1------------------------------------------------------------------------------ 2 T H E /proc F I L E S Y S T E M 3------------------------------------------------------------------------------ 4/proc/sys Terrehon Bowden <terrehon@pacbell.net> October 7 1999 5 Bodo Bauer <bb@ricochet.net> 6 72.4.x update Jorge Nerin <comandante@zaralinux.com> November 14 2000 8move /proc/sys Shen Feng <shen@cn.fujitsu.com> April 1 2009 9------------------------------------------------------------------------------ 10Version 1.3 Kernel version 2.2.12 11 Kernel version 2.4.0-test11-pre4 12------------------------------------------------------------------------------ 13fixes/update part 1.1 Stefani Seibold <stefani@seibold.net> June 9 2009 14 15Table of Contents 16----------------- 17 18 0 Preface 19 0.1 Introduction/Credits 20 0.2 Legal Stuff 21 22 1 Collecting System Information 23 1.1 Process-Specific Subdirectories 24 1.2 Kernel data 25 1.3 IDE devices in /proc/ide 26 1.4 Networking info in /proc/net 27 1.5 SCSI info 28 1.6 Parallel port info in /proc/parport 29 1.7 TTY info in /proc/tty 30 1.8 Miscellaneous kernel statistics in /proc/stat 31 1.9 Ext4 file system parameters 32 33 2 Modifying System Parameters 34 35 3 Per-Process Parameters 36 3.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj - Adjust the oom-killer 37 score 38 3.2 /proc/<pid>/oom_score - Display current oom-killer score 39 3.3 /proc/<pid>/io - Display the IO accounting fields 40 3.4 /proc/<pid>/coredump_filter - Core dump filtering settings 41 3.5 /proc/<pid>/mountinfo - Information about mounts 42 3.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 43 44 45------------------------------------------------------------------------------ 46Preface 47------------------------------------------------------------------------------ 48 490.1 Introduction/Credits 50------------------------ 51 52This documentation is part of a soon (or so we hope) to be released book on 53the SuSE Linux distribution. As there is no complete documentation for the 54/proc file system and we've used many freely available sources to write these 55chapters, it seems only fair to give the work back to the Linux community. 56This work is based on the 2.2.* kernel version and the upcoming 2.4.*. I'm 57afraid it's still far from complete, but we hope it will be useful. As far as 58we know, it is the first 'all-in-one' document about the /proc file system. It 59is focused on the Intel x86 hardware, so if you are looking for PPC, ARM, 60SPARC, AXP, etc., features, you probably won't find what you are looking for. 61It also only covers IPv4 networking, not IPv6 nor other protocols - sorry. But 62additions and patches are welcome and will be added to this document if you 63mail them to Bodo. 64 65We'd like to thank Alan Cox, Rik van Riel, and Alexey Kuznetsov and a lot of 66other people for help compiling this documentation. We'd also like to extend a 67special thank you to Andi Kleen for documentation, which we relied on heavily 68to create this document, as well as the additional information he provided. 69Thanks to everybody else who contributed source or docs to the Linux kernel 70and helped create a great piece of software... :) 71 72If you have any comments, corrections or additions, please don't hesitate to 73contact Bodo Bauer at bb@ricochet.net. We'll be happy to add them to this 74document. 75 76The latest version of this document is available online at 77http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/proc.html 78 79If the above direction does not works for you, you could try the kernel 80mailing list at linux-kernel@vger.kernel.org and/or try to reach me at 81comandante@zaralinux.com. 82 830.2 Legal Stuff 84--------------- 85 86We don't guarantee the correctness of this document, and if you come to us 87complaining about how you screwed up your system because of incorrect 88documentation, we won't feel responsible... 89 90------------------------------------------------------------------------------ 91CHAPTER 1: COLLECTING SYSTEM INFORMATION 92------------------------------------------------------------------------------ 93 94------------------------------------------------------------------------------ 95In This Chapter 96------------------------------------------------------------------------------ 97* Investigating the properties of the pseudo file system /proc and its 98 ability to provide information on the running Linux system 99* Examining /proc's structure 100* Uncovering various information about the kernel and the processes running 101 on the system 102------------------------------------------------------------------------------ 103 104 105The proc file system acts as an interface to internal data structures in the 106kernel. It can be used to obtain information about the system and to change 107certain kernel parameters at runtime (sysctl). 108 109First, we'll take a look at the read-only parts of /proc. In Chapter 2, we 110show you how you can use /proc/sys to change settings. 111 1121.1 Process-Specific Subdirectories 113----------------------------------- 114 115The directory /proc contains (among other things) one subdirectory for each 116process running on the system, which is named after the process ID (PID). 117 118The link self points to the process reading the file system. Each process 119subdirectory has the entries listed in Table 1-1. 120 121 122Table 1-1: Process specific entries in /proc 123.............................................................................. 124 File Content 125 clear_refs Clears page referenced bits shown in smaps output 126 cmdline Command line arguments 127 cpu Current and last cpu in which it was executed (2.4)(smp) 128 cwd Link to the current working directory 129 environ Values of environment variables 130 exe Link to the executable of this process 131 fd Directory, which contains all file descriptors 132 maps Memory maps to executables and library files (2.4) 133 mem Memory held by this process 134 root Link to the root directory of this process 135 stat Process status 136 statm Process memory status information 137 status Process status in human readable form 138 wchan If CONFIG_KALLSYMS is set, a pre-decoded wchan 139 pagemap Page table 140 stack Report full stack trace, enable via CONFIG_STACKTRACE 141 smaps a extension based on maps, showing the memory consumption of 142 each mapping 143.............................................................................. 144 145For example, to get the status information of a process, all you have to do is 146read the file /proc/PID/status: 147 148 >cat /proc/self/status 149 Name: cat 150 State: R (running) 151 Tgid: 5452 152 Pid: 5452 153 PPid: 743 154 TracerPid: 0 (2.4) 155 Uid: 501 501 501 501 156 Gid: 100 100 100 100 157 FDSize: 256 158 Groups: 100 14 16 159 VmPeak: 5004 kB 160 VmSize: 5004 kB 161 VmLck: 0 kB 162 VmHWM: 476 kB 163 VmRSS: 476 kB 164 VmData: 156 kB 165 VmStk: 88 kB 166 VmExe: 68 kB 167 VmLib: 1412 kB 168 VmPTE: 20 kb 169 VmSwap: 0 kB 170 Threads: 1 171 SigQ: 0/28578 172 SigPnd: 0000000000000000 173 ShdPnd: 0000000000000000 174 SigBlk: 0000000000000000 175 SigIgn: 0000000000000000 176 SigCgt: 0000000000000000 177 CapInh: 00000000fffffeff 178 CapPrm: 0000000000000000 179 CapEff: 0000000000000000 180 CapBnd: ffffffffffffffff 181 voluntary_ctxt_switches: 0 182 nonvoluntary_ctxt_switches: 1 183 184This shows you nearly the same information you would get if you viewed it with 185the ps command. In fact, ps uses the proc file system to obtain its 186information. But you get a more detailed view of the process by reading the 187file /proc/PID/status. It fields are described in table 1-2. 188 189The statm file contains more detailed information about the process 190memory usage. Its seven fields are explained in Table 1-3. The stat file 191contains details information about the process itself. Its fields are 192explained in Table 1-4. 193 194(for SMP CONFIG users) 195For making accounting scalable, RSS related information are handled in 196asynchronous manner and the vaule may not be very precise. To see a precise 197snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. 198It's slow but very precise. 199 200Table 1-2: Contents of the status files (as of 2.6.30-rc7) 201.............................................................................. 202 Field Content 203 Name filename of the executable 204 State state (R is running, S is sleeping, D is sleeping 205 in an uninterruptible wait, Z is zombie, 206 T is traced or stopped) 207 Tgid thread group ID 208 Pid process id 209 PPid process id of the parent process 210 TracerPid PID of process tracing this process (0 if not) 211 Uid Real, effective, saved set, and file system UIDs 212 Gid Real, effective, saved set, and file system GIDs 213 FDSize number of file descriptor slots currently allocated 214 Groups supplementary group list 215 VmPeak peak virtual memory size 216 VmSize total program size 217 VmLck locked memory size 218 VmHWM peak resident set size ("high water mark") 219 VmRSS size of memory portions 220 VmData size of data, stack, and text segments 221 VmStk size of data, stack, and text segments 222 VmExe size of text segment 223 VmLib size of shared library code 224 VmPTE size of page table entries 225 VmSwap size of swap usage (the number of referred swapents) 226 Threads number of threads 227 SigQ number of signals queued/max. number for queue 228 SigPnd bitmap of pending signals for the thread 229 ShdPnd bitmap of shared pending signals for the process 230 SigBlk bitmap of blocked signals 231 SigIgn bitmap of ignored signals 232 SigCgt bitmap of catched signals 233 CapInh bitmap of inheritable capabilities 234 CapPrm bitmap of permitted capabilities 235 CapEff bitmap of effective capabilities 236 CapBnd bitmap of capabilities bounding set 237 Cpus_allowed mask of CPUs on which this process may run 238 Cpus_allowed_list Same as previous, but in "list format" 239 Mems_allowed mask of memory nodes allowed to this process 240 Mems_allowed_list Same as previous, but in "list format" 241 voluntary_ctxt_switches number of voluntary context switches 242 nonvoluntary_ctxt_switches number of non voluntary context switches 243.............................................................................. 244 245Table 1-3: Contents of the statm files (as of 2.6.8-rc3) 246.............................................................................. 247 Field Content 248 size total program size (pages) (same as VmSize in status) 249 resident size of memory portions (pages) (same as VmRSS in status) 250 shared number of pages that are shared (i.e. backed by a file) 251 trs number of pages that are 'code' (not including libs; broken, 252 includes data segment) 253 lrs number of pages of library (always 0 on 2.6) 254 drs number of pages of data/stack (including libs; broken, 255 includes library text) 256 dt number of dirty pages (always 0 on 2.6) 257.............................................................................. 258 259 260Table 1-4: Contents of the stat files (as of 2.6.30-rc7) 261.............................................................................. 262 Field Content 263 pid process id 264 tcomm filename of the executable 265 state state (R is running, S is sleeping, D is sleeping in an 266 uninterruptible wait, Z is zombie, T is traced or stopped) 267 ppid process id of the parent process 268 pgrp pgrp of the process 269 sid session id 270 tty_nr tty the process uses 271 tty_pgrp pgrp of the tty 272 flags task flags 273 min_flt number of minor faults 274 cmin_flt number of minor faults with child's 275 maj_flt number of major faults 276 cmaj_flt number of major faults with child's 277 utime user mode jiffies 278 stime kernel mode jiffies 279 cutime user mode jiffies with child's 280 cstime kernel mode jiffies with child's 281 priority priority level 282 nice nice level 283 num_threads number of threads 284 it_real_value (obsolete, always 0) 285 start_time time the process started after system boot 286 vsize virtual memory size 287 rss resident set memory size 288 rsslim current limit in bytes on the rss 289 start_code address above which program text can run 290 end_code address below which program text can run 291 start_stack address of the start of the stack 292 esp current value of ESP 293 eip current value of EIP 294 pending bitmap of pending signals 295 blocked bitmap of blocked signals 296 sigign bitmap of ignored signals 297 sigcatch bitmap of catched signals 298 wchan address where process went to sleep 299 0 (place holder) 300 0 (place holder) 301 exit_signal signal to send to parent thread on exit 302 task_cpu which CPU the task is scheduled on 303 rt_priority realtime priority 304 policy scheduling policy (man sched_setscheduler) 305 blkio_ticks time spent waiting for block IO 306 gtime guest time of the task in jiffies 307 cgtime guest time of the task children in jiffies 308.............................................................................. 309 310The /proc/PID/maps file containing the currently mapped memory regions and 311their access permissions. 312 313The format is: 314 315address perms offset dev inode pathname 316 31708048000-08049000 r-xp 00000000 03:00 8312 /opt/test 31808049000-0804a000 rw-p 00001000 03:00 8312 /opt/test 3190804a000-0806b000 rw-p 00000000 00:00 0 [heap] 320a7cb1000-a7cb2000 ---p 00000000 00:00 0 321a7cb2000-a7eb2000 rw-p 00000000 00:00 0 322a7eb2000-a7eb3000 ---p 00000000 00:00 0 323a7eb3000-a7ed5000 rw-p 00000000 00:00 0 324a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 325a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 326a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 327a800b000-a800e000 rw-p 00000000 00:00 0 328a800e000-a8022000 r-xp 00000000 03:00 14462 /lib/libpthread.so.0 329a8022000-a8023000 r--p 00013000 03:00 14462 /lib/libpthread.so.0 330a8023000-a8024000 rw-p 00014000 03:00 14462 /lib/libpthread.so.0 331a8024000-a8027000 rw-p 00000000 00:00 0 332a8027000-a8043000 r-xp 00000000 03:00 8317 /lib/ld-linux.so.2 333a8043000-a8044000 r--p 0001b000 03:00 8317 /lib/ld-linux.so.2 334a8044000-a8045000 rw-p 0001c000 03:00 8317 /lib/ld-linux.so.2 335aff35000-aff4a000 rw-p 00000000 00:00 0 [stack] 336ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] 337 338where "address" is the address space in the process that it occupies, "perms" 339is a set of permissions: 340 341 r = read 342 w = write 343 x = execute 344 s = shared 345 p = private (copy on write) 346 347"offset" is the offset into the mapping, "dev" is the device (major:minor), and 348"inode" is the inode on that device. 0 indicates that no inode is associated 349with the memory region, as the case would be with BSS (uninitialized data). 350The "pathname" shows the name associated file for this mapping. If the mapping 351is not associated with a file: 352 353 [heap] = the heap of the program 354 [stack] = the stack of the main process 355 [vdso] = the "virtual dynamic shared object", 356 the kernel system call handler 357 358 or if empty, the mapping is anonymous. 359 360 361The /proc/PID/smaps is an extension based on maps, showing the memory 362consumption for each of the process's mappings. For each of mappings there 363is a series of lines such as the following: 364 36508048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash 366Size: 1084 kB 367Rss: 892 kB 368Pss: 374 kB 369Shared_Clean: 892 kB 370Shared_Dirty: 0 kB 371Private_Clean: 0 kB 372Private_Dirty: 0 kB 373Referenced: 892 kB 374Anonymous: 0 kB 375Swap: 0 kB 376KernelPageSize: 4 kB 377MMUPageSize: 4 kB 378Locked: 374 kB 379 380The first of these lines shows the same information as is displayed for the 381mapping in /proc/PID/maps. The remaining lines show the size of the mapping 382(size), the amount of the mapping that is currently resident in RAM (RSS), the 383process' proportional share of this mapping (PSS), the number of clean and 384dirty private pages in the mapping. Note that even a page which is part of a 385MAP_SHARED mapping, but has only a single pte mapped, i.e. is currently used 386by only one process, is accounted as private and not as shared. "Referenced" 387indicates the amount of memory currently marked as referenced or accessed. 388"Anonymous" shows the amount of memory that does not belong to any file. Even 389a mapping associated with a file may contain anonymous pages: when MAP_PRIVATE 390and a page is modified, the file page is replaced by a private anonymous copy. 391"Swap" shows how much would-be-anonymous memory is also used, but out on 392swap. 393 394This file is only present if the CONFIG_MMU kernel configuration option is 395enabled. 396 397The /proc/PID/clear_refs is used to reset the PG_Referenced and ACCESSED/YOUNG 398bits on both physical and virtual pages associated with a process. 399To clear the bits for all the pages associated with the process 400 > echo 1 > /proc/PID/clear_refs 401 402To clear the bits for the anonymous pages associated with the process 403 > echo 2 > /proc/PID/clear_refs 404 405To clear the bits for the file mapped pages associated with the process 406 > echo 3 > /proc/PID/clear_refs 407Any other value written to /proc/PID/clear_refs will have no effect. 408 409The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags 410using /proc/kpageflags and number of times a page is mapped using 411/proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt. 412 4131.2 Kernel data 414--------------- 415 416Similar to the process entries, the kernel data files give information about 417the running kernel. The files used to obtain this information are contained in 418/proc and are listed in Table 1-5. Not all of these will be present in your 419system. It depends on the kernel configuration and the loaded modules, which 420files are there, and which are missing. 421 422Table 1-5: Kernel info in /proc 423.............................................................................. 424 File Content 425 apm Advanced power management info 426 buddyinfo Kernel memory allocator information (see text) (2.5) 427 bus Directory containing bus specific information 428 cmdline Kernel command line 429 cpuinfo Info about the CPU 430 devices Available devices (block and character) 431 dma Used DMS channels 432 filesystems Supported filesystems 433 driver Various drivers grouped here, currently rtc (2.4) 434 execdomains Execdomains, related to security (2.4) 435 fb Frame Buffer devices (2.4) 436 fs File system parameters, currently nfs/exports (2.4) 437 ide Directory containing info about the IDE subsystem 438 interrupts Interrupt usage 439 iomem Memory map (2.4) 440 ioports I/O port usage 441 irq Masks for irq to cpu affinity (2.4)(smp?) 442 isapnp ISA PnP (Plug&Play) Info (2.4) 443 kcore Kernel core image (can be ELF or A.OUT(deprecated in 2.4)) 444 kmsg Kernel messages 445 ksyms Kernel symbol table 446 loadavg Load average of last 1, 5 & 15 minutes 447 locks Kernel locks 448 meminfo Memory info 449 misc Miscellaneous 450 modules List of loaded modules 451 mounts Mounted filesystems 452 net Networking info (see text) 453 pagetypeinfo Additional page allocator information (see text) (2.5) 454 partitions Table of partitions known to the system 455 pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, 456 decoupled by lspci (2.4) 457 rtc Real time clock 458 scsi SCSI info (see text) 459 slabinfo Slab pool info 460 softirqs softirq usage 461 stat Overall statistics 462 swaps Swap space utilization 463 sys See chapter 2 464 sysvipc Info of SysVIPC Resources (msg, sem, shm) (2.4) 465 tty Info of tty drivers 466 uptime System uptime 467 version Kernel version 468 video bttv info of video resources (2.4) 469 vmallocinfo Show vmalloced areas 470.............................................................................. 471 472You can, for example, check which interrupts are currently in use and what 473they are used for by looking in the file /proc/interrupts: 474 475 > cat /proc/interrupts 476 CPU0 477 0: 8728810 XT-PIC timer 478 1: 895 XT-PIC keyboard 479 2: 0 XT-PIC cascade 480 3: 531695 XT-PIC aha152x 481 4: 2014133 XT-PIC serial 482 5: 44401 XT-PIC pcnet_cs 483 8: 2 XT-PIC rtc 484 11: 8 XT-PIC i82365 485 12: 182918 XT-PIC PS/2 Mouse 486 13: 1 XT-PIC fpu 487 14: 1232265 XT-PIC ide0 488 15: 7 XT-PIC ide1 489 NMI: 0 490 491In 2.4.* a couple of lines where added to this file LOC & ERR (this time is the 492output of a SMP machine): 493 494 > cat /proc/interrupts 495 496 CPU0 CPU1 497 0: 1243498 1214548 IO-APIC-edge timer 498 1: 8949 8958 IO-APIC-edge keyboard 499 2: 0 0 XT-PIC cascade 500 5: 11286 10161 IO-APIC-edge soundblaster 501 8: 1 0 IO-APIC-edge rtc 502 9: 27422 27407 IO-APIC-edge 3c503 503 12: 113645 113873 IO-APIC-edge PS/2 Mouse 504 13: 0 0 XT-PIC fpu 505 14: 22491 24012 IO-APIC-edge ide0 506 15: 2183 2415 IO-APIC-edge ide1 507 17: 30564 30414 IO-APIC-level eth0 508 18: 177 164 IO-APIC-level bttv 509 NMI: 2457961 2457959 510 LOC: 2457882 2457881 511 ERR: 2155 512 513NMI is incremented in this case because every timer interrupt generates a NMI 514(Non Maskable Interrupt) which is used by the NMI Watchdog to detect lockups. 515 516LOC is the local interrupt counter of the internal APIC of every CPU. 517 518ERR is incremented in the case of errors in the IO-APIC bus (the bus that 519connects the CPUs in a SMP system. This means that an error has been detected, 520the IO-APIC automatically retry the transmission, so it should not be a big 521problem, but you should read the SMP-FAQ. 522 523In 2.6.2* /proc/interrupts was expanded again. This time the goal was for 524/proc/interrupts to display every IRQ vector in use by the system, not 525just those considered 'most important'. The new vectors are: 526 527 THR -- interrupt raised when a machine check threshold counter 528 (typically counting ECC corrected errors of memory or cache) exceeds 529 a configurable threshold. Only available on some systems. 530 531 TRM -- a thermal event interrupt occurs when a temperature threshold 532 has been exceeded for the CPU. This interrupt may also be generated 533 when the temperature drops back to normal. 534 535 SPU -- a spurious interrupt is some interrupt that was raised then lowered 536 by some IO device before it could be fully processed by the APIC. Hence 537 the APIC sees the interrupt but does not know what device it came from. 538 For this case the APIC will generate the interrupt with a IRQ vector 539 of 0xff. This might also be generated by chipset bugs. 540 541 RES, CAL, TLB -- rescheduling, call and TLB flush interrupts are 542 sent from one CPU to another per the needs of the OS. Typically, 543 their statistics are used by kernel developers and interested users to 544 determine the occurrence of interrupts of the given type. 545 546The above IRQ vectors are displayed only when relevant. For example, 547the threshold vector does not exist on x86_64 platforms. Others are 548suppressed when the system is a uniprocessor. As of this writing, only 549i386 and x86_64 platforms support the new IRQ vector displays. 550 551Of some interest is the introduction of the /proc/irq directory to 2.4. 552It could be used to set IRQ to CPU affinity, this means that you can "hook" an 553IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the 554irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and 555prof_cpu_mask. 556 557For example 558 > ls /proc/irq/ 559 0 10 12 14 16 18 2 4 6 8 prof_cpu_mask 560 1 11 13 15 17 19 3 5 7 9 default_smp_affinity 561 > ls /proc/irq/0/ 562 smp_affinity 563 564smp_affinity is a bitmask, in which you can specify which CPUs can handle the 565IRQ, you can set it by doing: 566 567 > echo 1 > /proc/irq/10/smp_affinity 568 569This means that only the first CPU will handle the IRQ, but you can also echo 5705 which means that only the first and fourth CPU can handle the IRQ. 571 572The contents of each smp_affinity file is the same by default: 573 574 > cat /proc/irq/0/smp_affinity 575 ffffffff 576 577The default_smp_affinity mask applies to all non-active IRQs, which are the 578IRQs which have not yet been allocated/activated, and hence which lack a 579/proc/irq/[0-9]* directory. 580 581The node file on an SMP system shows the node to which the device using the IRQ 582reports itself as being attached. This hardware locality information does not 583include information about any possible driver locality preference. 584 585prof_cpu_mask specifies which CPUs are to be profiled by the system wide 586profiler. Default value is ffffffff (all cpus). 587 588The way IRQs are routed is handled by the IO-APIC, and it's Round Robin 589between all the CPUs which are allowed to handle it. As usual the kernel has 590more info than you and does a better job than you, so the defaults are the 591best choice for almost everyone. 592 593There are three more important subdirectories in /proc: net, scsi, and sys. 594The general rule is that the contents, or even the existence of these 595directories, depend on your kernel configuration. If SCSI is not enabled, the 596directory scsi may not exist. The same is true with the net, which is there 597only when networking support is present in the running kernel. 598 599The slabinfo file gives information about memory usage at the slab level. 600Linux uses slab pools for memory management above page level in version 2.2. 601Commonly used objects have their own slab pool (such as network buffers, 602directory cache, and so on). 603 604.............................................................................. 605 606> cat /proc/buddyinfo 607 608Node 0, zone DMA 0 4 5 4 4 3 ... 609Node 0, zone Normal 1 0 0 1 101 8 ... 610Node 0, zone HighMem 2 0 0 1 1 0 ... 611 612External fragmentation is a problem under some workloads, and buddyinfo is a 613useful tool for helping diagnose these problems. Buddyinfo will give you a 614clue as to how big an area you can safely allocate, or why a previous 615allocation failed. 616 617Each column represents the number of pages of a certain order which are 618available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in 619ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE 620available in ZONE_NORMAL, etc... 621 622More information relevant to external fragmentation can be found in 623pagetypeinfo. 624 625> cat /proc/pagetypeinfo 626Page block order: 9 627Pages per block: 512 628 629Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 630Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 631Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 632Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 633Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 634Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 635Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 636Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 637Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 638Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 639Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 640 641Number of blocks type Unmovable Reclaimable Movable Reserve Isolate 642Node 0, zone DMA 2 0 5 1 0 643Node 0, zone DMA32 41 6 967 2 0 644 645Fragmentation avoidance in the kernel works by grouping pages of different 646migrate types into the same contiguous regions of memory called page blocks. 647A page block is typically the size of the default hugepage size e.g. 2MB on 648X86-64. By keeping pages grouped based on their ability to move, the kernel 649can reclaim pages within a page block to satisfy a high-order allocation. 650 651The pagetypinfo begins with information on the size of a page block. It 652then gives the same type of information as buddyinfo except broken down 653by migrate-type and finishes with details on how many page blocks of each 654type exist. 655 656If min_free_kbytes has been tuned correctly (recommendations made by hugeadm 657from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can 658make an estimate of the likely number of huge pages that can be allocated 659at a given point in time. All the "Movable" blocks should be allocatable 660unless memory has been mlock()'d. Some of the Reclaimable blocks should 661also be allocatable although a lot of filesystem metadata may have to be 662reclaimed to achieve this. 663 664.............................................................................. 665 666meminfo: 667 668Provides information about distribution and utilization of memory. This 669varies by architecture and compile options. The following is from a 67016GB PIII, which has highmem enabled. You may not have all of these fields. 671 672> cat /proc/meminfo 673 674The "Locked" indicates whether the mapping is locked in memory or not. 675 676 677MemTotal: 16344972 kB 678MemFree: 13634064 kB 679Buffers: 3656 kB 680Cached: 1195708 kB 681SwapCached: 0 kB 682Active: 891636 kB 683Inactive: 1077224 kB 684HighTotal: 15597528 kB 685HighFree: 13629632 kB 686LowTotal: 747444 kB 687LowFree: 4432 kB 688SwapTotal: 0 kB 689SwapFree: 0 kB 690Dirty: 968 kB 691Writeback: 0 kB 692AnonPages: 861800 kB 693Mapped: 280372 kB 694Slab: 284364 kB 695SReclaimable: 159856 kB 696SUnreclaim: 124508 kB 697PageTables: 24448 kB 698NFS_Unstable: 0 kB 699Bounce: 0 kB 700WritebackTmp: 0 kB 701CommitLimit: 7669796 kB 702Committed_AS: 100056 kB 703VmallocTotal: 112216 kB 704VmallocUsed: 428 kB 705VmallocChunk: 111088 kB 706 707 MemTotal: Total usable ram (i.e. physical ram minus a few reserved 708 bits and the kernel binary code) 709 MemFree: The sum of LowFree+HighFree 710 Buffers: Relatively temporary storage for raw disk blocks 711 shouldn't get tremendously large (20MB or so) 712 Cached: in-memory cache for files read from the disk (the 713 pagecache). Doesn't include SwapCached 714 SwapCached: Memory that once was swapped out, is swapped back in but 715 still also is in the swapfile (if memory is needed it 716 doesn't need to be swapped out AGAIN because it is already 717 in the swapfile. This saves I/O) 718 Active: Memory that has been used more recently and usually not 719 reclaimed unless absolutely necessary. 720 Inactive: Memory which has been less recently used. It is more 721 eligible to be reclaimed for other purposes 722 HighTotal: 723 HighFree: Highmem is all memory above ~860MB of physical memory 724 Highmem areas are for use by userspace programs, or 725 for the pagecache. The kernel must use tricks to access 726 this memory, making it slower to access than lowmem. 727 LowTotal: 728 LowFree: Lowmem is memory which can be used for everything that 729 highmem can be used for, but it is also available for the 730 kernel's use for its own data structures. Among many 731 other things, it is where everything from the Slab is 732 allocated. Bad things happen when you're out of lowmem. 733 SwapTotal: total amount of swap space available 734 SwapFree: Memory which has been evicted from RAM, and is temporarily 735 on the disk 736 Dirty: Memory which is waiting to get written back to the disk 737 Writeback: Memory which is actively being written back to the disk 738 AnonPages: Non-file backed pages mapped into userspace page tables 739 Mapped: files which have been mmaped, such as libraries 740 Slab: in-kernel data structures cache 741SReclaimable: Part of Slab, that might be reclaimed, such as caches 742 SUnreclaim: Part of Slab, that cannot be reclaimed on memory pressure 743 PageTables: amount of memory dedicated to the lowest level of page 744 tables. 745NFS_Unstable: NFS pages sent to the server, but not yet committed to stable 746 storage 747 Bounce: Memory used for block device "bounce buffers" 748WritebackTmp: Memory used by FUSE for temporary writeback buffers 749 CommitLimit: Based on the overcommit ratio ('vm.overcommit_ratio'), 750 this is the total amount of memory currently available to 751 be allocated on the system. This limit is only adhered to 752 if strict overcommit accounting is enabled (mode 2 in 753 'vm.overcommit_memory'). 754 The CommitLimit is calculated with the following formula: 755 CommitLimit = ('vm.overcommit_ratio' * Physical RAM) + Swap 756 For example, on a system with 1G of physical RAM and 7G 757 of swap with a `vm.overcommit_ratio` of 30 it would 758 yield a CommitLimit of 7.3G. 759 For more details, see the memory overcommit documentation 760 in vm/overcommit-accounting. 761Committed_AS: The amount of memory presently allocated on the system. 762 The committed memory is a sum of all of the memory which 763 has been allocated by processes, even if it has not been 764 "used" by them as of yet. A process which malloc()'s 1G 765 of memory, but only touches 300M of it will only show up 766 as using 300M of memory even if it has the address space 767 allocated for the entire 1G. This 1G is memory which has 768 been "committed" to by the VM and can be used at any time 769 by the allocating application. With strict overcommit 770 enabled on the system (mode 2 in 'vm.overcommit_memory'), 771 allocations which would exceed the CommitLimit (detailed 772 above) will not be permitted. This is useful if one needs 773 to guarantee that processes will not fail due to lack of 774 memory once that memory has been successfully allocated. 775VmallocTotal: total size of vmalloc memory area 776 VmallocUsed: amount of vmalloc area which is used 777VmallocChunk: largest contiguous block of vmalloc area which is free 778 779.............................................................................. 780 781vmallocinfo: 782 783Provides information about vmalloced/vmaped areas. One line per area, 784containing the virtual address range of the area, size in bytes, 785caller information of the creator, and optional information depending 786on the kind of area : 787 788 pages=nr number of pages 789 phys=addr if a physical address was specified 790 ioremap I/O mapping (ioremap() and friends) 791 vmalloc vmalloc() area 792 vmap vmap()ed pages 793 user VM_USERMAP area 794 vpages buffer for pages pointers was vmalloced (huge area) 795 N<node>=nr (Only on NUMA kernels) 796 Number of pages allocated on memory node <node> 797 798> cat /proc/vmallocinfo 7990xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... 800 /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 8010xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... 802 /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 8030xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... 804 phys=7fee8000 ioremap 8050xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... 806 phys=7fee7000 ioremap 8070xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 8080xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... 809 /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 8100xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... 811 pages=2 vmalloc N1=2 8120xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... 813 /0x130 [x_tables] pages=4 vmalloc N0=4 8140xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... 815 pages=14 vmalloc N2=14 8160xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... 817 pages=4 vmalloc N1=4 8180xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... 819 pages=2 vmalloc N1=2 8200xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... 821 pages=10 vmalloc N0=10 822 823.............................................................................. 824 825softirqs: 826 827Provides counts of softirq handlers serviced since boot time, for each cpu. 828 829> cat /proc/softirqs 830 CPU0 CPU1 CPU2 CPU3 831 HI: 0 0 0 0 832 TIMER: 27166 27120 27097 27034 833 NET_TX: 0 0 0 17 834 NET_RX: 42 0 0 39 835 BLOCK: 0 0 107 1121 836 TASKLET: 0 0 0 290 837 SCHED: 27035 26983 26971 26746 838 HRTIMER: 0 0 0 0 839 RCU: 1678 1769 2178 2250 840 841 8421.3 IDE devices in /proc/ide 843---------------------------- 844 845The subdirectory /proc/ide contains information about all IDE devices of which 846the kernel is aware. There is one subdirectory for each IDE controller, the 847file drivers and a link for each IDE device, pointing to the device directory 848in the controller specific subtree. 849 850The file drivers contains general information about the drivers used for the 851IDE devices: 852 853 > cat /proc/ide/drivers 854 ide-cdrom version 4.53 855 ide-disk version 1.08 856 857More detailed information can be found in the controller specific 858subdirectories. These are named ide0, ide1 and so on. Each of these 859directories contains the files shown in table 1-6. 860 861 862Table 1-6: IDE controller info in /proc/ide/ide? 863.............................................................................. 864 File Content 865 channel IDE channel (0 or 1) 866 config Configuration (only for PCI/IDE bridge) 867 mate Mate name 868 model Type/Chipset of IDE controller 869.............................................................................. 870 871Each device connected to a controller has a separate subdirectory in the 872controllers directory. The files listed in table 1-7 are contained in these 873directories. 874 875 876Table 1-7: IDE device information 877.............................................................................. 878 File Content 879 cache The cache 880 capacity Capacity of the medium (in 512Byte blocks) 881 driver driver and version 882 geometry physical and logical geometry 883 identify device identify block 884 media media type 885 model device identifier 886 settings device setup 887 smart_thresholds IDE disk management thresholds 888 smart_values IDE disk management values 889.............................................................................. 890 891The most interesting file is settings. This file contains a nice overview of 892the drive parameters: 893 894 # cat /proc/ide/ide0/hda/settings 895 name value min max mode 896 ---- ----- --- --- ---- 897 bios_cyl 526 0 65535 rw 898 bios_head 255 0 255 rw 899 bios_sect 63 0 63 rw 900 breada_readahead 4 0 127 rw 901 bswap 0 0 1 r 902 file_readahead 72 0 2097151 rw 903 io_32bit 0 0 3 rw 904 keepsettings 0 0 1 rw 905 max_kb_per_request 122 1 127 rw 906 multcount 0 0 8 rw 907 nice1 1 0 1 rw 908 nowerr 0 0 1 rw 909 pio_mode write-only 0 255 w 910 slow 0 0 1 rw 911 unmaskirq 0 0 1 rw 912 using_dma 0 0 1 rw 913 914 9151.4 Networking info in /proc/net 916-------------------------------- 917 918The subdirectory /proc/net follows the usual pattern. Table 1-8 shows the 919additional values you get for IP version 6 if you configure the kernel to 920support this. Table 1-9 lists the files and their meaning. 921 922 923Table 1-8: IPv6 info in /proc/net 924.............................................................................. 925 File Content 926 udp6 UDP sockets (IPv6) 927 tcp6 TCP sockets (IPv6) 928 raw6 Raw device statistics (IPv6) 929 igmp6 IP multicast addresses, which this host joined (IPv6) 930 if_inet6 List of IPv6 interface addresses 931 ipv6_route Kernel routing table for IPv6 932 rt6_stats Global IPv6 routing tables statistics 933 sockstat6 Socket statistics (IPv6) 934 snmp6 Snmp data (IPv6) 935.............................................................................. 936 937 938Table 1-9: Network info in /proc/net 939.............................................................................. 940 File Content 941 arp Kernel ARP table 942 dev network devices with statistics 943 dev_mcast the Layer2 multicast groups a device is listening too 944 (interface index, label, number of references, number of bound 945 addresses). 946 dev_stat network device status 947 ip_fwchains Firewall chain linkage 948 ip_fwnames Firewall chain names 949 ip_masq Directory containing the masquerading tables 950 ip_masquerade Major masquerading table 951 netstat Network statistics 952 raw raw device statistics 953 route Kernel routing table 954 rpc Directory containing rpc info 955 rt_cache Routing cache 956 snmp SNMP data 957 sockstat Socket statistics 958 tcp TCP sockets 959 tr_rif Token ring RIF routing table 960 udp UDP sockets 961 unix UNIX domain sockets 962 wireless Wireless interface data (Wavelan etc) 963 igmp IP multicast addresses, which this host joined 964 psched Global packet scheduler parameters. 965 netlink List of PF_NETLINK sockets 966 ip_mr_vifs List of multicast virtual interfaces 967 ip_mr_cache List of multicast routing cache 968.............................................................................. 969 970You can use this information to see which network devices are available in 971your system and how much traffic was routed over those devices: 972 973 > cat /proc/net/dev 974 Inter-|Receive |[... 975 face |bytes packets errs drop fifo frame compressed multicast|[... 976 lo: 908188 5596 0 0 0 0 0 0 [... 977 ppp0:15475140 20721 410 0 0 410 0 0 [... 978 eth0: 614530 7085 0 0 0 0 0 1 [... 979 980 ...] Transmit 981 ...] bytes packets errs drop fifo colls carrier compressed 982 ...] 908188 5596 0 0 0 0 0 0 983 ...] 1375103 17405 0 0 0 0 0 0 984 ...] 1703981 5535 0 0 0 3 0 0 985 986In addition, each Channel Bond interface has its own directory. For 987example, the bond0 device will have a directory called /proc/net/bond0/. 988It will contain information that is specific to that bond, such as the 989current slaves of the bond, the link status of the slaves, and how 990many times the slaves link has failed. 991 9921.5 SCSI info 993------------- 994 995If you have a SCSI host adapter in your system, you'll find a subdirectory 996named after the driver for this adapter in /proc/scsi. You'll also see a list 997of all recognized SCSI devices in /proc/scsi: 998 999 >cat /proc/scsi/scsi 1000 Attached devices:
1001 Host: scsi0 Channel: 00 Id: 00 Lun: 00 1002 Vendor: IBM Model: DGHS09U Rev: 03E0 1003 Type: Direct-Access ANSI SCSI revision: 03 1004 Host: scsi0 Channel: 00 Id: 06 Lun: 00 1005 Vendor: PIONEER Model: CD-ROM DR-U06S Rev: 1.04 1006 Type: CD-ROM ANSI SCSI revision: 02 1007 1008 1009The directory named after the driver has one file for each adapter found in 1010the system. These files contain information about the controller, including 1011the used IRQ and the IO address range. The amount of information shown is 1012dependent on the adapter you use. The example shows the output for an Adaptec 1013AHA-2940 SCSI adapter: 1014 1015 > cat /proc/scsi/aic7xxx/0 1016 1017 Adaptec AIC7xxx driver version: 5.1.19/3.2.4 1018 Compile Options: 1019 TCQ Enabled By Default : Disabled 1020 AIC7XXX_PROC_STATS : Disabled 1021 AIC7XXX_RESET_DELAY : 5 1022 Adapter Configuration: 1023 SCSI Adapter: Adaptec AHA-294X Ultra SCSI host adapter 1024 Ultra Wide Controller 1025 PCI MMAPed I/O Base: 0xeb001000 1026 Adapter SEEPROM Config: SEEPROM found and used. 1027 Adaptec SCSI BIOS: Enabled 1028 IRQ: 10 1029 SCBs: Active 0, Max Active 2, 1030 Allocated 15, HW 16, Page 255 1031 Interrupts: 160328 1032 BIOS Control Word: 0x18b6 1033 Adapter Control Word: 0x005b 1034 Extended Translation: Enabled 1035 Disconnect Enable Flags: 0xffff 1036 Ultra Enable Flags: 0x0001 1037 Tag Queue Enable Flags: 0x0000 1038 Ordered Queue Tag Flags: 0x0000 1039 Default Tag Queue Depth: 8 1040 Tagged Queue By Device array for aic7xxx host instance 0: 1041 {255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255} 1042 Actual queue depth per device for aic7xxx host instance 0: 1043 {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1} 1044 Statistics: 1045 (scsi0:0:0:0) 1046 Device using Wide/Sync transfers at 40.0 MByte/sec, offset 8 1047 Transinfo settings: current(12/8/1/0), goal(12/8/1/0), user(12/15/1/0) 1048 Total transfers 160151 (74577 reads and 85574 writes) 1049 (scsi0:0:6:0) 1050 Device using Narrow/Sync transfers at 5.0 MByte/sec, offset 15 1051 Transinfo settings: current(50/15/0/0), goal(50/15/0/0), user(50/15/0/0) 1052 Total transfers 0 (0 reads and 0 writes) 1053 1054 10551.6 Parallel port info in /proc/parport 1056--------------------------------------- 1057 1058The directory /proc/parport contains information about the parallel ports of 1059your system. It has one subdirectory for each port, named after the port 1060number (0,1,2,...). 1061 1062These directories contain the four files shown in Table 1-10. 1063 1064 1065Table 1-10: Files in /proc/parport 1066.............................................................................. 1067 File Content 1068 autoprobe Any IEEE-1284 device ID information that has been acquired. 1069 devices list of the device drivers using that port. A + will appear by the 1070 name of the device currently using the port (it might not appear 1071 against any). 1072 hardware Parallel port's base address, IRQ line and DMA channel. 1073 irq IRQ that parport is using for that port. This is in a separate 1074 file to allow you to alter it by writing a new value in (IRQ 1075 number or none). 1076.............................................................................. 1077 10781.7 TTY info in /proc/tty 1079------------------------- 1080 1081Information about the available and actually used tty's can be found in the 1082directory /proc/tty.You'll find entries for drivers and line disciplines in 1083this directory, as shown in Table 1-11. 1084 1085 1086Table 1-11: Files in /proc/tty 1087.............................................................................. 1088 File Content 1089 drivers list of drivers and their usage 1090 ldiscs registered line disciplines 1091 driver/serial usage statistic and status of single tty lines 1092.............................................................................. 1093 1094To see which tty's are currently in use, you can simply look into the file 1095/proc/tty/drivers: 1096 1097 > cat /proc/tty/drivers 1098 pty_slave /dev/pts 136 0-255 pty:slave 1099 pty_master /dev/ptm 128 0-255 pty:master 1100 pty_slave /dev/ttyp 3 0-255 pty:slave 1101 pty_master /dev/pty 2 0-255 pty:master 1102 serial /dev/cua 5 64-67 serial:callout 1103 serial /dev/ttyS 4 64-67 serial 1104 /dev/tty0 /dev/tty0 4 0 system:vtmaster 1105 /dev/ptmx /dev/ptmx 5 2 system 1106 /dev/console /dev/console 5 1 system:console 1107 /dev/tty /dev/tty 5 0 system:/dev/tty 1108 unknown /dev/tty 4 1-63 console 1109 1110 11111.8 Miscellaneous kernel statistics in /proc/stat 1112------------------------------------------------- 1113 1114Various pieces of information about kernel activity are available in the 1115/proc/stat file. All of the numbers reported in this file are aggregates 1116since the system first booted. For a quick look, simply cat the file: 1117 1118 > cat /proc/stat 1119 cpu 2255 34 2290 22625563 6290 127 456 0 0 1120 cpu0 1132 34 1441 11311718 3675 127 438 0 0 1121 cpu1 1123 0 849 11313845 2614 0 18 0 0 1122 intr 114930548 113199788 3 0 5 263 0 4 [... lots more numbers ...] 1123 ctxt 1990473 1124 btime 1062191376 1125 processes 2915 1126 procs_running 1 1127 procs_blocked 0 1128 softirq 183433 0 21755 12 39 1137 231 21459 2263 1129 1130The very first "cpu" line aggregates the numbers in all of the other "cpuN" 1131lines. These numbers identify the amount of time the CPU has spent performing 1132different kinds of work. Time units are in USER_HZ (typically hundredths of a 1133second). The meanings of the columns are as follows, from left to right: 1134 1135- user: normal processes executing in user mode 1136- nice: niced processes executing in user mode 1137- system: processes executing in kernel mode 1138- idle: twiddling thumbs 1139- iowait: waiting for I/O to complete 1140- irq: servicing interrupts 1141- softirq: servicing softirqs 1142- steal: involuntary wait 1143- guest: running a normal guest 1144- guest_nice: running a niced guest 1145 1146The "intr" line gives counts of interrupts serviced since boot time, for each 1147of the possible system interrupts. The first column is the total of all 1148interrupts serviced; each subsequent column is the total for that particular 1149interrupt. 1150 1151The "ctxt" line gives the total number of context switches across all CPUs. 1152 1153The "btime" line gives the time at which the system booted, in seconds since 1154the Unix epoch. 1155 1156The "processes" line gives the number of processes and threads created, which 1157includes (but is not limited to) those created by calls to the fork() and 1158clone() system calls. 1159 1160The "procs_running" line gives the total number of threads that are 1161running or ready to run (i.e., the total number of runnable threads). 1162 1163The "procs_blocked" line gives the number of processes currently blocked, 1164waiting for I/O to complete. 1165 1166The "softirq" line gives counts of softirqs serviced since boot time, for each 1167of the possible system softirqs. The first column is the total of all 1168softirqs serviced; each subsequent column is the total for that particular 1169softirq. 1170 1171 11721.9 Ext4 file system parameters 1173------------------------------ 1174 1175Information about mounted ext4 file systems can be found in 1176/proc/fs/ext4. Each mounted filesystem will have a directory in 1177/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or 1178/proc/fs/ext4/dm-0). The files in each per-device directory are shown 1179in Table 1-12, below. 1180 1181Table 1-12: Files in /proc/fs/ext4/<devname> 1182.............................................................................. 1183 File Content 1184 mb_groups details of multiblock allocator buddy cache of free blocks 1185.............................................................................. 1186 11872.0 /proc/consoles 1188------------------ 1189Shows registered system console lines. 1190 1191To see which character device lines are currently used for the system console 1192/dev/console, you may simply look into the file /proc/consoles: 1193 1194 > cat /proc/consoles 1195 tty0 -WU (ECp) 4:7 1196 ttyS0 -W- (Ep) 4:64 1197 1198The columns are: 1199 1200 device name of the device 1201 operations R = can do read operations 1202 W = can do write operations 1203 U = can do unblank 1204 flags E = it is enabled 1205 C = it is preferred console 1206 B = it is primary boot console 1207 p = it is used for printk buffer 1208 b = it is not a TTY but a Braille device 1209 a = it is safe to use when cpu is offline 1210 major:minor major and minor number of the device separated by a colon 1211 1212------------------------------------------------------------------------------ 1213Summary 1214------------------------------------------------------------------------------ 1215The /proc file system serves information about the running system. It not only 1216allows access to process data but also allows you to request the kernel status 1217by reading files in the hierarchy. 1218 1219The directory structure of /proc reflects the types of information and makes 1220it easy, if not obvious, where to look for specific data. 1221------------------------------------------------------------------------------ 1222 1223------------------------------------------------------------------------------ 1224CHAPTER 2: MODIFYING SYSTEM PARAMETERS 1225------------------------------------------------------------------------------ 1226 1227------------------------------------------------------------------------------ 1228In This Chapter 1229------------------------------------------------------------------------------ 1230* Modifying kernel parameters by writing into files found in /proc/sys 1231* Exploring the files which modify certain parameters 1232* Review of the /proc/sys file tree 1233------------------------------------------------------------------------------ 1234 1235 1236A very interesting part of /proc is the directory /proc/sys. This is not only 1237a source of information, it also allows you to change parameters within the 1238kernel. Be very careful when attempting this. You can optimize your system, 1239but you can also cause it to crash. Never alter kernel parameters on a 1240production system. Set up a development machine and test to make sure that 1241everything works the way you want it to. You may have no alternative but to 1242reboot the machine once an error has been made. 1243 1244To change a value, simply echo the new value into the file. An example is 1245given below in the section on the file system data. You need to be root to do 1246this. You can create your own boot script to perform this every time your 1247system boots. 1248 1249The files in /proc/sys can be used to fine tune and monitor miscellaneous and 1250general things in the operation of the Linux kernel. Since some of the files 1251can inadvertently disrupt your system, it is advisable to read both 1252documentation and source before actually making adjustments. In any case, be 1253very careful when writing to any of these files. The entries in /proc may 1254change slightly between the 2.1.* and the 2.2 kernel, so if there is any doubt 1255review the kernel documentation in the directory /usr/src/linux/Documentation. 1256This chapter is heavily based on the documentation included in the pre 2.2 1257kernels, and became part of it in version 2.2.1 of the Linux kernel. 1258 1259Please see: Documentation/sysctls/ directory for descriptions of these 1260entries. 1261 1262------------------------------------------------------------------------------ 1263Summary 1264------------------------------------------------------------------------------ 1265Certain aspects of kernel behavior can be modified at runtime, without the 1266need to recompile the kernel, or even to reboot the system. The files in the 1267/proc/sys tree can not only be read, but also modified. You can use the echo 1268command to write value into these files, thereby changing the default settings 1269of the kernel. 1270------------------------------------------------------------------------------ 1271 1272------------------------------------------------------------------------------ 1273CHAPTER 3: PER-PROCESS PARAMETERS 1274------------------------------------------------------------------------------ 1275 12763.1 /proc/<pid>/oom_adj & /proc/<pid>/oom_score_adj- Adjust the oom-killer score 1277-------------------------------------------------------------------------------- 1278 1279These file can be used to adjust the badness heuristic used to select which 1280process gets killed in out of memory conditions. 1281 1282The badness heuristic assigns a value to each candidate task ranging from 0 1283(never kill) to 1000 (always kill) to determine which process is targeted. The 1284units are roughly a proportion along that range of allowed memory the process 1285may allocate from based on an estimation of its current memory and swap use. 1286For example, if a task is using all allowed memory, its badness score will be 12871000. If it is using half of its allowed memory, its score will be 500. 1288 1289There is an additional factor included in the badness score: root 1290processes are given 3% extra memory over other tasks. 1291 1292The amount of "allowed" memory depends on the context in which the oom killer 1293was called. If it is due to the memory assigned to the allocating task's cpuset 1294being exhausted, the allowed memory represents the set of mems assigned to that 1295cpuset. If it is due to a mempolicy's node(s) being exhausted, the allowed 1296memory represents the set of mempolicy nodes. If it is due to a memory 1297limit (or swap limit) being reached, the allowed memory is that configured 1298limit. Finally, if it is due to the entire system being out of memory, the 1299allowed memory represents all allocatable resources. 1300 1301The value of /proc/<pid>/oom_score_adj is added to the badness score before it 1302is used to determine which task to kill. Acceptable values range from -1000 1303(OOM_SCORE_ADJ_MIN) to +1000 (OOM_SCORE_ADJ_MAX). This allows userspace to 1304polarize the preference for oom killing either by always preferring a certain 1305task or completely disabling it. The lowest possible value, -1000, is 1306equivalent to disabling oom killing entirely for that task since it will always 1307report a badness score of 0. 1308 1309Consequently, it is very simple for userspace to define the amount of memory to 1310consider for each task. Setting a /proc/<pid>/oom_score_adj value of +500, for 1311example, is roughly equivalent to allowing the remainder of tasks sharing the 1312same system, cpuset, mempolicy, or memory controller resources to use at least 131350% more memory. A value of -500, on the other hand, would be roughly 1314equivalent to discounting 50% of the task's allowed memory from being considered 1315as scoring against the task. 1316 1317For backwards compatibility with previous kernels, /proc/<pid>/oom_adj may also 1318be used to tune the badness score. Its acceptable values range from -16 1319(OOM_ADJUST_MIN) to +15 (OOM_ADJUST_MAX) and a special value of -17 1320(OOM_DISABLE) to disable oom killing entirely for that task. Its value is 1321scaled linearly with /proc/<pid>/oom_score_adj. 1322 1323Writing to /proc/<pid>/oom_score_adj or /proc/<pid>/oom_adj will change the 1324other with its scaled value. 1325 1326The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last 1327value set by a CAP_SYS_RESOURCE process. To reduce the value any lower 1328requires CAP_SYS_RESOURCE. 1329 1330NOTICE: /proc/<pid>/oom_adj is deprecated and will be removed, please see 1331Documentation/feature-removal-schedule.txt. 1332 1333Caveat: when a parent task is selected, the oom killer will sacrifice any first 1334generation children with separate address spaces instead, if possible. This 1335avoids servers and important system daemons from being killed and loses the 1336minimal amount of work. 1337 1338 13393.2 /proc/<pid>/oom_score - Display current oom-killer score 1340------------------------------------------------------------- 1341 1342This file can be used to check the current score used by the oom-killer is for 1343any given <pid>. Use it together with /proc/<pid>/oom_adj to tune which 1344process should be killed in an out-of-memory situation. 1345 1346 13473.3 /proc/<pid>/io - Display the IO accounting fields 1348------------------------------------------------------- 1349 1350This file contains IO statistics for each running process 1351 1352Example 1353------- 1354 1355test:/tmp # dd if=/dev/zero of=/tmp/test.dat & 1356[1] 3828 1357 1358test:/tmp # cat /proc/3828/io 1359rchar: 323934931 1360wchar: 323929600 1361syscr: 632687 1362syscw: 632675 1363read_bytes: 0 1364write_bytes: 323932160 1365cancelled_write_bytes: 0 1366 1367 1368Description 1369----------- 1370 1371rchar 1372----- 1373 1374I/O counter: chars read 1375The number of bytes which this task has caused to be read from storage. This 1376is simply the sum of bytes which this process passed to read() and pread(). 1377It includes things like tty IO and it is unaffected by whether or not actual 1378physical disk IO was required (the read might have been satisfied from 1379pagecache) 1380 1381 1382wchar 1383----- 1384 1385I/O counter: chars written 1386The number of bytes which this task has caused, or shall cause to be written 1387to disk. Similar caveats apply here as with rchar. 1388 1389 1390syscr 1391----- 1392 1393I/O counter: read syscalls 1394Attempt to count the number of read I/O operations, i.e. syscalls like read() 1395and pread(). 1396 1397 1398syscw 1399----- 1400 1401I/O counter: write syscalls 1402Attempt to count the number of write I/O operations, i.e. syscalls like 1403write() and pwrite(). 1404 1405 1406read_bytes 1407---------- 1408 1409I/O counter: bytes read 1410Attempt to count the number of bytes which this process really did cause to 1411be fetched from the storage layer. Done at the submit_bio() level, so it is 1412accurate for block-backed filesystems. <please add status regarding NFS and 1413CIFS at a later time> 1414 1415 1416write_bytes 1417----------- 1418 1419I/O counter: bytes written 1420Attempt to count the number of bytes which this process caused to be sent to 1421the storage layer. This is done at page-dirtying time. 1422 1423 1424cancelled_write_bytes 1425--------------------- 1426 1427The big inaccuracy here is truncate. If a process writes 1MB to a file and 1428then deletes the file, it will in fact perform no writeout. But it will have 1429been accounted as having caused 1MB of write. 1430In other words: The number of bytes which this process caused to not happen, 1431by truncating pagecache. A task can cause "negative" IO too. If this task 1432truncates some dirty pagecache, some IO which another task has been accounted 1433for (in its write_bytes) will not be happening. We _could_ just subtract that 1434from the truncating task's write_bytes, but there is information loss in doing 1435that. 1436 1437 1438Note 1439---- 1440 1441At its current implementation state, this is a bit racy on 32-bit machines: if 1442process A reads process B's /proc/pid/io while process B is updating one of 1443those 64-bit counters, process A could see an intermediate result. 1444 1445 1446More information about this can be found within the taskstats documentation in 1447Documentation/accounting. 1448 14493.4 /proc/<pid>/coredump_filter - Core dump filtering settings 1450--------------------------------------------------------------- 1451When a process is dumped, all anonymous memory is written to a core file as 1452long as the size of the core file isn't limited. But sometimes we don't want 1453to dump some memory segments, for example, huge shared memory. Conversely, 1454sometimes we want to save file-backed memory segments into a core file, not 1455only the individual files. 1456 1457/proc/<pid>/coredump_filter allows you to customize which memory segments 1458will be dumped when the <pid> process is dumped. coredump_filter is a bitmask 1459of memory types. If a bit of the bitmask is set, memory segments of the 1460corresponding memory type are dumped, otherwise they are not dumped. 1461 1462The following 7 memory types are supported: 1463 - (bit 0) anonymous private memory 1464 - (bit 1) anonymous shared memory 1465 - (bit 2) file-backed private memory 1466 - (bit 3) file-backed shared memory 1467 - (bit 4) ELF header pages in file-backed private memory areas (it is 1468 effective only if the bit 2 is cleared) 1469 - (bit 5) hugetlb private memory 1470 - (bit 6) hugetlb shared memory 1471 1472 Note that MMIO pages such as frame buffer are never dumped and vDSO pages 1473 are always dumped regardless of the bitmask status. 1474 1475 Note bit 0-4 doesn't effect any hugetlb memory. hugetlb memory are only 1476 effected by bit 5-6. 1477 1478Default value of coredump_filter is 0x23; this means all anonymous memory 1479segments and hugetlb private memory are dumped. 1480 1481If you don't want to dump all shared memory segments attached to pid 1234, 1482write 0x21 to the process's proc file. 1483 1484 $ echo 0x21 > /proc/1234/coredump_filter 1485 1486When a new process is created, the process inherits the bitmask status from its 1487parent. It is useful to set up coredump_filter before the program runs. 1488For example: 1489 1490 $ echo 0x7 > /proc/self/coredump_filter 1491 $ ./some_program 1492 14933.5 /proc/<pid>/mountinfo - Information about mounts 1494-------------------------------------------------------- 1495 1496This file contains lines of the form: 1497 149836 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue 1499(1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11) 1500 1501(1) mount ID: unique identifier of the mount (may be reused after umount) 1502(2) parent ID: ID of parent (or of self for the top of the mount tree) 1503(3) major:minor: value of st_dev for files on filesystem 1504(4) root: root of the mount within the filesystem 1505(5) mount point: mount point relative to the process's root 1506(6) mount options: per mount options 1507(7) optional fields: zero or more fields of the form "tag[:value]" 1508(8) separator: marks the end of the optional fields 1509(9) filesystem type: name of filesystem of the form "type[.subtype]" 1510(10) mount source: filesystem specific information or "none" 1511(11) super options: per super block options 1512 1513Parsers should ignore all unrecognised optional fields. Currently the 1514possible optional fields are: 1515 1516shared:X mount is shared in peer group X 1517master:X mount is slave to peer group X 1518propagate_from:X mount is slave and receives propagation from peer group X (*) 1519unbindable mount is unbindable 1520 1521(*) X is the closest dominant peer group under the process's root. If 1522X is the immediate master of the mount, or if there's no dominant peer 1523group under the same root, then only the "master:X" field is present 1524and not the "propagate_from:X" field. 1525 1526For more information on mount propagation see: 1527 1528 Documentation/filesystems/sharedsubtree.txt 1529 1530 15313.6 /proc/<pid>/comm & /proc/<pid>/task/<tid>/comm 1532-------------------------------------------------------- 1533These files provide a method to access a tasks comm value. It also allows for 1534a task to set its own or one of its thread siblings comm value. The comm value 1535is limited in size compared to the cmdline value, so writing anything longer 1536then the kernel's TASK_COMM_LEN (currently 16 chars) will result in a truncated 1537comm value. 1538