x86, UV: Clean up uv_tlb.c

SGI UV's uv_tlb.c driver has become rather hard to read, with overly large
functions, non-standard coding style and (way) too long variable, constant
and function names and non-obvious code flow sequences.

This patch improves the readability and maintainability of the driver
significantly, by doing the following strict code cleanups with no side
effects:

 - Split long functions into shorter logical functions.

 - Shortened some variable and structure member names.

 - Added special functions for reads and writes of MMR regs with
   very long names.

 - Added the 'tunables' table to shortened tunables_write().

 - Added the 'stat_description' table to shorten uv_ptc_proc_write().

 - Pass fewer 'stat' arguments where it can be derived from the 'bcp'
   argument.

 - Function definitions consistent on one line, and inline in few (short) cases.

 - Moved some small structures and an atomic inline function to the header file.

 - Moved some local variables to the blocks where they are used.

 - Updated the copyright date.

 - Shortened uv_write_global_mmr64() etc. using some aliasing; no
   line breaks. Renamed many uv_.. functions that are not exported.

 - Aligned structure fields.
    [ note that not all structures are aligned the same way though; I'd like
      to keep the extensive commenting in some of them. ]

 - Shortened some long structure names.

 - Standard pass/fail exit from init_per_cpu()

 - Vertical alignment for mass initializations.

 - More separation between blocks of code.

Tested on a 16-processor Altix UV.

Signed-off-by: Cliff Wickman <cpw@sgi.com>
Cc: penberg@kernel.org
Link: http://lkml.kernel.org/r/E1QOw12-0004MN-Lp@eag09.americas.sgi.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Cliff Wickman 2011-05-24 13:07:36 -05:00 committed by Ingo Molnar
parent 2a919596c1
commit f073cc8f39
2 changed files with 1106 additions and 906 deletions

View File

@ -5,7 +5,7 @@
*
* SGI UV Broadcast Assist Unit definitions
*
* Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved.
* Copyright (C) 2008-2011 Silicon Graphics, Inc. All rights reserved.
*/
#ifndef _ASM_X86_UV_UV_BAU_H
@ -35,9 +35,9 @@
#define MAX_CPUS_PER_UVHUB 64
#define MAX_CPUS_PER_SOCKET 32
#define UV_ADP_SIZE 64 /* hardware-provided max. */
#define UV_CPUS_PER_ACT_STATUS 32 /* hardware-provided max. */
#define UV_ITEMS_PER_DESCRIPTOR 8
#define ADP_SZ 64 /* hardware-provided max. */
#define UV_CPUS_PER_AS 32 /* hardware-provided max. */
#define ITEMS_PER_DESC 8
/* the 'throttle' to prevent the hardware stay-busy bug */
#define MAX_BAU_CONCURRENT 3
#define UV_ACT_STATUS_MASK 0x3
@ -48,7 +48,7 @@
#define UV2_NET_ENDPOINT_INTD 0x28
#define UV_NET_ENDPOINT_INTD (is_uv1_hub() ? \
UV1_NET_ENDPOINT_INTD : UV2_NET_ENDPOINT_INTD)
#define UV_DESC_BASE_PNODE_SHIFT 49
#define UV_DESC_PSHIFT 49
#define UV_PAYLOADQ_PNODE_SHIFT 49
#define UV_PTC_BASENAME "sgi_uv/ptc_statistics"
#define UV_BAU_BASENAME "sgi_uv/bau_tunables"
@ -56,7 +56,8 @@
#define UV_BAU_TUNABLES_FILE "bau_tunables"
#define WHITESPACE " \t\n"
#define uv_physnodeaddr(x) ((__pa((unsigned long)(x)) & uv_mmask))
#define cpubit_isset(cpu, bau_local_cpumask) \
test_bit((cpu), (bau_local_cpumask).bits)
/* [19:16] SOFT_ACK timeout period 19: 1 is urgency 7 17:16 1 is multiplier */
/*
@ -84,13 +85,25 @@
#define BAU_TRANS_SHIFT 40
#define BAU_TRANS_MASK 0x3f
/*
* shorten some awkward names
*/
#define AS_PUSH_SHIFT UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_SHFT
#define SOFTACK_MSHIFT UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT
#define SOFTACK_PSHIFT UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT
#define SOFTACK_TIMEOUT_PERIOD UV_INTD_SOFT_ACK_TIMEOUT_PERIOD
#define write_gmmr uv_write_global_mmr64
#define write_lmmr uv_write_local_mmr
#define read_lmmr uv_read_local_mmr
#define read_gmmr uv_read_global_mmr64
/*
* bits in UVH_LB_BAU_SB_ACTIVATION_STATUS_0/1
*/
#define DESC_STATUS_IDLE 0
#define DESC_STATUS_ACTIVE 1
#define DESC_STATUS_DESTINATION_TIMEOUT 2
#define DESC_STATUS_SOURCE_TIMEOUT 3
#define DS_IDLE 0
#define DS_ACTIVE 1
#define DS_DESTINATION_TIMEOUT 2
#define DS_SOURCE_TIMEOUT 3
/*
* bits put together from HRP_LB_BAU_SB_ACTIVATION_STATUS_0/1/2
* values 1 and 5 will not occur
@ -149,9 +162,16 @@
/*
* tuning the action when the numalink network is extremely delayed
*/
#define CONGESTED_RESPONSE_US 1000 /* 'long' response time, in microseconds */
#define CONGESTED_REPS 10 /* long delays averaged over this many broadcasts */
#define CONGESTED_PERIOD 30 /* time for the bau to be disabled, in seconds */
#define CONGESTED_RESPONSE_US 1000 /* 'long' response time, in
microseconds */
#define CONGESTED_REPS 10 /* long delays averaged over
this many broadcasts */
#define CONGESTED_PERIOD 30 /* time for the bau to be
disabled, in seconds */
/* see msg_type: */
#define MSG_NOOP 0
#define MSG_REGULAR 1
#define MSG_RETRY 2
/*
* Distribution: 32 bytes (256 bits) (bytes 0-0x1f of descriptor)
@ -163,7 +183,7 @@
* 'base_dest_nasid' field of the header corresponds to the
* destination nodeID associated with that specified bit.
*/
struct bau_target_uvhubmask {
struct bau_targ_hubmask {
unsigned long bits[BITS_TO_LONGS(UV_DISTRIBUTION_SIZE)];
};
@ -194,8 +214,8 @@ struct bau_local_cpumask {
* The payload is software-defined for INTD transactions
*/
struct bau_msg_payload {
unsigned long address; /* signifies a page or all TLB's
of the cpu */
unsigned long address; /* signifies a page or all
TLB's of the cpu */
/* 64 bits */
unsigned short sending_cpu; /* filled in by sender */
/* 16 bits */
@ -212,8 +232,8 @@ struct bau_msg_payload {
struct bau_msg_header {
unsigned int dest_subnodeid:6; /* must be 0x10, for the LB */
/* bits 5:0 */
unsigned int base_dest_nasid:15; /* nasid of the */
/* bits 20:6 */ /* first bit in uvhub map */
unsigned int base_dest_nasid:15; /* nasid of the first bit */
/* bits 20:6 */ /* in uvhub map */
unsigned int command:8; /* message type */
/* bits 28:21 */
/* 0x38: SN3net EndPoint Message */
@ -225,20 +245,24 @@ struct bau_msg_header {
/* Suppl_A is 56-41 */
unsigned int sequence:16; /* message sequence number */
/* bits 56:41 */ /* becomes bytes 16-17 of msg */
/* Address field (96:57) is never used as an
address (these are address bits 42:3) */
/* Address field (96:57) is
never used as an address
(these are address bits
42:3) */
unsigned int rsvd_3:1; /* must be zero */
/* bit 57 */
/* address bits 27:4 are payload */
/* these next 24 (58-81) bits become bytes 12-14 of msg */
/* bits 65:58 land in byte 12 */
unsigned int replied_to:1;/* sent as 0 by the source to byte 12 */
unsigned int replied_to:1; /* sent as 0 by the source to
byte 12 */
/* bit 58 */
unsigned int msg_type:3; /* software type of the message*/
unsigned int msg_type:3; /* software type of the
message */
/* bits 61:59 */
unsigned int canceled:1; /* message canceled, resource to be freed*/
unsigned int canceled:1; /* message canceled, resource
is to be freed*/
/* bit 62 */
unsigned int payload_1a:1; /* not currently used */
/* bit 63 */
@ -259,39 +283,38 @@ struct bau_msg_header {
unsigned int rsvd_4:7; /* must be zero */
/* bits 88:82 */
unsigned int sw_ack_flag:1;/* software acknowledge flag */
unsigned int swack_flag:1; /* software acknowledge flag */
/* bit 89 */
/* INTD trasactions at destination are to
wait for software acknowledge */
/* INTD trasactions at
destination are to wait for
software acknowledge */
unsigned int rsvd_5:6; /* must be zero */
/* bits 95:90 */
unsigned int rsvd_6:5; /* must be zero */
/* bits 100:96 */
unsigned int int_both:1;/* if 1, interrupt both sockets on the uvhub */
unsigned int int_both:1; /* if 1, interrupt both sockets
on the uvhub */
/* bit 101*/
unsigned int fairness:3; /* usually zero */
/* bits 104:102 */
unsigned int multilevel:1; /* multi-level multicast format */
unsigned int multilevel:1; /* multi-level multicast
format */
/* bit 105 */
/* 0 for TLB: endpoint multi-unicast messages */
unsigned int chaining:1;/* next descriptor is part of this activation*/
unsigned int chaining:1; /* next descriptor is part of
this activation*/
/* bit 106 */
unsigned int rsvd_7:21; /* must be zero */
/* bits 127:107 */
};
/* see msg_type: */
#define MSG_NOOP 0
#define MSG_REGULAR 1
#define MSG_RETRY 2
/*
* The activation descriptor:
* The format of the message to send, plus all accompanying control
* Should be 64 bytes
*/
struct bau_desc {
struct bau_target_uvhubmask distribution;
struct bau_targ_hubmask distribution;
/*
* message template, consisting of header and payload:
*/
@ -315,55 +338,47 @@ struct bau_desc {
* are 32 bytes (2 micropackets) (256 bits) in length, but contain only 17
* bytes of usable data, including the sw ack vector in byte 15 (bits 127:120)
* (12 bytes come from bau_msg_payload, 3 from payload_1, 2 from
* sw_ack_vector and payload_2)
* swack_vec and payload_2)
* "Enabling Software Acknowledgment mode (see Section 4.3.3 Software
* Acknowledge Processing) also selects 32 byte (17 bytes usable) payload
* operation."
*/
struct bau_payload_queue_entry {
struct bau_pq_entry {
unsigned long address; /* signifies a page or all TLB's
of the cpu */
/* 64 bits, bytes 0-7 */
unsigned short sending_cpu; /* cpu that sent the message */
/* 16 bits, bytes 8-9 */
unsigned short acknowledge_count; /* filled in by destination */
/* 16 bits, bytes 10-11 */
/* these next 3 bytes come from bits 58-81 of the message header */
unsigned short replied_to:1; /* sent as 0 by the source */
unsigned short msg_type:3; /* software message type */
unsigned short canceled:1; /* sent as 0 by the source */
unsigned short unused1:3; /* not currently using */
/* byte 12 */
unsigned char unused2a; /* not currently using */
/* byte 13 */
unsigned char unused2; /* not currently using */
/* byte 14 */
unsigned char sw_ack_vector; /* filled in by the hardware */
unsigned char swack_vec; /* filled in by the hardware */
/* byte 15 (bits 127:120) */
unsigned short sequence; /* message sequence number */
/* bytes 16-17 */
unsigned char unused4[2]; /* not currently using bytes 18-19 */
/* bytes 18-19 */
int number_of_cpus; /* filled in at destination */
/* 32 bits, bytes 20-23 (aligned) */
unsigned char unused5[8]; /* not using */
/* bytes 24-31 */
};
struct msg_desc {
struct bau_payload_queue_entry *msg;
struct bau_pq_entry *msg;
int msg_slot;
int sw_ack_slot;
struct bau_payload_queue_entry *va_queue_first;
struct bau_payload_queue_entry *va_queue_last;
int swack_slot;
struct bau_pq_entry *queue_first;
struct bau_pq_entry *queue_last;
};
struct reset_args {
@ -375,57 +390,101 @@ struct reset_args {
*/
struct ptc_stats {
/* sender statistics */
unsigned long s_giveup; /* number of fall backs to IPI-style flushes */
unsigned long s_requestor; /* number of shootdown requests */
unsigned long s_giveup; /* number of fall backs to
IPI-style flushes */
unsigned long s_requestor; /* number of shootdown
requests */
unsigned long s_stimeout; /* source side timeouts */
unsigned long s_dtimeout; /* destination side timeouts */
unsigned long s_time; /* time spent in sending side */
unsigned long s_retriesok; /* successful retries */
unsigned long s_ntargcpu; /* total number of cpu's targeted */
unsigned long s_ntargself; /* times the sending cpu was targeted */
unsigned long s_ntarglocals; /* targets of cpus on the local blade */
unsigned long s_ntargremotes; /* targets of cpus on remote blades */
unsigned long s_ntargcpu; /* total number of cpu's
targeted */
unsigned long s_ntargself; /* times the sending cpu was
targeted */
unsigned long s_ntarglocals; /* targets of cpus on the local
blade */
unsigned long s_ntargremotes; /* targets of cpus on remote
blades */
unsigned long s_ntarglocaluvhub; /* targets of the local hub */
unsigned long s_ntargremoteuvhub; /* remotes hubs targeted */
unsigned long s_ntarguvhub; /* total number of uvhubs targeted */
unsigned long s_ntarguvhub16; /* number of times target hubs >= 16*/
unsigned long s_ntarguvhub8; /* number of times target hubs >= 8 */
unsigned long s_ntarguvhub4; /* number of times target hubs >= 4 */
unsigned long s_ntarguvhub2; /* number of times target hubs >= 2 */
unsigned long s_ntarguvhub1; /* number of times target hubs == 1 */
unsigned long s_resets_plug; /* ipi-style resets from plug state */
unsigned long s_resets_timeout; /* ipi-style resets from timeouts */
unsigned long s_busy; /* status stayed busy past s/w timer */
unsigned long s_ntarguvhub; /* total number of uvhubs
targeted */
unsigned long s_ntarguvhub16; /* number of times target
hubs >= 16*/
unsigned long s_ntarguvhub8; /* number of times target
hubs >= 8 */
unsigned long s_ntarguvhub4; /* number of times target
hubs >= 4 */
unsigned long s_ntarguvhub2; /* number of times target
hubs >= 2 */
unsigned long s_ntarguvhub1; /* number of times target
hubs == 1 */
unsigned long s_resets_plug; /* ipi-style resets from plug
state */
unsigned long s_resets_timeout; /* ipi-style resets from
timeouts */
unsigned long s_busy; /* status stayed busy past
s/w timer */
unsigned long s_throttles; /* waits in throttle */
unsigned long s_retry_messages; /* retry broadcasts */
unsigned long s_bau_reenabled; /* for bau enable/disable */
unsigned long s_bau_disabled; /* for bau enable/disable */
/* destination statistics */
unsigned long d_alltlb; /* times all tlb's on this cpu were flushed */
unsigned long d_onetlb; /* times just one tlb on this cpu was flushed */
unsigned long d_multmsg; /* interrupts with multiple messages */
unsigned long d_alltlb; /* times all tlb's on this
cpu were flushed */
unsigned long d_onetlb; /* times just one tlb on this
cpu was flushed */
unsigned long d_multmsg; /* interrupts with multiple
messages */
unsigned long d_nomsg; /* interrupts with no message */
unsigned long d_time; /* time spent on destination side */
unsigned long d_requestee; /* number of messages processed */
unsigned long d_retries; /* number of retry messages processed */
unsigned long d_canceled; /* number of messages canceled by retries */
unsigned long d_nocanceled; /* retries that found nothing to cancel */
unsigned long d_resets; /* number of ipi-style requests processed */
unsigned long d_rcanceled; /* number of messages canceled by resets */
unsigned long d_time; /* time spent on destination
side */
unsigned long d_requestee; /* number of messages
processed */
unsigned long d_retries; /* number of retry messages
processed */
unsigned long d_canceled; /* number of messages canceled
by retries */
unsigned long d_nocanceled; /* retries that found nothing
to cancel */
unsigned long d_resets; /* number of ipi-style requests
processed */
unsigned long d_rcanceled; /* number of messages canceled
by resets */
};
struct tunables {
int *tunp;
int deflt;
};
struct hub_and_pnode {
short uvhub;
short pnode;
};
struct socket_desc {
short num_cpus;
short cpu_number[MAX_CPUS_PER_SOCKET];
};
struct uvhub_desc {
unsigned short socket_mask;
short num_cpus;
short uvhub;
short pnode;
struct socket_desc socket[2];
};
/*
* one per-cpu; to locate the software tables
*/
struct bau_control {
struct bau_desc *descriptor_base;
struct bau_payload_queue_entry *va_queue_first;
struct bau_payload_queue_entry *va_queue_last;
struct bau_payload_queue_entry *bau_msg_head;
struct bau_pq_entry *queue_first;
struct bau_pq_entry *queue_last;
struct bau_pq_entry *bau_msg_head;
struct bau_control *uvhub_master;
struct bau_control *socket_master;
struct ptc_stats *statp;
@ -452,35 +511,105 @@ struct bau_control {
spinlock_t uvhub_lock;
spinlock_t queue_lock;
/* tunables */
int max_bau_concurrent;
int max_bau_concurrent_constant;
int max_concurr;
int max_concurr_const;
int plugged_delay;
int plugsb4reset;
int timeoutsb4reset;
int ipi_reset_limit;
int complete_threshold;
int congested_response_us;
int congested_reps;
int congested_period;
int cong_response_us;
int cong_reps;
int cong_period;
cycles_t period_time;
long period_requests;
struct hub_and_pnode *target_hub_and_pnode;
struct hub_and_pnode *thp;
};
static inline int bau_uvhub_isset(int uvhub, struct bau_target_uvhubmask *dstp)
static unsigned long read_mmr_uv2_status(void)
{
return read_lmmr(UV2H_LB_BAU_SB_ACTIVATION_STATUS_2);
}
static void write_mmr_data_broadcast(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_BAU_DATA_BROADCAST, mmr_image);
}
static void write_mmr_descriptor_base(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, mmr_image);
}
static void write_mmr_activation(unsigned long index)
{
write_lmmr(UVH_LB_BAU_SB_ACTIVATION_CONTROL, index);
}
static void write_gmmr_activation(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_LB_BAU_SB_ACTIVATION_CONTROL, mmr_image);
}
static void write_mmr_payload_first(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST, mmr_image);
}
static void write_mmr_payload_tail(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL, mmr_image);
}
static void write_mmr_payload_last(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST, mmr_image);
}
static void write_mmr_misc_control(int pnode, unsigned long mmr_image)
{
write_gmmr(pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image);
}
static unsigned long read_mmr_misc_control(int pnode)
{
return read_gmmr(pnode, UVH_LB_BAU_MISC_CONTROL);
}
static void write_mmr_sw_ack(unsigned long mr)
{
uv_write_local_mmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, mr);
}
static unsigned long read_mmr_sw_ack(void)
{
return read_lmmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE);
}
static unsigned long read_gmmr_sw_ack(int pnode)
{
return read_gmmr(pnode, UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE);
}
static void write_mmr_data_config(int pnode, unsigned long mr)
{
uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, mr);
}
static inline int bau_uvhub_isset(int uvhub, struct bau_targ_hubmask *dstp)
{
return constant_test_bit(uvhub, &dstp->bits[0]);
}
static inline void bau_uvhub_set(int pnode, struct bau_target_uvhubmask *dstp)
static inline void bau_uvhub_set(int pnode, struct bau_targ_hubmask *dstp)
{
__set_bit(pnode, &dstp->bits[0]);
}
static inline void bau_uvhubs_clear(struct bau_target_uvhubmask *dstp,
static inline void bau_uvhubs_clear(struct bau_targ_hubmask *dstp,
int nbits)
{
bitmap_zero(&dstp->bits[0], nbits);
}
static inline int bau_uvhub_weight(struct bau_target_uvhubmask *dstp)
static inline int bau_uvhub_weight(struct bau_targ_hubmask *dstp)
{
return bitmap_weight((unsigned long *)&dstp->bits[0],
UV_DISTRIBUTION_SIZE);
@ -491,9 +620,6 @@ static inline void bau_cpubits_clear(struct bau_local_cpumask *dstp, int nbits)
bitmap_zero(&dstp->bits, nbits);
}
#define cpubit_isset(cpu, bau_local_cpumask) \
test_bit((cpu), (bau_local_cpumask).bits)
extern void uv_bau_message_intr1(void);
extern void uv_bau_timeout_intr1(void);
@ -501,7 +627,7 @@ struct atomic_short {
short counter;
};
/**
/*
* atomic_read_short - read a short atomic variable
* @v: pointer of type atomic_short
*
@ -512,14 +638,14 @@ static inline int atomic_read_short(const struct atomic_short *v)
return v->counter;
}
/**
* atomic_add_short_return - add and return a short int
/*
* atom_asr - add and return a short int
* @i: short value to add
* @v: pointer of type atomic_short
*
* Atomically adds @i to @v and returns @i + @v
*/
static inline int atomic_add_short_return(short i, struct atomic_short *v)
static inline int atom_asr(short i, struct atomic_short *v)
{
short __i = i;
asm volatile(LOCK_PREFIX "xaddw %0, %1"
@ -528,4 +654,26 @@ static inline int atomic_add_short_return(short i, struct atomic_short *v)
return i + __i;
}
/*
* conditionally add 1 to *v, unless *v is >= u
* return 0 if we cannot add 1 to *v because it is >= u
* return 1 if we can add 1 to *v because it is < u
* the add is atomic
*
* This is close to atomic_add_unless(), but this allows the 'u' value
* to be lowered below the current 'v'. atomic_add_unless can only stop
* on equal.
*/
static inline int atomic_inc_unless_ge(spinlock_t *lock, atomic_t *v, int u)
{
spin_lock(lock);
if (atomic_read(v) >= u) {
spin_unlock(lock);
return 0;
}
atomic_inc(v);
spin_unlock(lock);
return 1;
}
#endif /* _ASM_X86_UV_UV_BAU_H */

File diff suppressed because it is too large Load Diff