Re: [PATCH V6 01/18] x86: Make page cache mode a real type

[Steven Noonan]

On Mon, Nov 3, 2014 at 5:01 AM, Juergen Gross <jgross@xxxxxxxx> wrote:
> At the moment there are a lot of places that handle setting or getting
> the page cache mode by treating the pgprot bits equal to the cache mode.
> This is only true because there are a lot of assumptions about the setup
> of the PAT MSR. Otherwise the cache type needs to get translated into
> pgprot bits and vice versa.
>
> This patch tries to prepare for that by introducing a separate type
> for the cache mode and adding functions to translate between those and
> pgprot values.
>
> To avoid too much performance penalty the translation between cache mode
> and pgprot values is done via tables which contain the relevant
> information.  Write-back cache mode is hard-wired to be 0, all other
> modes are configurable via those tables. For large pages there are
> translation functions as the PAT bit is located at different positions
> in the ptes of 4k and large pages.
>
> Based-on-patch-by: Stefan Bader <stefan.bader@xxxxxxxxxxxxx>
> Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
> Reviewed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> ---
>  arch/x86/include/asm/pgtable_types.h | 73 +++++++++++++++++++++++++++++++++++-
>  arch/x86/mm/init.c                   | 29 ++++++++++++++
>  2 files changed, 101 insertions(+), 1 deletion(-)
>
> diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
> index 0778964..5124642 100644
> --- a/arch/x86/include/asm/pgtable_types.h
> +++ b/arch/x86/include/asm/pgtable_types.h
> @@ -128,12 +128,34 @@
>                          _PAGE_SOFT_DIRTY | _PAGE_NUMA)
>  #define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE | _PAGE_NUMA)
>
> -#define _PAGE_CACHE_MASK       (_PAGE_PCD | _PAGE_PWT)
>  #define _PAGE_CACHE_WB         (0)
>  #define _PAGE_CACHE_WC         (_PAGE_PWT)
>  #define _PAGE_CACHE_UC_MINUS   (_PAGE_PCD)
>  #define _PAGE_CACHE_UC         (_PAGE_PCD | _PAGE_PWT)
>
> +/*
> + * The cache modes defined here are used to translate between pure SW usage
> + * and the HW defined cache mode bits and/or PAT entries.
> + *
> + * The resulting bits for PWT, PCD and PAT should be chosen in a way
> + * to have the WB mode at index 0 (all bits clear). This is the default
> + * right now and likely would break too much if changed.
> + */
> +#ifndef __ASSEMBLY__
> +enum page_cache_mode {
> +       _PAGE_CACHE_MODE_WB = 0,
> +       _PAGE_CACHE_MODE_WC = 1,
> +       _PAGE_CACHE_MODE_UC_MINUS = 2,
> +       _PAGE_CACHE_MODE_UC = 3,
> +       _PAGE_CACHE_MODE_WT = 4,
> +       _PAGE_CACHE_MODE_WP = 5,
> +       _PAGE_CACHE_MODE_NUM = 8
> +};
> +#endif
> +
> +#define _PAGE_CACHE_MASK       (_PAGE_PAT | _PAGE_PCD | _PAGE_PWT)
> +#define _PAGE_NOCACHE          (cachemode2protval(_PAGE_CACHE_MODE_UC))
> +
>  #define PAGE_NONE      __pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
>  #define PAGE_SHARED    __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | \
>                                  _PAGE_ACCESSED | _PAGE_NX)
> @@ -341,6 +363,55 @@ static inline pmdval_t pmdnuma_flags(pmd_t pmd)
>  #define pgprot_val(x)  ((x).pgprot)
>  #define __pgprot(x)    ((pgprot_t) { (x) } )
>
> +extern uint16_t __cachemode2pte_tbl[_PAGE_CACHE_MODE_NUM];
> +extern uint8_t __pte2cachemode_tbl[8];
> +
> +#define __pte2cm_idx(cb)                               \
> +       ((((cb) >> (_PAGE_BIT_PAT - 2)) & 4) |          \
> +        (((cb) >> (_PAGE_BIT_PCD - 1)) & 2) |          \
> +        (((cb) >> _PAGE_BIT_PWT) & 1))
> +
> +static inline unsigned long cachemode2protval(enum page_cache_mode pcm)
> +{
> +       if (likely(pcm == 0))
> +               return 0;
> +       return __cachemode2pte_tbl[pcm];
> +}
> +static inline pgprot_t cachemode2pgprot(enum page_cache_mode pcm)
> +{
> +       return __pgprot(cachemode2protval(pcm));
> +}
> +static inline enum page_cache_mode pgprot2cachemode(pgprot_t pgprot)
> +{
> +       unsigned long masked;
> +
> +       masked = pgprot_val(pgprot) & _PAGE_CACHE_MASK;
> +       if (likely(masked == 0))
> +               return 0;
> +       return __pte2cachemode_tbl[__pte2cm_idx(masked)];
> +}
> +static inline pgprot_t pgprot_4k_2_large(pgprot_t pgprot)
> +{
> +       pgprot_t new;
> +       unsigned long val;
> +
> +       val = pgprot_val(pgprot);
> +       pgprot_val(new) = (val & ~(_PAGE_PAT | _PAGE_PAT_LARGE)) |
> +               ((val & _PAGE_PAT) << (_PAGE_BIT_PAT_LARGE - _PAGE_BIT_PAT));
> +       return new;
> +}
> +static inline pgprot_t pgprot_large_2_4k(pgprot_t pgprot)
> +{
> +       pgprot_t new;
> +       unsigned long val;
> +
> +       val = pgprot_val(pgprot);
> +       pgprot_val(new) = (val & ~(_PAGE_PAT | _PAGE_PAT_LARGE)) |
> +                         ((val & _PAGE_PAT_LARGE) >>
> +                          (_PAGE_BIT_PAT_LARGE - _PAGE_BIT_PAT));
> +       return new;
> +}
> +
>
>  typedef struct page *pgtable_t;
>
> diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
> index 66dba36..a9776ba 100644
> --- a/arch/x86/mm/init.c
> +++ b/arch/x86/mm/init.c
> @@ -27,6 +27,35 @@
>
>  #include "mm_internal.h"
>
> +/*
> + * Tables translating between page_cache_type_t and pte encoding.
> + * Minimal supported modes are defined statically, modified if more supported
> + * cache modes are available.
> + * Index into __cachemode2pte_tbl is the cachemode.
> + * Index into __pte2cachemode_tbl are the caching attribute bits of the pte
> + * (_PAGE_PWT, _PAGE_PCD, _PAGE_PAT) at index bit positions 0, 1, 2.
> + */
> +uint16_t __cachemode2pte_tbl[_PAGE_CACHE_MODE_NUM] = {
> +       [_PAGE_CACHE_MODE_WB]           = 0,
> +       [_PAGE_CACHE_MODE_WC]           = _PAGE_PWT,
> +       [_PAGE_CACHE_MODE_UC_MINUS]     = _PAGE_PCD,
> +       [_PAGE_CACHE_MODE_UC]           = _PAGE_PCD | _PAGE_PWT,
> +       [_PAGE_CACHE_MODE_WT]           = _PAGE_PCD,
> +       [_PAGE_CACHE_MODE_WP]           = _PAGE_PCD,
> +};
> +EXPORT_SYMBOL_GPL(__cachemode2pte_tbl);
> +uint8_t __pte2cachemode_tbl[8] = {
> +       [__pte2cm_idx(0)] = _PAGE_CACHE_MODE_WB,
> +       [__pte2cm_idx(_PAGE_PWT)] = _PAGE_CACHE_MODE_WC,
> +       [__pte2cm_idx(_PAGE_PCD)] = _PAGE_CACHE_MODE_UC_MINUS,
> +       [__pte2cm_idx(_PAGE_PWT | _PAGE_PCD)] = _PAGE_CACHE_MODE_UC,
> +       [__pte2cm_idx(_PAGE_PAT)] = _PAGE_CACHE_MODE_WB,
> +       [__pte2cm_idx(_PAGE_PWT | _PAGE_PAT)] = _PAGE_CACHE_MODE_WC,
> +       [__pte2cm_idx(_PAGE_PCD | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC_MINUS,
> +       [__pte2cm_idx(_PAGE_PWT | _PAGE_PCD | _PAGE_PAT)] = _PAGE_CACHE_MODE_UC,
> +};
> +EXPORT_SYMBOL_GPL(__pte2cachemode_tbl);
> +

I notice these two symbols are exported GPL-only. This breaks builds
of several out-of-tree non-GPL modules such as the NVIDIA driver, and
VMware modules, etc. What is the appropriate code path for proprietary
modules to use when setting page cache mode flags? Alternatively, is
it possible for these EXPORT_SYMBOL_GPLs to be changed to
EXPORT_SYMBOL?

>  static unsigned long __initdata pgt_buf_start;
>  static unsigned long __initdata pgt_buf_end;
>  static unsigned long __initdata pgt_buf_top;
> --
> 1.8.4.5
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/