[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH 2/8] target/arm: Optimize cpu_mmu_index
|
From: |
Richard Henderson |
|
Subject: |
Re: [PATCH 2/8] target/arm: Optimize cpu_mmu_index |
|
Date: |
Mon, 2 Mar 2020 08:24:05 -0800 |
|
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.4.1 |
On 3/2/20 4:03 AM, Peter Maydell wrote:
> On Tue, 25 Feb 2020 at 03:14, Richard Henderson
> <address@hidden> wrote:
>>
>> We now cache the core mmu_idx in env->hflags. Rather than recompute
>> from scratch, extract the field. All of the uses of cpu_mmu_index
>> within target/arm are within helpers where env->hflags is stable.
>
> Do you mean "within helpers, and env->hflags is always stable in
> a helper", or "within helpers, and env->hflags is stable for the
> particular set of helpers where we use cpu_mmu_index, though it might
> not be in other helpers" ?
The former.
With the caveat that it's pretty clear when a helper is doing things that make
it the exception to that rule. E.g. helper_exception_return, which itself
invokes rebuild_hflags.
r~