qemu-ppc
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 1/2] target/ppc: Add privileged message send facilities


From: David Gibson
Subject: Re: [PATCH 1/2] target/ppc: Add privileged message send facilities
Date: Fri, 17 Jan 2020 19:46:45 +1000

On Thu, Jan 09, 2020 at 05:33:45PM +0100, Cédric Le Goater wrote:
> The Processor Control facility POWER8 processors and later provides
> a mechanism for the hypervisor to send messages to other threads
> in the system (msgsnd instruction) and cause hypervisor-level
> exceptions. Privileged non-hypervisor programs are also able to
> send messages (msgsndp instruction) but are restricted to the
> threads of the same core and cause privileged-level exceptions.
> 
> The Directed Privileged Doorbell Exception State (DPDES) register
> reflects the state of pending privileged doorbell exceptions and can
> also be used to modify that state. The register can be used to read
> and modify the state of privileged doorbell exceptions for all threads
> of a subprocessor and thus is a shared facility for that subprocessor.
> The register can be read/written by the hypervisor and read by the
> supervisor if enabled in the HFSCR, otherwise a hypervisor facility
> unavailable exception is generated.
> 
> The privileged message send and clear instructions (msgsndp & msgclrp)
> are used to generate and clear the presence of a directed privileged
> doorbell exception, respectively. The msgsndp instruction can be used
> to target any thread of the current subprocessor, msgclrp acts on the
> thread issuing the instruction. These instructions are privileged, but
> will generate a hypervisor facility unavailable exception if not
> enabled in the HFSCR and executed in privileged non-hypervisor
> state. The HV facility unavailable exception will be addressed in
> other patch.
> 
> Add and implement this register and instructions by reading or
> modifying the pending interrupt state of the cpu.
> 
> Note that TCG only supports one thread per core and so we only need to
> worry about the cpu making the access.
> 
> Based on previous work from Suraj Jitindar Singh.
> 
> Cc: Suraj Jitindar Singh <address@hidden>
> [clg: took ownership due to the amount of changes ]

Hrm, I think this need's Suraj's S-o-b as well.  AIUI the primary
purpose of S-o-b lines isn't about credit or ownership, but about
tracking where code came from in case of questions of providence.  So
even if you've taken ownership and reworked substantially, it would
retain the original author's S-o-b.

> Signed-off-by: Cédric Le Goater <address@hidden>
> ---
>  target/ppc/cpu.h                |  1 +
>  target/ppc/helper.h             |  4 ++
>  target/ppc/excp_helper.c        | 68 +++++++++++++++++++++++++++------
>  target/ppc/misc_helper.c        | 36 +++++++++++++++++
>  target/ppc/translate.c          | 26 +++++++++++++
>  target/ppc/translate_init.inc.c | 20 ++++++++--
>  6 files changed, 140 insertions(+), 15 deletions(-)
> 
> diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
> index 103bfe9dc274..d175ec9a641d 100644
> --- a/target/ppc/cpu.h
> +++ b/target/ppc/cpu.h
> @@ -935,6 +935,7 @@ enum {
>  #define DBELL_PIRTAG_MASK              0x3fff
>  
>  #define DBELL_PROCIDTAG_MASK           PPC_BITMASK(44, 63)
> +#define DBELL_TIRTAG_MASK              PPC_BITMASK(57, 63)
>  
>  #define PPC_PAGE_SIZES_MAX_SZ   8
>  
> diff --git a/target/ppc/helper.h b/target/ppc/helper.h
> index cd0dfe383a2a..cfb4c07085ca 100644
> --- a/target/ppc/helper.h
> +++ b/target/ppc/helper.h
> @@ -657,6 +657,10 @@ DEF_HELPER_FLAGS_1(load_601_rtcu, TCG_CALL_NO_RWG, tl, 
> env)
>  DEF_HELPER_FLAGS_1(load_purr, TCG_CALL_NO_RWG, tl, env)
>  DEF_HELPER_FLAGS_2(store_purr, TCG_CALL_NO_RWG, void, env, tl)
>  DEF_HELPER_2(store_ptcr, void, env, tl)
> +DEF_HELPER_FLAGS_1(load_dpdes, TCG_CALL_NO_RWG, tl, env)
> +DEF_HELPER_FLAGS_2(store_dpdes, TCG_CALL_NO_RWG, void, env, tl)
> +DEF_HELPER_2(book3s_msgsndp, void, env, tl)
> +DEF_HELPER_2(book3s_msgclrp, void, env, tl)
>  #endif
>  DEF_HELPER_2(store_sdr1, void, env, tl)
>  DEF_HELPER_2(store_pidr, void, env, tl)
> diff --git a/target/ppc/excp_helper.c b/target/ppc/excp_helper.c
> index 5752ed4a4d83..343f3a6b30c4 100644
> --- a/target/ppc/excp_helper.c
> +++ b/target/ppc/excp_helper.c
> @@ -900,7 +900,11 @@ static void ppc_hw_interrupt(CPUPPCState *env)
>          }
>          if (env->pending_interrupts & (1 << PPC_INTERRUPT_DOORBELL)) {
>              env->pending_interrupts &= ~(1 << PPC_INTERRUPT_DOORBELL);
> -            powerpc_excp(cpu, env->excp_model, POWERPC_EXCP_DOORI);
> +            if (env->insns_flags & PPC_SEGMENT_64B) {

I don't love detecting this based on insns_flags this way, but there
are lots of places we do similarly ugly things, so it's probably the
most expedient way forward in the short term.

> +                powerpc_excp(cpu, env->excp_model, POWERPC_EXCP_SDOOR);
> +            } else {
> +                powerpc_excp(cpu, env->excp_model, POWERPC_EXCP_DOORI);
> +            }
>              return;
>          }
>          if (env->pending_interrupts & (1 << PPC_INTERRUPT_HDOORBELL)) {
> @@ -1221,7 +1225,7 @@ void helper_msgsnd(target_ulong rb)
>  }
>  
>  /* Server Processor Control */
> -static int book3s_dbell2irq(target_ulong rb)
> +static int book3s_dbell2irq(target_ulong rb, bool hv_dbell)
>  {
>      int msg = rb & DBELL_TYPE_MASK;
>  
> @@ -1230,12 +1234,16 @@ static int book3s_dbell2irq(target_ulong rb)
>       * message type is 5. All other types are reserved and the
>       * instruction is a no-op
>       */
> -    return msg == DBELL_TYPE_DBELL_SERVER ? PPC_INTERRUPT_HDOORBELL : -1;
> +    if (msg == DBELL_TYPE_DBELL_SERVER) {
> +        return hv_dbell ? PPC_INTERRUPT_HDOORBELL : PPC_INTERRUPT_DOORBELL;
> +    }
> +
> +    return -1;
>  }

This function kind of seems like overkill, and also doesn't have a
great name.  Mostly it just tests if we're dealing with a doorbell at
all.  Selecting the right irq number here is a bit weird, since its
based only on hv_dbell, which is a literal parameter for all callers,
so they might as well just use the right doorbell irq inline.

>  
>  void helper_book3s_msgclr(CPUPPCState *env, target_ulong rb)
>  {
> -    int irq = book3s_dbell2irq(rb);
> +    int irq = book3s_dbell2irq(rb, true);
>  
>      if (irq < 0) {
>          return;
> @@ -1244,16 +1252,10 @@ void helper_book3s_msgclr(CPUPPCState *env, 
> target_ulong rb)
>      env->pending_interrupts &= ~(1 << irq);
>  }
>  
> -void helper_book3s_msgsnd(target_ulong rb)
> +static void book3s_msgsnd_common(int pir, int irq)
>  {
> -    int irq = book3s_dbell2irq(rb);
> -    int pir = rb & DBELL_PROCIDTAG_MASK;
>      CPUState *cs;
>  
> -    if (irq < 0) {
> -        return;
> -    }
> -
>      qemu_mutex_lock_iothread();
>      CPU_FOREACH(cs) {
>          PowerPCCPU *cpu = POWERPC_CPU(cs);
> @@ -1267,6 +1269,50 @@ void helper_book3s_msgsnd(target_ulong rb)
>      }
>      qemu_mutex_unlock_iothread();
>  }
> +
> +void helper_book3s_msgsnd(target_ulong rb)
> +{
> +    int pir = rb & DBELL_PROCIDTAG_MASK;
> +    int irq = book3s_dbell2irq(rb, true);
> +
> +    if (irq < 0) {
> +        return;
> +    }
> +
> +    book3s_msgsnd_common(pir, irq);
> +}
> +
> +#if defined(TARGET_PPC64)
> +void helper_book3s_msgclrp(CPUPPCState *env, target_ulong rb)
> +{
> +    int irq = book3s_dbell2irq(rb, false);
> +
> +    if (irq < 0) {
> +        return;
> +    }
> +
> +    env->pending_interrupts &= ~(1 << irq);
> +}
> +
> +/*
> + * sends a message to other threads that are on the same
> + * multi-threaded processor
> + */
> +void helper_book3s_msgsndp(CPUPPCState *env, target_ulong rb)
> +{
> +    int irq = book3s_dbell2irq(rb, false);
> +    int pir = env->spr_cb[SPR_PIR].default_value;
> +
> +    if (irq < 0) {
> +        return;
> +    }
> +
> +    pir &= ~DBELL_TIRTAG_MASK;
> +    pir |= rb & DBELL_TIRTAG_MASK;

This seems overkill since we don't actually support > 1 thread/core.
Won't the answer always be equal to pir?

> +
> +    book3s_msgsnd_common(pir, irq);
> +}
> +#endif
>  #endif
>  
>  void ppc_cpu_do_unaligned_access(CPUState *cs, vaddr vaddr,
> diff --git a/target/ppc/misc_helper.c b/target/ppc/misc_helper.c
> index 2318f3ab45b2..66b5b0824208 100644
> --- a/target/ppc/misc_helper.c
> +++ b/target/ppc/misc_helper.c
> @@ -105,6 +105,42 @@ void helper_store_pcr(CPUPPCState *env, target_ulong 
> value)
>  
>      env->spr[SPR_PCR] = value & pcc->pcr_mask;
>  }
> +
> +/*
> + * DPDES register is shared. Each bit reflects the state of the
> + * doorbell interrupt of a thread of the same core.
> + */
> +target_ulong helper_load_dpdes(CPUPPCState *env)
> +{
> +    target_ulong dpdes = 0;
> +
> +    /* TODO: TCG supports only one thread */
> +    if (env->pending_interrupts & (1 << PPC_INTERRUPT_DOORBELL)) {
> +        dpdes |= (env->spr_cb[SPR_PIR].default_value & DBELL_TIRTAG_MASK);

Likewise, won't this just be 0 or 1?  I dislike half-measures to
achieve something we don't actually support at this stage.

> +    }
> +
> +    return dpdes;
> +}
> +
> +void helper_store_dpdes(CPUPPCState *env, target_ulong val)
> +{
> +    PowerPCCPU *cpu = env_archcpu(env);
> +    CPUState *cs = CPU(cpu);
> +
> +    /* TODO: TCG supports only one thread */
> +    if (val & ~0x1) {
> +        qemu_log_mask(LOG_GUEST_ERROR, "Invalid DPDES register value "
> +                      TARGET_FMT_lx"\n", val);
> +        return;
> +    }
> +
> +    if (val & 0x1) {
> +        env->pending_interrupts |= 1 << PPC_INTERRUPT_DOORBELL;
> +        cpu_interrupt(cs, CPU_INTERRUPT_HARD);
> +    } else {
> +        env->pending_interrupts &= ~(1 << PPC_INTERRUPT_DOORBELL);
> +    }
> +}
>  #endif /* defined(TARGET_PPC64) */
>  
>  void helper_store_pidr(CPUPPCState *env, target_ulong val)
> diff --git a/target/ppc/translate.c b/target/ppc/translate.c
> index f5fe5d06118a..a3a4a95cdf53 100644
> --- a/target/ppc/translate.c
> +++ b/target/ppc/translate.c
> @@ -6645,6 +6645,28 @@ static void gen_msgsnd(DisasContext *ctx)
>  #endif /* defined(CONFIG_USER_ONLY) */
>  }
>  
> +#if defined(TARGET_PPC64)
> +static void gen_msgclrp(DisasContext *ctx)
> +{
> +#if defined(CONFIG_USER_ONLY)
> +    GEN_PRIV;
> +#else
> +    CHK_SV;
> +    gen_helper_book3s_msgclrp(cpu_env, cpu_gpr[rB(ctx->opcode)]);
> +#endif /* defined(CONFIG_USER_ONLY) */
> +}
> +
> +static void gen_msgsndp(DisasContext *ctx)
> +{
> +#if defined(CONFIG_USER_ONLY)
> +    GEN_PRIV;
> +#else
> +    CHK_SV;
> +    gen_helper_book3s_msgsndp(cpu_env, cpu_gpr[rB(ctx->opcode)]);
> +#endif /* defined(CONFIG_USER_ONLY) */
> +}
> +#endif
> +
>  static void gen_msgsync(DisasContext *ctx)
>  {
>  #if defined(CONFIG_USER_ONLY)
> @@ -7187,6 +7209,10 @@ GEN_HANDLER(vmladduhm, 0x04, 0x11, 0xFF, 0x00000000, 
> PPC_ALTIVEC),
>  GEN_HANDLER_E(maddhd_maddhdu, 0x04, 0x18, 0xFF, 0x00000000, PPC_NONE,
>                PPC2_ISA300),
>  GEN_HANDLER_E(maddld, 0x04, 0x19, 0xFF, 0x00000000, PPC_NONE, PPC2_ISA300),
> +GEN_HANDLER2_E(msgsndp, "msgsndp", 0x1F, 0x0E, 0x04, 0x03ff0001,
> +               PPC_NONE, PPC2_ISA207S),
> +GEN_HANDLER2_E(msgclrp, "msgclrp", 0x1F, 0x0E, 0x05, 0x03ff0001,
> +               PPC_NONE, PPC2_ISA207S),
>  #endif
>  
>  #undef GEN_INT_ARITH_ADD
> diff --git a/target/ppc/translate_init.inc.c b/target/ppc/translate_init.inc.c
> index d33d65dff702..9e2396a7b5a1 100644
> --- a/target/ppc/translate_init.inc.c
> +++ b/target/ppc/translate_init.inc.c
> @@ -464,6 +464,17 @@ static void spr_write_pcr(DisasContext *ctx, int sprn, 
> int gprn)
>  {
>      gen_helper_store_pcr(cpu_env, cpu_gpr[gprn]);
>  }
> +
> +/* DPDES */
> +static void spr_read_dpdes(DisasContext *ctx, int gprn, int sprn)
> +{
> +    gen_helper_load_dpdes(cpu_gpr[gprn], cpu_env);
> +}
> +
> +static void spr_write_dpdes(DisasContext *ctx, int sprn, int gprn)
> +{
> +    gen_helper_store_dpdes(cpu_env, cpu_gpr[gprn]);
> +}
>  #endif
>  #endif
>  
> @@ -8238,10 +8249,11 @@ static void gen_spr_power8_dpdes(CPUPPCState *env)
>  {
>  #if !defined(CONFIG_USER_ONLY)
>      /* Directed Privileged Door-bell Exception State, used for IPI */
> -    spr_register(env, SPR_DPDES, "DPDES",
> -                 SPR_NOACCESS, SPR_NOACCESS,
> -                 &spr_read_generic, SPR_NOACCESS,
> -                 0x00000000);
> +    spr_register_kvm_hv(env, SPR_DPDES, "DPDES",
> +                        SPR_NOACCESS, SPR_NOACCESS,
> +                        &spr_read_dpdes, SPR_NOACCESS,
> +                        &spr_read_dpdes, &spr_write_dpdes,
> +                        KVM_REG_PPC_DPDES, 0x00000000);
>  #endif
>  }
>  

-- 
David Gibson                    | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
                                | _way_ _around_!
http://www.ozlabs.org/~dgibson

Attachment: signature.asc
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]