Re: [PATCH 08/11] x86, fpu: add C structures for AVX-512 state components

[Ingo Molnar <mingo@kernel.org>]

* Dave Hansen <dave@sr71.net> wrote:

> 
> From: Dave Hansen <dave.hansen@linux.intel.com>
> 
> AVX-512 has 3 separate state components:
> 1. opmask registers
> 2. zmm upper half of registers 0-15
> 3. new zmm registers (16-31)
> 
> This patch adds C structures for the three components along with
> a few comments mostly lifted from the SDM to explain what they
> do.  This will allow us to check our structures against what the
> hardware tells us about the sizes of the components.
> 
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: x86@kernel.org
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: Fenghua Yu <fenghua.yu@intel.com>
> Cc: Tim Chen <tim.c.chen@linux.intel.com>
> Cc: linux-kernel@vger.kernel.org
> ---
> 
>  b/arch/x86/include/asm/fpu/types.h |   43 ++++++++++++++++++++++++++++++++++---
>  1 file changed, 40 insertions(+), 3 deletions(-)
> 
> diff -puN arch/x86/include/asm/fpu/types.h~avx-512-structs arch/x86/include/asm/fpu/types.h
> --- a/arch/x86/include/asm/fpu/types.h~avx-512-structs	2015-08-27 10:08:03.909740783 -0700
> +++ b/arch/x86/include/asm/fpu/types.h	2015-08-27 10:08:03.912740919 -0700
> @@ -129,6 +129,12 @@ enum xfeature_nr {
>  struct reg_128_bit {
>  	u8      regbytes[128/8];
>  };
> +struct reg_256_bit {
> +	u8	regbytes[256/8];
> +};
> +struct reg_512_bit {
> +	u8	regbytes[512/8];
> +};
>  
>  /*
>   * State component 2:
> @@ -177,6 +183,33 @@ struct mpx_bndcsr_state {
>  	};
>  } __packed;
>  
> +/* AVX-512 Components: */
> +
> +/*
> + * State component 5 is used for the 8 64-bit opmask registers
> + * k0–k7 (opmask state).
> + */
> +struct avx_512_opmask_state {
> +	u64 				opmask_reg[8];
> +} __packed;
> +
> +/*
> + * State component 6 is used for the upper 256 bits of the
> + * registers ZMM0–ZMM15. These 16 256-bit values are denoted
> + * ZMM0_H–ZMM15_H (ZMM_Hi256 state).
> + */
> +struct avx_512_zmm_uppers_state {
> +	struct reg_256_bit		zmm_upper[16];
> +} __packed;
> +
> +/*
> + * State component 7 is used for the 16 512-bit registers
> + * ZMM16–ZMM31 (Hi16_ZMM state).
> + */
> +struct avx_512_hi16_state {
> +	struct reg_512_bit		hi16_zmm[16];
> +} __packed;
> +
>  struct xstate_header {
>  	u64				xfeatures;
>  	u64				xcomp_bv;
> @@ -184,9 +217,13 @@ struct xstate_header {
>  } __attribute__((packed));
>  
>  /* New processor state extensions should be added here: */
> -#define XSTATE_RESERVE			(sizeof(struct ymmh_struct) + \
> -					 sizeof(struct mpx_bndreg_state) + \
> -					 sizeof(struct mpx_bndcsr_state)  )
> +#define XSTATE_RESERVE		(sizeof(struct ymmh_struct) 		+ \
> +				 sizeof(struct mpx_bndreg_state) 	+ \
> +				 sizeof(struct mpx_bndcsr_state) 	+ \
> +				 sizeof(struct avx_512_opmask_state) 	+ \
> +				 sizeof(struct avx_512_zmm_uppers_state) + \
> +				 sizeof(struct avx_512_hi16_state))

So in a previous mail I suggested getting rid of XSTATE_RESERVE, which removes the 
usage of the C structures..

What we could use these C structures for is to double check that the xstate size 
given by CPUID for that particular xstate feature is equal to the C structure size 
- or if it's larger, it's a multiple of the natural cache line size, or so?

Any 'failure' in such a check should be print-once and non-fatal, as in 90% of the 
cases I'd expect the kernel assumptions/checks to be buggy, not the hardware 
itself.

We should perhaps also print a gentle (non-warning) kernel message if we find an 
xfeature that the kernel doesn't know about, with its essential parameters: the 
bit number, the size and the offset position within the xstate buffer.

Thanks,

	Ingo