Re: [PATCH V3 01/23] perf/x86: Support outputting XMM registers

[Peter Zijlstra]

On Fri, Mar 22, 2019 at 09:36:56AM -0700, kan.liang@xxxxxxxxxxxxxxx wrote:
> From: Kan Liang <kan.liang@xxxxxxxxxxxxxxx>
> 
> Starting from Icelake, XMM registers can be collected in PEBS record.
> But current code only output the pt_regs.
> 
> Add a new struct x86_perf_regs for both pt_regs and xmm_regs.
> XMM registers are 128 bit. To simplify the code, they are handled like
> two different registers, which means setting two bits in the register
> bitmap. This also allows only sampling the lower 64bit bits in XMM.
> The index of XMM registers starts from 32. There are 16 XMM registers.
> So all reserved space for regs are used.
> PERF_REG_X86_MAX stands for the max number of all x86 regs include XMM.
> PERF_REG_GPR_X86_MAX stands for the max number of all x86 general
> purpose registers, which not include XMM.
> PERF_REG_GPR_X86_32_MAX and PERF_REG_GPR_X86_64_MAX are introduced to
> replace PERF_REG_X86_32_MAX and PERF_REG_X86_64_MAX for x86 general
> purpose registers.
> 
> The REG_RESERVED is also updated to allow the XMM registers.
> XMM is not supported on all platforms. Adding has_xmm_regs to indicate
> the specific platform. Also add checks in x86_pmu_hw_config() to reject
> invalid config of regs_user and regs_intr.

> diff --git a/arch/x86/include/uapi/asm/perf_regs.h b/arch/x86/include/uapi/asm/perf_regs.h
> index f3329cabce5c..b33995313d17 100644
> --- a/arch/x86/include/uapi/asm/perf_regs.h
> +++ b/arch/x86/include/uapi/asm/perf_regs.h
> @@ -28,7 +28,29 @@ enum perf_event_x86_regs {
>  	PERF_REG_X86_R14,
>  	PERF_REG_X86_R15,
>  
> -	PERF_REG_X86_32_MAX = PERF_REG_X86_GS + 1,
> -	PERF_REG_X86_64_MAX = PERF_REG_X86_R15 + 1,

So this changes UAPI visible symbols... did we think about that?

> +	/* These all need two bits set because they are 128bit */
> +	PERF_REG_X86_XMM0  = 32,
> +	PERF_REG_X86_XMM1  = 34,
> +	PERF_REG_X86_XMM2  = 36,
> +	PERF_REG_X86_XMM3  = 38,
> +	PERF_REG_X86_XMM4  = 40,
> +	PERF_REG_X86_XMM5  = 42,
> +	PERF_REG_X86_XMM6  = 44,
> +	PERF_REG_X86_XMM7  = 46,
> +	PERF_REG_X86_XMM8  = 48,
> +	PERF_REG_X86_XMM9  = 50,
> +	PERF_REG_X86_XMM10 = 52,
> +	PERF_REG_X86_XMM11 = 54,
> +	PERF_REG_X86_XMM12 = 56,
> +	PERF_REG_X86_XMM13 = 58,
> +	PERF_REG_X86_XMM14 = 60,
> +	PERF_REG_X86_XMM15 = 62,
> +
> +	/* This does not include the XMMX registers */
> +	PERF_REG_GPR_X86_32_MAX = PERF_REG_X86_GS + 1,
> +	PERF_REG_GPR_X86_64_MAX = PERF_REG_X86_R15 + 1,
> +
> +	/* All registers include the XMMX registers */
> +	PERF_REG_X86_MAX = PERF_REG_X86_XMM15 + 2,
>  };
>  #endif /* _ASM_X86_PERF_REGS_H */

Also, what happens if we run a 32bit kernel or 32bit compat task?

Then the register dump will report PERF_SAMPLE_REGS_ABI_32, should we
then still interpret the XMM registers as 2x64bit?

Are they still at the same offset?

Do we need additional PERF_SAMPLE_REGS_ABI_* definitions for this?