procps/proc/diskstats.h

130 lines
3.9 KiB
C
Raw Normal View History

/*
library: removed all the 'PROCPS_' enumerator prefixes Many of our item enumerator identifiers are very long, especially in that <VMSTAT> module. Additionally, they all contain the exact same universal 'PROCPS_' prefix. The origins for this are likely found in the desire to avoid name clashes with other potential include files. But with procps-ng newlib, we've probably gone way too far. Did 'PROCPS_PIDS_TICS_SYSTEM' actually offer more protection against clash than 'PIDS_TICS_SYSTEM' does? I don't think so. Besides, no matter how big that name becomes, one can never guarantee they'll never be some clash. And, conversely, extremely short names will not always create conflict. Of course, in either case when some clash occurs, one can always #undef that problem. Thus, this commit will eliminate that 'PROCPS_' prefix making all of those enum identifiers a little shorter. And, we'll still be well above some ridiculously short (criminally short) names found in some common headers: - - - - - - - - - - <term.h> - 'tab', 'TTY', etc - - - - - - - - - - - - - - - - <search.h> - 'ENTER', ENTRY', 'FIND', etc ------------------------------------------------------ Finally, with this as a last of the wholesale changes, we will have established the naming conventions below: . only functions will begin with that 'procps_' prefix . exposed structures begin with the module/header name . item enumerators begin like structs, but capitalized . other enumerators work exactly like item enumerators . macros and constants begin just like the enumerators ------------------------------------------------------ Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-21 10:30:00 +05:30
* libprocps - Library to read proc filesystem
*
* Copyright (c) 2003 Fabian Frederick
* Copyright (C) 2003 Albert Cahalan
* Copyright (C) 2015 Craig Small <csmall@enc.com.au>
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
* Copyright (C) 2016 Jim Warner <james.warner@comcast.net>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library 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
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
#ifndef PROC_DISKSTATS_H
#define PROC_DISKSTATS_H
#include <sys/cdefs.h>
__BEGIN_DECLS
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
enum diskstats_item {
library: removed all the 'PROCPS_' enumerator prefixes Many of our item enumerator identifiers are very long, especially in that <VMSTAT> module. Additionally, they all contain the exact same universal 'PROCPS_' prefix. The origins for this are likely found in the desire to avoid name clashes with other potential include files. But with procps-ng newlib, we've probably gone way too far. Did 'PROCPS_PIDS_TICS_SYSTEM' actually offer more protection against clash than 'PIDS_TICS_SYSTEM' does? I don't think so. Besides, no matter how big that name becomes, one can never guarantee they'll never be some clash. And, conversely, extremely short names will not always create conflict. Of course, in either case when some clash occurs, one can always #undef that problem. Thus, this commit will eliminate that 'PROCPS_' prefix making all of those enum identifiers a little shorter. And, we'll still be well above some ridiculously short (criminally short) names found in some common headers: - - - - - - - - - - <term.h> - 'tab', 'TTY', etc - - - - - - - - - - - - - - - - <search.h> - 'ENTER', ENTRY', 'FIND', etc ------------------------------------------------------ Finally, with this as a last of the wholesale changes, we will have established the naming conventions below: . only functions will begin with that 'procps_' prefix . exposed structures begin with the module/header name . item enumerators begin like structs, but capitalized . other enumerators work exactly like item enumerators . macros and constants begin just like the enumerators ------------------------------------------------------ Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-21 10:30:00 +05:30
DISKSTATS_noop, // ( never altered )
DISKSTATS_extra, // ( reset to zero )
DISKSTATS_NAME, // str
DISKSTATS_TYPE, // s_int
DISKSTATS_MAJOR, // s_int
DISKSTATS_MINOR, // s_int
DISKSTATS_READS, // ul_int
DISKSTATS_READS_MERGED, // ul_int
DISKSTATS_READ_SECTORS, // ul_int
DISKSTATS_READ_TIME, // ul_int
DISKSTATS_WRITES, // ul_int
DISKSTATS_WRITES_MERGED, // ul_int
DISKSTATS_WRITE_SECTORS, // ul_int
DISKSTATS_WRITE_TIME, // ul_int
DISKSTATS_IO_TIME, // ul_int
DISKSTATS_IO_WTIME, // ul_int
DISKSTATS_IO_INPROGRESS, // s_int
DISKSTATS_DELTA_READS, // s_int
DISKSTATS_DELTA_READS_MERGED, // s_int
DISKSTATS_DELTA_READ_SECTORS, // s_int
DISKSTATS_DELTA_READ_TIME, // s_int
DISKSTATS_DELTA_WRITES, // s_int
DISKSTATS_DELTA_WRITES_MERGED, // s_int
DISKSTATS_DELTA_WRITE_SECTORS, // s_int
DISKSTATS_DELTA_WRITE_TIME, // s_int
DISKSTATS_DELTA_IO_TIME, // s_int
DISKSTATS_DELTA_IO_WTIME // s_int
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
};
enum diskstats_sort_order {
library: removed all the 'PROCPS_' enumerator prefixes Many of our item enumerator identifiers are very long, especially in that <VMSTAT> module. Additionally, they all contain the exact same universal 'PROCPS_' prefix. The origins for this are likely found in the desire to avoid name clashes with other potential include files. But with procps-ng newlib, we've probably gone way too far. Did 'PROCPS_PIDS_TICS_SYSTEM' actually offer more protection against clash than 'PIDS_TICS_SYSTEM' does? I don't think so. Besides, no matter how big that name becomes, one can never guarantee they'll never be some clash. And, conversely, extremely short names will not always create conflict. Of course, in either case when some clash occurs, one can always #undef that problem. Thus, this commit will eliminate that 'PROCPS_' prefix making all of those enum identifiers a little shorter. And, we'll still be well above some ridiculously short (criminally short) names found in some common headers: - - - - - - - - - - <term.h> - 'tab', 'TTY', etc - - - - - - - - - - - - - - - - <search.h> - 'ENTER', ENTRY', 'FIND', etc ------------------------------------------------------ Finally, with this as a last of the wholesale changes, we will have established the naming conventions below: . only functions will begin with that 'procps_' prefix . exposed structures begin with the module/header name . item enumerators begin like structs, but capitalized . other enumerators work exactly like item enumerators . macros and constants begin just like the enumerators ------------------------------------------------------ Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-21 10:30:00 +05:30
DISKSTATS_SORT_ASCEND = +1,
DISKSTATS_SORT_DESCEND = -1
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
};
struct diskstats_result {
enum diskstats_item item;
union {
signed int s_int;
unsigned long ul_int;
char *str;
} result;
};
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
struct diskstats_stack {
struct diskstats_result *head;
};
struct diskstats_reap {
int total;
struct diskstats_stack **stacks;
};
library: removed all the 'PROCPS_' enumerator prefixes Many of our item enumerator identifiers are very long, especially in that <VMSTAT> module. Additionally, they all contain the exact same universal 'PROCPS_' prefix. The origins for this are likely found in the desire to avoid name clashes with other potential include files. But with procps-ng newlib, we've probably gone way too far. Did 'PROCPS_PIDS_TICS_SYSTEM' actually offer more protection against clash than 'PIDS_TICS_SYSTEM' does? I don't think so. Besides, no matter how big that name becomes, one can never guarantee they'll never be some clash. And, conversely, extremely short names will not always create conflict. Of course, in either case when some clash occurs, one can always #undef that problem. Thus, this commit will eliminate that 'PROCPS_' prefix making all of those enum identifiers a little shorter. And, we'll still be well above some ridiculously short (criminally short) names found in some common headers: - - - - - - - - - - <term.h> - 'tab', 'TTY', etc - - - - - - - - - - - - - - - - <search.h> - 'ENTER', ENTRY', 'FIND', etc ------------------------------------------------------ Finally, with this as a last of the wholesale changes, we will have established the naming conventions below: . only functions will begin with that 'procps_' prefix . exposed structures begin with the module/header name . item enumerators begin like structs, but capitalized . other enumerators work exactly like item enumerators . macros and constants begin just like the enumerators ------------------------------------------------------ Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-21 10:30:00 +05:30
#define DISKSTATS_TYPE_DISK -11111
#define DISKSTATS_TYPE_PARTITION -22222
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
#define DISKSTATS_GET( info, name, actual_enum, type ) \
procps_diskstats_get( info, name, actual_enum ) -> result . type
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
#define DISKSTATS_VAL( relative_enum, type, stack, info ) \
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
stack -> head [ relative_enum ] . result . type
struct diskstats_info;
int procps_diskstats_new (struct diskstats_info **info);
int procps_diskstats_ref (struct diskstats_info *info);
int procps_diskstats_unref (struct diskstats_info **info);
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
struct diskstats_result *procps_diskstats_get (
struct diskstats_info *info,
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
const char *name,
enum diskstats_item item);
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
struct diskstats_reap *procps_diskstats_reap (
struct diskstats_info *info,
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
enum diskstats_item *items,
int numitems);
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
struct diskstats_stack *procps_diskstats_select (
struct diskstats_info *info,
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
const char *name,
enum diskstats_item *items,
int numitems);
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
struct diskstats_stack **procps_diskstats_sort (
struct diskstats_info *info,
library: normalize/standardize an i/f, <DISKSTATS> api This patch will bring this interface up to our 3rd gen standards. The following summarizes the major changes. * New delta provisions have been added to most fields. There are, of course, some fields for which a delta is inappropriate. They include the identifying items such as name, type, major and minor. Plus the io_inprogress field which already acts, in effect, as a delta value. * To provide delta support, dev_node historical values have become persistent. By the same token, the library must provide for future removal of disks/partitions. A timestamp is used to detect 'stale' data which will be deleted so as not to satisfy some get, select or reap. * Such persistent support is provided by a linked list which, by default, grows from the bottom down so as to maintain compatibility with the /proc/diskstats order. Initially, I was tempted to use the GNU tsearch (tree) provisions until I discovered the overhead of building that tree plus costs of a subsequent 'twalk'. Besides, walking such a tree means retrieval order would differ from an order required/expected by the vmstat program. * The '/sys/block' directory is no longer scanned with every refresh cycle. Rather, it's only accessed when a node is first encountered. Then, that node's 'type' is persistent for its lifetime like several other fields. * A sort provision was included, at virtually no cost, even though such a provision was not currently needed. Signed-off-by: Jim Warner <james.warner@comcast.net>
2016-07-20 10:30:00 +05:30
struct diskstats_stack *stacks[],
int numstacked,
enum diskstats_item sortitem,
enum diskstats_sort_order order);
__END_DECLS
#endif