From: Nicholas Piggin <npiggin@gmail.com>
To: kvm-ppc@vger.kernel.org
Cc: linuxppc-dev@lists.ozlabs.org,
Nicholas Piggin <npiggin@gmail.com>,
Fabiano Rosas <farosas@linux.ibm.com>
Subject: [PATCH v2 44/60] KVM: PPC: Book3S HV P9: Demand fault EBB facility registers
Date: Thu, 12 Aug 2021 02:01:18 +1000 [thread overview]
Message-ID: <20210811160134.904987-45-npiggin@gmail.com> (raw)
In-Reply-To: <20210811160134.904987-1-npiggin@gmail.com>
Use HFSCR facility disabling to implement demand faulting for EBB, with
a hysteresis counter similar to the load_fp etc counters in context
switching that implement the equivalent demand faulting for userspace
facilities.
This speeds up guest entry/exit by avoiding the register save/restore
when a guest is not frequently using them. When a guest does use them
often, there will be some additional demand fault overhead, but these
are not commonly used facilities.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_host.h | 1 +
arch/powerpc/kvm/book3s_hv.c | 16 +++++++++++++--
arch/powerpc/kvm/book3s_hv_p9_entry.c | 28 +++++++++++++++++++++------
3 files changed, 37 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 89216e247e8b..c8d8bd90bf14 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -580,6 +580,7 @@ struct kvm_vcpu_arch {
ulong cfar;
ulong ppr;
u32 pspb;
+ u8 load_ebb;
ulong fscr;
ulong shadow_fscr;
ulong ebbhr;
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 8ca081a32d91..99d4fd84f0b2 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1436,6 +1436,16 @@ static int kvmppc_pmu_unavailable(struct kvm_vcpu *vcpu)
return RESUME_GUEST;
}
+static int kvmppc_ebb_unavailable(struct kvm_vcpu *vcpu)
+{
+ if (!(vcpu->arch.hfscr_permitted & HFSCR_EBB))
+ return EMULATE_FAIL;
+
+ vcpu->arch.hfscr |= HFSCR_EBB;
+
+ return RESUME_GUEST;
+}
+
static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
struct task_struct *tsk)
{
@@ -1730,6 +1740,8 @@ XXX benchmark guest exits
r = kvmppc_emulate_doorbell_instr(vcpu);
if (cause == FSCR_PM_LG)
r = kvmppc_pmu_unavailable(vcpu);
+ if (cause == FSCR_EBB_LG)
+ r = kvmppc_ebb_unavailable(vcpu);
}
if (r == EMULATE_FAIL) {
kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
@@ -2774,9 +2786,9 @@ static int kvmppc_core_vcpu_create_hv(struct kvm_vcpu *vcpu)
vcpu->arch.hfscr_permitted = vcpu->arch.hfscr;
/*
- * PM is demand-faulted so start with it clear.
+ * PM, EBB is demand-faulted so start with it clear.
*/
- vcpu->arch.hfscr &= ~HFSCR_PM;
+ vcpu->arch.hfscr &= ~(HFSCR_PM | HFSCR_EBB);
kvmppc_mmu_book3s_hv_init(vcpu);
diff --git a/arch/powerpc/kvm/book3s_hv_p9_entry.c b/arch/powerpc/kvm/book3s_hv_p9_entry.c
index 3a7e242887aa..d196642061aa 100644
--- a/arch/powerpc/kvm/book3s_hv_p9_entry.c
+++ b/arch/powerpc/kvm/book3s_hv_p9_entry.c
@@ -232,9 +232,12 @@ static void load_spr_state(struct kvm_vcpu *vcpu,
struct p9_host_os_sprs *host_os_sprs)
{
mtspr(SPRN_TAR, vcpu->arch.tar);
- mtspr(SPRN_EBBHR, vcpu->arch.ebbhr);
- mtspr(SPRN_EBBRR, vcpu->arch.ebbrr);
- mtspr(SPRN_BESCR, vcpu->arch.bescr);
+
+ if (vcpu->arch.hfscr & HFSCR_EBB) {
+ mtspr(SPRN_EBBHR, vcpu->arch.ebbhr);
+ mtspr(SPRN_EBBRR, vcpu->arch.ebbrr);
+ mtspr(SPRN_BESCR, vcpu->arch.bescr);
+ }
if (!cpu_has_feature(CPU_FTR_ARCH_31))
mtspr(SPRN_TIDR, vcpu->arch.tid);
@@ -265,9 +268,22 @@ static void load_spr_state(struct kvm_vcpu *vcpu,
static void store_spr_state(struct kvm_vcpu *vcpu)
{
vcpu->arch.tar = mfspr(SPRN_TAR);
- vcpu->arch.ebbhr = mfspr(SPRN_EBBHR);
- vcpu->arch.ebbrr = mfspr(SPRN_EBBRR);
- vcpu->arch.bescr = mfspr(SPRN_BESCR);
+
+ if (vcpu->arch.hfscr & HFSCR_EBB) {
+ vcpu->arch.ebbhr = mfspr(SPRN_EBBHR);
+ vcpu->arch.ebbrr = mfspr(SPRN_EBBRR);
+ vcpu->arch.bescr = mfspr(SPRN_BESCR);
+ /*
+ * This is like load_fp in context switching, turn off the
+ * facility after it wraps the u8 to try avoiding saving
+ * and restoring the registers each partition switch.
+ */
+ if (!vcpu->arch.nested) {
+ vcpu->arch.load_ebb++;
+ if (!vcpu->arch.load_ebb)
+ vcpu->arch.hfscr &= ~HFSCR_EBB;
+ }
+ }
if (!cpu_has_feature(CPU_FTR_ARCH_31))
vcpu->arch.tid = mfspr(SPRN_TIDR);
--
2.23.0
next prev parent reply other threads:[~2021-08-11 16:31 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-11 16:00 [PATCH v2 00/60] KVM: PPC: Book3S HV P9: entry/exit optimisations Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 01/60] KVM: PPC: Book3S HV: Initialise vcpu MSR with MSR_ME Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 02/60] KVM: PPC: Book3S HV: Remove TM emulation from POWER7/8 path Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 03/60] KVM: PPC: Book3S HV P9: Fixes for TM softpatch interrupt NIP Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 04/60] KVM: PPC: Book3S HV Nested: Fix TM softpatch HFAC interrupt emulation Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 05/60] KVM: PPC: Book3S HV Nested: Sanitise vcpu registers Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 06/60] KVM: PPC: Book3S HV Nested: Make nested HFSCR state accessible Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 07/60] KVM: PPC: Book3S HV Nested: Stop forwarding all HFUs to L1 Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 08/60] KVM: PPC: Book3S HV Nested: save_hv_return_state does not require trap argument Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 09/60] KVM: PPC: Book3S HV Nested: Reflect guest PMU in-use to L0 when guest SPRs are live Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 10/60] powerpc/64s: Remove WORT SPR from POWER9/10 Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 11/60] KMV: PPC: Book3S HV P9: Use set_dec to set decrementer to host Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 12/60] KVM: PPC: Book3S HV P9: Use host timer accounting to avoid decrementer read Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 13/60] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 14/60] KVM: PPC: Book3S HV P9: Reduce mftb per guest entry/exit Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 15/60] powerpc/time: add API for KVM to re-arm the host timer/decrementer Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 16/60] KVM: PPC: Book3S HV: POWER10 enable HAIL when running radix guests Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 17/60] powerpc/64s: Keep AMOR SPR a constant ~0 at runtime Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 18/60] KVM: PPC: Book3S HV: Don't always save PMU for guest capable of nesting Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 19/60] powerpc/64s: Always set PMU control registers to frozen/disabled when not in use Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 20/60] powerpc/64s: Implement PMU override command line option Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 21/60] KVM: PPC: Book3S HV P9: Implement PMU save/restore in C Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 22/60] KVM: PPC: Book3S HV P9: Factor PMU save/load into context switch functions Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 23/60] KVM: PPC: Book3S HV P9: Demand fault PMU SPRs when marked not inuse Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 24/60] KVM: PPC: Book3S HV P9: Factor out yield_count increment Nicholas Piggin
2021-08-11 16:00 ` [PATCH v2 25/60] KVM: PPC: Book3S HV: CTRL SPR does not require read-modify-write Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 26/60] KVM: PPC: Book3S HV P9: Move SPRG restore to restore_p9_host_os_sprs Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 27/60] KVM: PPC: Book3S HV P9: Reduce mtmsrd instructions required to save host SPRs Nicholas Piggin
2021-08-11 19:18 ` kernel test robot
2021-08-11 19:49 ` kernel test robot
2021-08-11 16:01 ` [PATCH v2 28/60] KVM: PPC: Book3S HV P9: Improve mtmsrd scheduling by delaying MSR[EE] disable Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 29/60] KVM: PPC: Book3S HV P9: Add kvmppc_stop_thread to match kvmppc_start_thread Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 30/60] KVM: PPC: Book3S HV: Change dec_expires to be relative to guest timebase Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 31/60] KVM: PPC: Book3S HV P9: Move TB updates Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 32/60] KVM: PPC: Book3S HV P9: Optimise timebase reads Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 33/60] KVM: PPC: Book3S HV P9: Avoid SPR scoreboard stalls Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 34/60] KVM: PPC: Book3S HV P9: Only execute mtSPR if the value changed Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 35/60] KVM: PPC: Book3S HV P9: Juggle SPR switching around Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 36/60] KVM: PPC: Book3S HV P9: Move vcpu register save/restore into functions Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 37/60] KVM: PPC: Book3S HV P9: Move host OS save/restore functions to built-in Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 38/60] KVM: PPC: Book3S HV P9: Move nested guest entry into its own function Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 39/60] KVM: PPC: Book3S HV P9: Move remaining SPR and MSR access into low level entry Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 40/60] KVM: PPC: Book3S HV P9: Implement TM fastpath for guest entry/exit Nicholas Piggin
2021-08-11 20:40 ` kernel test robot
2021-08-11 16:01 ` [PATCH v2 41/60] KVM: PPC: Book3S HV P9: Switch PMU to guest as late as possible Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 42/60] KVM: PPC: Book3S HV P9: Restrict DSISR canary workaround to processors that require it Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 43/60] KVM: PPC: Book3S HV P9: More SPR speed improvements Nicholas Piggin
2021-08-11 16:01 ` Nicholas Piggin [this message]
2021-08-11 16:01 ` [PATCH v2 45/60] KVM: PPC: Book3S HV P9: Demand fault TM facility registers Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 46/60] KVM: PPC: Book3S HV P9: Use Linux SPR save/restore to manage some host SPRs Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 47/60] KVM: PPC: Book3S HV P9: Comment and fix MMU context switching code Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 48/60] KVM: PPC: Book3S HV P9: Test dawr_enabled() before saving host DAWR SPRs Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 49/60] KVM: PPC: Book3S HV P9: Don't restore PSSCR if not needed Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 50/60] KVM: PPC: Book3S HV P9: Avoid tlbsync sequence on radix guest exit Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 51/60] KVM: PPC: Book3S HV Nested: Avoid extra mftb() in nested entry Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 52/60] KVM: PPC: Book3S HV P9: Improve mfmsr performance on entry Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 53/60] KVM: PPC: Book3S HV P9: Optimise hash guest SLB saving Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 54/60] KVM: PPC: Book3S HV P9: Avoid changing MSR[RI] in entry and exit Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 55/60] KVM: PPC: Book3S HV P9: Add unlikely annotation for !mmu_ready Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 56/60] KVM: PPC: Book3S HV P9: Avoid cpu_in_guest atomics on entry and exit Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 57/60] KVM: PPC: Book3S HV P9: Remove most of the vcore logic Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 58/60] KVM: PPC: Book3S HV P9: Tidy kvmppc_create_dtl_entry Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 59/60] KVM: PPC: Book3S HV P9: Stop using vc->dpdes Nicholas Piggin
2021-08-11 16:01 ` [PATCH v2 60/60] KVM: PPC: Book3S HV P9: Remove subcore HMI handling Nicholas Piggin
2021-08-16 8:41 ` [PATCH v2 00/60] KVM: PPC: Book3S HV P9: entry/exit optimisations Athira Rajeev
2021-08-27 13:16 ` (subset) " Michael Ellerman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210811160134.904987-45-npiggin@gmail.com \
--to=npiggin@gmail.com \
--cc=farosas@linux.ibm.com \
--cc=kvm-ppc@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).