qemu-e2k/include/exec/exec-all.h
Emilio G. Cota 909eaac9bb tb hash: track translated blocks with qht
Having a fixed-size hash table for keeping track of all translation blocks
is suboptimal: some workloads are just too big or too small to get maximum
performance from the hash table. The MRU promotion policy helps improve
performance when the hash table is a little undersized, but it cannot
make up for severely undersized hash tables.

Furthermore, frequent MRU promotions result in writes that are a scalability
bottleneck. For scalability, lookups should only perform reads, not writes.
This is not a big deal for now, but it will become one once MTTCG matures.

The appended fixes these issues by using qht as the implementation of
the TB hash table. This solution is superior to other alternatives considered,
namely:

- master: implementation in QEMU before this patchset
- xxhash: before this patch, i.e. fixed buckets + xxhash hashing + MRU.
- xxhash-rcu: fixed buckets + xxhash + RCU list + MRU.
              MRU is implemented here by adding an intermediate struct
              that contains the u32 hash and a pointer to the TB; this
              allows us, on an MRU promotion, to copy said struct (that is not
              at the head), and put this new copy at the head. After a grace
              period, the original non-head struct can be eliminated, and
              after another grace period, freed.
- qht-fixed-nomru: fixed buckets + xxhash + qht without auto-resize +
                   no MRU for lookups; MRU for inserts.
The appended solution is the following:
- qht-dyn-nomru: dynamic number of buckets + xxhash + qht w/ auto-resize +
                 no MRU for lookups; MRU for inserts.

The plots below compare the considered solutions. The Y axis shows the
boot time (in seconds) of a debian jessie image with arm-softmmu; the X axis
sweeps the number of buckets (or initial number of buckets for qht-autoresize).
The plots in PNG format (and with errorbars) can be seen here:
  http://imgur.com/a/Awgnq

Each test runs 5 times, and the entire QEMU process is pinned to a
single core for repeatability of results.

                            Host: Intel Xeon E5-2690

  28 ++------------+-------------+-------------+-------------+------------++
     A*****        +             +             +             master **A*** +
  27 ++    *                                                 xxhash ##B###++
     |      A******A******                               xxhash-rcu $$C$$$ |
  26 C$$                  A******A******            qht-fixed-nomru*%%D%%%++
     D%%$$                              A******A******A*qht-dyn-mru A*E****A
  25 ++ %%$$                                          qht-dyn-nomru &&F&&&++
     B#####%                                                               |
  24 ++    #C$$$$$                                                        ++
     |      B###  $                                                        |
     |          ## C$$$$$$                                                 |
  23 ++           #       C$$$$$$                                         ++
     |             B######       C$$$$$$                                %%%D
  22 ++                  %B######       C$$$$$$C$$$$$$C$$$$$$C$$$$$$C$$$$$$C
     |                    D%%%%%%B######      @E@@@@@@    %%%D%%%@@@E@@@@@@E
  21 E@@@@@@E@@@@@@F&&&@@@E@@@&&&D%%%%%%B######B######B######B######B######B
     +             E@@@   F&&&   +      E@     +      F&&&   +             +
  20 ++------------+-------------+-------------+-------------+------------++
     14            16            18            20            22            24
                             log2 number of buckets

                                 Host: Intel i7-4790K

  14.5 ++------------+------------+-------------+------------+------------++
       A**           +            +             +            master **A*** +
    14 ++ **                                                 xxhash ##B###++
  13.5 ++   **                                           xxhash-rcu $$C$$$++
       |                                            qht-fixed-nomru %%D%%% |
    13 ++     A******                                   qht-dyn-mru @@E@@@++
       |             A*****A******A******             qht-dyn-nomru &&F&&& |
  12.5 C$$                               A******A******A*****A******    ***A
    12 ++ $$                                                        A***  ++
       D%%% $$                                                             |
  11.5 ++  %%                                                             ++
       B###  %C$$$$$$                                                      |
    11 ++  ## D%%%%% C$$$$$                                               ++
       |     #      %      C$$$$$$                                         |
  10.5 F&&&&&&B######D%%%%%       C$$$$$$C$$$$$$C$$$$$$C$$$$$C$$$$$$    $$$C
    10 E@@@@@@E@@@@@@B#####B######B######E@@@@@@E@@@%%%D%%%%%D%%%###B######B
       +             F&&          D%%%%%%B######B######B#####B###@@@D%%%   +
   9.5 ++------------+------------+-------------+------------+------------++
       14            16           18            20           22            24
                              log2 number of buckets

Note that the original point before this patch series is X=15 for "master";
the little sensitivity to the increased number of buckets is due to the
poor hashing function in master.

xxhash-rcu has significant overhead due to the constant churn of allocating
and deallocating intermediate structs for implementing MRU. An alternative
would be do consider failed lookups as "maybe not there", and then
acquire the external lock (tb_lock in this case) to really confirm that
there was indeed a failed lookup. This, however, would not be enough
to implement dynamic resizing--this is more complex: see
"Resizable, Scalable, Concurrent Hash Tables via Relativistic
Programming" by Triplett, McKenney and Walpole. This solution was
discarded due to the very coarse RCU read critical sections that we have
in MTTCG; resizing requires waiting for readers after every pointer update,
and resizes require many pointer updates, so this would quickly become
prohibitive.

qht-fixed-nomru shows that MRU promotion is advisable for undersized
hash tables.

However, qht-dyn-mru shows that MRU promotion is not important if the
hash table is properly sized: there is virtually no difference in
performance between qht-dyn-nomru and qht-dyn-mru.

Before this patch, we're at X=15 on "xxhash"; after this patch, we're at
X=15 @ qht-dyn-nomru. This patch thus matches the best performance that we
can achieve with optimum sizing of the hash table, while keeping the hash
table scalable for readers.

The improvement we get before and after this patch for booting debian jessie
with arm-softmmu is:

- Intel Xeon E5-2690: 10.5% less time
- Intel i7-4790K: 5.2% less time

We could get this same improvement _for this particular workload_ by
statically increasing the size of the hash table. But this would hurt
workloads that do not need a large hash table. The dynamic (upward)
resizing allows us to start small and enlarge the hash table as needed.

A quick note on downsizing: the table is resized back to 2**15 buckets
on every tb_flush; this makes sense because it is not guaranteed that the
table will reach the same number of TBs later on (e.g. most bootup code is
thrown away after boot); it makes sense to grow the hash table as
more code blocks are translated. This also avoids the complication of
having to build downsizing hysteresis logic into qht.

Reviewed-by: Sergey Fedorov <serge.fedorov@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <1465412133-3029-15-git-send-email-cota@braap.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
2016-06-11 17:11:16 -07:00

411 lines
15 KiB
C

/*
* internal execution defines for qemu
*
* Copyright (c) 2003 Fabrice Bellard
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, see <http://www.gnu.org/licenses/>.
*/
#ifndef _EXEC_ALL_H_
#define _EXEC_ALL_H_
#include "qemu-common.h"
#include "exec/tb-context.h"
/* allow to see translation results - the slowdown should be negligible, so we leave it */
#define DEBUG_DISAS
/* Page tracking code uses ram addresses in system mode, and virtual
addresses in userspace mode. Define tb_page_addr_t to be an appropriate
type. */
#if defined(CONFIG_USER_ONLY)
typedef abi_ulong tb_page_addr_t;
#else
typedef ram_addr_t tb_page_addr_t;
#endif
/* is_jmp field values */
#define DISAS_NEXT 0 /* next instruction can be analyzed */
#define DISAS_JUMP 1 /* only pc was modified dynamically */
#define DISAS_UPDATE 2 /* cpu state was modified dynamically */
#define DISAS_TB_JUMP 3 /* only pc was modified statically */
#include "qemu/log.h"
void gen_intermediate_code(CPUArchState *env, struct TranslationBlock *tb);
void restore_state_to_opc(CPUArchState *env, struct TranslationBlock *tb,
target_ulong *data);
void cpu_gen_init(void);
bool cpu_restore_state(CPUState *cpu, uintptr_t searched_pc);
void QEMU_NORETURN cpu_loop_exit_noexc(CPUState *cpu);
void QEMU_NORETURN cpu_io_recompile(CPUState *cpu, uintptr_t retaddr);
TranslationBlock *tb_gen_code(CPUState *cpu,
target_ulong pc, target_ulong cs_base,
uint32_t flags,
int cflags);
void cpu_exec_init(CPUState *cpu, Error **errp);
void QEMU_NORETURN cpu_loop_exit(CPUState *cpu);
void QEMU_NORETURN cpu_loop_exit_restore(CPUState *cpu, uintptr_t pc);
#if !defined(CONFIG_USER_ONLY)
void cpu_reloading_memory_map(void);
/**
* cpu_address_space_init:
* @cpu: CPU to add this address space to
* @as: address space to add
* @asidx: integer index of this address space
*
* Add the specified address space to the CPU's cpu_ases list.
* The address space added with @asidx 0 is the one used for the
* convenience pointer cpu->as.
* The target-specific code which registers ASes is responsible
* for defining what semantics address space 0, 1, 2, etc have.
*
* Before the first call to this function, the caller must set
* cpu->num_ases to the total number of address spaces it needs
* to support.
*
* Note that with KVM only one address space is supported.
*/
void cpu_address_space_init(CPUState *cpu, AddressSpace *as, int asidx);
/* cputlb.c */
/**
* tlb_flush_page:
* @cpu: CPU whose TLB should be flushed
* @addr: virtual address of page to be flushed
*
* Flush one page from the TLB of the specified CPU, for all
* MMU indexes.
*/
void tlb_flush_page(CPUState *cpu, target_ulong addr);
/**
* tlb_flush:
* @cpu: CPU whose TLB should be flushed
* @flush_global: ignored
*
* Flush the entire TLB for the specified CPU.
* The flush_global flag is in theory an indicator of whether the whole
* TLB should be flushed, or only those entries not marked global.
* In practice QEMU does not implement any global/not global flag for
* TLB entries, and the argument is ignored.
*/
void tlb_flush(CPUState *cpu, int flush_global);
/**
* tlb_flush_page_by_mmuidx:
* @cpu: CPU whose TLB should be flushed
* @addr: virtual address of page to be flushed
* @...: list of MMU indexes to flush, terminated by a negative value
*
* Flush one page from the TLB of the specified CPU, for the specified
* MMU indexes.
*/
void tlb_flush_page_by_mmuidx(CPUState *cpu, target_ulong addr, ...);
/**
* tlb_flush_by_mmuidx:
* @cpu: CPU whose TLB should be flushed
* @...: list of MMU indexes to flush, terminated by a negative value
*
* Flush all entries from the TLB of the specified CPU, for the specified
* MMU indexes.
*/
void tlb_flush_by_mmuidx(CPUState *cpu, ...);
/**
* tlb_set_page_with_attrs:
* @cpu: CPU to add this TLB entry for
* @vaddr: virtual address of page to add entry for
* @paddr: physical address of the page
* @attrs: memory transaction attributes
* @prot: access permissions (PAGE_READ/PAGE_WRITE/PAGE_EXEC bits)
* @mmu_idx: MMU index to insert TLB entry for
* @size: size of the page in bytes
*
* Add an entry to this CPU's TLB (a mapping from virtual address
* @vaddr to physical address @paddr) with the specified memory
* transaction attributes. This is generally called by the target CPU
* specific code after it has been called through the tlb_fill()
* entry point and performed a successful page table walk to find
* the physical address and attributes for the virtual address
* which provoked the TLB miss.
*
* At most one entry for a given virtual address is permitted. Only a
* single TARGET_PAGE_SIZE region is mapped; the supplied @size is only
* used by tlb_flush_page.
*/
void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
hwaddr paddr, MemTxAttrs attrs,
int prot, int mmu_idx, target_ulong size);
/* tlb_set_page:
*
* This function is equivalent to calling tlb_set_page_with_attrs()
* with an @attrs argument of MEMTXATTRS_UNSPECIFIED. It's provided
* as a convenience for CPUs which don't use memory transaction attributes.
*/
void tlb_set_page(CPUState *cpu, target_ulong vaddr,
hwaddr paddr, int prot,
int mmu_idx, target_ulong size);
void tb_invalidate_phys_addr(AddressSpace *as, hwaddr addr);
void probe_write(CPUArchState *env, target_ulong addr, int mmu_idx,
uintptr_t retaddr);
#else
static inline void tlb_flush_page(CPUState *cpu, target_ulong addr)
{
}
static inline void tlb_flush(CPUState *cpu, int flush_global)
{
}
static inline void tlb_flush_page_by_mmuidx(CPUState *cpu,
target_ulong addr, ...)
{
}
static inline void tlb_flush_by_mmuidx(CPUState *cpu, ...)
{
}
#endif
#define CODE_GEN_ALIGN 16 /* must be >= of the size of a icache line */
/* Estimated block size for TB allocation. */
/* ??? The following is based on a 2015 survey of x86_64 host output.
Better would seem to be some sort of dynamically sized TB array,
adapting to the block sizes actually being produced. */
#if defined(CONFIG_SOFTMMU)
#define CODE_GEN_AVG_BLOCK_SIZE 400
#else
#define CODE_GEN_AVG_BLOCK_SIZE 150
#endif
#if defined(__arm__) || defined(_ARCH_PPC) \
|| defined(__x86_64__) || defined(__i386__) \
|| defined(__sparc__) || defined(__aarch64__) \
|| defined(__s390x__) || defined(__mips__) \
|| defined(CONFIG_TCG_INTERPRETER)
/* NOTE: Direct jump patching must be atomic to be thread-safe. */
#define USE_DIRECT_JUMP
#endif
struct TranslationBlock {
target_ulong pc; /* simulated PC corresponding to this block (EIP + CS base) */
target_ulong cs_base; /* CS base for this block */
uint32_t flags; /* flags defining in which context the code was generated */
uint16_t size; /* size of target code for this block (1 <=
size <= TARGET_PAGE_SIZE) */
uint16_t icount;
uint32_t cflags; /* compile flags */
#define CF_COUNT_MASK 0x7fff
#define CF_LAST_IO 0x8000 /* Last insn may be an IO access. */
#define CF_NOCACHE 0x10000 /* To be freed after execution */
#define CF_USE_ICOUNT 0x20000
#define CF_IGNORE_ICOUNT 0x40000 /* Do not generate icount code */
void *tc_ptr; /* pointer to the translated code */
uint8_t *tc_search; /* pointer to search data */
/* original tb when cflags has CF_NOCACHE */
struct TranslationBlock *orig_tb;
/* first and second physical page containing code. The lower bit
of the pointer tells the index in page_next[] */
struct TranslationBlock *page_next[2];
tb_page_addr_t page_addr[2];
/* The following data are used to directly call another TB from
* the code of this one. This can be done either by emitting direct or
* indirect native jump instructions. These jumps are reset so that the TB
* just continue its execution. The TB can be linked to another one by
* setting one of the jump targets (or patching the jump instruction). Only
* two of such jumps are supported.
*/
uint16_t jmp_reset_offset[2]; /* offset of original jump target */
#define TB_JMP_RESET_OFFSET_INVALID 0xffff /* indicates no jump generated */
#ifdef USE_DIRECT_JUMP
uint16_t jmp_insn_offset[2]; /* offset of native jump instruction */
#else
uintptr_t jmp_target_addr[2]; /* target address for indirect jump */
#endif
/* Each TB has an assosiated circular list of TBs jumping to this one.
* jmp_list_first points to the first TB jumping to this one.
* jmp_list_next is used to point to the next TB in a list.
* Since each TB can have two jumps, it can participate in two lists.
* jmp_list_first and jmp_list_next are 4-byte aligned pointers to a
* TranslationBlock structure, but the two least significant bits of
* them are used to encode which data field of the pointed TB should
* be used to traverse the list further from that TB:
* 0 => jmp_list_next[0], 1 => jmp_list_next[1], 2 => jmp_list_first.
* In other words, 0/1 tells which jump is used in the pointed TB,
* and 2 means that this is a pointer back to the target TB of this list.
*/
uintptr_t jmp_list_next[2];
uintptr_t jmp_list_first;
};
void tb_free(TranslationBlock *tb);
void tb_flush(CPUState *cpu);
void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr);
#if defined(USE_DIRECT_JUMP)
#if defined(CONFIG_TCG_INTERPRETER)
static inline void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr)
{
/* patch the branch destination */
atomic_set((int32_t *)jmp_addr, addr - (jmp_addr + 4));
/* no need to flush icache explicitly */
}
#elif defined(_ARCH_PPC)
void ppc_tb_set_jmp_target(uintptr_t jmp_addr, uintptr_t addr);
#define tb_set_jmp_target1 ppc_tb_set_jmp_target
#elif defined(__i386__) || defined(__x86_64__)
static inline void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr)
{
/* patch the branch destination */
atomic_set((int32_t *)jmp_addr, addr - (jmp_addr + 4));
/* no need to flush icache explicitly */
}
#elif defined(__s390x__)
static inline void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr)
{
/* patch the branch destination */
intptr_t disp = addr - (jmp_addr - 2);
atomic_set((int32_t *)jmp_addr, disp / 2);
/* no need to flush icache explicitly */
}
#elif defined(__aarch64__)
void aarch64_tb_set_jmp_target(uintptr_t jmp_addr, uintptr_t addr);
#define tb_set_jmp_target1 aarch64_tb_set_jmp_target
#elif defined(__arm__)
void arm_tb_set_jmp_target(uintptr_t jmp_addr, uintptr_t addr);
#define tb_set_jmp_target1 arm_tb_set_jmp_target
#elif defined(__sparc__) || defined(__mips__)
void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr);
#else
#error tb_set_jmp_target1 is missing
#endif
static inline void tb_set_jmp_target(TranslationBlock *tb,
int n, uintptr_t addr)
{
uint16_t offset = tb->jmp_insn_offset[n];
tb_set_jmp_target1((uintptr_t)(tb->tc_ptr + offset), addr);
}
#else
/* set the jump target */
static inline void tb_set_jmp_target(TranslationBlock *tb,
int n, uintptr_t addr)
{
tb->jmp_target_addr[n] = addr;
}
#endif
static inline void tb_add_jump(TranslationBlock *tb, int n,
TranslationBlock *tb_next)
{
if (tb->jmp_list_next[n]) {
/* Another thread has already done this while we were
* outside of the lock; nothing to do in this case */
return;
}
qemu_log_mask_and_addr(CPU_LOG_EXEC, tb->pc,
"Linking TBs %p [" TARGET_FMT_lx
"] index %d -> %p [" TARGET_FMT_lx "]\n",
tb->tc_ptr, tb->pc, n,
tb_next->tc_ptr, tb_next->pc);
/* patch the native jump address */
tb_set_jmp_target(tb, n, (uintptr_t)tb_next->tc_ptr);
/* add in TB jmp circular list */
tb->jmp_list_next[n] = tb_next->jmp_list_first;
tb_next->jmp_list_first = (uintptr_t)tb | n;
}
/* GETRA is the true target of the return instruction that we'll execute,
defined here for simplicity of defining the follow-up macros. */
#if defined(CONFIG_TCG_INTERPRETER)
extern uintptr_t tci_tb_ptr;
# define GETRA() tci_tb_ptr
#else
# define GETRA() \
((uintptr_t)__builtin_extract_return_addr(__builtin_return_address(0)))
#endif
/* The true return address will often point to a host insn that is part of
the next translated guest insn. Adjust the address backward to point to
the middle of the call insn. Subtracting one would do the job except for
several compressed mode architectures (arm, mips) which set the low bit
to indicate the compressed mode; subtracting two works around that. It
is also the case that there are no host isas that contain a call insn
smaller than 4 bytes, so we don't worry about special-casing this. */
#define GETPC_ADJ 2
#define GETPC() (GETRA() - GETPC_ADJ)
#if !defined(CONFIG_USER_ONLY)
struct MemoryRegion *iotlb_to_region(CPUState *cpu,
hwaddr index, MemTxAttrs attrs);
void tlb_fill(CPUState *cpu, target_ulong addr, int is_write, int mmu_idx,
uintptr_t retaddr);
#endif
#if defined(CONFIG_USER_ONLY)
void mmap_lock(void);
void mmap_unlock(void);
static inline tb_page_addr_t get_page_addr_code(CPUArchState *env1, target_ulong addr)
{
return addr;
}
#else
static inline void mmap_lock(void) {}
static inline void mmap_unlock(void) {}
/* cputlb.c */
tb_page_addr_t get_page_addr_code(CPUArchState *env1, target_ulong addr);
void tlb_reset_dirty(CPUState *cpu, ram_addr_t start1, ram_addr_t length);
void tlb_set_dirty(CPUState *cpu, target_ulong vaddr);
/* exec.c */
void tb_flush_jmp_cache(CPUState *cpu, target_ulong addr);
MemoryRegionSection *
address_space_translate_for_iotlb(CPUState *cpu, int asidx, hwaddr addr,
hwaddr *xlat, hwaddr *plen);
hwaddr memory_region_section_get_iotlb(CPUState *cpu,
MemoryRegionSection *section,
target_ulong vaddr,
hwaddr paddr, hwaddr xlat,
int prot,
target_ulong *address);
bool memory_region_is_unassigned(MemoryRegion *mr);
#endif
/* vl.c */
extern int singlestep;
/* cpu-exec.c, accessed with atomic_mb_read/atomic_mb_set */
extern CPUState *tcg_current_cpu;
extern bool exit_request;
#endif