From de067814d6b69030d0030e1c5b3dbaf0385aae41 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 15 May 2008 13:24:52 +0100 Subject: x86/xen: fix arbitrary_virt_to_machine() While I realize that the function isn't currently being used, I still think an obvious mistake like this should be corrected. Signed-off-by: Jan Beulich Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- arch/x86/xen/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 126766d43aea..3525ef523a74 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -60,7 +60,7 @@ xmaddr_t arbitrary_virt_to_machine(unsigned long address) { unsigned int level; pte_t *pte = lookup_address(address, &level); - unsigned offset = address & PAGE_MASK; + unsigned offset = address & ~PAGE_MASK; BUG_ON(pte == NULL); -- cgit v1.2.3 From f595ec964daf7f99668039d7303ddedd09a75142 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 12 Jun 2008 10:47:56 +0200 Subject: common implementation of iterative div/mod We have a few instances of the open-coded iterative div/mod loop, used when we don't expcet the dividend to be much bigger than the divisor. Unfortunately modern gcc's have the tendency to strength "reduce" this into a full mod operation, which isn't necessarily any faster, and even if it were, doesn't exist if gcc implements it in libgcc. The workaround is to put a dummy asm statement in the loop to prevent gcc from performing the transformation. This patch creates a single implementation of this loop, and uses it to replace the open-coded versions I know about. Signed-off-by: Jeremy Fitzhardinge Cc: Andrew Morton Cc: john stultz Cc: Segher Boessenkool Cc: Christian Kujau Cc: Robert Hancock Signed-off-by: Ingo Molnar --- arch/x86/xen/time.c | 13 +++---------- include/linux/math64.h | 2 ++ include/linux/time.h | 11 ++--------- lib/div64.c | 23 +++++++++++++++++++++++ 4 files changed, 30 insertions(+), 19 deletions(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index c39e1a5aa241..52b2e3856980 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -150,11 +151,7 @@ static void do_stolen_accounting(void) if (stolen < 0) stolen = 0; - ticks = 0; - while (stolen >= NS_PER_TICK) { - ticks++; - stolen -= NS_PER_TICK; - } + ticks = iter_div_u64_rem(stolen, NS_PER_TICK, &stolen); __get_cpu_var(residual_stolen) = stolen; account_steal_time(NULL, ticks); @@ -166,11 +163,7 @@ static void do_stolen_accounting(void) if (blocked < 0) blocked = 0; - ticks = 0; - while (blocked >= NS_PER_TICK) { - ticks++; - blocked -= NS_PER_TICK; - } + ticks = iter_div_u64_rem(blocked, NS_PER_TICK, &blocked); __get_cpu_var(residual_blocked) = blocked; account_steal_time(idle_task(smp_processor_id()), ticks); } diff --git a/include/linux/math64.h b/include/linux/math64.h index c1a5f81501ff..177785e1e4a3 100644 --- a/include/linux/math64.h +++ b/include/linux/math64.h @@ -81,4 +81,6 @@ static inline s64 div_s64(s64 dividend, s32 divisor) } #endif +u32 iter_div_u64_rem(u64 dividend, u32 divisor, u64 *remainder); + #endif /* _LINUX_MATH64_H */ diff --git a/include/linux/time.h b/include/linux/time.h index d32ef0ad4c0a..05f9517a8ed1 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -6,6 +6,7 @@ #ifdef __KERNEL__ # include # include +# include #endif #ifndef _STRUCT_TIMESPEC @@ -172,15 +173,7 @@ extern struct timeval ns_to_timeval(const s64 nsec); */ static inline void timespec_add_ns(struct timespec *a, u64 ns) { - ns += a->tv_nsec; - while(unlikely(ns >= NSEC_PER_SEC)) { - /* The following asm() prevents the compiler from - * optimising this loop into a modulo operation. */ - asm("" : "+r"(ns)); - - ns -= NSEC_PER_SEC; - a->tv_sec++; - } + a->tv_sec += iter_div_u64_rem(a->tv_nsec + ns, NSEC_PER_SEC, &ns); a->tv_nsec = ns; } #endif /* __KERNEL__ */ diff --git a/lib/div64.c b/lib/div64.c index bb5bd0c0f030..76c01542d3e1 100644 --- a/lib/div64.c +++ b/lib/div64.c @@ -98,3 +98,26 @@ EXPORT_SYMBOL(div64_u64); #endif #endif /* BITS_PER_LONG == 32 */ + +/* + * Iterative div/mod for use when dividend is not expected to be much + * bigger than divisor. + */ +u32 iter_div_u64_rem(u64 dividend, u32 divisor, u64 *remainder) +{ + u32 ret = 0; + + while (dividend >= divisor) { + /* The following asm() prevents the compiler from + optimising this loop into a modulo operation. */ + asm("" : "+rm"(dividend)); + + dividend -= divisor; + ret++; + } + + *remainder = dividend; + + return ret; +} +EXPORT_SYMBOL(iter_div_u64_rem); -- cgit v1.2.3 From 05345b0f006ac226d0d25d48fcb2d792ac44a071 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 16 Jun 2008 15:01:53 -0700 Subject: xen: mask unwanted pte bits in __supported_pte_mask [ Stable: this isn't a bugfix in itself, but it's a pre-requiste for "xen: don't drop NX bit" ] Signed-off-by: Jeremy Fitzhardinge Cc: Stable Kernel Cc: the arch/x86 maintainers Signed-off-by: Ingo Molnar --- arch/x86/xen/enlighten.c | 5 +++++ arch/x86/xen/mmu.c | 4 +--- 2 files changed, 6 insertions(+), 3 deletions(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index c8a56e457d61..c048de34d6a1 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1228,6 +1228,11 @@ asmlinkage void __init xen_start_kernel(void) if (xen_feature(XENFEAT_supervisor_mode_kernel)) pv_info.kernel_rpl = 0; + /* Prevent unwanted bits from being set in PTEs. */ + __supported_pte_mask &= ~_PAGE_GLOBAL; + if (!is_initial_xendomain()) + __supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD); + /* set the limit of our address space */ xen_reserve_top(); diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 3525ef523a74..3f2a67fe6ad6 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -199,10 +199,8 @@ pgdval_t xen_pgd_val(pgd_t pgd) pte_t xen_make_pte(pteval_t pte) { - if (pte & _PAGE_PRESENT) { + if (pte & _PAGE_PRESENT) pte = phys_to_machine(XPADDR(pte)).maddr; - pte &= ~(_PAGE_PCD | _PAGE_PWT); - } return (pte_t){ .pte = pte }; } -- cgit v1.2.3 From ebb9cfe20fe167f29960a5e913193a684fac50bf Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 16 Jun 2008 15:01:56 -0700 Subject: xen: don't drop NX bit Because NX is now enforced properly, we must put the hypercall page into the .text segment so that it is executable. Signed-off-by: Jeremy Fitzhardinge Cc: Stable Kernel Cc: the arch/x86 maintainers Signed-off-by: Ingo Molnar --- arch/x86/xen/mmu.c | 54 +++++++++++++++++++++++++++---------------------- arch/x86/xen/xen-head.S | 2 +- 2 files changed, 31 insertions(+), 25 deletions(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 3f2a67fe6ad6..265601d5a6ae 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -179,46 +179,54 @@ out: preempt_enable(); } -pteval_t xen_pte_val(pte_t pte) +/* Assume pteval_t is equivalent to all the other *val_t types. */ +static pteval_t pte_mfn_to_pfn(pteval_t val) { - pteval_t ret = pte.pte; + if (val & _PAGE_PRESENT) { + unsigned long mfn = (val & PTE_MASK) >> PAGE_SHIFT; + pteval_t flags = val & ~PTE_MASK; + val = (mfn_to_pfn(mfn) << PAGE_SHIFT) | flags; + } - if (ret & _PAGE_PRESENT) - ret = machine_to_phys(XMADDR(ret)).paddr | _PAGE_PRESENT; + return val; +} - return ret; +static pteval_t pte_pfn_to_mfn(pteval_t val) +{ + if (val & _PAGE_PRESENT) { + unsigned long pfn = (val & PTE_MASK) >> PAGE_SHIFT; + pteval_t flags = val & ~PTE_MASK; + val = (pfn_to_mfn(pfn) << PAGE_SHIFT) | flags; + } + + return val; +} + +pteval_t xen_pte_val(pte_t pte) +{ + return pte_mfn_to_pfn(pte.pte); } pgdval_t xen_pgd_val(pgd_t pgd) { - pgdval_t ret = pgd.pgd; - if (ret & _PAGE_PRESENT) - ret = machine_to_phys(XMADDR(ret)).paddr | _PAGE_PRESENT; - return ret; + return pte_mfn_to_pfn(pgd.pgd); } pte_t xen_make_pte(pteval_t pte) { - if (pte & _PAGE_PRESENT) - pte = phys_to_machine(XPADDR(pte)).maddr; - - return (pte_t){ .pte = pte }; + pte = pte_pfn_to_mfn(pte); + return native_make_pte(pte); } pgd_t xen_make_pgd(pgdval_t pgd) { - if (pgd & _PAGE_PRESENT) - pgd = phys_to_machine(XPADDR(pgd)).maddr; - - return (pgd_t){ pgd }; + pgd = pte_pfn_to_mfn(pgd); + return native_make_pgd(pgd); } pmdval_t xen_pmd_val(pmd_t pmd) { - pmdval_t ret = native_pmd_val(pmd); - if (ret & _PAGE_PRESENT) - ret = machine_to_phys(XMADDR(ret)).paddr | _PAGE_PRESENT; - return ret; + return pte_mfn_to_pfn(pmd.pmd); } #ifdef CONFIG_X86_PAE void xen_set_pud(pud_t *ptr, pud_t val) @@ -265,9 +273,7 @@ void xen_pmd_clear(pmd_t *pmdp) pmd_t xen_make_pmd(pmdval_t pmd) { - if (pmd & _PAGE_PRESENT) - pmd = phys_to_machine(XPADDR(pmd)).maddr; - + pmd = pte_pfn_to_mfn(pmd); return native_make_pmd(pmd); } #else /* !PAE */ diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S index 288d587ce73c..3175e973fd0d 100644 --- a/arch/x86/xen/xen-head.S +++ b/arch/x86/xen/xen-head.S @@ -17,7 +17,7 @@ ENTRY(startup_xen) __FINIT -.pushsection .bss.page_aligned +.pushsection .text .align PAGE_SIZE_asm ENTRY(hypercall_page) .skip 0x1000 -- cgit v1.2.3 From 28499143933f19b28008a556ed59255d6009391a Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 9 May 2008 12:05:57 +0100 Subject: xen: remove support for non-PAE 32-bit Non-PAE operation has been deprecated in Xen for a while, and is rarely tested or used. xen-unstable has now officially dropped non-PAE support. Since Xen/pvops' non-PAE support has also been broken for a while, we may as well completely drop it altogether. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- arch/x86/xen/Kconfig | 2 +- arch/x86/xen/enlighten.c | 51 ++++++++++++++++------------------------------ arch/x86/xen/mmu.c | 19 ++--------------- arch/x86/xen/mmu.h | 24 ++++++---------------- arch/x86/xen/xen-head.S | 4 ---- include/asm-x86/xen/page.h | 4 ---- 6 files changed, 27 insertions(+), 77 deletions(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig index 2e641be2737e..525b108411bd 100644 --- a/arch/x86/xen/Kconfig +++ b/arch/x86/xen/Kconfig @@ -6,7 +6,7 @@ config XEN bool "Xen guest support" select PARAVIRT depends on X86_32 - depends on X86_CMPXCHG && X86_TSC && !(X86_VISWS || X86_VOYAGER) + depends on X86_CMPXCHG && X86_TSC && X86_PAE && !(X86_VISWS || X86_VOYAGER) help This is the Linux Xen port. Enabling this will allow the kernel to boot in a paravirtualized environment under the diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index c048de34d6a1..f09c1c69c37a 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -785,38 +785,35 @@ static __init void xen_set_pte_init(pte_t *ptep, pte_t pte) static __init void xen_pagetable_setup_start(pgd_t *base) { pgd_t *xen_pgd = (pgd_t *)xen_start_info->pt_base; + int i; /* special set_pte for pagetable initialization */ pv_mmu_ops.set_pte = xen_set_pte_init; init_mm.pgd = base; /* - * copy top-level of Xen-supplied pagetable into place. For - * !PAE we can use this as-is, but for PAE it is a stand-in - * while we copy the pmd pages. + * copy top-level of Xen-supplied pagetable into place. This + * is a stand-in while we copy the pmd pages. */ memcpy(base, xen_pgd, PTRS_PER_PGD * sizeof(pgd_t)); - if (PTRS_PER_PMD > 1) { - int i; - /* - * For PAE, need to allocate new pmds, rather than - * share Xen's, since Xen doesn't like pmd's being - * shared between address spaces. - */ - for (i = 0; i < PTRS_PER_PGD; i++) { - if (pgd_val_ma(xen_pgd[i]) & _PAGE_PRESENT) { - pmd_t *pmd = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE); + /* + * For PAE, need to allocate new pmds, rather than + * share Xen's, since Xen doesn't like pmd's being + * shared between address spaces. + */ + for (i = 0; i < PTRS_PER_PGD; i++) { + if (pgd_val_ma(xen_pgd[i]) & _PAGE_PRESENT) { + pmd_t *pmd = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE); - memcpy(pmd, (void *)pgd_page_vaddr(xen_pgd[i]), - PAGE_SIZE); + memcpy(pmd, (void *)pgd_page_vaddr(xen_pgd[i]), + PAGE_SIZE); - make_lowmem_page_readonly(pmd); + make_lowmem_page_readonly(pmd); - set_pgd(&base[i], __pgd(1 + __pa(pmd))); - } else - pgd_clear(&base[i]); - } + set_pgd(&base[i], __pgd(1 + __pa(pmd))); + } else + pgd_clear(&base[i]); } /* make sure zero_page is mapped RO so we can use it in pagetables */ @@ -873,17 +870,7 @@ static __init void xen_pagetable_setup_done(pgd_t *base) /* Actually pin the pagetable down, but we can't set PG_pinned yet because the page structures don't exist yet. */ - { - unsigned level; - -#ifdef CONFIG_X86_PAE - level = MMUEXT_PIN_L3_TABLE; -#else - level = MMUEXT_PIN_L2_TABLE; -#endif - - pin_pagetable_pfn(level, PFN_DOWN(__pa(base))); - } + pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(base))); } /* This is called once we have the cpu_possible_map */ @@ -1093,7 +1080,6 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = { .make_pte = xen_make_pte, .make_pgd = xen_make_pgd, -#ifdef CONFIG_X86_PAE .set_pte_atomic = xen_set_pte_atomic, .set_pte_present = xen_set_pte_at, .set_pud = xen_set_pud, @@ -1102,7 +1088,6 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = { .make_pmd = xen_make_pmd, .pmd_val = xen_pmd_val, -#endif /* PAE */ .activate_mm = xen_activate_mm, .dup_mmap = xen_dup_mmap, diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 265601d5a6ae..df40bf74ea75 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -228,7 +228,7 @@ pmdval_t xen_pmd_val(pmd_t pmd) { return pte_mfn_to_pfn(pmd.pmd); } -#ifdef CONFIG_X86_PAE + void xen_set_pud(pud_t *ptr, pud_t val) { struct multicall_space mcs; @@ -276,12 +276,6 @@ pmd_t xen_make_pmd(pmdval_t pmd) pmd = pte_pfn_to_mfn(pmd); return native_make_pmd(pmd); } -#else /* !PAE */ -void xen_set_pte(pte_t *ptep, pte_t pte) -{ - *ptep = pte; -} -#endif /* CONFIG_X86_PAE */ /* (Yet another) pagetable walker. This one is intended for pinning a @@ -434,8 +428,6 @@ static int pin_page(struct page *page, enum pt_level level) read-only, and can be pinned. */ void xen_pgd_pin(pgd_t *pgd) { - unsigned level; - xen_mc_batch(); if (pgd_walk(pgd, pin_page, TASK_SIZE)) { @@ -445,14 +437,7 @@ void xen_pgd_pin(pgd_t *pgd) xen_mc_batch(); } -#ifdef CONFIG_X86_PAE - level = MMUEXT_PIN_L3_TABLE; -#else - level = MMUEXT_PIN_L2_TABLE; -#endif - - xen_do_pin(level, PFN_DOWN(__pa(pgd))); - + xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd))); xen_mc_issue(0); } diff --git a/arch/x86/xen/mmu.h b/arch/x86/xen/mmu.h index b5e189b1519d..5fe961caffd4 100644 --- a/arch/x86/xen/mmu.h +++ b/arch/x86/xen/mmu.h @@ -37,14 +37,13 @@ void xen_exit_mmap(struct mm_struct *mm); void xen_pgd_pin(pgd_t *pgd); //void xen_pgd_unpin(pgd_t *pgd); -#ifdef CONFIG_X86_PAE -unsigned long long xen_pte_val(pte_t); -unsigned long long xen_pmd_val(pmd_t); -unsigned long long xen_pgd_val(pgd_t); +pteval_t xen_pte_val(pte_t); +pmdval_t xen_pmd_val(pmd_t); +pgdval_t xen_pgd_val(pgd_t); -pte_t xen_make_pte(unsigned long long); -pmd_t xen_make_pmd(unsigned long long); -pgd_t xen_make_pgd(unsigned long long); +pte_t xen_make_pte(pteval_t); +pmd_t xen_make_pmd(pmdval_t); +pgd_t xen_make_pgd(pgdval_t); void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pteval); @@ -53,15 +52,4 @@ void xen_set_pud(pud_t *ptr, pud_t val); void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep); void xen_pmd_clear(pmd_t *pmdp); - -#else -unsigned long xen_pte_val(pte_t); -unsigned long xen_pmd_val(pmd_t); -unsigned long xen_pgd_val(pgd_t); - -pte_t xen_make_pte(unsigned long); -pmd_t xen_make_pmd(unsigned long); -pgd_t xen_make_pgd(unsigned long); -#endif - #endif /* _XEN_MMU_H */ diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S index 3175e973fd0d..6ec3b4f7719b 100644 --- a/arch/x86/xen/xen-head.S +++ b/arch/x86/xen/xen-head.S @@ -30,11 +30,7 @@ ENTRY(hypercall_page) ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .long startup_xen) ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .long hypercall_page) ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .asciz "!writable_page_tables|pae_pgdir_above_4gb") -#ifdef CONFIG_X86_PAE ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz "yes") -#else - ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz "no") -#endif ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz "generic") #endif /*CONFIG_XEN */ diff --git a/include/asm-x86/xen/page.h b/include/asm-x86/xen/page.h index baf3a4dce28c..e11f24038b1d 100644 --- a/include/asm-x86/xen/page.h +++ b/include/asm-x86/xen/page.h @@ -150,13 +150,9 @@ static inline pte_t __pte_ma(pteval_t x) return (pte_t) { .pte = x }; } -#ifdef CONFIG_X86_PAE #define pmd_val_ma(v) ((v).pmd) #define pud_val_ma(v) ((v).pgd.pgd) #define __pmd_ma(x) ((pmd_t) { (x) } ) -#else /* !X86_PAE */ -#define pmd_val_ma(v) ((v).pud.pgd.pgd) -#endif /* CONFIG_X86_PAE */ #define pgd_val_ma(x) ((x).pgd) -- cgit v1.2.3 From 1c7b67f7576c4ca2a344379a4a29eec8fe8e7935 Mon Sep 17 00:00:00 2001 From: Gerd Hoffmann Date: Tue, 3 Jun 2008 16:17:30 +0200 Subject: x86: Make xen use the paravirt clocksource structs and functions This patch updates the xen guest to use the pvclock structs and helper functions. Signed-off-by: Gerd Hoffmann Acked-by: Jeremy Fitzhardinge Signed-off-by: Avi Kivity --- arch/x86/xen/Kconfig | 1 + arch/x86/xen/time.c | 132 ++++---------------------------------------- include/xen/interface/xen.h | 7 +-- 3 files changed, 16 insertions(+), 124 deletions(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig index 2e641be2737e..3a4f16aea4bf 100644 --- a/arch/x86/xen/Kconfig +++ b/arch/x86/xen/Kconfig @@ -5,6 +5,7 @@ config XEN bool "Xen guest support" select PARAVIRT + select PARAVIRT_CLOCK depends on X86_32 depends on X86_CMPXCHG && X86_TSC && !(X86_VISWS || X86_VOYAGER) help diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 52b2e3856980..41e217503c96 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -14,6 +14,7 @@ #include #include +#include #include #include @@ -31,17 +32,6 @@ static cycle_t xen_clocksource_read(void); -/* These are perodically updated in shared_info, and then copied here. */ -struct shadow_time_info { - u64 tsc_timestamp; /* TSC at last update of time vals. */ - u64 system_timestamp; /* Time, in nanosecs, since boot. */ - u32 tsc_to_nsec_mul; - int tsc_shift; - u32 version; -}; - -static DEFINE_PER_CPU(struct shadow_time_info, shadow_time); - /* runstate info updated by Xen */ static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate); @@ -211,7 +201,7 @@ unsigned long long xen_sched_clock(void) unsigned long xen_cpu_khz(void) { u64 xen_khz = 1000000ULL << 32; - const struct vcpu_time_info *info = + const struct pvclock_vcpu_time_info *info = &HYPERVISOR_shared_info->vcpu_info[0].time; do_div(xen_khz, info->tsc_to_system_mul); @@ -223,121 +213,26 @@ unsigned long xen_cpu_khz(void) return xen_khz; } -/* - * Reads a consistent set of time-base values from Xen, into a shadow data - * area. - */ -static unsigned get_time_values_from_xen(void) -{ - struct vcpu_time_info *src; - struct shadow_time_info *dst; - - /* src is shared memory with the hypervisor, so we need to - make sure we get a consistent snapshot, even in the face of - being preempted. */ - src = &__get_cpu_var(xen_vcpu)->time; - dst = &__get_cpu_var(shadow_time); - - do { - dst->version = src->version; - rmb(); /* fetch version before data */ - dst->tsc_timestamp = src->tsc_timestamp; - dst->system_timestamp = src->system_time; - dst->tsc_to_nsec_mul = src->tsc_to_system_mul; - dst->tsc_shift = src->tsc_shift; - rmb(); /* test version after fetching data */ - } while ((src->version & 1) | (dst->version ^ src->version)); - - return dst->version; -} - -/* - * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction, - * yielding a 64-bit result. - */ -static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift) -{ - u64 product; -#ifdef __i386__ - u32 tmp1, tmp2; -#endif - - if (shift < 0) - delta >>= -shift; - else - delta <<= shift; - -#ifdef __i386__ - __asm__ ( - "mul %5 ; " - "mov %4,%%eax ; " - "mov %%edx,%4 ; " - "mul %5 ; " - "xor %5,%5 ; " - "add %4,%%eax ; " - "adc %5,%%edx ; " - : "=A" (product), "=r" (tmp1), "=r" (tmp2) - : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) ); -#elif __x86_64__ - __asm__ ( - "mul %%rdx ; shrd $32,%%rdx,%%rax" - : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) ); -#else -#error implement me! -#endif - - return product; -} - -static u64 get_nsec_offset(struct shadow_time_info *shadow) -{ - u64 now, delta; - now = native_read_tsc(); - delta = now - shadow->tsc_timestamp; - return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift); -} - static cycle_t xen_clocksource_read(void) { - struct shadow_time_info *shadow = &get_cpu_var(shadow_time); + struct pvclock_vcpu_time_info *src; cycle_t ret; - unsigned version; - - do { - version = get_time_values_from_xen(); - barrier(); - ret = shadow->system_timestamp + get_nsec_offset(shadow); - barrier(); - } while (version != __get_cpu_var(xen_vcpu)->time.version); - - put_cpu_var(shadow_time); + src = &get_cpu_var(xen_vcpu)->time; + ret = pvclock_clocksource_read(src); + put_cpu_var(xen_vcpu); return ret; } static void xen_read_wallclock(struct timespec *ts) { - const struct shared_info *s = HYPERVISOR_shared_info; - u32 version; - u64 delta; - struct timespec now; - - /* get wallclock at system boot */ - do { - version = s->wc_version; - rmb(); /* fetch version before time */ - now.tv_sec = s->wc_sec; - now.tv_nsec = s->wc_nsec; - rmb(); /* fetch time before checking version */ - } while ((s->wc_version & 1) | (version ^ s->wc_version)); + struct shared_info *s = HYPERVISOR_shared_info; + struct pvclock_wall_clock *wall_clock = &(s->wc); + struct pvclock_vcpu_time_info *vcpu_time; - delta = xen_clocksource_read(); /* time since system boot */ - delta += now.tv_sec * (u64)NSEC_PER_SEC + now.tv_nsec; - - now.tv_nsec = do_div(delta, NSEC_PER_SEC); - now.tv_sec = delta; - - set_normalized_timespec(ts, now.tv_sec, now.tv_nsec); + vcpu_time = &get_cpu_var(xen_vcpu)->time; + pvclock_read_wallclock(wall_clock, vcpu_time, ts); + put_cpu_var(xen_vcpu); } unsigned long xen_get_wallclock(void) @@ -345,7 +240,6 @@ unsigned long xen_get_wallclock(void) struct timespec ts; xen_read_wallclock(&ts); - return ts.tv_sec; } @@ -569,8 +463,6 @@ __init void xen_time_init(void) { int cpu = smp_processor_id(); - get_time_values_from_xen(); - clocksource_register(&xen_clocksource); if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL) == 0) { diff --git a/include/xen/interface/xen.h b/include/xen/interface/xen.h index 9b018da48cf3..819a0331cda9 100644 --- a/include/xen/interface/xen.h +++ b/include/xen/interface/xen.h @@ -10,6 +10,7 @@ #define __XEN_PUBLIC_XEN_H__ #include +#include /* * XEN "SYSTEM CALLS" (a.k.a. HYPERCALLS). @@ -336,7 +337,7 @@ struct vcpu_info { uint8_t evtchn_upcall_mask; unsigned long evtchn_pending_sel; struct arch_vcpu_info arch; - struct vcpu_time_info time; + struct pvclock_vcpu_time_info time; }; /* 64 bytes (x86) */ /* @@ -384,9 +385,7 @@ struct shared_info { * Wallclock time: updated only by control software. Guests should base * their gettimeofday() syscall on this wallclock-base value. */ - uint32_t wc_version; /* Version counter: see vcpu_time_info_t. */ - uint32_t wc_sec; /* Secs 00:00:00 UTC, Jan 1, 1970. */ - uint32_t wc_nsec; /* Nsecs 00:00:00 UTC, Jan 1, 1970. */ + struct pvclock_wall_clock wc; struct arch_shared_info arch; -- cgit v1.2.3 From d8355aca23863be659ec5b7e0393cfbfa91ec221 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 3 Jul 2008 22:10:18 -0700 Subject: xen: fix address truncation in pte mfn<->pfn conversion When converting the page number in a pte/pmd/pud/pgd between machine and pseudo-physical addresses, the converted result was being truncated at 32-bits. This caused failures on machines with more than 4G of physical memory. Signed-off-by: Jeremy Fitzhardinge Cc: "Christopher S. Aker" Cc: Ian Campbell Signed-off-by: Ingo Molnar --- arch/x86/xen/mmu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86/xen') diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index df40bf74ea75..4e527e7893a8 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -185,7 +185,7 @@ static pteval_t pte_mfn_to_pfn(pteval_t val) if (val & _PAGE_PRESENT) { unsigned long mfn = (val & PTE_MASK) >> PAGE_SHIFT; pteval_t flags = val & ~PTE_MASK; - val = (mfn_to_pfn(mfn) << PAGE_SHIFT) | flags; + val = ((pteval_t)mfn_to_pfn(mfn) << PAGE_SHIFT) | flags; } return val; @@ -196,7 +196,7 @@ static pteval_t pte_pfn_to_mfn(pteval_t val) if (val & _PAGE_PRESENT) { unsigned long pfn = (val & PTE_MASK) >> PAGE_SHIFT; pteval_t flags = val & ~PTE_MASK; - val = (pfn_to_mfn(pfn) << PAGE_SHIFT) | flags; + val = ((pteval_t)pfn_to_mfn(pfn) << PAGE_SHIFT) | flags; } return val; -- cgit v1.2.3