linux/arch/s390/appldata/appldata_mem.c

165 lines
4.1 KiB
C
Raw Normal View History

/*
* Data gathering module for Linux-VM Monitor Stream, Stage 1.
* Collects data related to memory management.
*
* Copyright IBM Corp. 2003, 2006
*
* Author: Gerald Schaefer <gerald.schaefer@de.ibm.com>
*/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/kernel_stat.h>
#include <linux/pagemap.h>
#include <linux/swap.h>
#include <linux/slab.h>
#include <asm/io.h>
#include "appldata.h"
#define P2K(x) ((x) << (PAGE_SHIFT - 10)) /* Converts #Pages to KB */
/*
* Memory data
*
* This is accessed as binary data by z/VM. If changes to it can't be avoided,
* the structure version (product ID, see appldata_base.c) needs to be changed
* as well and all documentation and z/VM applications using it must be
* updated.
*
* The record layout is documented in the Linux for zSeries Device Drivers
* book:
* http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml
*/
struct appldata_mem_data {
u64 timestamp;
u32 sync_count_1; /* after VM collected the record data, */
u32 sync_count_2; /* sync_count_1 and sync_count_2 should be the
same. If not, the record has been updated on
the Linux side while VM was collecting the
(possibly corrupt) data */
u64 pgpgin; /* data read from disk */
u64 pgpgout; /* data written to disk */
u64 pswpin; /* pages swapped in */
u64 pswpout; /* pages swapped out */
u64 sharedram; /* sharedram is currently set to 0 */
u64 totalram; /* total main memory size */
u64 freeram; /* free main memory size */
u64 totalhigh; /* total high memory size */
u64 freehigh; /* free high memory size */
u64 bufferram; /* memory reserved for buffers, free cache */
u64 cached; /* size of (used) cache, w/o buffers */
u64 totalswap; /* total swap space size */
u64 freeswap; /* free swap space */
// New in 2.6 -->
u64 pgalloc; /* page allocations */
u64 pgfault; /* page faults (major+minor) */
u64 pgmajfault; /* page faults (major only) */
// <-- New in 2.6
} __packed;
/*
* appldata_get_mem_data()
*
* gather memory data
*/
static void appldata_get_mem_data(void *data)
{
/*
* don't put large structures on the stack, we are
* serialized through the appldata_ops_mutex and can use static
*/
static struct sysinfo val;
[PATCH] Light weight event counters The remaining counters in page_state after the zoned VM counter patches have been applied are all just for show in /proc/vmstat. They have no essential function for the VM. We use a simple increment of per cpu variables. In order to avoid the most severe races we disable preempt. Preempt does not prevent the race between an increment and an interrupt handler incrementing the same statistics counter. However, that race is exceedingly rare, we may only loose one increment or so and there is no requirement (at least not in kernel) that the vm event counters have to be accurate. In the non preempt case this results in a simple increment for each counter. For many architectures this will be reduced by the compiler to a single instruction. This single instruction is atomic for i386 and x86_64. And therefore even the rare race condition in an interrupt is avoided for both architectures in most cases. The patchset also adds an off switch for embedded systems that allows a building of linux kernels without these counters. The implementation of these counters is through inline code that hopefully results in only a single instruction increment instruction being emitted (i386, x86_64) or in the increment being hidden though instruction concurrency (EPIC architectures such as ia64 can get that done). Benefits: - VM event counter operations usually reduce to a single inline instruction on i386 and x86_64. - No interrupt disable, only preempt disable for the preempt case. Preempt disable can also be avoided by moving the counter into a spinlock. - Handling is similar to zoned VM counters. - Simple and easily extendable. - Can be omitted to reduce memory use for embedded use. References: RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=113512330605497&w=2 RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=114988082814934&w=2 local_t http://marc.theaimsgroup.com/?l=linux-kernel&m=114991748606690&w=2 V2 http://marc.theaimsgroup.com/?t=115014808400007&r=1&w=2 V3 http://marc.theaimsgroup.com/?l=linux-kernel&m=115024767022346&w=2 V4 http://marc.theaimsgroup.com/?l=linux-kernel&m=115047968808926&w=2 Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-30 16:55:45 +08:00
unsigned long ev[NR_VM_EVENT_ITEMS];
struct appldata_mem_data *mem_data;
mem_data = data;
mem_data->sync_count_1++;
[PATCH] Light weight event counters The remaining counters in page_state after the zoned VM counter patches have been applied are all just for show in /proc/vmstat. They have no essential function for the VM. We use a simple increment of per cpu variables. In order to avoid the most severe races we disable preempt. Preempt does not prevent the race between an increment and an interrupt handler incrementing the same statistics counter. However, that race is exceedingly rare, we may only loose one increment or so and there is no requirement (at least not in kernel) that the vm event counters have to be accurate. In the non preempt case this results in a simple increment for each counter. For many architectures this will be reduced by the compiler to a single instruction. This single instruction is atomic for i386 and x86_64. And therefore even the rare race condition in an interrupt is avoided for both architectures in most cases. The patchset also adds an off switch for embedded systems that allows a building of linux kernels without these counters. The implementation of these counters is through inline code that hopefully results in only a single instruction increment instruction being emitted (i386, x86_64) or in the increment being hidden though instruction concurrency (EPIC architectures such as ia64 can get that done). Benefits: - VM event counter operations usually reduce to a single inline instruction on i386 and x86_64. - No interrupt disable, only preempt disable for the preempt case. Preempt disable can also be avoided by moving the counter into a spinlock. - Handling is similar to zoned VM counters. - Simple and easily extendable. - Can be omitted to reduce memory use for embedded use. References: RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=113512330605497&w=2 RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=114988082814934&w=2 local_t http://marc.theaimsgroup.com/?l=linux-kernel&m=114991748606690&w=2 V2 http://marc.theaimsgroup.com/?t=115014808400007&r=1&w=2 V3 http://marc.theaimsgroup.com/?l=linux-kernel&m=115024767022346&w=2 V4 http://marc.theaimsgroup.com/?l=linux-kernel&m=115047968808926&w=2 Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-30 16:55:45 +08:00
all_vm_events(ev);
mem_data->pgpgin = ev[PGPGIN] >> 1;
mem_data->pgpgout = ev[PGPGOUT] >> 1;
mem_data->pswpin = ev[PSWPIN];
mem_data->pswpout = ev[PSWPOUT];
mem_data->pgalloc = ev[PGALLOC_NORMAL];
mem_data->pgalloc += ev[PGALLOC_DMA];
[PATCH] Light weight event counters The remaining counters in page_state after the zoned VM counter patches have been applied are all just for show in /proc/vmstat. They have no essential function for the VM. We use a simple increment of per cpu variables. In order to avoid the most severe races we disable preempt. Preempt does not prevent the race between an increment and an interrupt handler incrementing the same statistics counter. However, that race is exceedingly rare, we may only loose one increment or so and there is no requirement (at least not in kernel) that the vm event counters have to be accurate. In the non preempt case this results in a simple increment for each counter. For many architectures this will be reduced by the compiler to a single instruction. This single instruction is atomic for i386 and x86_64. And therefore even the rare race condition in an interrupt is avoided for both architectures in most cases. The patchset also adds an off switch for embedded systems that allows a building of linux kernels without these counters. The implementation of these counters is through inline code that hopefully results in only a single instruction increment instruction being emitted (i386, x86_64) or in the increment being hidden though instruction concurrency (EPIC architectures such as ia64 can get that done). Benefits: - VM event counter operations usually reduce to a single inline instruction on i386 and x86_64. - No interrupt disable, only preempt disable for the preempt case. Preempt disable can also be avoided by moving the counter into a spinlock. - Handling is similar to zoned VM counters. - Simple and easily extendable. - Can be omitted to reduce memory use for embedded use. References: RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=113512330605497&w=2 RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=114988082814934&w=2 local_t http://marc.theaimsgroup.com/?l=linux-kernel&m=114991748606690&w=2 V2 http://marc.theaimsgroup.com/?t=115014808400007&r=1&w=2 V3 http://marc.theaimsgroup.com/?l=linux-kernel&m=115024767022346&w=2 V4 http://marc.theaimsgroup.com/?l=linux-kernel&m=115047968808926&w=2 Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-30 16:55:45 +08:00
mem_data->pgfault = ev[PGFAULT];
mem_data->pgmajfault = ev[PGMAJFAULT];
si_meminfo(&val);
mem_data->sharedram = val.sharedram;
mem_data->totalram = P2K(val.totalram);
mem_data->freeram = P2K(val.freeram);
mem_data->totalhigh = P2K(val.totalhigh);
mem_data->freehigh = P2K(val.freehigh);
mem_data->bufferram = P2K(val.bufferram);
mem_data->cached = P2K(global_page_state(NR_FILE_PAGES)
- val.bufferram);
si_swapinfo(&val);
mem_data->totalswap = P2K(val.totalswap);
mem_data->freeswap = P2K(val.freeswap);
mem_data->timestamp = get_tod_clock();
mem_data->sync_count_2++;
}
static struct appldata_ops ops = {
.name = "mem",
.record_nr = APPLDATA_RECORD_MEM_ID,
.size = sizeof(struct appldata_mem_data),
.callback = &appldata_get_mem_data,
.owner = THIS_MODULE,
.mod_lvl = {0xF0, 0xF0}, /* EBCDIC "00" */
};
/*
* appldata_mem_init()
*
* init_data, register ops
*/
static int __init appldata_mem_init(void)
{
int ret;
ops.data = kzalloc(sizeof(struct appldata_mem_data), GFP_KERNEL);
if (!ops.data)
return -ENOMEM;
ret = appldata_register_ops(&ops);
if (ret)
kfree(ops.data);
return ret;
}
/*
* appldata_mem_exit()
*
* unregister ops
*/
static void __exit appldata_mem_exit(void)
{
appldata_unregister_ops(&ops);
kfree(ops.data);
}
module_init(appldata_mem_init);
module_exit(appldata_mem_exit);
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Gerald Schaefer");
MODULE_DESCRIPTION("Linux-VM Monitor Stream, MEMORY statistics");