diff options
| author | Hollis Blanchard <hollisb@us.ibm.com> | 2008-12-02 15:51:55 -0600 |
|---|---|---|
| committer | Avi Kivity <avi@redhat.com> | 2008-12-31 16:55:09 +0200 |
| commit | 7924bd41097ae8991c6d38cef8b1e4058e30d198 (patch) | |
| tree | b39629f81598739eb886126c5f3f8705656ce9cd /arch/powerpc/kvm/booke_interrupts.S | |
| parent | c0ca609c5f874f7d6ae8e180afe79317e1943d22 (diff) | |
KVM: ppc: directly insert shadow mappings into the hardware TLB
Formerly, we used to maintain a per-vcpu shadow TLB and on every entry to the
guest would load this array into the hardware TLB. This consumed 1280 bytes of
memory (64 entries of 16 bytes plus a struct page pointer each), and also
required some assembly to loop over the array on every entry.
Instead of saving a copy in memory, we can just store shadow mappings directly
into the hardware TLB, accepting that the host kernel will clobber these as
part of the normal 440 TLB round robin. When we do that we need less than half
the memory, and we have decreased the exit handling time for all guest exits,
at the cost of increased number of TLB misses because the host overwrites some
guest entries.
These savings will be increased on processors with larger TLBs or which
implement intelligent flush instructions like tlbivax (which will avoid the
need to walk arrays in software).
In addition to that and to the code simplification, we have a greater chance of
leaving other host userspace mappings in the TLB, instead of forcing all
subsequent tasks to re-fault all their mappings.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Diffstat (limited to 'arch/powerpc/kvm/booke_interrupts.S')
| -rw-r--r-- | arch/powerpc/kvm/booke_interrupts.S | 48 |
1 files changed, 0 insertions, 48 deletions
diff --git a/arch/powerpc/kvm/booke_interrupts.S b/arch/powerpc/kvm/booke_interrupts.S index 8d6929b7fdb6..eb2186823e4e 100644 --- a/arch/powerpc/kvm/booke_interrupts.S +++ b/arch/powerpc/kvm/booke_interrupts.S @@ -335,54 +335,6 @@ lightweight_exit: lwz r3, VCPU_SHADOW_PID(r4) mtspr SPRN_PID, r3 - /* Prevent all asynchronous TLB updates. */ - mfmsr r5 - lis r6, (MSR_EE|MSR_CE|MSR_ME|MSR_DE)@h - ori r6, r6, (MSR_EE|MSR_CE|MSR_ME|MSR_DE)@l - andc r6, r5, r6 - mtmsr r6 - - /* Load the guest mappings, leaving the host's "pinned" kernel mappings - * in place. */ - mfspr r10, SPRN_MMUCR /* Save host MMUCR. */ - li r5, PPC44x_TLB_SIZE - lis r5, tlb_44x_hwater@ha - lwz r5, tlb_44x_hwater@l(r5) - mtctr r5 - addi r9, r4, -VCPU_TO_44X + VCPU44x_SHADOW_TLB - addi r5, r4, -VCPU_TO_44X + VCPU44x_SHADOW_MOD - li r3, 0 -1: - lbzx r7, r3, r5 - cmpwi r7, 0 - beq 3f - - /* Load guest entry. */ - mulli r11, r3, TLBE_BYTES - add r11, r11, r9 - lwz r7, 0(r11) - mtspr SPRN_MMUCR, r7 - lwz r7, 4(r11) - tlbwe r7, r3, PPC44x_TLB_PAGEID - lwz r7, 8(r11) - tlbwe r7, r3, PPC44x_TLB_XLAT - lwz r7, 12(r11) - tlbwe r7, r3, PPC44x_TLB_ATTRIB -3: - addi r3, r3, 1 /* Increment index. */ - bdnz 1b - - mtspr SPRN_MMUCR, r10 /* Restore host MMUCR. */ - - /* Clear bitmap of modified TLB entries */ - li r5, PPC44x_TLB_SIZE>>2 - mtctr r5 - addi r5, r4, -VCPU_TO_44X + VCPU44x_SHADOW_MOD - 4 - li r6, 0 -1: - stwu r6, 4(r5) - bdnz 1b - iccci 0, 0 /* XXX hack */ /* Load some guest volatiles. */ |