[PATCH V9 00/10] arm64/perf: Enable branch stack sampling

[Anshuman Khandual]

On 3/23/23 18:24, Mark Brown wrote:
> On Thu, Mar 23, 2023 at 09:55:47AM +0530, Anshuman Khandual wrote:
>> On 3/22/23 00:32, Mark Brown wrote:
> 
>>> document a requirement for those traps to be disabled now in case we
>>> need them later, and do so during EL2 setup for KVM guests?  That could
>>> always be done incrementally.
> 
>> Unlike all other instruction trap enable fields in SYS_HFGITR_EL2, these BRBE
>> instructions ones are actually inverted in semantics i.e the particular fields
>> need to be set for these traps to be disabled in EL2.
> 
> Right, for backwards compatibility all newly added fields are trap by
> default.

Okay

> 
>> SYS_HFGITR_EL2.nBRBIALL
>> SYS_HFGITR_EL2.nBRBINJ
> 
>> By default entire SYS_HFGITR_EL2 is set as cleared during init and that would
>> prevent a guest from using BRBE.
> 
> It should prevent the host as well shouldn't it? 

In a EL2 host environment, BRBE is being enabled either in EL2 (kernel/hv) or
in EL0 (user space), it never gets enabled on EL1. Moreover BRBIALL/BRBINJ
instructions are always executed while being inside EL2 (kernel/hv). Hence how
could these instructions cause trap in EL2 ?

> 
>> I guess something like the following (untested) needs to be done, to enable
>> BRBE in guests.
> 
>> +       mrs     x1, id_aa64dfr0_el1
>> +       ubfx    x1, x1, #ID_AA64DFR0_EL1_BRBE_SHIFT, #4
>> +       cbz     x1, .Lskip_brbe_\@
>> +       mov     x0, xzr
>> +       orr     x0, x0, #HFGITR_EL2_nBRBIALL
>> +       orr     x0, x0, #HFGITR_EL2_nBRBINJ
>> +       msr_s   SYS_HFGITR_EL2, x0
>> +
>> +.Lskip_brbe_\@:
> 
> Yes, looks roughly what I'd expect.

I could send an stand alone patch after your latest series [1], which disables
BRBINJ/BRBIALL instruction trap in EL2 to enable BRBE usage in the guest.

https://lore.kernel.org/all/20230306-arm64-fgt-reg-gen-v3-2-decba93cbaab@kernel.org/T/

> 
>>> I've got a patch adding the definition of that register to sysreg which
>>> I should be sending shortly, no need to duplicate that effort.
> 
>> Sure, I assume you are moving the existing definition for SYS_HFGITR_EL2 along
>> with all its fields from ../include/asm/sysreg.h to ../tools/sysreg. Right, it
>> makes sense.
> 
> No fields at the minute but yes, like the other conversions.

Sure.