| Message ID | 20210305150638.2675513-24-npiggin@gmail.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | KVM: PPC: Book3S: C-ify the P9 entry/exit code | expand |
| Related | show |
| Context | Check | Description |
|---|---|---|
| snowpatch_ozlabs/apply_patch | warning | Failed to apply on branch powerpc/merge (91966823812efbd175f904599e5cf2a854b39809) |
| snowpatch_ozlabs/apply_patch | warning | Failed to apply on branch powerpc/next (fe07bfda2fb9cdef8a4d4008a409bb02f35f1bd8) |
| snowpatch_ozlabs/apply_patch | warning | Failed to apply on branch linus/master (280d542f6ffac0e6d65dc267f92191d509b13b64) |
| snowpatch_ozlabs/apply_patch | warning | Failed to apply on branch powerpc/fixes (5c88a17e15795226b56d83f579cbb9b7a4864f79) |
| snowpatch_ozlabs/apply_patch | warning | Failed to apply on branch linux-next (7a7fd0de4a9804299793e564a555a49c1fc924cb) |
| snowpatch_ozlabs/apply_patch | fail | Failed to apply to any branch |
Nicholas Piggin <npiggin@gmail.com> writes: > mftb is serialising (dispatch next-to-complete) so it is heavy weight > for a mfspr. Avoid reading it multiple times in the entry or exit paths. > A small number of cycles delay to timers is tolerable. > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com> > --- > arch/powerpc/kvm/book3s_hv.c | 9 +++++---- > 1 file changed, 5 insertions(+), 4 deletions(-) > > diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c > index c1965a9d8d00..6f3e3aed99aa 100644 > --- a/arch/powerpc/kvm/book3s_hv.c > +++ b/arch/powerpc/kvm/book3s_hv.c > @@ -3505,12 +3505,13 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit, > host_dawrx1 = mfspr(SPRN_DAWRX1); > } > > - hdec = time_limit - mftb(); > + tb = mftb(); > + hdec = time_limit - tb; > if (hdec < 0) > return BOOK3S_INTERRUPT_HV_DECREMENTER; > > if (vc->tb_offset) { > - u64 new_tb = mftb() + vc->tb_offset; > + u64 new_tb = tb + vc->tb_offset; > mtspr(SPRN_TBU40, new_tb); > tb = mftb(); > if ((tb & 0xffffff) < (new_tb & 0xffffff)) > @@ -3703,7 +3704,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit, > if (!(vcpu->arch.ctrl & 1)) > mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1); > > - mtspr(SPRN_DEC, vcpu->arch.dec_expires - mftb()); > + mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb); > > if (kvmhv_on_pseries()) { > /* > @@ -3837,7 +3838,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit, > vc->entry_exit_map = 0x101; > vc->in_guest = 0; > > - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - mftb()); > + mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb); > mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso); > > kvmhv_load_host_pmu();
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index c1965a9d8d00..6f3e3aed99aa 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -3505,12 +3505,13 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit, host_dawrx1 = mfspr(SPRN_DAWRX1); } - hdec = time_limit - mftb(); + tb = mftb(); + hdec = time_limit - tb; if (hdec < 0) return BOOK3S_INTERRUPT_HV_DECREMENTER; if (vc->tb_offset) { - u64 new_tb = mftb() + vc->tb_offset; + u64 new_tb = tb + vc->tb_offset; mtspr(SPRN_TBU40, new_tb); tb = mftb(); if ((tb & 0xffffff) < (new_tb & 0xffffff)) @@ -3703,7 +3704,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit, if (!(vcpu->arch.ctrl & 1)) mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1); - mtspr(SPRN_DEC, vcpu->arch.dec_expires - mftb()); + mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb); if (kvmhv_on_pseries()) { /* @@ -3837,7 +3838,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit, vc->entry_exit_map = 0x101; vc->in_guest = 0; - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - mftb()); + mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb); mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso); kvmhv_load_host_pmu();
mftb is serialising (dispatch next-to-complete) so it is heavy weight for a mfspr. Avoid reading it multiple times in the entry or exit paths. A small number of cycles delay to timers is tolerable. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> --- arch/powerpc/kvm/book3s_hv.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)