Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit f244d910 authored by Paolo Bonzini's avatar Paolo Bonzini
Browse files

Merge tag 'kvm-s390-20140130' of... 

Merge tag 'kvm-s390-20140130' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD

Two new features are added by this patch set:
- The floating interrupt controller (flic) that allows us to inject,
  clear and inspect non-vcpu local interrupts. This also gives us an
  opportunity to fix deficiencies in our existing interrupt definitions.
- Support for asynchronous page faults via the pfault mechanism. Testing
  show significant guest performance improvements under host swap.
parents 4f34d683 536336c2

Documentation/virtual/kvm/devices/s390_flic.txt

0 → 100644
+46 −0
Original line number Diff line number Diff line 
FLIC (floating interrupt controller)

====================================



FLIC handles floating (non per-cpu) interrupts, i.e. I/O, service and some

machine check interruptions. All interrupts are stored in a per-vm list of

pending interrupts. FLIC performs operations on this list.



Only one FLIC instance may be instantiated.



FLIC provides support to

- add interrupts (KVM_DEV_FLIC_ENQUEUE)

- inspect currently pending interrupts (KVM_FLIC_GET_ALL_IRQS)

- purge all pending floating interrupts (KVM_DEV_FLIC_CLEAR_IRQS)

- enable/disable for the guest transparent async page faults



Groups:

  KVM_DEV_FLIC_ENQUEUE

    Passes a buffer and length into the kernel which are then injected into

    the list of pending interrupts.

    attr->addr contains the pointer to the buffer and attr->attr contains

    the length of the buffer.

    The format of the data structure kvm_s390_irq as it is copied from userspace

    is defined in usr/include/linux/kvm.h.



  KVM_DEV_FLIC_GET_ALL_IRQS

    Copies all floating interrupts into a buffer provided by userspace.

    When the buffer is too small it returns -ENOMEM, which is the indication

    for userspace to try again with a bigger buffer.

    All interrupts remain pending, i.e. are not deleted from the list of

    currently pending interrupts.

    attr->addr contains the userspace address of the buffer into which all

    interrupt data will be copied.

    attr->attr contains the size of the buffer in bytes.



  KVM_DEV_FLIC_CLEAR_IRQS

    Simply deletes all elements from the list of currently pending floating

    interrupts.  No interrupts are injected into the guest.



  KVM_DEV_FLIC_APF_ENABLE

    Enables async page faults for the guest. So in case of a major page fault

    the host is allowed to handle this async and continues the guest.



  KVM_DEV_FLIC_APF_DISABLE_WAIT

    Disables async page faults for the guest and waits until already pending

    async page faults are done. This is necessary to trigger a completion interrupt

    for every init interrupt before migrating the interrupt list.

arch/s390/include/asm/kvm_host.h

+25 −33
Original line number Diff line number Diff line
@@ -16,6 +16,7 @@ 
#include <linux/hrtimer.h>

#include <linux/interrupt.h>

#include <linux/kvm_host.h>

#include <linux/kvm.h>

#include <asm/debug.h>

#include <asm/cpu.h>
@@ -168,18 +169,6 @@ struct kvm_vcpu_stat { 
	u32 diagnose_9c;

};



struct kvm_s390_io_info {

	__u16        subchannel_id;            /* 0x0b8 */

	__u16        subchannel_nr;            /* 0x0ba */

	__u32        io_int_parm;              /* 0x0bc */

	__u32        io_int_word;              /* 0x0c0 */

};



struct kvm_s390_ext_info {

	__u32 ext_params;

	__u64 ext_params2;

};



#define PGM_OPERATION            0x01

#define PGM_PRIVILEGED_OP	 0x02

#define PGM_EXECUTE              0x03
@@ -188,27 +177,6 @@ struct kvm_s390_ext_info { 
#define PGM_SPECIFICATION        0x06

#define PGM_DATA                 0x07



struct kvm_s390_pgm_info {

	__u16 code;

};



struct kvm_s390_prefix_info {

	__u32 address;

};



struct kvm_s390_extcall_info {

	__u16 code;

};



struct kvm_s390_emerg_info {

	__u16 code;

};



struct kvm_s390_mchk_info {

	__u64 cr14;

	__u64 mcic;

};



struct kvm_s390_interrupt_info {

	struct list_head list;

	u64	type;
@@ -246,6 +214,7 @@ struct kvm_s390_float_interrupt { 
	unsigned long idle_mask[(KVM_MAX_VCPUS + sizeof(long) - 1)

				/ sizeof(long)];

	struct kvm_s390_local_interrupt *local_int[KVM_MAX_VCPUS];

	unsigned int irq_count;

};
@@ -262,6 +231,10 @@ struct kvm_vcpu_arch { 
		u64		stidp_data;

	};

	struct gmap *gmap;

#define KVM_S390_PFAULT_TOKEN_INVALID	(-1UL)

	unsigned long pfault_token;

	unsigned long pfault_select;

	unsigned long pfault_compare;

};



struct kvm_vm_stat {
@@ -275,6 +248,7 @@ struct kvm_arch{ 
	struct sca_block *sca;

	debug_info_t *dbf;

	struct kvm_s390_float_interrupt float_int;

	struct kvm_device *flic;

	struct gmap *gmap;

	int css_support;

};
@@ -287,6 +261,24 @@ static inline bool kvm_is_error_hva(unsigned long addr) 
	return IS_ERR_VALUE(addr);

}



#define ASYNC_PF_PER_VCPU	64

struct kvm_vcpu;

struct kvm_async_pf;

struct kvm_arch_async_pf {

	unsigned long pfault_token;

};



bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu);



void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,

			       struct kvm_async_pf *work);



void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu,

				     struct kvm_async_pf *work);



void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,

				 struct kvm_async_pf *work);



extern int sie64a(struct kvm_s390_sie_block *, u64 *);

extern char sie_exit;

#endif

arch/s390/include/asm/pgtable.h

+2 −0
Original line number Diff line number Diff line
@@ -767,6 +767,7 @@ static inline void pgste_set_pte(pte_t *ptep, pte_t entry) 
 * @table: pointer to the page directory

 * @asce: address space control element for gmap page table

 * @crst_list: list of all crst tables used in the guest address space

 * @pfault_enabled: defines if pfaults are applicable for the guest

 */

struct gmap {

	struct list_head list;
@@ -775,6 +776,7 @@ struct gmap { 
	unsigned long asce;

	void *private;

	struct list_head crst_list;

	bool pfault_enabled;

};



/**

arch/s390/include/asm/processor.h

+1 −0
Original line number Diff line number Diff line
@@ -79,6 +79,7 @@ struct thread_struct { 
        unsigned long ksp;              /* kernel stack pointer             */

	mm_segment_t mm_segment;

	unsigned long gmap_addr;	/* address of last gmap fault. */

	unsigned int gmap_pfault;	/* signal of a pending guest pfault */

	struct per_regs per_user;	/* User specified PER registers */

	struct per_event per_event;	/* Cause of the last PER trap */

	unsigned long per_flags;	/* Flags to control debug behavior */

arch/s390/include/uapi/asm/kvm.h

+19 −0
Original line number Diff line number Diff line
@@ -16,6 +16,22 @@ 


#define __KVM_S390



/* Device control API: s390-specific devices */

#define KVM_DEV_FLIC_GET_ALL_IRQS	1

#define KVM_DEV_FLIC_ENQUEUE		2

#define KVM_DEV_FLIC_CLEAR_IRQS		3

#define KVM_DEV_FLIC_APF_ENABLE		4

#define KVM_DEV_FLIC_APF_DISABLE_WAIT	5

/*

 * We can have up to 4*64k pending subchannels + 8 adapter interrupts,

 * as well as up  to ASYNC_PF_PER_VCPU*KVM_MAX_VCPUS pfault done interrupts.

 * There are also sclp and machine checks. This gives us

 * sizeof(kvm_s390_irq)*(4*65536+8+64*64+1+1) = 72 * 266250 = 19170000

 * Lets round up to 8192 pages.

 */

#define KVM_S390_MAX_FLOAT_IRQS	266250

#define KVM_S390_FLIC_MAX_BUFFER	0x2000000



/* for KVM_GET_REGS and KVM_SET_REGS */

struct kvm_regs {

	/* general purpose regs for s390 */
@@ -57,4 +73,7 @@ struct kvm_sync_regs { 
#define KVM_REG_S390_EPOCHDIFF	(KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x2)

#define KVM_REG_S390_CPU_TIMER  (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x3)

#define KVM_REG_S390_CLOCK_COMP (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x4)

#define KVM_REG_S390_PFTOKEN	(KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x5)

#define KVM_REG_S390_PFCOMPARE	(KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x6)

#define KVM_REG_S390_PFSELECT	(KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x7)

#endif