From d0af9eed5aa91b6b7b5049cae69e5ea956fd85c3 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 19 Aug 2009 18:05:36 -0700 Subject: x86, pat/mtrr: Rendezvous all the cpus for MTRR/PAT init SDM Vol 3a section titled "MTRR considerations in MP systems" specifies the need for synchronizing the logical cpu's while initializing/updating MTRR. Currently Linux kernel does the synchronization of all cpu's only when a single MTRR register is programmed/updated. During an AP online (during boot/cpu-online/resume) where we initialize all the MTRR/PAT registers, we don't follow this synchronization algorithm. This can lead to scenarios where during a dynamic cpu online, that logical cpu is initializing MTRR/PAT with cache disabled (cr0.cd=1) etc while other logical HT sibling continue to run (also with cache disabled because of cr0.cd=1 on its sibling). Starting from Westmere, VMX transitions with cr0.cd=1 don't work properly (because of some VMX performance optimizations) and the above scenario (with one logical cpu doing VMX activity and another logical cpu coming online) can result in system crash. Fix the MTRR initialization by doing rendezvous of all the cpus. During boot and resume, we delay the MTRR/PAT init for APs till all the logical cpu's come online and the rendezvous process at the end of AP's bringup, will initialize the MTRR/PAT for all AP's. For dynamic single cpu online, we synchronize all the logical cpus and do the MTRR/PAT init on the AP that is coming online. Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/mtrr.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h index a51ada8467d..d5366ec5cb8 100644 --- a/arch/x86/include/asm/mtrr.h +++ b/arch/x86/include/asm/mtrr.h @@ -121,8 +121,12 @@ extern int mtrr_del_page(int reg, unsigned long base, unsigned long size); extern void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi); extern void mtrr_ap_init(void); extern void mtrr_bp_init(void); +extern void set_mtrr_aps_delayed_init(void); +extern void mtrr_aps_init(void); +extern void mtrr_bp_restore(void); extern int mtrr_trim_uncached_memory(unsigned long end_pfn); extern int amd_special_default_mtrr(void); +extern u32 mtrr_aps_delayed_init; # else static inline u8 mtrr_type_lookup(u64 addr, u64 end) { @@ -161,6 +165,9 @@ static inline void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi) #define mtrr_ap_init() do {} while (0) #define mtrr_bp_init() do {} while (0) +#define set_mtrr_aps_delayed_init() do {} while (0) +#define mtrr_aps_init() do {} while (0) +#define mtrr_bp_restore() do {} while (0) # endif #ifdef CONFIG_COMPAT -- cgit v1.2.3 From 5400743db5a06a4e6e298725a2044c40edcb27b9 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 21 Aug 2009 17:00:02 -0700 Subject: x86, mtrr: make mtrr_aps_delayed_init static bool mtr_aps_delayed_init was declared u32 and made global, but it only ever takes boolean values and is only ever used in arch/x86/kernel/cpu/mtrr/main.c. Declare it "static bool" and remove external references. Signed-off-by: H. Peter Anvin Cc: Suresh Siddha --- arch/x86/include/asm/mtrr.h | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h index d5366ec5cb8..4365ffdb461 100644 --- a/arch/x86/include/asm/mtrr.h +++ b/arch/x86/include/asm/mtrr.h @@ -126,7 +126,6 @@ extern void mtrr_aps_init(void); extern void mtrr_bp_restore(void); extern int mtrr_trim_uncached_memory(unsigned long end_pfn); extern int amd_special_default_mtrr(void); -extern u32 mtrr_aps_delayed_init; # else static inline u8 mtrr_type_lookup(u64 addr, u64 end) { -- cgit v1.2.3 From 9fd126bc742f74a95d2ba610247712ff05da02fe Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Fri, 10 Jul 2009 09:57:34 -0700 Subject: x86, pat: New i/f for driver to request memtype for IO regions Add new routines to request memtype for IO regions. This will currently be a backend for io_mapping_* routines. But, it can also be made available to drivers directly in future, in case it is needed. reserve interface reserves the memory, makes sure we have a compatible memory type available and keeps the identity map in sync when needed. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pat.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/pat.h b/arch/x86/include/asm/pat.h index 7af14e512f9..e2c1668dde7 100644 --- a/arch/x86/include/asm/pat.h +++ b/arch/x86/include/asm/pat.h @@ -19,4 +19,9 @@ extern int free_memtype(u64 start, u64 end); extern int kernel_map_sync_memtype(u64 base, unsigned long size, unsigned long flag); +int io_reserve_memtype(resource_size_t start, resource_size_t end, + unsigned long *type); + +void io_free_memtype(resource_size_t start, resource_size_t end); + #endif /* _ASM_X86_PAT_H */ -- cgit v1.2.3 From 9e36fda0b359d2a6ae039c3d7e71a04502a77898 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Fri, 10 Jul 2009 09:57:35 -0700 Subject: x86, pat: Add PAT reserve free to io_mapping* APIs io_mapping_* interfaces were added, mainly for graphics drivers. Make this interface go through the PAT reserve/free, instead of hardcoding WC mapping. This makes sure that there are no aliases due to unconditional WC setting. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/iomap.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/iomap.h b/arch/x86/include/asm/iomap.h index 0e9fe1d9d97..f35eb45d657 100644 --- a/arch/x86/include/asm/iomap.h +++ b/arch/x86/include/asm/iomap.h @@ -26,13 +26,16 @@ #include #include -int -is_io_mapping_possible(resource_size_t base, unsigned long size); - void * iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); void iounmap_atomic(void *kvaddr, enum km_type type); +int +iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot); + +void +iomap_free(resource_size_t base, unsigned long size); + #endif /* _ASM_X86_IOMAP_H */ -- cgit v1.2.3 From f58417409603d62f2eb23db4d2cf6853d84a1698 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Fri, 10 Jul 2009 09:57:38 -0700 Subject: x86, pat: Use page flags to track memtypes of RAM pages Change reserve_ram_pages_type and free_ram_pages_type to use 2 page flags to track UC_MINUS, WC, WB and default types. Previous RAM tracking just tracked WB or NonWB, which was not complete and did not allow tracking of RAM fully and there was no way to get the actual type reserved by looking at the page flags. We use the memtype_lock spinlock for atomicity in dealing with memtype tracking in struct page. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/cacheflush.h | 54 +++++++++++++++++++++++++++++++++++++-- 1 file changed, 52 insertions(+), 2 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/cacheflush.h b/arch/x86/include/asm/cacheflush.h index e55dfc1ad45..b54f6afe7ec 100644 --- a/arch/x86/include/asm/cacheflush.h +++ b/arch/x86/include/asm/cacheflush.h @@ -43,8 +43,58 @@ static inline void copy_from_user_page(struct vm_area_struct *vma, memcpy(dst, src, len); } -#define PG_non_WB PG_arch_1 -PAGEFLAG(NonWB, non_WB) +#define PG_WC PG_arch_1 +PAGEFLAG(WC, WC) + +#ifdef CONFIG_X86_PAT +/* + * X86 PAT uses page flags WC and Uncached together to keep track of + * memory type of pages that have backing page struct. X86 PAT supports 3 + * different memory types, _PAGE_CACHE_WB, _PAGE_CACHE_WC and + * _PAGE_CACHE_UC_MINUS and fourth state where page's memory type has not + * been changed from its default (value of -1 used to denote this). + * Note we do not support _PAGE_CACHE_UC here. + * + * Caller must hold memtype_lock for atomicity. + */ +static inline unsigned long get_page_memtype(struct page *pg) +{ + if (!PageUncached(pg) && !PageWC(pg)) + return -1; + else if (!PageUncached(pg) && PageWC(pg)) + return _PAGE_CACHE_WC; + else if (PageUncached(pg) && !PageWC(pg)) + return _PAGE_CACHE_UC_MINUS; + else + return _PAGE_CACHE_WB; +} + +static inline void set_page_memtype(struct page *pg, unsigned long memtype) +{ + switch (memtype) { + case _PAGE_CACHE_WC: + ClearPageUncached(pg); + SetPageWC(pg); + break; + case _PAGE_CACHE_UC_MINUS: + SetPageUncached(pg); + ClearPageWC(pg); + break; + case _PAGE_CACHE_WB: + SetPageUncached(pg); + SetPageWC(pg); + break; + default: + case -1: + ClearPageUncached(pg); + ClearPageWC(pg); + break; + } +} +#else +static inline unsigned long get_page_memtype(struct page *pg) { return -1; } +static inline void set_page_memtype(struct page *pg, unsigned long memtype) { } +#endif /* * The set_memory_* API can be used to change various attributes of a virtual -- cgit v1.2.3