Re: [Qemu-devel] [Qemu-arm] [PATCH v5 17/17] target/arm: Rely on hflags correct in cpu_get_tb_cpu_state

[Alex Bennée]

Richard Henderson <address@hidden> writes:

> This is the payoff.
>
> From perf record -g data of ubuntu 18 boot and shutdown:
>
> BEFORE:
>
> -   23.02%     2.82%  qemu-system-aar  [.] helper_lookup_tb_ptr
>    - 20.22% helper_lookup_tb_ptr
>       + 10.05% tb_htable_lookup
>       - 9.13% cpu_get_tb_cpu_state
>            3.20% aa64_va_parameters_both
>            0.55% fp_exception_el
>
> -   11.66%     4.74%  qemu-system-aar  [.] cpu_get_tb_cpu_state
>    - 6.96% cpu_get_tb_cpu_state
>         3.63% aa64_va_parameters_both
>         0.60% fp_exception_el
>         0.53% sve_exception_el
>
> AFTER:
>
> -   16.40%     3.40%  qemu-system-aar  [.] helper_lookup_tb_ptr
>    - 13.03% helper_lookup_tb_ptr
>       + 11.19% tb_htable_lookup
>         0.55% cpu_get_tb_cpu_state
>
>      0.98%     0.71%  qemu-system-aar  [.] cpu_get_tb_cpu_state
>
>      0.87%     0.24%  qemu-system-aar  [.] rebuild_hflags_a64
>
> Before, helper_lookup_tb_ptr is the second hottest function in the
> application, consuming almost a quarter of the runtime.  Within the
> entire execution, cpu_get_tb_cpu_state consumes about 12%.
>
> After, helper_lookup_tb_ptr has dropped to the fourth hottest function,
> with consumption dropping to a sixth of the runtime.  Within the
> entire execution, cpu_get_tb_cpu_state has dropped below 1%, and the
> supporting function to rebuild hflags also consumes about 1%.
>
> Assertions are retained for --enable-debug-tcg.
>
> Tested-by: Alex Bennée <address@hidden>

Hmm something must have been missed for M-profile because:

  make run-tcg-tests-arm-softmmu V=1

Leads to:

  timeout 15  
/home/alex/lsrc/qemu.git/builds/all.debug/arm-softmmu/qemu-system-arm -monitor 
none -display none -chardev file,path=test-armv6m-undef.out,id=output 
-semihosting -M microbit -kernel test-armv6m-undef
  qemu: fatal: Lockup: can't escalate 3 to HardFault (current priority -1)

  R00=00000000 R01=00000000 R02=00000000 R03=00000000
  R04=00000000 R05=00000000 R06=00000000 R07=00000000
  R08=00000000 R09=00000000 R10=00000000 R11=00000000
  R12=00000000 R13=20003fe0 R14=fffffff9 R15=000000c0
  XPSR=41000003 -Z-- T handler
  FPSCR: 00000000
  timeout: the monitored command dumped core

But annoyingly not shown up by the debug-tcg verification. The commit
before works fine.

> Reviewed-by: Alex Bennée <address@hidden>
> Signed-off-by: Richard Henderson <address@hidden>
> ---
> v2: Retain asserts for future debugging.
> ---
>  target/arm/helper.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/target/arm/helper.c b/target/arm/helper.c
> index d1bf71a260..5e4f996882 100644
> --- a/target/arm/helper.c
> +++ b/target/arm/helper.c
> @@ -11211,12 +11211,15 @@ void HELPER(rebuild_hflags_a64)(CPUARMState *env, 
> int el)
>  void cpu_get_tb_cpu_state(CPUARMState *env, target_ulong *pc,
>                            target_ulong *cs_base, uint32_t *pflags)
>  {
> -    uint32_t flags, pstate_for_ss;
> +    uint32_t flags = env->hflags;
> +    uint32_t pstate_for_ss;
>
>      *cs_base = 0;
> -    flags = rebuild_hflags_internal(env);
> +#ifdef CONFIG_TCG_DEBUG
> +    assert(flags == rebuild_hflags_internal(env));
> +#endif
>
> -    if (is_a64(env)) {
> +    if (FIELD_EX32(flags, TBFLAG_ANY, AARCH64_STATE)) {
>          *pc = env->pc;
>          if (cpu_isar_feature(aa64_bti, env_archcpu(env))) {
>              flags = FIELD_DP32(flags, TBFLAG_A64, BTYPE, env->btype);


--
Alex Bennée