book

Linux® Kernel Primer, The: A Top-Down Approach for x86 and PowerPC Architectures

by Claudia Salzberg Rodriguez, Gordon Fischer, Steven Smolski

September 2005

Intermediate to advanced

648 pages

12h 50m

English

Pearson

Read now

Unlock full access

Dedication
Intended AudienceOrganization of MaterialOur ApproachConventions
1.1. History of UNIX1.2. Standards and Common Interfaces1.3. Free Software and Open Source1.4. A Quick Survey of Linux Distributions1.4.1. Debian1.4.2. Red Hat/Fedora1.4.3. Mandriva1.4.4. SUSE1.4.5. Gentoo1.4.6. Yellow Dog1.4.7. Other Distros1.5. Kernel Release Information1.6. Linux on Power1.7. What Is an Operating System?1.8. Kernel Organization1.9. Overview of the Linux Kernel1.9.1. User Interface1.9.2. User Identification1.9.3. Files and Filesystems1.9.3.1. Directories, Files, and Pathnames1.9.3.2. Filesystem Mounting1.9.3.3. File Protection and Access Rights1.9.3.4. File Modesstickysuidsgid1.9.3.5. File Metadata1.9.3.6. Types of FilesRegular FileDirectoryBlock DevicesCharacter DevicesLinkNamed PipesSockets1.9.3.7. Types of Filesystems1.9.3.8. File Control1.9.4. Processes1.9.4.1. Process Creation and Control1.9.4.2. Process IDs1.9.4.3. Process Groups1.9.4.4. Process States1.9.4.5. Process Descriptor1.9.4.6. Process Priority1.9.5. System Calls1.9.6. Linux Scheduler1.9.7. Linux Device Drivers1.10. Portability and Architecture DependenceSummaryExercises
2.1. Common Kernel Datatypes2.1.1. Linked ListsLine 34Line 37Line 492Line 4932.1.2. Searching2.1.3. Trees2.1.3.1. Binary Trees2.1.3.2. Red Black Trees2.2. Assembly2.2.1. PowerPC2.2.1.1. Branch Instructions2.2.1.2. Fixed-Point Instructions2.2.1.3. Floating-Point Instructions2.2.2. x862.2.2.1. Control InstructionsLine 100Line 101Line 102Line 103Line 1042.2.2.2. Arithmetic Instructions2.2.2.3. Data InstructionsLine 100Line 101Line 102Line 103Line 1042.3. Assembly Language ExampleLine 1Line 3Line 5Line 82.3.1. x86 Assembly ExampleLine 9Line 10Line 11Line 12Line 13Line 14Line 15Line 16Line 17Line 18Line 19Line 20Line 21Line 22Line 23Line 24Line 25Line 26Line 27Line 28Line 29Line 30Line 312.3.2. PowerPC Assembly ExampleLine 7Line 8Line 9Line 10Line 11Line 12Line 13Line 14Line 15Line 16Line 17Line 18Line 19Line 20Line 21Line 22Line 23Line 24Line 25Line 26Line 27Line 28Line 29Line 30Line 31Line 32Line 33Line 342.4. Inline Assembly2.4.1. Ouput Operands2.4.2. Input Operands2.4.3. Clobbered Registers (or Clobber List)2.4.4. Parameter Numbering2.4.5. Constraints2.4.6. asm2.4.7. __volatile__Line 6Line 8Line 9Line 10Line 11Line 12Line 13Line 14Line 15Line 16Line 12Line 15Line 16Line 17Line 23Lines 17–24Line 26Line 27Line 28Line 29Line 103Line 104Line 106Line 108Lines 109–111Line 112Line 113Line 114Line 1152.5. Quirky C Language Usage2.5.1. asmlinkage2.5.2. UL2.5.3. inline2.5.4. const and volatile2.6. A Quick Tour of Kernel Exploration Tools2.6.1. objdump/readelf2.6.2. hexdump2.6.3. nm2.6.4. objcopy2.6.5. ar2.7. Kernel Speak: Listening to Kernel Messages2.7.1. printk()2.7.2. dmesg2.7.3. /var/log/messages2.8. Miscellaneous Quirks2.8.1. __init2.8.2. likely() and unlikely()2.8.3. IS_ERR and PTR_ERR2.8.4. Notifier ChainsSummaryProject: HellomodStep 1: Writing the Linux Module SkeletonLine 4Line 5Line 6Line 7Lines 9–12Lines 15–18Line 20Line 21Step 2: Compiling the ModuleLine 1Step 3: Running the CodeExercises
3.1. Introducing Our Program3.2. Process Descriptor3.2.1. Process Attribute–Related Fields3.2.1.1. state3.2.1.2. pid3.2.1.3. flags3.2.1.4. binfmt3.2.1.5. exit_code and exit_signal3.2.1.6. pdeath_signal3.2.1.7. comm3.2.1.8. ptrace3.2.2. Scheduling Related Fields3.2.2.1. prio3.2.2.2. static_prio3.2.2.3. run_list3.2.2.4. array3.2.2.5. sleep_avg3.2.2.6. timestamp3.2.2.7. interactive_credit3.2.2.8. policy3.2.2.9. cpus_allowed3.2.2.10. time_slice3.2.2.11. first_time_slice3.2.2.12. activated3.2.2.13. rt_priority3.2.2.14. nivcsw and nvcsw3.2.3. Process Relations–Related Fields3.2.3.1. real_parent3.2.3.2. parent3.2.3.3. children3.2.3.4. sibling3.2.3.5. group_leader3.2.4. Process Credentials–Related Fields3.2.4.1. uid and gid3.2.4.2. euid and egid3.2.4.3. suid and sgid3.2.4.4. fsuid and fsgid3.2.4.5. group_info3.2.5. Process Capabilities–Related Fields3.2.5.1. cap_effective, cap_inheritable, cap_permitted, and keep_capabilities3.2.6. Process Limitations–Related Fields3.2.6.1. rlim3.2.7. Filesystem- and Address Space–Related Fields3.2.7.1. fs3.2.7.2. files3.2.7.3. mm3.2.7.4. active_mm3.3. Process Creation: fork(), vfork(), and clone() System Calls3.3.1. fork() Function3.3.2. vfork() Function3.3.3. clone() Function3.3.4. do_fork() FunctionLines 1178–1183Line 1184Line 1189Lines 1194–1197Lines 1199–1205Lines 1226–1229Lines 1231–1234Lines 1236–1239Line 12483.4. Process Lifespan3.4.1. Process States3.4.2. Process State Transitions3.4.2.1. Ready to RunningTASK_RUNNING to TASK_RUNNING3.4.2.2. Running to ReadyTASK_RUNNING to TASK_RUNNING3.4.2.3. Running to BlockedTASK_RUNNING to TASK_INTERRUPTIBLETASK_RUNNING to TASK_UNINTERRUPTIBLETASK_RUNNING to TASK_ZOMBIETASK_RUNNING to TASK_STOPPEDTASK_UNINTERRUPTIBLE or TASK_INTERRUPTIBLE to TASK_STOPPED3.4.2.4. Blocked to Ready3.5. Process Termination3.5.1. sys_exit() Function3.5.2. do_exit() FunctionLine 707Lines 711–716Line 719Lines 729–732Lines 735–742Lines 744–745Lines 747–749Line 751Line 752Line 754Line 7573.5.3. Parent Notification and sys_wait4()Line 1031Lines 1033 and 1040Line 1037–1038Line 1042Line 1043Lines 1046 and 1094Line 1051Line 1054Line 1058–1079Line 1106Line 1107–11113.6. Keeping Track of Processes: Basic Scheduler Construction3.6.1. Basic Structure3.6.2. Waking Up from Waiting or Activation3.6.2.1. sched_fork(): Scheduler Initialization for Newly Forked ProcessLine 727Line 728–730Lines 740–753Lines 754–767Lines 930–934Lines 940–947Line 948Line 749Lines 951–960Lines 311–312Line 313Line 314Line 315Line 3163.7. Wait Queues3.7.1. Adding to the Wait Queue3.7.2. Waiting on the EventLine 124–126Lines 127–132Lines 133–1343.7.3. Waking UpLine 2336Lines 2340, 2342Line 2341Line 2313Line 2315Line 2317Line 2320Line 2322Lines 2322–23253.8. Asynchronous Execution Flow3.8.1. Exceptions3.8.1.1. System CallsLine 485Line 4883.8.2. Interrupts3.8.2.1. Interrupt Handlers3.8.2.2. IRQ StructuresStruct irq_desc_tLine 61Line 62Line 63Line 64Lines 65–66Line 67Struct irqactionLine 36Line 37Line 39Struct hw_interrupt_typeLine 41Lines 42–483.8.2.3. An Interrupt Example: System TimerLine 413Line 902Lines 908–910Lines 912–935Lines 941–942Line 947Line 949Line 415Line 410Lines 342–348Lines 422–426Line 431Line 437Line 419Line 317Line 351Line 72Lines 81–84Interrupt TimeProcessing the PowerPC External Interrupt VectorLines 513–530Line 523Line 524Line 432Line 435Line 436Lines 441–443Line 450Lines 451–462Lines 463–465Lines 489–497Processing the PowerPC System Timer InterruptLine 152Line 159Lines 165 and 195Line 167Line 189Line 198Line 208Processing the x86 System Timer InterruptLine 274Line 287Line 227Line 18Line 25SummaryProject: current System VariableProject Source CodeLine 4Line 5Line 12Line 15Line 16Line 18Lines 32–38Running the CodeExercises
4.1. Pages4.1.1. flags4.1.1.1. count4.1.1.2. list4.1.1.3. mapping4.1.1.4. lru4.1.1.5. virtual4.2. Memory Zones4.2.1. Memory Zone Descriptor4.2.1.1. lock4.2.1.2. free_pages4.2.1.3. pages_min, pages_low, and pages_high4.2.1.4. lru_lock4.2.1.5. active_list and inactive_list4.2.1.6. all_unreclaimable4.2.1.7. pages_scanned, temp_priority, and prev_priority4.2.1.8. free_area4.2.1.9. wait_table, wait_table_size, and wait_table_bits4.2.2. Memory Zone Helper Functions4.2.2.1. for_each_zone()4.2.2.2. is_highmem() and is_normal()4.3. Page Frames4.3.1. Functions for Requesting Page Frames4.3.1.1. alloc_pages() and alloc_page()4.3.1.2. __get_free_page() and __get_dma_pages()4.3.2. Functions for Releasing Page Frames4.3.3. Buddy SystemLine 586Line 599Lines 608–628Lines 631–632Lines 635–653Lines 720–727Lines 728–730Lines 184–215Line 2164.4. Slab Allocator4.4.1. Cache Descriptor4.4.1.1. listslists.slabs_partiallists.slabs_fulllists.slabs_freelists.next_reap4.4.1.2. objsize4.4.1.3. flags4.4.1.4. num4.4.1.5. gfporder4.4.1.6. gfpflags4.4.1.7. slabp_cache4.4.1.8. ctor4.4.1.9. dtor4.4.1.10. name4.4.1.11. next4.4.2. General Purpose Cache Descriptor4.4.2.1. cs_size4.4.2.2. cs_cachep4.4.2.3. cs_dmacachep4.4.3. Slab Descriptor4.4.3.1. list4.4.3.2. s_mem4.4.3.3. inuse4.4.3.4. free4.5. Slab Allocator's Lifecycle4.5.1. Global Variables of the Slab Allocator4.5.2. Creating a Cache4.5.2.1. kmem_cache_init()Lines 661–663Lines 669–670Line 690Line 691Line 692Line 693Lines 695–698Line 708Lines 714–718Lines 726–730Lines 732–7334.5.2.2. kmem_cache_create()namesizeoffsetflagsctor and dtorLines 1079–1084Lines 1144–1174Lines 1200–1217Lines 1243–1244Lines 1247–12764.5.3. Slab Creation and cache_grow()4.5.3.1. cache_grow()Lines 1572–1573Lines 1582–1585Lines 1597–1598Lines 1601–1602Lines 1605–1613Line 1615Lines 1616–1619Lines 1622–1624Lines 1625–1626Lines 1627–1628Lines 1629–16324.5.4. Slab Destruction: Returning Memory and kmem_cache_destroy()4.5.4.1. kmem_cache_destroy()Lines 1425–1426Lines 1429–1434Lines 1436–1442Line 14504.6. Memory Request Path4.6.1. kmalloc()4.6.1.1. size4.6.1.2. flagsLines 2102–2104Lines 2112–21134.6.2. kmem_cache_alloc()4.6.2.1. cachep4.6.2.2. flags4.7. Linux Process Memory Structures4.7.1. mm_struct4.7.1.1. mmap4.7.1.2. mm_rb4.7.1.3. mmap_cache4.7.1.4. pgd4.7.1.5. mm_users4.7.1.6. mm_count4.7.1.7. map_count4.7.1.8. mm_list4.7.1.9. start_code and end_code4.7.1.10. start_data and end_data4.7.1.11. start_brk and brk4.7.1.12. start_stack4.7.1.13. arg_start and arg_end4.7.1.14. env_start and env_end4.7.2. vm_area_struct4.7.2.1. vm_mm4.7.2.2. vm_start and vm_end4.7.2.3. vm_next4.7.2.4. vm_flags4.7.2.5. vm_rb4.7.2.6. vm_ops4.8. Process Image Layout and Linear Address Space4.9. Page Tables4.10. Page Fault4.10.1. x86 Page Fault Exception4.10.2. Page Fault HandlerLine 223Line 232Lines 246–248Line 253Line 257Lines 262–263Line 265Lines 267–269Lines 270–271Lines 272–273Lines 274–284Lines 473–509Lines 294–304Lines 305–309Lines 318–329Line 348Lines 351–353Lines 354–366Line 390Line 4324.10.3. PowerPC Page Fault ExceptionSummaryProject: Process Memory MapExercises

5.1. How Hardware Does It: Busses, Bridges, Ports, and Interfaces5.2. Devices5.2.1. Block Device Overview5.2.1.1. Generic Block Device Layer5.2.2. Request Queues and Scheduling I/O5.2.2.1. No-Op I/O Scheduler5.2.2.2. Deadline I/O Scheduler5.2.2.3. Anticipatory I/O Scheduling5.2.2.4. Request QueueLine 275Line 276Line 277Line 282Lines 283–293Lines 294–303Lines 304–311Lines 312–317Lines 318–321Lines 323–360Line 1402Line 1406Line 1414Lines 1417–1424Line 1426Line 1428Line 1429Line 1431Line 1432Line 1434Lines 1435–14395.2.3. Example: “Generic” Block DriverLine 82Line 83Line 85Line 87Line 88Line 89Line 90Line 92Line 196Line 198Line 1995.2.4. Device OperationsLine 2599Line 2601Line 2602Line 2603Line 26045.2.5. Character Device Overview5.2.6. A Note on Network Devices5.2.7. Clock Devices5.2.8. Terminal Devices5.2.9. Direct Memory Access (DMA)SummaryProject: Building a Parallel Port DriverParallel Port HardwareParallel Port Software1. Setting Up the File Operations (fops)2. Setting Up the Module Initialization Routine3. Setting Up the Module Cleanup Routine4. Inserting the Module5. Application CodeLine 4Line 12Line 18Line 20Line 21Line 22Lines 24–27Lines 28–33Line 38Exercises
6.1. General Filesystem Concepts6.1.1. File and Filenames6.1.2. File Types6.1.3. Additional File Attributes6.1.4. Directories and Pathnames6.1.5. File Operations6.1.6. File Descriptors6.1.7. Disk Blocks, Partitions, and Implementation6.1.8. Performance6.2. Linux Virtual Filesystem6.2.1. VFS Data Structures6.2.1.1. superblock StructureLine 667Line 672Line 673Line 674Line 675Line 681Line 690Line 693Line 696Line 7026.2.1.2. superblock Operations6.2.1.3. inode StructureLine 369Line 370Line 371Line 372Line 373Line 392Line 4076.2.1.4. dentry StructureLine 85Lines 85–88Line 91Line 92Line 1006.2.1.5. file StructureLine 507Line 508Line 509Line 510Line 511Line 512Line 514Line 516Line 517Line 5276.2.2. Global and Local List References6.3. Structures Associated with VFS6.3.1. fs_struct Structure6.3.1.1. count6.3.1.2. umask6.3.1.3. root, pwd, and altroot6.3.1.4. rootmnt, pwdmnt, and altrootmnt6.3.2. files_struct StructureLine 23Line 25Line 26Line 27Line 28Lines 30–326.3.2.1. close_on_exec6.3.2.2. open_fds6.3.2.3. close_on_exec6.3.2.4. open_fds_init6.3.2.5. fd_array6.4. Page Cache6.4.1. address_space Structure6.4.2. buffer_head Structure6.5. VFS System Calls and the Filesystem Layer6.5.1. open ()Lines 932–934Line 935Line 938Line 940Lines 745–749Line 751Line 753Line 852Lines 855–856Lines 866–869Line 870Lines 872–877Line 944Line 947Line 949Line 9526.5.2. close ()Line 1023Lines 1025–1030Lines 1031–1032Line 1033Lines 890–891Lines 1034–1035Lines 991–993Lines 995–997Lines 1000–1001Line 10086.5.3. read()Line 272Lines 273–282Line 200Line 202Lines 205–208Lines 210–214Lines 215–2176.5.3.1. Moving from the Generic to the SpecificLines 924–925Line 927Lines 928 and 931Line 932Lines 933–935Lines 835–842Lines 845–862Lines 863–886Lines 889–890Lines 891–898Line 838Line 839Line 840Line 841Lines 899Lines 900–9096.5.3.2. Tracing the Page CacheLines 1420–1430Line 652Lines 658–660Lines 662–681Lines 682–689Lines 690–723Lines 724–740Lines 741–743Lines 746–750Lines 751–758Lines 698–772Lines 773–775Lines 776–777Lines 779–781Line 786Lines 787–788Line 789Lines 360–361Lines 363–364Lines 365–367Line 90Lines 92–94Lines 95–96Lines 2433–2443Lines 2445–2452Line 24546.5.4. write()Lines 44–45Line 6326.5.4.1. Flushing Dirty PagesSummaryExercises
7.1. Linux Scheduler7.1.1. Choosing the Next TaskLines 2213–2218Lines 2224–2225Line 2227Lines 2233–2241Line 2243Lines 2244–2252Lines 2255–2264Lines 2266–2268Lines 2270–2273Lines 2275–2285Line 2286Line 2288Line 2290Lines 2292–2298Lines 2300–2304Lines 2306–2308Lines 2314–23187.1.2. Context SwitchLine 39Line 42Line 477.1.2.1. Following the x86 Trail of switch_to()Line 12Lines 15–16Lines 17 and 30Lines 17–18Line 19Line 20Line 21Line 22Line 23Lines 24–25Lines 26–29Line 297.1.2.2. Following the PPC context_switch()Line 157Line 158Line 159Lines 1437–1440Lines 1461–14657.1.2.3. Following the PPC Trail of switch_to()Line 205Lines 247–248Line 249Line 250Tracing the PPC Code for _switch()Lines 438–460Line 461Lines 463–465Line 466Line 468Line 469Lines 478–4867.1.3. Yielding the CPULines 1981–1986Line 1988Lines 1990–1991Lines 1994–2000Lines 2002–2011Lines 2012–2016Lines 2019–2022Line 2023Lines 2031–2046Lines 2047–2061Lines 2079–2089Lines 2090–20947.1.3.1. Dynamic Priority CalculationLines 386–389Lines 391–404Lines 405–410Lines 411–418Lines 419–432Lines 434–450Line 4527.1.3.2. DeactivationLine 509Lines 510–511Line 512–513Line 305Lines 306–3087.2. Preemption7.2.1. Explicit Kernel Preemption7.2.2. Implicit User PreemptionLines 988–996Lines 1003–10067.2.3. Implicit Kernel PreemptionLines 46–50Lines 40–44Line 2336–2337Line 2340–23477.3. Spinlocks and Semaphores7.4. System Clock: Of Time and Timers7.4.1. Real-Time Clock: What Time Is It?7.4.2. Reading the PPC Real-Time Clock7.4.3. Reading the x86 Real-Time ClockSummaryExercises
8.1. BIOS and Open Firmware8.2. Boot Loaders8.2.1. GRUB8.2.1.1. Multiboot Specification8.2.2. LILO8.2.3. PowerPC and Yaboot8.3. Architecture-Dependent Memory Initialization8.3.1. PowerPC Hardware Memory Management8.3.1.1. Real Addressing Mode8.3.1.2. Address TranslationSegmented Address Translation Direct Store Segment TBlock Address TranslationTranslation Lookaside BuffersStorage Access Mode Control8.3.1.3. How Linux Uses PPC Address TranslationLine 40Line 54Line 64Line 68Line 73Line 89Line 131Line 150Line 170Lines 173–176Lines 181–182Lines 188–198Line 210Lines 224–232Line 1337Line 1364Line 1365Line 1385Line 1395Lines 1412–14198.3.2. x86 Intel-Based Hardware Memory ManagementLines 307–345Lines 595–628Lines 723–739Lines 790–797Lines 981–1006Lines 830–831Lines 834–839Line 8528.3.2.1. Protected ModeLine 57Line 63Lines 64–68Lines 91–111Lines 113–157Lines 162–164Lines 165–168Line 170Lines 177–178Lines 180–185Line 191Lines 197–214Lines 215–3038.3.3. PowerPC and x86 Code Convergence8.4. Initial RAM Disk8.5. The Beginning: start_kernel()8.5.1. The Call to lock_kernel()Line 405Lines 44–48Lines 10–15Lines 59–638.5.2. The Call to page_address_init()Line 406Line 597Lines 598–599Lines 600–603Line 6048.5.3. The Call to printk(linux_banner)Line 4078.5.4. The Call to setup_archLine 408Line 1087Line 1088Line 1089Lines 1103–1116Lines 1118–1122Lines 1124–1127Line 1129Lines 1133–1141Line 1143Line 1145Lines 1153–1155Lines 1157–1167Line 1170Lines 1172–1174Lines 1175–1176Line 1181Lines 1183–1186Line 1188Lines 1190–11978.5.5. The Call to setup_per_cpu_areas()Line 409Lines 329–332Lines 334–341Lines 343–3468.5.6. The Call to smp_prepare_boot_cpu()Line 4158.5.7. The Call to sched_init()Line 422Lines 3919–3926Lines 3938–3947Lines 3952–3956Lines 3961–39628.5.8. The Call to build_all_zonelists()Line 4248.5.9. The Call to page_alloc_initLine 425Line 1809Line 1788Lines 1794–18028.5.10. The Call to parse_args()Line 427Lines 116–125Lines 126–153Lines 46–54Lines 55–61Lines 62–668.5.11. The Call to trap_init()Line 4318.5.12. The Call to rcu_init()Line 4328.5.13. The Call to init_IRQ()Line 433Lines 422–432Line 437Line 443Lines 449–450Line 704Line 7078.5.14. The Call to softirq_init()Line 436Lines 319–3208.5.15. The Call to time_init()Line 437Lines 379–387Lines 388–391Line 3938.5.16. The Call to console_init()Line 444Line 2352Line 2359Line 23668.5.17. The Call to profile_init()Line 447Lines 34–35Lines 38–39Lines 41–428.5.18. The Call to local_irq_enable()Line 448Lines 446–4478.5.19. initrd ConfigurationLines 449–4568.5.20. The Call to mem_init()Line 457Line 459Line 469Line 477Lines 480–486Line 488Lines 490–506Lines 521–523Lines 399–410Lines 412–419Lines 421–432Lines 433–450Lines 452–467Lines 469–480Lines 482–4928.5.21. The Call to late_time_init()Lines 459–4608.5.22. The Call to calibrate_delay()Line 461Line 186Lines 189–200Line 204Lines 206–214Lines 217–2198.5.23. The Call to pgtable_cache_init()Line 463Lines 532–542Lines 544–5518.5.24. The Call to buffer_init()Line 472Line 3036Line 3039Line 30448.5.25. The Call to security_scaffolding_startup()Line 4748.5.26. The Call to vfs_caches_init()Line 475Line 1623Lines 1625–1629Lines 1631–1635Line 1637Line 1638Line 1639Line 299Lines 301–302Line 1640Lines 1139–1144Lines 1161–1173Lines 1179–1185Lines 218–222Lines 237–240Lines 1116–1123Line 1125Lines 1134–1135Lines 293–298Line 301Lines 304–3088.5.27. The Call to radix_tree_init()Line 4768.5.28. The Call to signals_init()Line 477Lines 2567–25718.5.29. The Call to page_writeback_init()Line 479Lines 495–507Line 508Line 509Line 510Lines 416–4178.5.30. The Call to proc_root_init()Lines 480–482Line 42Line 45Line 48Lines 54–788.5.31. The Call to init_idle()Line 490Line 2645Line 2648–2649Line 2651Lines 2652–2656Lines 2657–26598.5.32. The Call to rest_init()Line 493Line 388Line 390Line 391Line 3928.6. The init Thread (or Process 1)Line 612Line 629Line 553Lines 555–557Line 560Line 562Line 563Lines 635–638Line 387Lines 391–396Line 398Lines 400–401Line 406Lines 355–358Lines 364–375Line 377Line 645Lines 649–650Lines 662–668Line 670SummaryExercises
9.1. Toolchain9.1.1. Compilers9.1.2. Cross Compilers9.1.3. Linker9.1.4. ELF Object Files9.1.4.1. ELF HeaderLine 237Line 238Line 239Line 240Line 241Line 242Line 243Line 244Line 245Line 246Line 247Line 248Line 249Line 2509.1.4.2. Section Header TableLine 333Line 334Line 335Line 336Line 337Line 338Line 339Line 340Line 341Line 3429.1.4.3. Non-Executable ELF File Sections9.1.4.4. Program Header TableLine 277Line 278Line 279Line 280Line 281Line 282Line 283Line 2849.2. Kernel Source Build9.2.1. Source Explained9.2.1.1. Architecture-Independent Codefs/init/kernel/mm/9.2.1.2. Architecture-Dependent Code9.2.1.3. Miscellaneous Files and Directories9.2.2. Building the Kernel Image9.2.2.1. Kernel Configuration Tool9.2.2.2. Sub-Makefiles9.2.2.3. Linux Kernel MakefilesSummaryExercises
10.1. Traversing the Source10.1.1. Getting Familiar with the FilesystemLines 655–68110.1.2. Filps and FopsLines 1824–1829Lines 1831–183510.1.3. User Memory and Kernel MemoryLines 1454–1455Lines 1460–146110.1.4. Wait QueuesLine 1591Lines 1608–1611Lines 1618–1626Lines 1631–1633Lines 1635–1636Lines 1638–1639Line 2209Line 2233Line 2234Lines 2235–2241Lines 2494–2497Lines 2499–2502Lines 2504–251310.1.5. Work Queues and Interrupts10.1.6. System Calls10.1.7. Other Types of DriversLine 83Lines 84–85Line 86Line 87Line 88Line 89Line 90Line 91Line 576Line 495Line 55710.1.8. Device Model and sysfsLine 103Line 104Lines 105–115Line 135Line 136Line 13710.2. Writing the Code10.2.1. Device Basics10.2.2. Symbol Exporting10.2.3. IOCTL10.2.4. Polling and InterruptsLine 8Lines 9–12Line 2710.2.5. Work Queues and Tasklets10.2.6. Adding Code for a System Call10.3. Building and Debugging10.3.1. Debugging Device DriversSummaryExercises

Content preview from Linux® Kernel Primer, The: A Top-Down Approach for x86 and PowerPC Architectures

Chapter 9. Building the Linux Kernel

In this chapter

9.1 Toolchain 526
9.2 Kernel Source Build 536
Summary 551
Exercises 551

Until now, we've seen the subsystems within the Linux kernel and we've explored the system initialization functions. It is also important to understand how the image gets created. This chapter explores the process of compiling and linking the kernel image. We also look at the internals of the Linux build process.

Toolchain

A toolchain is the set of programs necessary to create a Linux kernel image. The concept of the chain is that the output of one tool becomes the input for the next. Our toolchain includes a compiler, an assembler, and a linker. Technically, it needs to also include your text editor, but this section covers the ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Start your free trial

Mobile 3D Graphics SoC: From Algorithm to Chip

Publisher Resources

ISBN: 0131181637Purchase book

Linux® Kernel Primer, The: A Top-Down Approach for x86 and PowerPC Architectures

by Claudia Salzberg Rodriguez, Gordon Fischer, Steven Smolski

Chapter 9. Building the Linux Kernel

Toolchain

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

You might also like

Mobile 3D Graphics SoC: From Algorithm to Chip

An Introduction to Software Defined Networking (SDN) LiveLessons—Networking Talks

What Employees Want Most in Uncertain Times

How I Built a Personal Board of Directors With GenAI

Publisher Resources

Chapter 9. Building the Linux Kernel

Toolchain

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,and much more.

You might also like

Mobile 3D Graphics SoC: From Algorithm to Chip

An Introduction to Software Defined Networking (SDN) LiveLessons—Networking Talks

What Employees Want Most in Uncertain Times

How I Built a Personal Board of Directors With GenAI

Publisher Resources

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.