2005-04-17 05:20:36 +07:00
|
|
|
/*
|
|
|
|
* lppaca.h
|
|
|
|
* Copyright (C) 2001 Mike Corrigan IBM Corporation
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation; either version 2 of the License, or
|
|
|
|
* (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program; if not, write to the Free Software
|
|
|
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
|
|
|
*/
|
2005-11-09 09:38:01 +07:00
|
|
|
#ifndef _ASM_POWERPC_LPPACA_H
|
|
|
|
#define _ASM_POWERPC_LPPACA_H
|
2005-12-17 04:43:46 +07:00
|
|
|
#ifdef __KERNEL__
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-06-03 04:17:45 +07:00
|
|
|
/* These definitions relate to hypervisors that only exist when using
|
|
|
|
* a server type processor
|
|
|
|
*/
|
|
|
|
#ifdef CONFIG_PPC_BOOK3S
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
//=============================================================================
|
|
|
|
//
|
|
|
|
// This control block contains the data that is shared between the
|
|
|
|
// hypervisor (PLIC) and the OS.
|
|
|
|
//
|
|
|
|
//
|
|
|
|
//----------------------------------------------------------------------------
|
2006-08-07 13:19:19 +07:00
|
|
|
#include <linux/cache.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <asm/types.h>
|
2006-08-07 13:19:19 +07:00
|
|
|
#include <asm/mmu.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2006-01-13 06:26:42 +07:00
|
|
|
/* The Hypervisor barfs if the lppaca crosses a page boundary. A 1k
|
|
|
|
* alignment is sufficient to prevent this */
|
2006-01-18 08:00:05 +07:00
|
|
|
struct lppaca {
|
2005-04-17 05:20:36 +07:00
|
|
|
//=============================================================================
|
|
|
|
// CACHE_LINE_1 0x0000 - 0x007F Contains read-only data
|
|
|
|
// NOTE: The xDynXyz fields are fields that will be dynamically changed by
|
|
|
|
// PLIC when preparing to bring a processor online or when dispatching a
|
|
|
|
// virtual processor!
|
|
|
|
//=============================================================================
|
|
|
|
u32 desc; // Eye catcher 0xD397D781 x00-x03
|
|
|
|
u16 size; // Size of this struct x04-x05
|
|
|
|
u16 reserved1; // Reserved x06-x07
|
|
|
|
u16 reserved2:14; // Reserved x08-x09
|
|
|
|
u8 shared_proc:1; // Shared processor indicator ...
|
|
|
|
u8 secondary_thread:1; // Secondary thread indicator ...
|
|
|
|
volatile u8 dyn_proc_status:8; // Dynamic Status of this proc x0A-x0A
|
|
|
|
u8 secondary_thread_count; // Secondary thread count x0B-x0B
|
|
|
|
volatile u16 dyn_hv_phys_proc_index;// Dynamic HV Physical Proc Index0C-x0D
|
|
|
|
volatile u16 dyn_hv_log_proc_index;// Dynamic HV Logical Proc Indexx0E-x0F
|
|
|
|
u32 decr_val; // Value for Decr programming x10-x13
|
|
|
|
u32 pmc_val; // Value for PMC regs x14-x17
|
|
|
|
volatile u32 dyn_hw_node_id; // Dynamic Hardware Node id x18-x1B
|
|
|
|
volatile u32 dyn_hw_proc_id; // Dynamic Hardware Proc Id x1C-x1F
|
|
|
|
volatile u32 dyn_pir; // Dynamic ProcIdReg value x20-x23
|
|
|
|
u32 dsei_data; // DSEI data x24-x27
|
|
|
|
u64 sprg3; // SPRG3 value x28-x2F
|
2010-12-01 19:31:15 +07:00
|
|
|
u8 reserved3[40]; // Reserved x30-x57
|
|
|
|
volatile u8 vphn_assoc_counts[8]; // Virtual processor home node
|
|
|
|
// associativity change counters x58-x5F
|
|
|
|
u8 reserved4[32]; // Reserved x60-x7F
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
//=============================================================================
|
|
|
|
// CACHE_LINE_2 0x0080 - 0x00FF Contains local read-write data
|
|
|
|
//=============================================================================
|
|
|
|
// This Dword contains a byte for each type of interrupt that can occur.
|
|
|
|
// The IPI is a count while the others are just a binary 1 or 0.
|
|
|
|
union {
|
|
|
|
u64 any_int;
|
|
|
|
struct {
|
|
|
|
u16 reserved; // Reserved - cleared by #mpasmbl
|
|
|
|
u8 xirr_int; // Indicates xXirrValue is valid or Immed IO
|
|
|
|
u8 ipi_cnt; // IPI Count
|
|
|
|
u8 decr_int; // DECR interrupt occurred
|
|
|
|
u8 pdc_int; // PDC interrupt occurred
|
|
|
|
u8 quantum_int; // Interrupt quantum reached
|
|
|
|
u8 old_plic_deferred_ext_int; // Old PLIC has a deferred XIRR pending
|
|
|
|
} fields;
|
|
|
|
} int_dword;
|
|
|
|
|
|
|
|
// Whenever any fields in this Dword are set then PLIC will defer the
|
|
|
|
// processing of external interrupts. Note that PLIC will store the
|
|
|
|
// XIRR directly into the xXirrValue field so that another XIRR will
|
|
|
|
// not be presented until this one clears. The layout of the low
|
|
|
|
// 4-bytes of this Dword is upto SLIC - PLIC just checks whether the
|
|
|
|
// entire Dword is zero or not. A non-zero value in the low order
|
|
|
|
// 2-bytes will result in SLIC being granted the highest thread
|
|
|
|
// priority upon return. A 0 will return to SLIC as medium priority.
|
|
|
|
u64 plic_defer_ints_area; // Entire Dword
|
|
|
|
|
|
|
|
// Used to pass the real SRR0/1 from PLIC to SLIC as well as to
|
|
|
|
// pass the target SRR0/1 from SLIC to PLIC on a SetAsrAndRfid.
|
|
|
|
u64 saved_srr0; // Saved SRR0 x10-x17
|
|
|
|
u64 saved_srr1; // Saved SRR1 x18-x1F
|
|
|
|
|
|
|
|
// Used to pass parms from the OS to PLIC for SetAsrAndRfid
|
|
|
|
u64 saved_gpr3; // Saved GPR3 x20-x27
|
|
|
|
u64 saved_gpr4; // Saved GPR4 x28-x2F
|
2009-10-30 02:22:48 +07:00
|
|
|
union {
|
|
|
|
u64 saved_gpr5; /* Saved GPR5 x30-x37 */
|
|
|
|
struct {
|
|
|
|
u8 cede_latency_hint; /* x30 */
|
|
|
|
u8 reserved[7]; /* x31-x36 */
|
|
|
|
} fields;
|
|
|
|
} gpr5_dword;
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-03-12 00:55:52 +07:00
|
|
|
u8 dtl_enable_mask; // Dispatch Trace Log mask x38-x38
|
2007-06-08 04:27:11 +07:00
|
|
|
u8 donate_dedicated_cpu; // Donate dedicated CPU cycles x39-x39
|
2005-04-17 05:20:36 +07:00
|
|
|
u8 fpregs_in_use; // FP regs in use x3A-x3A
|
|
|
|
u8 pmcregs_in_use; // PMC regs in use x3B-x3B
|
|
|
|
volatile u32 saved_decr; // Saved Decr Value x3C-x3F
|
|
|
|
volatile u64 emulated_time_base;// Emulated TB for this thread x40-x47
|
|
|
|
volatile u64 cur_plic_latency; // Unaccounted PLIC latency x48-x4F
|
|
|
|
u64 tot_plic_latency; // Accumulated PLIC latency x50-x57
|
|
|
|
u64 wait_state_cycles; // Wait cycles for this proc x58-x5F
|
|
|
|
u64 end_of_quantum; // TB at end of quantum x60-x67
|
|
|
|
u64 pdc_saved_sprg1; // Saved SPRG1 for PMC int x68-x6F
|
|
|
|
u64 pdc_saved_srr0; // Saved SRR0 for PMC int x70-x77
|
|
|
|
volatile u32 virtual_decr; // Virtual DECR for shared procsx78-x7B
|
|
|
|
u16 slb_count; // # of SLBs to maintain x7C-x7D
|
|
|
|
u8 idle; // Indicate OS is idle x7E
|
2005-09-04 05:55:59 +07:00
|
|
|
u8 vmxregs_in_use; // VMX registers in use x7F
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
|
|
|
|
//=============================================================================
|
2006-08-07 14:34:50 +07:00
|
|
|
// CACHE_LINE_3 0x0100 - 0x017F: This line is shared with other processors
|
2005-04-17 05:20:36 +07:00
|
|
|
//=============================================================================
|
|
|
|
// This is the yield_count. An "odd" value (low bit on) means that
|
|
|
|
// the processor is yielded (either because of an OS yield or a PLIC
|
|
|
|
// preempt). An even value implies that the processor is currently
|
|
|
|
// executing.
|
|
|
|
// NOTE: This value will ALWAYS be zero for dedicated processors and
|
|
|
|
// will NEVER be zero for shared processors (ie, initialized to a 1).
|
|
|
|
volatile u32 yield_count; // PLIC increments each dispatchx00-x03
|
2009-04-01 03:12:44 +07:00
|
|
|
volatile u32 dispersion_count; // dispatch changed phys cpu x04-x07
|
2008-07-24 01:30:58 +07:00
|
|
|
volatile u64 cmo_faults; // CMO page fault count x08-x0F
|
|
|
|
volatile u64 cmo_fault_time; // CMO page fault time x10-x17
|
|
|
|
u8 reserved7[104]; // Reserved x18-x7F
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
//=============================================================================
|
2006-08-07 14:34:50 +07:00
|
|
|
// CACHE_LINE_4-5 0x0180 - 0x027F Contains PMC interrupt data
|
2005-04-17 05:20:36 +07:00
|
|
|
//=============================================================================
|
2009-03-12 00:55:52 +07:00
|
|
|
u32 page_ins; // CMO Hint - # page ins by OS x00-x03
|
2009-03-12 00:55:52 +07:00
|
|
|
u8 reserved8[148]; // Reserved x04-x97
|
|
|
|
volatile u64 dtl_idx; // Dispatch Trace Log head idx x98-x9F
|
|
|
|
u8 reserved9[96]; // Reserved xA0-xFF
|
2006-01-18 08:00:05 +07:00
|
|
|
} __attribute__((__aligned__(0x400)));
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2006-01-13 06:26:42 +07:00
|
|
|
extern struct lppaca lppaca[];
|
|
|
|
|
2010-08-13 03:18:48 +07:00
|
|
|
#define lppaca_of(cpu) (*paca[cpu].lppaca_ptr)
|
2010-08-13 03:18:15 +07:00
|
|
|
|
2006-08-07 13:19:19 +07:00
|
|
|
/*
|
|
|
|
* SLB shadow buffer structure as defined in the PAPR. The save_area
|
|
|
|
* contains adjacent ESID and VSID pairs for each shadowed SLB. The
|
|
|
|
* ESID is stored in the lower 64bits, then the VSID.
|
|
|
|
*/
|
|
|
|
struct slb_shadow {
|
|
|
|
u32 persistent; // Number of persistent SLBs x00-x03
|
|
|
|
u32 buffer_length; // Total shadow buffer length x04-x07
|
|
|
|
u64 reserved; // Alignment x08-x0f
|
|
|
|
struct {
|
|
|
|
u64 esid;
|
|
|
|
u64 vsid;
|
|
|
|
} save_area[SLB_NUM_BOLTED]; // x10-x40
|
|
|
|
} ____cacheline_aligned;
|
|
|
|
|
|
|
|
extern struct slb_shadow slb_shadow[];
|
|
|
|
|
powerpc: Account time using timebase rather than PURR
Currently, when CONFIG_VIRT_CPU_ACCOUNTING is enabled, we use the
PURR register for measuring the user and system time used by
processes, as well as other related times such as hardirq and
softirq times. This turns out to be quite confusing for users
because it means that a program will often be measured as taking
less time when run on a multi-threaded processor (SMT2 or SMT4 mode)
than it does when run on a single-threaded processor (ST mode), even
though the program takes longer to finish. The discrepancy is
accounted for as stolen time, which is also confusing, particularly
when there are no other partitions running.
This changes the accounting to use the timebase instead, meaning that
the reported user and system times are the actual number of real-time
seconds that the program was executing on the processor thread,
regardless of which SMT mode the processor is in. Thus a program will
generally show greater user and system times when run on a
multi-threaded processor than on a single-threaded processor.
On pSeries systems on POWER5 or later processors, we measure the
stolen time (time when this partition wasn't running) using the
hypervisor dispatch trace log. We check for new entries in the
log on every entry from user mode and on every transition from
kernel process context to soft or hard IRQ context (i.e. when
account_system_vtime() gets called). So that we can correctly
distinguish time stolen from user time and time stolen from system
time, without having to check the log on every exit to user mode,
we store separate timestamps for exit to user mode and entry from
user mode.
On systems that have a SPURR (POWER6 and POWER7), we read the SPURR
in account_system_vtime() (as before), and then apportion the SPURR
ticks since the last time we read it between scaled user time and
scaled system time according to the relative proportions of user
time and system time over the same interval. This avoids having to
read the SPURR on every kernel entry and exit. On systems that have
PURR but not SPURR (i.e., POWER5), we do the same using the PURR
rather than the SPURR.
This disables the DTL user interface in /sys/debug/kernel/powerpc/dtl
for now since it conflicts with the use of the dispatch trace log
by the time accounting code.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2010-08-27 02:56:43 +07:00
|
|
|
/*
|
|
|
|
* Layout of entries in the hypervisor's dispatch trace log buffer.
|
|
|
|
*/
|
|
|
|
struct dtl_entry {
|
|
|
|
u8 dispatch_reason;
|
|
|
|
u8 preempt_reason;
|
|
|
|
u16 processor_id;
|
|
|
|
u32 enqueue_to_dispatch_time;
|
|
|
|
u32 ready_to_enqueue_time;
|
|
|
|
u32 waiting_to_ready_time;
|
|
|
|
u64 timebase;
|
|
|
|
u64 fault_addr;
|
|
|
|
u64 srr0;
|
|
|
|
u64 srr1;
|
|
|
|
};
|
|
|
|
|
|
|
|
#define DISPATCH_LOG_BYTES 4096 /* bytes per cpu */
|
|
|
|
#define N_DISPATCH_LOG (DISPATCH_LOG_BYTES / sizeof(struct dtl_entry))
|
|
|
|
|
2010-08-31 08:59:53 +07:00
|
|
|
/*
|
|
|
|
* When CONFIG_VIRT_CPU_ACCOUNTING = y, the cpu accounting code controls
|
|
|
|
* reading from the dispatch trace log. If other code wants to consume
|
|
|
|
* DTL entries, it can set this pointer to a function that will get
|
|
|
|
* called once for each DTL entry that gets processed.
|
|
|
|
*/
|
|
|
|
extern void (*dtl_consumer)(struct dtl_entry *entry, u64 index);
|
|
|
|
|
2009-06-03 04:17:45 +07:00
|
|
|
#endif /* CONFIG_PPC_BOOK3S */
|
2005-12-17 04:43:46 +07:00
|
|
|
#endif /* __KERNEL__ */
|
2005-11-09 09:38:01 +07:00
|
|
|
#endif /* _ASM_POWERPC_LPPACA_H */
|