Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH v2] ACPI S3 to work under Xen.
Attached is an [v2] set of patches to enable S3 to work with the Xen hypervisor. Changes since the RFC posting [http://comments.gmane.org/gmane.linux.acpi.devel/50701] by Liang Tang: - Per review comments added: __unused__ attribute, support for PM1A/B if more than 16-bit, copyright/license. - Added support for PHYSDEVOP_restore_msi_ext call. The relationship that Xen has with Linux kernel is symbiotic. The Linux kernel does the ACPI "stuff" and tells the hypervisor to do the low-level stuff (such as program the IOAPIC, setup vectors, etc). The realm of ACPI S3 is more complex as we need to save the CPU state (and Intel TXT values - which the hypervisor has to do). The major difficulties we hit was with ''acpi_suspend_lowlevel'' - which tweaks a lot of lowlevel values and some of them are not properly handled by Xen. Liang Tang has figured which ones of them we trip over (read below) - and he suggested that perhaps we can provide a registration mechanism to abstract this away. The reason for all of this is that Linux does not talk to the BIOS directly - instead it simply walks through the necessary ACPI methods and then issues hypercall to Xen which then further completes the remaining suspend steps. So the attached patches do exactly that - there are two entry points in the ACPI. 1). For S3: acpi_suspend_lowlevel -> .. lots of code -> acpi_enter_sleep_state 2). For S1/S4/S5: acpi_enter_sleep_state The first naive idea was of abstracting away in the ''acpi_enter_sleep_state'' function the tboot_sleep code so that we can use it too. And low-behold - it worked splendidly for powering off (S5 I believe) For S3 that did not work - during suspend the hypervisor tripped over when saving cr8. During resume it tripped over at restoring the cr3, cr8, idt, and gdt values. When I posted the RFC, the feedback I got was to use a higher upper interface to make the call to the hypervisor. Instead of doing it at the lower pv-ops case for cr3, cr8, idt, gdt, etc. The code I''ve to say - is much nicer than doing it via pv-ops. Anyhow, please take a look! Konrad Rzeszutek Wilk (5): x86: Expand the x86_msi_ops to have a restore MSIs. x86, acpi, tboot: Have a ACPI sleep override instead of calling tboot_sleep. xen: Utilize the restore_msi_irqs hook. xen/acpi/sleep: Enable ACPI sleep via the __acpi_override_sleep xen/acpi/sleep: Register to the acpi_suspend_lowlevel a callback. Liang Tang (2): x86/acpi/sleep: Provide registration for acpi_suspend_lowlevel. xen/pci:use hypercall PHYSDEVOP_restore_msi_ext to restore MSI/MSI-X vectors Yu Ke (1): xen/acpi: Domain0 acpi parser related platform hypercall arch/ia64/include/asm/xen/interface.h | 1 + arch/x86/include/asm/acpi.h | 6 +- arch/x86/include/asm/pci.h | 9 + arch/x86/include/asm/x86_init.h | 1 + arch/x86/include/asm/xen/hypercall.h | 8 + arch/x86/include/asm/xen/interface.h | 1 + arch/x86/kernel/acpi/boot.c | 6 + arch/x86/kernel/acpi/sleep.c | 4 +- arch/x86/kernel/acpi/sleep.h | 2 + arch/x86/kernel/tboot.c | 14 +- arch/x86/kernel/x86_init.c | 1 + arch/x86/pci/xen.c | 29 +++ arch/x86/xen/enlighten.c | 3 + drivers/acpi/acpica/hwsleep.c | 12 +- drivers/acpi/sleep.c | 2 + drivers/pci/msi.c | 29 +++- drivers/xen/Makefile | 2 +- drivers/xen/acpi.c | 65 +++++++ include/linux/tboot.h | 3 +- include/xen/acpi.h | 70 +++++++ include/xen/interface/physdev.h | 15 ++ include/xen/interface/platform.h | 320 +++++++++++++++++++++++++++++++++ include/xen/interface/xen.h | 1 + 23 files changed, 591 insertions(+), 13 deletions(-) _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 1/8] x86: Expand the x86_msi_ops to have a restore MSIs.
The MSI restore function will become a function pointer in an x86_msi_ops struct. It defaults to the implementation in the io_apic.c and msi.c. We piggyback on the indirection mechanism introduced by "x86: Introduce x86_msi_ops". c: x86@kernel.org Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- arch/x86/include/asm/pci.h | 9 +++++++++ arch/x86/include/asm/x86_init.h | 1 + arch/x86/kernel/x86_init.c | 1 + drivers/pci/msi.c | 29 +++++++++++++++++++++++++++-- 4 files changed, 38 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h index d498943..df75d07 100644 --- a/arch/x86/include/asm/pci.h +++ b/arch/x86/include/asm/pci.h @@ -112,19 +112,28 @@ static inline void x86_teardown_msi_irq(unsigned int irq) { x86_msi.teardown_msi_irq(irq); } +static inline void x86_restore_msi_irqs(struct pci_dev *dev, int irq) +{ + x86_msi.restore_msi_irqs(dev, irq); +} #define arch_setup_msi_irqs x86_setup_msi_irqs #define arch_teardown_msi_irqs x86_teardown_msi_irqs #define arch_teardown_msi_irq x86_teardown_msi_irq +#define arch_restore_msi_irqs x86_restore_msi_irqs /* implemented in arch/x86/kernel/apic/io_apic. */ int native_setup_msi_irqs(struct pci_dev *dev, int nvec, int type); void native_teardown_msi_irq(unsigned int irq); +void native_restore_msi_irqs(struct pci_dev *dev, int irq); /* default to the implementation in drivers/lib/msi.c */ #define HAVE_DEFAULT_MSI_TEARDOWN_IRQS +#define HAVE_DEFAULT_MSI_RESTORE_IRQS void default_teardown_msi_irqs(struct pci_dev *dev); +void default_restore_msi_irqs(struct pci_dev *dev, int irq); #else #define native_setup_msi_irqs NULL #define native_teardown_msi_irq NULL #define default_teardown_msi_irqs NULL +#define default_restore_msi_irqs NULL #endif #define PCI_DMA_BUS_IS_PHYS (dma_ops->is_phys) diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index d3d8590..7af18be 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -174,6 +174,7 @@ struct x86_msi_ops { int (*setup_msi_irqs)(struct pci_dev *dev, int nvec, int type); void (*teardown_msi_irq)(unsigned int irq); void (*teardown_msi_irqs)(struct pci_dev *dev); + void (*restore_msi_irqs)(struct pci_dev *dev, int irq); }; extern struct x86_init_ops x86_init; diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index 6f164bd..bd1fe10 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -110,4 +110,5 @@ struct x86_msi_ops x86_msi = { .setup_msi_irqs = native_setup_msi_irqs, .teardown_msi_irq = native_teardown_msi_irq, .teardown_msi_irqs = default_teardown_msi_irqs, + .restore_msi_irqs = default_restore_msi_irqs, }; diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 2f10328..f1fd801 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -85,6 +85,31 @@ void default_teardown_msi_irqs(struct pci_dev *dev) } #endif +#ifndef arch_restore_msi_irqs +# define arch_restore_msi_irqs default_restore_msi_irqs +# define HAVE_DEFAULT_MSI_RESTORE_IRQS +#endif + +#ifdef HAVE_DEFAULT_MSI_RESTORE_IRQS +void default_restore_msi_irqs(struct pci_dev *dev, int irq) +{ + struct msi_desc *entry; + + entry = NULL; + if (dev->msix_enabled) { + list_for_each_entry(entry, &dev->msi_list, list) { + if (irq == entry->irq) + break; + } + } else if (dev->msi_enabled) { + entry = irq_get_msi_desc(irq); + } + + if (entry) + write_msi_msg(irq, &entry->msg); +} +#endif + static void msi_set_enable(struct pci_dev *dev, int pos, int enable) { u16 control; @@ -359,7 +384,7 @@ static void __pci_restore_msi_state(struct pci_dev *dev) pci_intx_for_msi(dev, 0); msi_set_enable(dev, pos, 0); - write_msi_msg(dev->irq, &entry->msg); + arch_restore_msi_irqs(dev, dev->irq); pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control); msi_mask_irq(entry, msi_capable_mask(control), entry->masked); @@ -387,7 +412,7 @@ static void __pci_restore_msix_state(struct pci_dev *dev) pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); list_for_each_entry(entry, &dev->msi_list, list) { - write_msi_msg(entry->irq, &entry->msg); + arch_restore_msi_irqs(dev, entry->irq); msix_mask_irq(entry, entry->masked); } -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 2/8] x86, acpi, tboot: Have a ACPI sleep override instead of calling tboot_sleep.
The ACPI suspend path makes a call to tboot_sleep right before it writes the PM1A, PM1B values. We replace the direct call to tboot via an registration callback similar to __acpi_register_gsi. CC: Thomas Gleixner <tglx@linutronix.de> CC: "H. Peter Anvin" <hpa@zytor.com> CC: x86@kernel.org CC: Len Brown <len.brown@intel.com> CC: Joseph Cihula <joseph.cihula@intel.com> CC: Shane Wang <shane.wang@intel.com> CC: xen-devel@lists.xensource.com CC: linux-pm@lists.linux-foundation.org CC: tboot-devel@lists.sourceforge.net CC: linux-acpi@vger.kernel.org [v1: Added __attribute__ ((unused))] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- arch/x86/include/asm/acpi.h | 4 ++++ arch/x86/kernel/acpi/boot.c | 4 ++++ arch/x86/kernel/tboot.c | 14 ++++++++++---- drivers/acpi/acpica/hwsleep.c | 12 ++++++++++-- include/linux/tboot.h | 3 ++- 5 files changed, 30 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 610001d..0a46696 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -98,6 +98,10 @@ void acpi_pic_sci_set_trigger(unsigned int, u16); extern int (*__acpi_register_gsi)(struct device *dev, u32 gsi, int trigger, int polarity); +extern int (*__acpi_override_sleep)(u8 sleep_state, u32 pm1a_ctrl, + u32 pm1b_ctrl, bool *skip_rest) \ + __attribute__ ((unused)); + static inline void disable_acpi(void) { acpi_disabled = 1; diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 4558f0d..7f30806 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -552,6 +552,10 @@ static int acpi_register_gsi_ioapic(struct device *dev, u32 gsi, int (*__acpi_register_gsi)(struct device *dev, u32 gsi, int trigger, int polarity) = acpi_register_gsi_pic; +int (*__acpi_override_sleep)(u8 sleep_state, u32 pm1a_ctrl, + u32 pm1b_ctrl, bool *skip_rest) \ + __attribute__ ((unused)) = NULL; + /* * success: return IRQ number (>=0) * failure: return < 0 diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c index e07a2fc..a6c0a30 100644 --- a/arch/x86/kernel/tboot.c +++ b/arch/x86/kernel/tboot.c @@ -42,7 +42,7 @@ #include <asm/setup.h> #include <asm/e820.h> #include <asm/io.h> - +#include <linux/acpi.h> #include "acpi/realmode/wakeup.h" /* Global pointer to shared data; NULL means no measured launch. */ @@ -271,7 +271,9 @@ static void tboot_copy_fadt(const struct acpi_table_fadt *fadt) offsetof(struct acpi_table_facs, firmware_waking_vector); } -void tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control) + +int tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control, + bool *skip_rest) { static u32 acpi_shutdown_map[ACPI_S_STATE_COUNT] = { /* S0,1,2: */ -1, -1, -1, @@ -280,7 +282,7 @@ void tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control) /* S5: */ TB_SHUTDOWN_S5 }; if (!tboot_enabled()) - return; + return AE_OK; tboot_copy_fadt(&acpi_gbl_FADT); tboot->acpi_sinfo.pm1a_cnt_val = pm1a_control; @@ -291,10 +293,12 @@ void tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control) if (sleep_state >= ACPI_S_STATE_COUNT || acpi_shutdown_map[sleep_state] == -1) { pr_warning("unsupported sleep state 0x%x\n", sleep_state); - return; + return AE_ERROR; } tboot_shutdown(acpi_shutdown_map[sleep_state]); + + return AE_OK; } static atomic_t ap_wfs_count; @@ -344,6 +348,8 @@ static __init int tboot_late_init(void) atomic_set(&ap_wfs_count, 0); register_hotcpu_notifier(&tboot_cpu_notifier); + + __acpi_override_sleep = tboot_sleep; return 0; } diff --git a/drivers/acpi/acpica/hwsleep.c b/drivers/acpi/acpica/hwsleep.c index 2ac28bb..31d1198 100644 --- a/drivers/acpi/acpica/hwsleep.c +++ b/drivers/acpi/acpica/hwsleep.c @@ -45,7 +45,6 @@ #include <acpi/acpi.h> #include "accommon.h" #include "actables.h" -#include <linux/tboot.h> #define _COMPONENT ACPI_HARDWARE ACPI_MODULE_NAME("hwsleep") @@ -343,8 +342,17 @@ acpi_status asmlinkage acpi_enter_sleep_state(u8 sleep_state) ACPI_FLUSH_CPU_CACHE(); - tboot_sleep(sleep_state, pm1a_control, pm1b_control); + if (__acpi_override_sleep) { + bool skip_rest = false; + status = __acpi_override_sleep(sleep_state, pm1a_control, + pm1b_control, &skip_rest); + + if (ACPI_FAILURE(status)) + return_ACPI_STATUS(status); + if (skip_rest) + return_ACPI_STATUS(AE_OK); + } /* Write #2: Write both SLP_TYP + SLP_EN */ status = acpi_hw_write_pm1_control(pm1a_control, pm1b_control); diff --git a/include/linux/tboot.h b/include/linux/tboot.h index 1dba6ee..1216698 100644 --- a/include/linux/tboot.h +++ b/include/linux/tboot.h @@ -143,7 +143,8 @@ static inline int tboot_enabled(void) extern void tboot_probe(void); extern void tboot_shutdown(u32 shutdown_type); -extern void tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control); +extern int tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control, + bool *skip_rest) __attribute__ ((unused)); extern struct acpi_table_header *tboot_get_dmar_table( struct acpi_table_header *dmar_tbl); extern int tboot_force_iommu(void); -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 3/8] x86/acpi/sleep: Provide registration for acpi_suspend_lowlevel.
From: Liang Tang <liang.tang@oracle.com> Which by default will be x86_acpi_suspend_lowlevel. This registration allows us to register another callback if there is a need to use another platform specific callback. CC: Thomas Gleixner <tglx@linutronix.de> CC: "H. Peter Anvin" <hpa@zytor.com> CC: x86@kernel.org CC: Len Brown <len.brown@intel.com> CC: Joseph Cihula <joseph.cihula@intel.com> CC: Shane Wang <shane.wang@intel.com> CC: linux-pm@lists.linux-foundation.org CC: linux-acpi@vger.kernel.org CC: Len Brown <len.brown@intel.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Liang Tang <liang.tang@oracle.com> --- arch/x86/include/asm/acpi.h | 2 +- arch/x86/kernel/acpi/boot.c | 2 ++ arch/x86/kernel/acpi/sleep.c | 4 ++-- arch/x86/kernel/acpi/sleep.h | 2 ++ drivers/acpi/sleep.c | 2 ++ 5 files changed, 9 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 0a46696..9b538dc 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -119,7 +119,7 @@ static inline void acpi_disable_pci(void) } /* Low-level suspend routine. */ -extern int acpi_suspend_lowlevel(void); +extern int (*acpi_suspend_lowlevel)(void); extern const unsigned char acpi_wakeup_code[]; #define acpi_wakeup_address (__pa(TRAMPOLINE_SYM(acpi_wakeup_code))) diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 7f30806..ddd081b 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -44,6 +44,7 @@ #include <asm/mpspec.h> #include <asm/smp.h> +#include "sleep.h" /* To include x86_acpi_suspend_lowlevel */ static int __initdata acpi_force = 0; u32 acpi_rsdt_forced; int acpi_disabled; @@ -556,6 +557,7 @@ int (*__acpi_override_sleep)(u8 sleep_state, u32 pm1a_ctrl, u32 pm1b_ctrl, bool *skip_rest) \ __attribute__ ((unused)) = NULL; +int (*acpi_suspend_lowlevel)(void) = x86_acpi_suspend_lowlevel; /* * success: return IRQ number (>=0) * failure: return < 0 diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 103b6ab..4d2d0b1 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -25,12 +25,12 @@ static char temp_stack[4096]; #endif /** - * acpi_suspend_lowlevel - save kernel state + * x86_acpi_suspend_lowlevel - save kernel state * * Create an identity mapped page table and copy the wakeup routine to * low memory. */ -int acpi_suspend_lowlevel(void) +int x86_acpi_suspend_lowlevel(void) { struct wakeup_header *header; /* address in low memory of the wakeup routine. */ diff --git a/arch/x86/kernel/acpi/sleep.h b/arch/x86/kernel/acpi/sleep.h index 416d4be..4d3feb5 100644 --- a/arch/x86/kernel/acpi/sleep.h +++ b/arch/x86/kernel/acpi/sleep.h @@ -13,3 +13,5 @@ extern unsigned long acpi_copy_wakeup_routine(unsigned long); extern void wakeup_long64(void); extern void do_suspend_lowlevel(void); + +extern int x86_acpi_suspend_lowlevel(void); diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index 3ed80b2..3570c00 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -254,6 +254,8 @@ static int acpi_suspend_enter(suspend_state_t pm_state) break; case ACPI_STATE_S3: + if (!acpi_suspend_lowlevel) + return -ENODEV; error = acpi_suspend_lowlevel(); if (error) return error; -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 4/8] xen: Utilize the restore_msi_irqs hook.
to make a hypercall to restore the vectors in the MSI/MSI-X configuration space. Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- arch/x86/pci/xen.c | 12 ++++++++++++ include/xen/interface/physdev.h | 7 +++++++ 2 files changed, 19 insertions(+), 0 deletions(-) diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c index 1017c7b..9eea4ed 100644 --- a/arch/x86/pci/xen.c +++ b/arch/x86/pci/xen.c @@ -304,6 +304,17 @@ static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) out: return ret; } + +static void xen_initdom_restore_msi_irqs(struct pci_dev *dev, int irq) +{ + int ret = 0; + struct physdev_restore_msi restore; + + restore.bus = dev->bus->number; + restore.devfn = dev->devfn; + ret = HYPERVISOR_physdev_op(PHYSDEVOP_restore_msi, &restore); + WARN(ret && ret != -ENOSYS, "restore_msi -> %d\n", ret); +} #endif static void xen_teardown_msi_irqs(struct pci_dev *dev) @@ -426,6 +437,7 @@ int __init pci_xen_initial_domain(void) #ifdef CONFIG_PCI_MSI x86_msi.setup_msi_irqs = xen_initdom_setup_msi_irqs; x86_msi.teardown_msi_irq = xen_teardown_msi_irq; + x86_msi.restore_msi_irqs = xen_initdom_restore_msi_irqs; #endif xen_setup_acpi_sci(); __acpi_register_gsi = acpi_register_gsi_xen; diff --git a/include/xen/interface/physdev.h b/include/xen/interface/physdev.h index 534cac8..44aefa9 100644 --- a/include/xen/interface/physdev.h +++ b/include/xen/interface/physdev.h @@ -144,6 +144,13 @@ struct physdev_manage_pci { uint8_t devfn; }; +#define PHYSDEVOP_restore_msi 19 +struct physdev_restore_msi { + /* IN */ + uint8_t bus; + uint8_t devfn; +}; + #define PHYSDEVOP_manage_pci_add_ext 20 struct physdev_manage_pci_ext { /* IN */ -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 5/8] xen/acpi: Domain0 acpi parser related platform hypercall
From: Yu Ke <ke.yu@intel.com> This patches implements the xen_platform_op hypercall, to pass the parsed ACPI info to hypervisor. Signed-off-by: Yu Ke <ke.yu@intel.com> Signed-off-by: Tian Kevin <kevin.tian@intel.com> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> [v1: Added DEFINE_GUEST.. in appropiate headers] [v2: Ripped out typedefs] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- arch/ia64/include/asm/xen/interface.h | 1 + arch/x86/include/asm/xen/interface.h | 1 + include/xen/interface/platform.h | 320 +++++++++++++++++++++++++++++++++ include/xen/interface/xen.h | 1 + 4 files changed, 323 insertions(+), 0 deletions(-) create mode 100644 include/xen/interface/platform.h diff --git a/arch/ia64/include/asm/xen/interface.h b/arch/ia64/include/asm/xen/interface.h index e951e74..1d2427d 100644 --- a/arch/ia64/include/asm/xen/interface.h +++ b/arch/ia64/include/asm/xen/interface.h @@ -76,6 +76,7 @@ DEFINE_GUEST_HANDLE(char); DEFINE_GUEST_HANDLE(int); DEFINE_GUEST_HANDLE(long); DEFINE_GUEST_HANDLE(void); +DEFINE_GUEST_HANDLE(uint64_t); typedef unsigned long xen_pfn_t; DEFINE_GUEST_HANDLE(xen_pfn_t); diff --git a/arch/x86/include/asm/xen/interface.h b/arch/x86/include/asm/xen/interface.h index 5d4922a..a1f2db5 100644 --- a/arch/x86/include/asm/xen/interface.h +++ b/arch/x86/include/asm/xen/interface.h @@ -55,6 +55,7 @@ DEFINE_GUEST_HANDLE(char); DEFINE_GUEST_HANDLE(int); DEFINE_GUEST_HANDLE(long); DEFINE_GUEST_HANDLE(void); +DEFINE_GUEST_HANDLE(uint64_t); #endif #ifndef HYPERVISOR_VIRT_START diff --git a/include/xen/interface/platform.h b/include/xen/interface/platform.h new file mode 100644 index 0000000..c168468 --- /dev/null +++ b/include/xen/interface/platform.h @@ -0,0 +1,320 @@ +/****************************************************************************** + * platform.h + * + * Hardware platform operations. Intended for use by domain-0 kernel. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + * + * Copyright (c) 2002-2006, K Fraser + */ + +#ifndef __XEN_PUBLIC_PLATFORM_H__ +#define __XEN_PUBLIC_PLATFORM_H__ + +#include "xen.h" + +#define XENPF_INTERFACE_VERSION 0x03000001 + +/* + * Set clock such that it would read <secs,nsecs> after 00:00:00 UTC, + * 1 January, 1970 if the current system time was <system_time>. + */ +#define XENPF_settime 17 +struct xenpf_settime { + /* IN variables. */ + uint32_t secs; + uint32_t nsecs; + uint64_t system_time; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_settime_t); + +/* + * Request memory range (@mfn, @mfn+@nr_mfns-1) to have type @type. + * On x86, @type is an architecture-defined MTRR memory type. + * On success, returns the MTRR that was used (@reg) and a handle that can + * be passed to XENPF_DEL_MEMTYPE to accurately tear down the new setting. + * (x86-specific). + */ +#define XENPF_add_memtype 31 +struct xenpf_add_memtype { + /* IN variables. */ + unsigned long mfn; + uint64_t nr_mfns; + uint32_t type; + /* OUT variables. */ + uint32_t handle; + uint32_t reg; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_add_memtype_t); + +/* + * Tear down an existing memory-range type. If @handle is remembered then it + * should be passed in to accurately tear down the correct setting (in case + * of overlapping memory regions with differing types). If it is not known + * then @handle should be set to zero. In all cases @reg must be set. + * (x86-specific). + */ +#define XENPF_del_memtype 32 +struct xenpf_del_memtype { + /* IN variables. */ + uint32_t handle; + uint32_t reg; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_del_memtype_t); + +/* Read current type of an MTRR (x86-specific). */ +#define XENPF_read_memtype 33 +struct xenpf_read_memtype { + /* IN variables. */ + uint32_t reg; + /* OUT variables. */ + unsigned long mfn; + uint64_t nr_mfns; + uint32_t type; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_read_memtype_t); + +#define XENPF_microcode_update 35 +struct xenpf_microcode_update { + /* IN variables. */ + GUEST_HANDLE(void) data; /* Pointer to microcode data */ + uint32_t length; /* Length of microcode data. */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_microcode_update_t); + +#define XENPF_platform_quirk 39 +#define QUIRK_NOIRQBALANCING 1 /* Do not restrict IO-APIC RTE targets */ +#define QUIRK_IOAPIC_BAD_REGSEL 2 /* IO-APIC REGSEL forgets its value */ +#define QUIRK_IOAPIC_GOOD_REGSEL 3 /* IO-APIC REGSEL behaves properly */ +struct xenpf_platform_quirk { + /* IN variables. */ + uint32_t quirk_id; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_platform_quirk_t); + +#define XENPF_firmware_info 50 +#define XEN_FW_DISK_INFO 1 /* from int 13 AH=08/41/48 */ +#define XEN_FW_DISK_MBR_SIGNATURE 2 /* from MBR offset 0x1b8 */ +#define XEN_FW_VBEDDC_INFO 3 /* from int 10 AX=4f15 */ +struct xenpf_firmware_info { + /* IN variables. */ + uint32_t type; + uint32_t index; + /* OUT variables. */ + union { + struct { + /* Int13, Fn48: Check Extensions Present. */ + uint8_t device; /* %dl: bios device number */ + uint8_t version; /* %ah: major version */ + uint16_t interface_support; /* %cx: support bitmap */ + /* Int13, Fn08: Legacy Get Device Parameters. */ + uint16_t legacy_max_cylinder; /* %cl[7:6]:%ch: max cyl # */ + uint8_t legacy_max_head; /* %dh: max head # */ + uint8_t legacy_sectors_per_track; /* %cl[5:0]: max sector # */ + /* Int13, Fn41: Get Device Parameters (as filled into %ds:%esi). */ + /* NB. First uint16_t of buffer must be set to buffer size. */ + GUEST_HANDLE(void) edd_params; + } disk_info; /* XEN_FW_DISK_INFO */ + struct { + uint8_t device; /* bios device number */ + uint32_t mbr_signature; /* offset 0x1b8 in mbr */ + } disk_mbr_signature; /* XEN_FW_DISK_MBR_SIGNATURE */ + struct { + /* Int10, AX=4F15: Get EDID info. */ + uint8_t capabilities; + uint8_t edid_transfer_time; + /* must refer to 128-byte buffer */ + GUEST_HANDLE(uchar) edid; + } vbeddc_info; /* XEN_FW_VBEDDC_INFO */ + } u; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_firmware_info_t); + +#define XENPF_enter_acpi_sleep 51 +struct xenpf_enter_acpi_sleep { + /* IN variables */ + uint16_t pm1a_cnt_val; /* PM1a control value. */ + uint16_t pm1b_cnt_val; /* PM1b control value. */ + uint32_t sleep_state; /* Which state to enter (Sn). */ + uint32_t flags; /* Must be zero. */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_enter_acpi_sleep_t); + +#define XENPF_change_freq 52 +struct xenpf_change_freq { + /* IN variables */ + uint32_t flags; /* Must be zero. */ + uint32_t cpu; /* Physical cpu. */ + uint64_t freq; /* New frequency (Hz). */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_change_freq_t); + +/* + * Get idle times (nanoseconds since boot) for physical CPUs specified in the + * @cpumap_bitmap with range [0..@cpumap_nr_cpus-1]. The @idletime array is + * indexed by CPU number; only entries with the corresponding @cpumap_bitmap + * bit set are written to. On return, @cpumap_bitmap is modified so that any + * non-existent CPUs are cleared. Such CPUs have their @idletime array entry + * cleared. + */ +#define XENPF_getidletime 53 +struct xenpf_getidletime { + /* IN/OUT variables */ + /* IN: CPUs to interrogate; OUT: subset of IN which are present */ + GUEST_HANDLE(uchar) cpumap_bitmap; + /* IN variables */ + /* Size of cpumap bitmap. */ + uint32_t cpumap_nr_cpus; + /* Must be indexable for every cpu in cpumap_bitmap. */ + GUEST_HANDLE(uint64_t) idletime; + /* OUT variables */ + /* System time when the idletime snapshots were taken. */ + uint64_t now; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_getidletime_t); + +#define XENPF_set_processor_pminfo 54 + +/* ability bits */ +#define XEN_PROCESSOR_PM_CX 1 +#define XEN_PROCESSOR_PM_PX 2 +#define XEN_PROCESSOR_PM_TX 4 + +/* cmd type */ +#define XEN_PM_CX 0 +#define XEN_PM_PX 1 +#define XEN_PM_TX 2 + +/* Px sub info type */ +#define XEN_PX_PCT 1 +#define XEN_PX_PSS 2 +#define XEN_PX_PPC 4 +#define XEN_PX_PSD 8 + +struct xen_power_register { + uint32_t space_id; + uint32_t bit_width; + uint32_t bit_offset; + uint32_t access_size; + uint64_t address; +}; + +struct xen_processor_csd { + uint32_t domain; /* domain number of one dependent group */ + uint32_t coord_type; /* coordination type */ + uint32_t num; /* number of processors in same domain */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_processor_csd); + +struct xen_processor_cx { + struct xen_power_register reg; /* GAS for Cx trigger register */ + uint8_t type; /* cstate value, c0: 0, c1: 1, ... */ + uint32_t latency; /* worst latency (ms) to enter/exit this cstate */ + uint32_t power; /* average power consumption(mW) */ + uint32_t dpcnt; /* number of dependency entries */ + GUEST_HANDLE(xen_processor_csd) dp; /* NULL if no dependency */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_processor_cx); + +struct xen_processor_flags { + uint32_t bm_control:1; + uint32_t bm_check:1; + uint32_t has_cst:1; + uint32_t power_setup_done:1; + uint32_t bm_rld_set:1; +}; + +struct xen_processor_power { + uint32_t count; /* number of C state entries in array below */ + struct xen_processor_flags flags; /* global flags of this processor */ + GUEST_HANDLE(xen_processor_cx) states; /* supported c states */ +}; + +struct xen_pct_register { + uint8_t descriptor; + uint16_t length; + uint8_t space_id; + uint8_t bit_width; + uint8_t bit_offset; + uint8_t reserved; + uint64_t address; +}; + +struct xen_processor_px { + uint64_t core_frequency; /* megahertz */ + uint64_t power; /* milliWatts */ + uint64_t transition_latency; /* microseconds */ + uint64_t bus_master_latency; /* microseconds */ + uint64_t control; /* control value */ + uint64_t status; /* success indicator */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_processor_px); + +struct xen_psd_package { + uint64_t num_entries; + uint64_t revision; + uint64_t domain; + uint64_t coord_type; + uint64_t num_processors; +}; + +struct xen_processor_performance { + uint32_t flags; /* flag for Px sub info type */ + uint32_t platform_limit; /* Platform limitation on freq usage */ + struct xen_pct_register control_register; + struct xen_pct_register status_register; + uint32_t state_count; /* total available performance states */ + GUEST_HANDLE(xen_processor_px) states; + struct xen_psd_package domain_info; + uint32_t shared_type; /* coordination type of this processor */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_processor_performance); + +struct xenpf_set_processor_pminfo { + /* IN variables */ + uint32_t id; /* ACPI CPU ID */ + uint32_t type; /* {XEN_PM_CX, XEN_PM_PX} */ + union { + struct xen_processor_power power;/* Cx: _CST/_CSD */ + struct xen_processor_performance perf; /* Px: _PPC/_PCT/_PSS/_PSD */ + }; +}; +DEFINE_GUEST_HANDLE_STRUCT(xenpf_set_processor_pminfo); + +struct xen_platform_op { + uint32_t cmd; + uint32_t interface_version; /* XENPF_INTERFACE_VERSION */ + union { + struct xenpf_settime settime; + struct xenpf_add_memtype add_memtype; + struct xenpf_del_memtype del_memtype; + struct xenpf_read_memtype read_memtype; + struct xenpf_microcode_update microcode; + struct xenpf_platform_quirk platform_quirk; + struct xenpf_firmware_info firmware_info; + struct xenpf_enter_acpi_sleep enter_acpi_sleep; + struct xenpf_change_freq change_freq; + struct xenpf_getidletime getidletime; + struct xenpf_set_processor_pminfo set_pminfo; + uint8_t pad[128]; + } u; +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_platform_op_t); + +#endif /* __XEN_PUBLIC_PLATFORM_H__ */ diff --git a/include/xen/interface/xen.h b/include/xen/interface/xen.h index 6acd9ce..6a6e914 100644 --- a/include/xen/interface/xen.h +++ b/include/xen/interface/xen.h @@ -492,6 +492,7 @@ struct dom0_vga_console_info { /* These flags are passed in the ''flags'' field of start_info_t. */ #define SIF_PRIVILEGED (1<<0) /* Is the domain privileged? */ #define SIF_INITDOMAIN (1<<1) /* Is this the initial control domain? */ +#define SIF_PM_MASK (0xFF<<8) /* reserve 1 byte for xen-pm options */ typedef uint64_t cpumap_t; -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 6/8] xen/acpi/sleep: Enable ACPI sleep via the __acpi_override_sleep
Provide the registration callback to call in the Xen''s ACPI sleep functionality. This means that during S3/S5 we make a hypercall XENPF_enter_acpi_sleep with the proper PM1A/PM1B registers. Based of Ke Yu''s <ke.yu@intel.com> initial idea. [ From http://xenbits.xensource.com/linux-2.6.18-xen.hg change c68699484a65 ] [v1: Added Copyright and license] [v2: Added check if PM1A/B the 16-bits MSB contain something. The spec only uses 16-bits but might have more in future] Signed-off-by: Liang Tang <liang.tang@oracle.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- arch/x86/include/asm/xen/hypercall.h | 8 ++++ arch/x86/xen/enlighten.c | 3 ++ drivers/xen/Makefile | 2 +- drivers/xen/acpi.c | 65 ++++++++++++++++++++++++++++++++++ include/xen/acpi.h | 58 ++++++++++++++++++++++++++++++ 5 files changed, 135 insertions(+), 1 deletions(-) create mode 100644 drivers/xen/acpi.c create mode 100644 include/xen/acpi.h diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h index 417777d..5728852 100644 --- a/arch/x86/include/asm/xen/hypercall.h +++ b/arch/x86/include/asm/xen/hypercall.h @@ -47,6 +47,7 @@ #include <xen/interface/xen.h> #include <xen/interface/sched.h> #include <xen/interface/physdev.h> +#include <xen/interface/platform.h> /* * The hypercall asms have to meet several constraints: @@ -301,6 +302,13 @@ HYPERVISOR_set_timer_op(u64 timeout) } static inline int +HYPERVISOR_dom0_op(struct xen_platform_op *platform_op) +{ + platform_op->interface_version = XENPF_INTERFACE_VERSION; + return _hypercall1(int, dom0_op, platform_op); +} + +static inline int HYPERVISOR_set_debugreg(int reg, unsigned long value) { return _hypercall2(int, set_debugreg, reg, value); diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 2d69617..9306320 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -42,6 +42,7 @@ #include <xen/page.h> #include <xen/hvm.h> #include <xen/hvc-console.h> +#include <xen/acpi.h> #include <asm/paravirt.h> #include <asm/apic.h> @@ -1276,6 +1277,8 @@ asmlinkage void __init xen_start_kernel(void) /* Make sure ACS will be enabled */ pci_request_acs(); + + xen_acpi_sleep_register(); } diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile index 72bbb27..6539673 100644 --- a/drivers/xen/Makefile +++ b/drivers/xen/Makefile @@ -17,7 +17,7 @@ obj-$(CONFIG_XEN_SYS_HYPERVISOR) += sys-hypervisor.o obj-$(CONFIG_XEN_PLATFORM_PCI) += xen-platform-pci.o obj-$(CONFIG_XEN_TMEM) += tmem.o obj-$(CONFIG_SWIOTLB_XEN) += swiotlb-xen.o -obj-$(CONFIG_XEN_DOM0) += pci.o +obj-$(CONFIG_XEN_DOM0) += pci.o acpi.o obj-$(CONFIG_XEN_PCIDEV_BACKEND) += xen-pciback/ xen-evtchn-y := evtchn.o diff --git a/drivers/xen/acpi.c b/drivers/xen/acpi.c new file mode 100644 index 0000000..ba9a5d2 --- /dev/null +++ b/drivers/xen/acpi.c @@ -0,0 +1,65 @@ +/****************************************************************************** + * acpi.c + * acpi file for domain 0 kernel + * + * Copyright (c) 2011 Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> + * Copyright (c) 2011 Yu Ke ke.yu@intel.com + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include <xen/acpi.h> +#include <xen/interface/platform.h> +#include <asm/xen/hypercall.h> +#include <asm/xen/hypervisor.h> + +int xen_acpi_notify_hypervisor_state(u8 sleep_state, + u32 pm1a_cnt, u32 pm1b_cnt, + bool *skip_rest) +{ + struct xen_platform_op op = { + .cmd = XENPF_enter_acpi_sleep, + .interface_version = XENPF_INTERFACE_VERSION, + .u = { + .enter_acpi_sleep = { + .pm1a_cnt_val = (u16)pm1a_cnt, + .pm1b_cnt_val = (u16)pm1b_cnt, + .sleep_state = sleep_state, + }, + }, + }; + + if ((pm1a_cnt & 0xffff0000) || (pm1b_cnt & 0xffff0000)) { + WARN(1, "Using more than 16bits of PM1A/B 0x%x/0x%x!" + "Email xen-devel@lists.xensource.com Thank you.\n", \ + pm1a_cnt, pm1b_cnt); + return AE_ERROR; + } + + if (skip_rest) + *skip_rest = true; + + return HYPERVISOR_dom0_op(&op); +} diff --git a/include/xen/acpi.h b/include/xen/acpi.h new file mode 100644 index 0000000..c981887 --- /dev/null +++ b/include/xen/acpi.h @@ -0,0 +1,58 @@ +/****************************************************************************** + * acpi.h + * acpi file for domain 0 kernel + * + * Copyright (c) 2011 Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> + * Copyright (c) 2011 Yu Ke <ke.yu@intel.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef _XEN_ACPI_H +#define _XEN_ACPI_H + +#include <linux/types.h> + +#ifdef CONFIG_XEN_DOM0 +#include <asm/xen/hypervisor.h> +#include <xen/xen.h> +#include <linux/acpi.h> + +int xen_acpi_notify_hypervisor_state(u8 sleep_state, + u32 pm1a_cnt, u32 pm1b_cnd, + bool *skip_rest); + +static inline void xen_acpi_sleep_register(void) +{ + if (xen_initial_domain()) + __acpi_override_sleep = xen_acpi_notify_hypervisor_state; +} +#else +static inline void xen_acpi_sleep_register(void) +{ +} +#endif + +#endif /* _XEN_ACPI_H */ -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 7/8] xen/acpi/sleep: Register to the acpi_suspend_lowlevel a callback.
We piggyback on "x86/acpi: Provide registration for acpi_suspend_lowlevel." to register a Xen version of the callback. The callback does not do anything special - except it omits the x86_acpi_suspend_lowlevel. It does that b/c during suspend it tries to save cr8 values (which the hypervisor does not support), and then on resume path the cr3, cr8, idt, and gdt are all resumed which clashes with what the hypervisor has set up for the guest. Signed-off-by: Liang Tang <liang.tang@oracle.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> --- include/xen/acpi.h | 14 +++++++++++++- 1 files changed, 13 insertions(+), 1 deletions(-) diff --git a/include/xen/acpi.h b/include/xen/acpi.h index c981887..18025e0 100644 --- a/include/xen/acpi.h +++ b/include/xen/acpi.h @@ -44,10 +44,22 @@ int xen_acpi_notify_hypervisor_state(u8 sleep_state, u32 pm1a_cnt, u32 pm1b_cnd, bool *skip_rest); +static inline int xen_acpi_suspend_lowlevel(void) +{ + /* + * Xen will save and restore CPU context, so + * we can skip that and just go straight to + * the suspend. + */ + acpi_enter_sleep_state(ACPI_STATE_S3); + return 0; +} static inline void xen_acpi_sleep_register(void) { - if (xen_initial_domain()) + if (xen_initial_domain()) { + acpi_suspend_lowlevel = xen_acpi_suspend_lowlevel; __acpi_override_sleep = xen_acpi_notify_hypervisor_state; + } } #else static inline void xen_acpi_sleep_register(void) -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Sep-29 20:16 UTC
[Xen-devel] [PATCH 8/8] xen/pci:use hypercall PHYSDEVOP_restore_msi_ext to restore MSI/MSI-X vectors
From: Liang Tang <liang.tang@oracle.com> .. to use the new hypercall to restore the vectors for MSI/MSI-X devices. If the new hypercall fail, we will call the old one (PHYSDEVOP_restore_msi). [v1: Attempt only once to make the new hypercall, not everytime] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Liang Tang <liang.tang@oracle.com> --- arch/x86/pci/xen.c | 27 ++++++++++++++++++++++----- include/xen/interface/physdev.h | 8 ++++++++ 2 files changed, 30 insertions(+), 5 deletions(-) diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c index 9eea4ed..4521b05 100644 --- a/arch/x86/pci/xen.c +++ b/arch/x86/pci/xen.c @@ -248,6 +248,8 @@ error: } #ifdef CONFIG_XEN_DOM0 +static bool __read_mostly pci_seg_supported = true; + static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { int ret = 0; @@ -308,12 +310,27 @@ out: static void xen_initdom_restore_msi_irqs(struct pci_dev *dev, int irq) { int ret = 0; - struct physdev_restore_msi restore; - restore.bus = dev->bus->number; - restore.devfn = dev->devfn; - ret = HYPERVISOR_physdev_op(PHYSDEVOP_restore_msi, &restore); - WARN(ret && ret != -ENOSYS, "restore_msi -> %d\n", ret); + if (pci_seg_supported) { + struct physdev_pci_device restore_ext; + + restore_ext.seg = pci_domain_nr(dev->bus); + restore_ext.bus = dev->bus->number; + restore_ext.devfn = dev->devfn; + ret = HYPERVISOR_physdev_op(PHYSDEVOP_restore_msi_ext, + &restore_ext); + if (ret == -ENOSYS) + pci_seg_supported = false; + WARN(ret && ret != -ENOSYS, "restore_msi_ext -> %d\n", ret); + } + if (!pci_seg_supported) { + struct physdev_restore_msi restore; + + restore.bus = dev->bus->number; + restore.devfn = dev->devfn; + ret = HYPERVISOR_physdev_op(PHYSDEVOP_restore_msi, &restore); + WARN(ret && ret != -ENOSYS, "restore_msi -> %d\n", ret); + } } #endif diff --git a/include/xen/interface/physdev.h b/include/xen/interface/physdev.h index 44aefa9..9818456 100644 --- a/include/xen/interface/physdev.h +++ b/include/xen/interface/physdev.h @@ -205,6 +205,14 @@ struct physdev_get_free_pirq { uint32_t pirq; }; +#define PHYSDEVOP_restore_msi_ext 27 +struct physdev_pci_device { + /* IN */ + uint16_t seg; + uint8_t bus; + uint8_t devfn; +}; + /* * Notify that some PIRQ-bound event channels have been unmasked. * ** This command is obsolete since interface version 0x00030202 and is ** -- 1.7.4.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Cihula, Joseph
2011-Sep-29 20:28 UTC
[Xen-devel] RE: [PATCH 2/8] x86, acpi, tboot: Have a ACPI sleep override instead of calling tboot_sleep.
ACK.> -----Original Message----- > From: Konrad Rzeszutek Wilk [mailto:konrad.wilk@oracle.com] > Sent: Thursday, September 29, 2011 1:17 PM > To: linux-kernel@vger.kernel.org; rjw@sisk.pl; tglx@linutronix.de; hpa@zytor.com; x86@kernel.org; > Brown, Len; Cihula, Joseph; Wang, Shane; xen-devel@lists.xensource.com; linux-pm@lists.linux- > foundation.org; tboot-devel@lists.sourceforge.net; linux-acpi@vger.kernel.org; > liang.tang@oracle.com; Yu, Ke; Tian, Kevin; jeremy@goop.org > Cc: Konrad Rzeszutek Wilk > Subject: [PATCH 2/8] x86, acpi, tboot: Have a ACPI sleep override instead of calling tboot_sleep. > > The ACPI suspend path makes a call to tboot_sleep right before it writes the PM1A, PM1B values. We > replace the direct call to tboot via an registration callback similar to __acpi_register_gsi. > > CC: Thomas Gleixner <tglx@linutronix.de> > CC: "H. Peter Anvin" <hpa@zytor.com> > CC: x86@kernel.org > CC: Len Brown <len.brown@intel.com> > CC: Joseph Cihula <joseph.cihula@intel.com> > CC: Shane Wang <shane.wang@intel.com> > CC: xen-devel@lists.xensource.com > CC: linux-pm@lists.linux-foundation.org > CC: tboot-devel@lists.sourceforge.net > CC: linux-acpi@vger.kernel.org > [v1: Added __attribute__ ((unused))] > Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> > --- > arch/x86/include/asm/acpi.h | 4 ++++ > arch/x86/kernel/acpi/boot.c | 4 ++++ > arch/x86/kernel/tboot.c | 14 ++++++++++---- > drivers/acpi/acpica/hwsleep.c | 12 ++++++++++-- > include/linux/tboot.h | 3 ++- > 5 files changed, 30 insertions(+), 7 deletions(-) > > diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 610001d..0a46696 > 100644 > --- a/arch/x86/include/asm/acpi.h > +++ b/arch/x86/include/asm/acpi.h > @@ -98,6 +98,10 @@ void acpi_pic_sci_set_trigger(unsigned int, u16); extern int > (*__acpi_register_gsi)(struct device *dev, u32 gsi, > int trigger, int polarity); > > +extern int (*__acpi_override_sleep)(u8 sleep_state, u32 pm1a_ctrl, > + u32 pm1b_ctrl, bool *skip_rest) \ > + __attribute__ ((unused)); > + > static inline void disable_acpi(void) > { > acpi_disabled = 1; > diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 4558f0d..7f30806 > 100644 > --- a/arch/x86/kernel/acpi/boot.c > +++ b/arch/x86/kernel/acpi/boot.c > @@ -552,6 +552,10 @@ static int acpi_register_gsi_ioapic(struct device *dev, u32 gsi, int > (*__acpi_register_gsi)(struct device *dev, u32 gsi, > int trigger, int polarity) = acpi_register_gsi_pic; > > +int (*__acpi_override_sleep)(u8 sleep_state, u32 pm1a_ctrl, > + u32 pm1b_ctrl, bool *skip_rest) \ > + __attribute__ ((unused)) = NULL; > + > /* > * success: return IRQ number (>=0) > * failure: return < 0 > diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c index e07a2fc..a6c0a30 100644 > --- a/arch/x86/kernel/tboot.c > +++ b/arch/x86/kernel/tboot.c > @@ -42,7 +42,7 @@ > #include <asm/setup.h> > #include <asm/e820.h> > #include <asm/io.h> > - > +#include <linux/acpi.h> > #include "acpi/realmode/wakeup.h" > > /* Global pointer to shared data; NULL means no measured launch. */ @@ -271,7 +271,9 @@ static > void tboot_copy_fadt(const struct acpi_table_fadt *fadt) > offsetof(struct acpi_table_facs, firmware_waking_vector); } > > -void tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control) > + > +int tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control, > + bool *skip_rest) > { > static u32 acpi_shutdown_map[ACPI_S_STATE_COUNT] = { > /* S0,1,2: */ -1, -1, -1, > @@ -280,7 +282,7 @@ void tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control) > /* S5: */ TB_SHUTDOWN_S5 }; > > if (!tboot_enabled()) > - return; > + return AE_OK; > > tboot_copy_fadt(&acpi_gbl_FADT); > tboot->acpi_sinfo.pm1a_cnt_val = pm1a_control; @@ -291,10 +293,12 @@ void tboot_sleep(u8 > sleep_state, u32 pm1a_control, u32 pm1b_control) > if (sleep_state >= ACPI_S_STATE_COUNT || > acpi_shutdown_map[sleep_state] == -1) { > pr_warning("unsupported sleep state 0x%x\n", sleep_state); > - return; > + return AE_ERROR; > } > > tboot_shutdown(acpi_shutdown_map[sleep_state]); > + > + return AE_OK; > } > > static atomic_t ap_wfs_count; > @@ -344,6 +348,8 @@ static __init int tboot_late_init(void) > > atomic_set(&ap_wfs_count, 0); > register_hotcpu_notifier(&tboot_cpu_notifier); > + > + __acpi_override_sleep = tboot_sleep; > return 0; > } > > diff --git a/drivers/acpi/acpica/hwsleep.c b/drivers/acpi/acpica/hwsleep.c index 2ac28bb..31d1198 > 100644 > --- a/drivers/acpi/acpica/hwsleep.c > +++ b/drivers/acpi/acpica/hwsleep.c > @@ -45,7 +45,6 @@ > #include <acpi/acpi.h> > #include "accommon.h" > #include "actables.h" > -#include <linux/tboot.h> > > #define _COMPONENT ACPI_HARDWARE > ACPI_MODULE_NAME("hwsleep") > @@ -343,8 +342,17 @@ acpi_status asmlinkage acpi_enter_sleep_state(u8 sleep_state) > > ACPI_FLUSH_CPU_CACHE(); > > - tboot_sleep(sleep_state, pm1a_control, pm1b_control); > + if (__acpi_override_sleep) { > + bool skip_rest = false; > > + status = __acpi_override_sleep(sleep_state, pm1a_control, > + pm1b_control, &skip_rest); > + > + if (ACPI_FAILURE(status)) > + return_ACPI_STATUS(status); > + if (skip_rest) > + return_ACPI_STATUS(AE_OK); > + } > /* Write #2: Write both SLP_TYP + SLP_EN */ > > status = acpi_hw_write_pm1_control(pm1a_control, pm1b_control); diff --git > a/include/linux/tboot.h b/include/linux/tboot.h index 1dba6ee..1216698 100644 > --- a/include/linux/tboot.h > +++ b/include/linux/tboot.h > @@ -143,7 +143,8 @@ static inline int tboot_enabled(void) > > extern void tboot_probe(void); > extern void tboot_shutdown(u32 shutdown_type); -extern void tboot_sleep(u8 sleep_state, u32 > pm1a_control, u32 pm1b_control); > +extern int tboot_sleep(u8 sleep_state, u32 pm1a_control, u32 pm1b_control, > + bool *skip_rest) __attribute__ ((unused)); > extern struct acpi_table_header *tboot_get_dmar_table( > struct acpi_table_header *dmar_tbl); extern int > tboot_force_iommu(void); > -- > 1.7.4.1_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Cihula, Joseph
2011-Sep-29 20:30 UTC
[Xen-devel] RE: [PATCH 3/8] x86/acpi/sleep: Provide registration for acpi_suspend_lowlevel.
ACK> -----Original Message----- > From: Konrad Rzeszutek Wilk [mailto:konrad.wilk@oracle.com] > Sent: Thursday, September 29, 2011 1:17 PM > To: linux-kernel@vger.kernel.org; rjw@sisk.pl; tglx@linutronix.de; hpa@zytor.com; x86@kernel.org; > Brown, Len; Cihula, Joseph; Wang, Shane; xen-devel@lists.xensource.com; linux-pm@lists.linux- > foundation.org; tboot-devel@lists.sourceforge.net; linux-acpi@vger.kernel.org; > liang.tang@oracle.com; Yu, Ke; Tian, Kevin; jeremy@goop.org > Cc: Konrad Rzeszutek Wilk > Subject: [PATCH 3/8] x86/acpi/sleep: Provide registration for acpi_suspend_lowlevel. > > From: Liang Tang <liang.tang@oracle.com> > > Which by default will be x86_acpi_suspend_lowlevel. > This registration allows us to register another callback if there is a need to use another > platform specific callback. > > CC: Thomas Gleixner <tglx@linutronix.de> > CC: "H. Peter Anvin" <hpa@zytor.com> > CC: x86@kernel.org > CC: Len Brown <len.brown@intel.com> > CC: Joseph Cihula <joseph.cihula@intel.com> > CC: Shane Wang <shane.wang@intel.com> > CC: linux-pm@lists.linux-foundation.org > CC: linux-acpi@vger.kernel.org > CC: Len Brown <len.brown@intel.com> > Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> > Signed-off-by: Liang Tang <liang.tang@oracle.com> > --- > arch/x86/include/asm/acpi.h | 2 +- > arch/x86/kernel/acpi/boot.c | 2 ++ > arch/x86/kernel/acpi/sleep.c | 4 ++-- > arch/x86/kernel/acpi/sleep.h | 2 ++ > drivers/acpi/sleep.c | 2 ++ > 5 files changed, 9 insertions(+), 3 deletions(-) > > diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 0a46696..9b538dc > 100644 > --- a/arch/x86/include/asm/acpi.h > +++ b/arch/x86/include/asm/acpi.h > @@ -119,7 +119,7 @@ static inline void acpi_disable_pci(void) } > > /* Low-level suspend routine. */ > -extern int acpi_suspend_lowlevel(void); > +extern int (*acpi_suspend_lowlevel)(void); > > extern const unsigned char acpi_wakeup_code[]; #define acpi_wakeup_address > (__pa(TRAMPOLINE_SYM(acpi_wakeup_code))) > diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 7f30806..ddd081b > 100644 > --- a/arch/x86/kernel/acpi/boot.c > +++ b/arch/x86/kernel/acpi/boot.c > @@ -44,6 +44,7 @@ > #include <asm/mpspec.h> > #include <asm/smp.h> > > +#include "sleep.h" /* To include x86_acpi_suspend_lowlevel */ > static int __initdata acpi_force = 0; > u32 acpi_rsdt_forced; > int acpi_disabled; > @@ -556,6 +557,7 @@ int (*__acpi_override_sleep)(u8 sleep_state, u32 pm1a_ctrl, > u32 pm1b_ctrl, bool *skip_rest) \ > __attribute__ ((unused)) = NULL; > > +int (*acpi_suspend_lowlevel)(void) = x86_acpi_suspend_lowlevel; > /* > * success: return IRQ number (>=0) > * failure: return < 0 > diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 103b6ab..4d2d0b1 > 100644 > --- a/arch/x86/kernel/acpi/sleep.c > +++ b/arch/x86/kernel/acpi/sleep.c > @@ -25,12 +25,12 @@ static char temp_stack[4096]; #endif > > /** > - * acpi_suspend_lowlevel - save kernel state > + * x86_acpi_suspend_lowlevel - save kernel state > * > * Create an identity mapped page table and copy the wakeup routine to > * low memory. > */ > -int acpi_suspend_lowlevel(void) > +int x86_acpi_suspend_lowlevel(void) > { > struct wakeup_header *header; > /* address in low memory of the wakeup routine. */ diff --git a/arch/x86/kernel/acpi/sleep.h > b/arch/x86/kernel/acpi/sleep.h index 416d4be..4d3feb5 100644 > --- a/arch/x86/kernel/acpi/sleep.h > +++ b/arch/x86/kernel/acpi/sleep.h > @@ -13,3 +13,5 @@ extern unsigned long acpi_copy_wakeup_routine(unsigned long); extern void > wakeup_long64(void); > > extern void do_suspend_lowlevel(void); > + > +extern int x86_acpi_suspend_lowlevel(void); > diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index 3ed80b2..3570c00 100644 > --- a/drivers/acpi/sleep.c > +++ b/drivers/acpi/sleep.c > @@ -254,6 +254,8 @@ static int acpi_suspend_enter(suspend_state_t pm_state) > break; > > case ACPI_STATE_S3: > + if (!acpi_suspend_lowlevel) > + return -ENODEV; > error = acpi_suspend_lowlevel(); > if (error) > return error; > -- > 1.7.4.1_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Oct-11 20:08 UTC
[Xen-devel] Re: [PATCH v2] ACPI S3 to work under Xen.
On Thu, Sep 29, 2011 at 04:16:46PM -0400, Konrad Rzeszutek Wilk wrote:> Attached is an [v2] set of patches to enable S3 to work with the Xen hypervisor. > > Changes since the RFC posting [http://comments.gmane.org/gmane.linux.acpi.devel/50701] by > Liang Tang: > - Per review comments added: __unused__ attribute, support for PM1A/B if more than 16-bit, > copyright/license. > - Added support for PHYSDEVOP_restore_msi_ext call.Rafael, Len, Any chance you could take look at the patches - the parts that touch the ACPI generic components are couple of lines? Much appreciated. Joseph was OK with the tboot sections. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel