[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 09/15] exec: Introduce AddressSpaceDispatch.mru_secti
From: |
Paolo Bonzini |
Subject: |
[Qemu-devel] [PULL 09/15] exec: Introduce AddressSpaceDispatch.mru_section |
Date: |
Mon, 7 Mar 2016 18:36:55 +0100 |
From: Fam Zheng <address@hidden>
Under heavy workloads the lookup will likely end up with the same
MemoryRegionSection from last time. Using a pointer to cache the result,
like ram_list.mru_block, significantly reduces cost of
address_space_translate.
During address space topology update, as->dispatch will be reallocated
so the pointer is invalidated automatically.
Perf reports a visible drop on the cpu usage, because phys_page_find is
not called. Before:
2.35% qemu-system-x86_64 [.] phys_page_find
0.97% qemu-system-x86_64 [.] address_space_translate_internal
0.95% qemu-system-x86_64 [.] address_space_translate
0.55% qemu-system-x86_64 [.] address_space_lookup_region
After:
0.97% qemu-system-x86_64 [.] address_space_translate_internal
0.97% qemu-system-x86_64 [.] address_space_lookup_region
0.84% qemu-system-x86_64 [.] address_space_translate
Signed-off-by: Fam Zheng <address@hidden>
Message-Id: <address@hidden>
Signed-off-by: Paolo Bonzini <address@hidden>
---
exec.c | 16 ++++++++++++++--
1 file changed, 14 insertions(+), 2 deletions(-)
diff --git a/exec.c b/exec.c
index 9279af5..f09dd4e 100644
--- a/exec.c
+++ b/exec.c
@@ -135,6 +135,7 @@ typedef struct PhysPageMap {
struct AddressSpaceDispatch {
struct rcu_head rcu;
+ MemoryRegionSection *mru_section;
/* This is a multi-level map on the physical address space.
* The bottom level has pointers to MemoryRegionSections.
*/
@@ -351,14 +352,25 @@ static MemoryRegionSection
*address_space_lookup_region(AddressSpaceDispatch *d,
hwaddr addr,
bool resolve_subpage)
{
- MemoryRegionSection *section;
+ MemoryRegionSection *section = atomic_read(&d->mru_section);
subpage_t *subpage;
+ bool update;
- section = phys_page_find(d->phys_map, addr, d->map.nodes, d->map.sections);
+ if (section && section != &d->map.sections[PHYS_SECTION_UNASSIGNED] &&
+ section_covers_addr(section, addr)) {
+ update = false;
+ } else {
+ section = phys_page_find(d->phys_map, addr, d->map.nodes,
+ d->map.sections);
+ update = true;
+ }
if (resolve_subpage && section->mr->subpage) {
subpage = container_of(section->mr, subpage_t, iomem);
section = &d->map.sections[subpage->sub_section[SUBPAGE_IDX(addr)]];
}
+ if (update) {
+ atomic_set(&d->mru_section, section);
+ }
return section;
}
--
2.5.0
- [Qemu-devel] [PULL 00/15] Memory changes and misc bug fixes for 2016-03-07, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 01/15] log: do not log if QEMU is daemonized but without -D, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 02/15] i8257: fix Terminal Count status, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 03/15] exec: Return RAMBlock pointer from allocating functions, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 04/15] memory: Move assignment to ram_block to memory_region_init_*, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 05/15] memory: Implement memory_region_get_ram_addr with mr->ram_block, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 06/15] memory: Drop MemoryRegion.ram_addr, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 07/15] exec: Pass RAMBlock pointer to qemu_ram_free, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 08/15] exec: Factor out section_covers_addr, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 11/15] doc/memory.txt: correct a logic error, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 09/15] exec: Introduce AddressSpaceDispatch.mru_section,
Paolo Bonzini <=
- [Qemu-devel] [PULL 12/15] doc/memory.txt: correct description of MemoryRegionOps fields, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 13/15] MAINTAINERS: Add entry for include/sysemu/kvm*.h, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 14/15] kvm/irqchip: use bitmap utility for gsi tracking, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 15/15] scsi-bus: Remove tape command from scsi_req_xfer, Paolo Bonzini, 2016/03/07
- [Qemu-devel] [PULL 10/15] icount: possible options for sleep are on or off, Paolo Bonzini, 2016/03/07
- Re: [Qemu-devel] [PULL 00/15] Memory changes and misc bug fixes for 2016-03-07, Peter Maydell, 2016/03/08