proc.rst - OpenGrok cross reference for /linux/Documentation/filesystems/proc.rst

Deleted Added

sdiffudifftextold (98817a84..)new (059db434..)

proc.rst (98817a84ff1c755c347ac633ff017a623a631fad)	proc.rst (059db4341303922b6181df4b58c05f22d33b2f4f)
1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 --- 109 unchanged lines hidden (view full) --- 118show you how you can use /proc/sys to change settings. 119 1201.1 Process-Specific Subdirectories 121----------------------------------- 122 123The directory /proc contains (among other things) one subdirectory for each 124process running on the system, which is named after the process ID (PID). 125	1.. SPDX-License-Identifier: GPL-2.0 2 3==================== 4The /proc Filesystem 5==================== 6 7===================== ======================================= ================ 8/proc/sys Terrehon Bowden <terrehon@pacbell.net>, October 7 1999 --- 109 unchanged lines hidden (view full) --- 118show you how you can use /proc/sys to change settings. 119 1201.1 Process-Specific Subdirectories 121----------------------------------- 122 123The directory /proc contains (among other things) one subdirectory for each 124process running on the system, which is named after the process ID (PID). 125
126The link self points to the process reading the file system. Each process	126The link 'self' points to the process reading the file system. Each process
127subdirectory has the entries listed in Table 1-1. 128	127subdirectory has the entries listed in Table 1-1. 128
129Note that an open a file descriptor to /proc/<pid> or to any of its	129Note that an open file descriptor to /proc/<pid> or to any of its
130contained files or subdirectories does not prevent <pid> being reused 131for some other process in the event that <pid> exits. Operations on 132open /proc/<pid> file descriptors corresponding to dead processes 133never act on any new process that the kernel may, through chance, have 134also assigned the process ID <pid>. Instead, operations on these FDs 135usually fail with ESRCH. 136 137.. table:: Table 1-1: Process specific entries in /proc --- 77 unchanged lines hidden (view full) --- 215 216This shows you nearly the same information you would get if you viewed it with 217the ps command. In fact, ps uses the proc file system to obtain its 218information. But you get a more detailed view of the process by reading the 219file /proc/PID/status. It fields are described in table 1-2. 220 221The statm file contains more detailed information about the process 222memory usage. Its seven fields are explained in Table 1-3. The stat file	130contained files or subdirectories does not prevent <pid> being reused 131for some other process in the event that <pid> exits. Operations on 132open /proc/<pid> file descriptors corresponding to dead processes 133never act on any new process that the kernel may, through chance, have 134also assigned the process ID <pid>. Instead, operations on these FDs 135usually fail with ESRCH. 136 137.. table:: Table 1-1: Process specific entries in /proc --- 77 unchanged lines hidden (view full) --- 215 216This shows you nearly the same information you would get if you viewed it with 217the ps command. In fact, ps uses the proc file system to obtain its 218information. But you get a more detailed view of the process by reading the 219file /proc/PID/status. It fields are described in table 1-2. 220 221The statm file contains more detailed information about the process 222memory usage. Its seven fields are explained in Table 1-3. The stat file
223contains details information about the process itself. Its fields are	223contains detailed information about the process itself. Its fields are
224explained in Table 1-4. 225 226(for SMP CONFIG users) 227 228For making accounting scalable, RSS related information are handled in an 229asynchronous manner and the value may not be very precise. To see a precise 230snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 231It's slow but very precise. --- 308 unchanged lines hidden (view full) --- 540 ht area uses huge tlb pages 541 ar architecture specific flag 542 dd do not include area into core dump 543 sd soft dirty flag 544 mm mixed map area 545 hg huge page advise flag 546 nh no huge page advise flag 547 mg mergable advise flag	224explained in Table 1-4. 225 226(for SMP CONFIG users) 227 228For making accounting scalable, RSS related information are handled in an 229asynchronous manner and the value may not be very precise. To see a precise 230snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 231It's slow but very precise. --- 308 unchanged lines hidden (view full) --- 540 ht area uses huge tlb pages 541 ar architecture specific flag 542 dd do not include area into core dump 543 sd soft dirty flag 544 mm mixed map area 545 hg huge page advise flag 546 nh no huge page advise flag 547 mg mergable advise flag
548 bt - arm64 BTI guarded page	548 bt arm64 BTI guarded page
549 == ======================================= 550 551Note that there is no guarantee that every flag and associated mnemonic will 552be present in all further kernel releases. Things get changed, the flags may 553be vanished or the reverse -- new added. Interpretation of their meaning 554might change in future as well. So each consumer of these flags has to 555follow each specific kernel version for the exact semantic. 556 --- 220 unchanged lines hidden (view full) --- 777 778SPU 779 a spurious interrupt is some interrupt that was raised then lowered 780 by some IO device before it could be fully processed by the APIC. Hence 781 the APIC sees the interrupt but does not know what device it came from. 782 For this case the APIC will generate the interrupt with a IRQ vector 783 of 0xff. This might also be generated by chipset bugs. 784	549 == ======================================= 550 551Note that there is no guarantee that every flag and associated mnemonic will 552be present in all further kernel releases. Things get changed, the flags may 553be vanished or the reverse -- new added. Interpretation of their meaning 554might change in future as well. So each consumer of these flags has to 555follow each specific kernel version for the exact semantic. 556 --- 220 unchanged lines hidden (view full) --- 777 778SPU 779 a spurious interrupt is some interrupt that was raised then lowered 780 by some IO device before it could be fully processed by the APIC. Hence 781 the APIC sees the interrupt but does not know what device it came from. 782 For this case the APIC will generate the interrupt with a IRQ vector 783 of 0xff. This might also be generated by chipset bugs. 784
785RES, CAL, TLB]	785RES, CAL, TLB
786 rescheduling, call and TLB flush interrupts are 787 sent from one CPU to another per the needs of the OS. Typically, 788 their statistics are used by kernel developers and interested users to 789 determine the occurrence of interrupts of the given type. 790 791The above IRQ vectors are displayed only when relevant. For example, 792the threshold vector does not exist on x86_64 platforms. Others are 793suppressed when the system is a uniprocessor. As of this writing, only 794i386 and x86_64 platforms support the new IRQ vector displays. 795 796Of some interest is the introduction of the /proc/irq directory to 2.4.	786 rescheduling, call and TLB flush interrupts are 787 sent from one CPU to another per the needs of the OS. Typically, 788 their statistics are used by kernel developers and interested users to 789 determine the occurrence of interrupts of the given type. 790 791The above IRQ vectors are displayed only when relevant. For example, 792the threshold vector does not exist on x86_64 platforms. Others are 793suppressed when the system is a uniprocessor. As of this writing, only 794i386 and x86_64 platforms support the new IRQ vector displays. 795 796Of some interest is the introduction of the /proc/irq directory to 2.4.
797It could be used to set IRQ to CPU affinity, this means that you can "hook" an	797It could be used to set IRQ to CPU affinity. This means that you can "hook" an
798IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 799irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 800prof_cpu_mask. 801 802For example:: 803 804 > ls /proc/irq/ 805 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 806 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 807 > ls /proc/irq/0/ 808 smp_affinity 809 810smp_affinity is a bitmask, in which you can specify which CPUs can handle the	798IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 799irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 800prof_cpu_mask. 801 802For example:: 803 804 > ls /proc/irq/ 805 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 806 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 807 > ls /proc/irq/0/ 808 smp_affinity 809 810smp_affinity is a bitmask, in which you can specify which CPUs can handle the
811IRQ, you can set it by doing::	811IRQ. You can set it by doing::
812 813 > echo 1 > /proc/irq/10/smp_affinity 814 815This means that only the first CPU will handle the IRQ, but you can also echo 8165 which means that only the first and third CPU can handle the IRQ. 817 818The contents of each smp_affinity file is the same by default:: 819 820 > cat /proc/irq/0/smp_affinity 821 ffffffff 822 823There is an alternate interface, smp_affinity_list which allows specifying	812 813 > echo 1 > /proc/irq/10/smp_affinity 814 815This means that only the first CPU will handle the IRQ, but you can also echo 8165 which means that only the first and third CPU can handle the IRQ. 817 818The contents of each smp_affinity file is the same by default:: 819 820 > cat /proc/irq/0/smp_affinity 821 ffffffff 822 823There is an alternate interface, smp_affinity_list which allows specifying
824a cpu range instead of a bitmask::	824a CPU range instead of a bitmask::
825 826 > cat /proc/irq/0/smp_affinity_list 827 1024-1031 828 829The default_smp_affinity mask applies to all non-active IRQs, which are the 830IRQs which have not yet been allocated/activated, and hence which lack a 831/proc/irq/[0-9]* directory. 832 833The node file on an SMP system shows the node to which the device using the IRQ 834reports itself as being attached. This hardware locality information does not 835include information about any possible driver locality preference. 836 837prof_cpu_mask specifies which CPUs are to be profiled by the system wide	825 826 > cat /proc/irq/0/smp_affinity_list 827 1024-1031 828 829The default_smp_affinity mask applies to all non-active IRQs, which are the 830IRQs which have not yet been allocated/activated, and hence which lack a 831/proc/irq/[0-9]* directory. 832 833The node file on an SMP system shows the node to which the device using the IRQ 834reports itself as being attached. This hardware locality information does not 835include information about any possible driver locality preference. 836 837prof_cpu_mask specifies which CPUs are to be profiled by the system wide
838profiler. Default value is ffffffff (all cpus if there are only 32 of them).	838profiler. Default value is ffffffff (all CPUs if there are only 32 of them).
839 840The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 841between all the CPUs which are allowed to handle it. As usual the kernel has 842more info than you and does a better job than you, so the defaults are the 843best choice for almost everyone. [Note this applies only to those IO-APIC's 844that support "Round Robin" interrupt distribution.] 845 846There are three more important subdirectories in /proc: net, scsi, and sys. --- 45 unchanged lines hidden (view full) --- 892 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 893 894 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 895 Node 0, zone DMA 2 0 5 1 0 896 Node 0, zone DMA32 41 6 967 2 0 897 898Fragmentation avoidance in the kernel works by grouping pages of different 899migrate types into the same contiguous regions of memory called page blocks.	839 840The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 841between all the CPUs which are allowed to handle it. As usual the kernel has 842more info than you and does a better job than you, so the defaults are the 843best choice for almost everyone. [Note this applies only to those IO-APIC's 844that support "Round Robin" interrupt distribution.] 845 846There are three more important subdirectories in /proc: net, scsi, and sys. --- 45 unchanged lines hidden (view full) --- 892 Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 893 894 Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 895 Node 0, zone DMA 2 0 5 1 0 896 Node 0, zone DMA32 41 6 967 2 0 897 898Fragmentation avoidance in the kernel works by grouping pages of different 899migrate types into the same contiguous regions of memory called page blocks.
900A page block is typically the size of the default hugepage size e.g. 2MB on	900A page block is typically the size of the default hugepage size, e.g. 2MB on
901X86-64. By keeping pages grouped based on their ability to move, the kernel 902can reclaim pages within a page block to satisfy a high-order allocation. 903 904The pagetypinfo begins with information on the size of a page block. It 905then gives the same type of information as buddyinfo except broken down 906by migrate-type and finishes with details on how many page blocks of each 907type exist. 908 --- 51 unchanged lines hidden (view full) --- 960 VmallocChunk: 111088 kB 961 Percpu: 62080 kB 962 HardwareCorrupted: 0 kB 963 AnonHugePages: 49152 kB 964 ShmemHugePages: 0 kB 965 ShmemPmdMapped: 0 kB 966 967MemTotal	901X86-64. By keeping pages grouped based on their ability to move, the kernel 902can reclaim pages within a page block to satisfy a high-order allocation. 903 904The pagetypinfo begins with information on the size of a page block. It 905then gives the same type of information as buddyinfo except broken down 906by migrate-type and finishes with details on how many page blocks of each 907type exist. 908 --- 51 unchanged lines hidden (view full) --- 960 VmallocChunk: 111088 kB 961 Percpu: 62080 kB 962 HardwareCorrupted: 0 kB 963 AnonHugePages: 49152 kB 964 ShmemHugePages: 0 kB 965 ShmemPmdMapped: 0 kB 966 967MemTotal
968 Total usable ram (i.e. physical ram minus a few reserved	968 Total usable RAM (i.e. physical RAM minus a few reserved
969 bits and the kernel binary code) 970MemFree 971 The sum of LowFree+HighFree 972MemAvailable 973 An estimate of how much memory is available for starting new 974 applications, without swapping. Calculated from MemFree, 975 SReclaimable, the size of the file LRU lists, and the low 976 watermarks in each zone. --- 14 unchanged lines hidden (view full) --- 991 in the swapfile. This saves I/O) 992Active 993 Memory that has been used more recently and usually not 994 reclaimed unless absolutely necessary. 995Inactive 996 Memory which has been less recently used. It is more 997 eligible to be reclaimed for other purposes 998HighTotal, HighFree	969 bits and the kernel binary code) 970MemFree 971 The sum of LowFree+HighFree 972MemAvailable 973 An estimate of how much memory is available for starting new 974 applications, without swapping. Calculated from MemFree, 975 SReclaimable, the size of the file LRU lists, and the low 976 watermarks in each zone. --- 14 unchanged lines hidden (view full) --- 991 in the swapfile. This saves I/O) 992Active 993 Memory that has been used more recently and usually not 994 reclaimed unless absolutely necessary. 995Inactive 996 Memory which has been less recently used. It is more 997 eligible to be reclaimed for other purposes 998HighTotal, HighFree
999 Highmem is all memory above ~860MB of physical memory	999 Highmem is all memory above ~860MB of physical memory.
1000 Highmem areas are for use by userspace programs, or 1001 for the pagecache. The kernel must use tricks to access 1002 this memory, making it slower to access than lowmem. 1003LowTotal, LowFree 1004 Lowmem is memory which can be used for everything that 1005 highmem can be used for, but it is also available for the 1006 kernel's use for its own data structures. Among many 1007 other things, it is where everything from the Slab is --- 65 unchanged lines hidden (view full) --- 1073 The amount of memory presently allocated on the system. 1074 The committed memory is a sum of all of the memory which 1075 has been allocated by processes, even if it has not been 1076 "used" by them as of yet. A process which malloc()'s 1G 1077 of memory, but only touches 300M of it will show up as 1078 using 1G. This 1G is memory which has been "committed" to 1079 by the VM and can be used at any time by the allocating 1080 application. With strict overcommit enabled on the system	1000 Highmem areas are for use by userspace programs, or 1001 for the pagecache. The kernel must use tricks to access 1002 this memory, making it slower to access than lowmem. 1003LowTotal, LowFree 1004 Lowmem is memory which can be used for everything that 1005 highmem can be used for, but it is also available for the 1006 kernel's use for its own data structures. Among many 1007 other things, it is where everything from the Slab is --- 65 unchanged lines hidden (view full) --- 1073 The amount of memory presently allocated on the system. 1074 The committed memory is a sum of all of the memory which 1075 has been allocated by processes, even if it has not been 1076 "used" by them as of yet. A process which malloc()'s 1G 1077 of memory, but only touches 300M of it will show up as 1078 using 1G. This 1G is memory which has been "committed" to 1079 by the VM and can be used at any time by the allocating 1080 application. With strict overcommit enabled on the system
1081 (mode 2 in 'vm.overcommit_memory'),allocations which would	1081 (mode 2 in 'vm.overcommit_memory'), allocations which would
1082 exceed the CommitLimit (detailed above) will not be permitted. 1083 This is useful if one needs to guarantee that processes will 1084 not fail due to lack of memory once that memory has been 1085 successfully allocated. 1086VmallocTotal 1087 total size of vmalloc memory area 1088VmallocUsed 1089 amount of vmalloc area which is used --- 4 unchanged lines hidden (view full) --- 1094 allocations. This stat excludes the cost of metadata. 1095 1096vmallocinfo 1097~~~~~~~~~~~ 1098 1099Provides information about vmalloced/vmaped areas. One line per area, 1100containing the virtual address range of the area, size in bytes, 1101caller information of the creator, and optional information depending	1082 exceed the CommitLimit (detailed above) will not be permitted. 1083 This is useful if one needs to guarantee that processes will 1084 not fail due to lack of memory once that memory has been 1085 successfully allocated. 1086VmallocTotal 1087 total size of vmalloc memory area 1088VmallocUsed 1089 amount of vmalloc area which is used --- 4 unchanged lines hidden (view full) --- 1094 allocations. This stat excludes the cost of metadata. 1095 1096vmallocinfo 1097~~~~~~~~~~~ 1098 1099Provides information about vmalloced/vmaped areas. One line per area, 1100containing the virtual address range of the area, size in bytes, 1101caller information of the creator, and optional information depending
1102on the kind of area :	1102on the kind of area:
1103 1104 ========== =================================================== 1105 pages=nr number of pages 1106 phys=addr if a physical address was specified 1107 ioremap I/O mapping (ioremap() and friends) 1108 vmalloc vmalloc() area 1109 vmap vmap()ed pages 1110 user VM_USERMAP area --- 28 unchanged lines hidden (view full) --- 1139 pages=2 vmalloc N1=2 1140 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1141 pages=10 vmalloc N0=10 1142 1143 1144softirqs 1145~~~~~~~~ 1146	1103 1104 ========== =================================================== 1105 pages=nr number of pages 1106 phys=addr if a physical address was specified 1107 ioremap I/O mapping (ioremap() and friends) 1108 vmalloc vmalloc() area 1109 vmap vmap()ed pages 1110 user VM_USERMAP area --- 28 unchanged lines hidden (view full) --- 1139 pages=2 vmalloc N1=2 1140 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 1141 pages=10 vmalloc N0=10 1142 1143 1144softirqs 1145~~~~~~~~ 1146
1147Provides counts of softirq handlers serviced since boot time, for each cpu.	1147Provides counts of softirq handlers serviced since boot time, for each CPU.
1148 1149:: 1150 1151 > cat /proc/softirqs	1148 1149:: 1150 1151 > cat /proc/softirqs
1152 CPU0 CPU1 CPU2 CPU3	1152 CPU0 CPU1 CPU2 CPU3
1153 HI: 0 0 0 0	1153 HI: 0 0 0 0
1154 TIMER: 27166 27120 27097 27034	1154 TIMER: 27166 27120 27097 27034
1155 NET_TX: 0 0 0 17 1156 NET_RX: 42 0 0 39	1155 NET_TX: 0 0 0 17 1156 NET_RX: 42 0 0 39
1157 BLOCK: 0 0 107 1121 1158 TASKLET: 0 0 0 290 1159 SCHED: 27035 26983 26971 26746 1160 HRTIMER: 0 0 0 0 1161 RCU: 1678 1769 2178 2250	1157 BLOCK: 0 0 107 1121 1158 TASKLET: 0 0 0 290 1159 SCHED: 27035 26983 26971 26746 1160 HRTIMER: 0 0 0 0 1161 RCU: 1678 1769 2178 2250
1162 1163 11641.3 IDE devices in /proc/ide 1165---------------------------- 1166 1167The subdirectory /proc/ide contains information about all IDE devices of which 1168the kernel is aware. There is one subdirectory for each IDE controller, the 1169file drivers and a link for each IDE device, pointing to the device directory 1170in the controller specific subtree. 1171	1162 1163 11641.3 IDE devices in /proc/ide 1165---------------------------- 1166 1167The subdirectory /proc/ide contains information about all IDE devices of which 1168the kernel is aware. There is one subdirectory for each IDE controller, the 1169file drivers and a link for each IDE device, pointing to the device directory 1170in the controller specific subtree. 1171
1172The file drivers contains general information about the drivers used for the	1172The file 'drivers' contains general information about the drivers used for the
1173IDE devices:: 1174 1175 > cat /proc/ide/drivers 1176 ide-cdrom version 4.53 1177 ide-disk version 1.08 1178 1179More detailed information can be found in the controller specific 1180subdirectories. These are named ide0, ide1 and so on. Each of these --- 223 unchanged lines hidden (view full) --- 1404 file to allow you to alter it by writing a new value in (IRQ 1405 number or none). 1406 ========= ==================================================================== 1407 14081.7 TTY info in /proc/tty 1409------------------------- 1410 1411Information about the available and actually used tty's can be found in the	1173IDE devices:: 1174 1175 > cat /proc/ide/drivers 1176 ide-cdrom version 4.53 1177 ide-disk version 1.08 1178 1179More detailed information can be found in the controller specific 1180subdirectories. These are named ide0, ide1 and so on. Each of these --- 223 unchanged lines hidden (view full) --- 1404 file to allow you to alter it by writing a new value in (IRQ 1405 number or none). 1406 ========= ==================================================================== 1407 14081.7 TTY info in /proc/tty 1409------------------------- 1410 1411Information about the available and actually used tty's can be found in the
1412directory /proc/tty.You'll find entries for drivers and line disciplines in	1412directory /proc/tty. You'll find entries for drivers and line disciplines in