sbi_pmu_fw_event_code_id fw_id);void sbi_pmu_ovf_irq (struct sbi_trap_regs *regs); +void sbi_pmu_irq_clear (void ); +#endif
lib/sbi/sbi_ecall_pmu.c +3 -0
@@ -73,6 +73,9 @@ static int sbi_ecall_pmu_handler(unsigned long extid, unsigned long funcid,case SBI_EXT_PMU_COUNTER_STOP : ret =sbi_pmu_ctr_stop (regs->a0, regs->a1, regs->a2);break ; +case SBI_EXT_PMU_IRQ_CLEAR : +sbi_pmu_irq_clear (); +break ;case SBI_EXT_PMU_SNAPSHOT_SET_SHMEM :/* fallthrough as OpenSBI doesn't support snapshot yet */ default :
Signature · 1 line
2.43.0
RFC PATCH v2 4/4 lib: sbi: add SBI_EXT_PMU_IRQ_CLEAR
Quoted 8 lines · On Tue, Jan 9, 2024 at 4:15?PM Cl?ment L?ger <[email protected]> wrote:
Now that we have sse support for such interrupt, the IRQ can only be cleared by the SBI. In some cases in S-mode OS, it might be needed to clear the pending IRQ. Add
SBI_EXT_PMU_IRQ_CLEARecall to clear such IRQ.NOTE: this is temporary until the counter delegation support from Atish lands in Linux/OpenSBI.
Why not clear the PMU IRQ when supervisor software completes the PMU SSE event ?
Regards,
Anup
Quoted 53 lines
Signed-off-by Cl?ment L?ger <[email protected]>--- include/sbi/sbi_ecall_interface.h | 1 + include/sbi/sbi_pmu.h | 2 ++ lib/sbi/sbi_ecall_pmu.c | 3 +++ 3 files changed, 6 insertions(+)include/sbi/sbi_ecall_interface.h +1
-0@@ -105,6 +105,7 @@#define SBI_EXT_PMU_COUNTER_FW_READ 0x5#define SBI_EXT_PMU_COUNTER_FW_READ_HI 0x6#define SBI_EXT_PMU_SNAPSHOT_SET_SHMEM 0x7 +#define SBI_EXT_PMU_IRQ_CLEAR 0x8/** General pmu event codes specified in SBI PMU extension */ enum sbi_pmu_hw_generic_events_t {include/sbi/sbi_pmu.h +2
-0@@ -153,4 +153,6 @@ int sbi_pmu_ctr_incr_fw(enum sbi_pmu_fw_event_code_id fw_id);void sbi_pmu_ovf_irq (struct sbi_trap_regs *regs); +void sbi_pmu_irq_clear (void ); +#endif lib/sbi/sbi_ecall_pmu.c +3
-1@@ -73,6 +73,9 @@ static int sbi_ecall_pmu_handler(unsigned long extid, unsigned long funcid,case SBI_EXT_PMU_COUNTER_STOP : ret =sbi_pmu_ctr_stop (regs->a0, regs->a1, regs->a2);break ; +case SBI_EXT_PMU_IRQ_CLEAR : +sbi_pmu_irq_clear (); +break ;case SBI_EXT_PMU_SNAPSHOT_SET_SHMEM :/* fallthrough as OpenSBI doesn't support snapshot yet */ default : --2.43.0
--
opensbi mailing list
opensbi at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi
RFC PATCH v2 4/4 lib: sbi: add SBI_EXT_PMU_IRQ_CLEAR
Quoted 7 lines · Anup Patel
Quoted 6 lines · On Tue, Jan 9, 2024 at 4:15?PM Cl?ment L?ger <[email protected]> wrote:
Now that we have sse support for such interrupt, the IRQ can only be cleared by the SBI. In some cases in S-mode OS, it might be needed to clear the pending IRQ. Add
SBI_EXT_PMU_IRQ_CLEARecall to clear such IRQ.
Quoted 5 lines
Quoted 2 lines
NOTE: this is temporary until the counter delegation support from Atish lands in Linux/OpenSBI.
Why not clear the PMU IRQ when supervisor software completes the PMU SSE event ?
Hi Anup,
Yeah indeed, I think I did that at a time when I did not had SSE events callbacks on SBI side. I'll fix that and remove this.
Cl?ment
Quoted 57 lines
Regards,
AnupQuoted 53 lines
Signed-off-by Cl?ment L?ger <[email protected]>--- include/sbi/sbi_ecall_interface.h | 1 + include/sbi/sbi_pmu.h | 2 ++ lib/sbi/sbi_ecall_pmu.c | 3 +++ 3 files changed, 6 insertions(+)include/sbi/sbi_ecall_interface.h +1
-0@@ -105,6 +105,7 @@#define SBI_EXT_PMU_COUNTER_FW_READ 0x5#define SBI_EXT_PMU_COUNTER_FW_READ_HI 0x6#define SBI_EXT_PMU_SNAPSHOT_SET_SHMEM 0x7 +#define SBI_EXT_PMU_IRQ_CLEAR 0x8/** General pmu event codes specified in SBI PMU extension */ enum sbi_pmu_hw_generic_events_t {include/sbi/sbi_pmu.h +2
-0@@ -153,4 +153,6 @@ int sbi_pmu_ctr_incr_fw(enum sbi_pmu_fw_event_code_id fw_id);void sbi_pmu_ovf_irq (struct sbi_trap_regs *regs); +void sbi_pmu_irq_clear (void ); +#endif lib/sbi/sbi_ecall_pmu.c +3
-1@@ -73,6 +73,9 @@ static int sbi_ecall_pmu_handler(unsigned long extid, unsigned long funcid,case SBI_EXT_PMU_COUNTER_STOP : ret =sbi_pmu_ctr_stop (regs->a0, regs->a1, regs->a2);break ; +case SBI_EXT_PMU_IRQ_CLEAR : +sbi_pmu_irq_clear (); +break ;case SBI_EXT_PMU_SNAPSHOT_SET_SHMEM :/* fallthrough as OpenSBI doesn't support snapshot yet */ default : --2.43.0
--
opensbi mailing list
opensbi at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi