| 1 | ============================================
|
| 2 | Unreliable Guide To Hacking The Linux Kernel
|
| 3 | ============================================
|
| 4 |
|
| 5 | :Author: Rusty Russell
|
| 6 |
|
| 7 | Introduction
|
| 8 | ============
|
| 9 |
|
| 10 | Welcome, gentle reader, to Rusty's Remarkably Unreliable Guide to Linux
|
| 11 | Kernel Hacking. This document describes the common routines and general
|
| 12 | requirements for kernel code: its goal is to serve as a primer for Linux
|
| 13 | kernel development for experienced C programmers. I avoid implementation
|
| 14 | details: that's what the code is for, and I ignore whole tracts of
|
| 15 | useful routines.
|
| 16 |
|
| 17 | Before you read this, please understand that I never wanted to write
|
| 18 | this document, being grossly under-qualified, but I always wanted to
|
| 19 | read it, and this was the only way. I hope it will grow into a
|
| 20 | compendium of best practice, common starting points and random
|
| 21 | information.
|
| 22 |
|
| 23 | The Players
|
| 24 | ===========
|
| 25 |
|
| 26 | At any time each of the CPUs in a system can be:
|
| 27 |
|
| 28 | - not associated with any process, serving a hardware interrupt;
|
| 29 |
|
| 30 | - not associated with any process, serving a softirq or tasklet;
|
| 31 |
|
| 32 | - running in kernel space, associated with a process (user context);
|
| 33 |
|
| 34 | - running a process in user space.
|
| 35 |
|
| 36 | There is an ordering between these. The bottom two can preempt each
|
| 37 | other, but above that is a strict hierarchy: each can only be preempted
|
| 38 | by the ones above it. For example, while a softirq is running on a CPU,
|
| 39 | no other softirq will preempt it, but a hardware interrupt can. However,
|
| 40 | any other CPUs in the system execute independently.
|
| 41 |
|
| 42 | We'll see a number of ways that the user context can block interrupts,
|
| 43 | to become truly non-preemptable.
|
| 44 |
|
| 45 | User Context
|
| 46 | ------------
|
| 47 |
|
| 48 | User context is when you are coming in from a system call or other trap:
|
| 49 | like userspace, you can be preempted by more important tasks and by
|
| 50 | interrupts. You can sleep, by calling :c:func:`schedule()`.
|
| 51 |
|
| 52 | .. note::
|
| 53 |
|
| 54 | You are always in user context on module load and unload, and on
|
| 55 | operations on the block device layer.
|
| 56 |
|
| 57 | In user context, the ``current`` pointer (indicating the task we are
|
| 58 | currently executing) is valid, and :c:func:`in_interrupt()`
|
| 59 | (``include/linux/preempt.h``) is false.
|
| 60 |
|
| 61 | .. warning::
|
| 62 |
|
| 63 | Beware that if you have preemption or softirqs disabled (see below),
|
| 64 | :c:func:`in_interrupt()` will return a false positive.
|
| 65 |
|
| 66 | Hardware Interrupts (Hard IRQs)
|
| 67 | -------------------------------
|
| 68 |
|
| 69 | Timer ticks, network cards and keyboard are examples of real hardware
|
| 70 | which produce interrupts at any time. The kernel runs interrupt
|
| 71 | handlers, which services the hardware. The kernel guarantees that this
|
| 72 | handler is never re-entered: if the same interrupt arrives, it is queued
|
| 73 | (or dropped). Because it disables interrupts, this handler has to be
|
| 74 | fast: frequently it simply acknowledges the interrupt, marks a 'software
|
| 75 | interrupt' for execution and exits.
|
| 76 |
|
| 77 | You can tell you are in a hardware interrupt, because
|
| 78 | :c:func:`in_irq()` returns true.
|
| 79 |
|
| 80 | .. warning::
|
| 81 |
|
| 82 | Beware that this will return a false positive if interrupts are
|
| 83 | disabled (see below).
|
| 84 |
|
| 85 | Software Interrupt Context: Softirqs and Tasklets
|
| 86 | -------------------------------------------------
|
| 87 |
|
| 88 | Whenever a system call is about to return to userspace, or a hardware
|
| 89 | interrupt handler exits, any 'software interrupts' which are marked
|
| 90 | pending (usually by hardware interrupts) are run (``kernel/softirq.c``).
|
| 91 |
|
| 92 | Much of the real interrupt handling work is done here. Early in the
|
| 93 | transition to SMP, there were only 'bottom halves' (BHs), which didn't
|
| 94 | take advantage of multiple CPUs. Shortly after we switched from wind-up
|
| 95 | computers made of match-sticks and snot, we abandoned this limitation
|
| 96 | and switched to 'softirqs'.
|
| 97 |
|
| 98 | ``include/linux/interrupt.h`` lists the different softirqs. A very
|
| 99 | important softirq is the timer softirq (``include/linux/timer.h``): you
|
| 100 | can register to have it call functions for you in a given length of
|
| 101 | time.
|
| 102 |
|
| 103 | Softirqs are often a pain to deal with, since the same softirq will run
|
| 104 | simultaneously on more than one CPU. For this reason, tasklets
|
| 105 | (``include/linux/interrupt.h``) are more often used: they are
|
| 106 | dynamically-registrable (meaning you can have as many as you want), and
|
| 107 | they also guarantee that any tasklet will only run on one CPU at any
|
| 108 | time, although different tasklets can run simultaneously.
|
| 109 |
|
| 110 | .. warning::
|
| 111 |
|
| 112 | The name 'tasklet' is misleading: they have nothing to do with
|
| 113 | 'tasks', and probably more to do with some bad vodka Alexey
|
| 114 | Kuznetsov had at the time.
|
| 115 |
|
| 116 | You can tell you are in a softirq (or tasklet) using the
|
| 117 | :c:func:`in_softirq()` macro (``include/linux/preempt.h``).
|
| 118 |
|
| 119 | .. warning::
|
| 120 |
|
| 121 | Beware that this will return a false positive if a
|
| 122 | :ref:`botton half lock <local_bh_disable>` is held.
|
| 123 |
|
| 124 | Some Basic Rules
|
| 125 | ================
|
| 126 |
|
| 127 | No memory protection
|
| 128 | If you corrupt memory, whether in user context or interrupt context,
|
| 129 | the whole machine will crash. Are you sure you can't do what you
|
| 130 | want in userspace?
|
| 131 |
|
| 132 | No floating point or MMX
|
| 133 | The FPU context is not saved; even in user context the FPU state
|
| 134 | probably won't correspond with the current process: you would mess
|
| 135 | with some user process' FPU state. If you really want to do this,
|
| 136 | you would have to explicitly save/restore the full FPU state (and
|
| 137 | avoid context switches). It is generally a bad idea; use fixed point
|
| 138 | arithmetic first.
|
| 139 |
|
| 140 | A rigid stack limit
|
| 141 | Depending on configuration options the kernel stack is about 3K to
|
| 142 | 6K for most 32-bit architectures: it's about 14K on most 64-bit
|
| 143 | archs, and often shared with interrupts so you can't use it all.
|
| 144 | Avoid deep recursion and huge local arrays on the stack (allocate
|
| 145 | them dynamically instead).
|
| 146 |
|
| 147 | The Linux kernel is portable
|
| 148 | Let's keep it that way. Your code should be 64-bit clean, and
|
| 149 | endian-independent. You should also minimize CPU specific stuff,
|
| 150 | e.g. inline assembly should be cleanly encapsulated and minimized to
|
| 151 | ease porting. Generally it should be restricted to the
|
| 152 | architecture-dependent part of the kernel tree.
|
| 153 |
|
| 154 | ioctls: Not writing a new system call
|
| 155 | =====================================
|
| 156 |
|
| 157 | A system call generally looks like this::
|
| 158 |
|
| 159 | asmlinkage long sys_mycall(int arg)
|
| 160 | {
|
| 161 | return 0;
|
| 162 | }
|
| 163 |
|
| 164 |
|
| 165 | First, in most cases you don't want to create a new system call. You
|
| 166 | create a character device and implement an appropriate ioctl for it.
|
| 167 | This is much more flexible than system calls, doesn't have to be entered
|
| 168 | in every architecture's ``include/asm/unistd.h`` and
|
| 169 | ``arch/kernel/entry.S`` file, and is much more likely to be accepted by
|
| 170 | Linus.
|
| 171 |
|
| 172 | If all your routine does is read or write some parameter, consider
|
| 173 | implementing a :c:func:`sysfs()` interface instead.
|
| 174 |
|
| 175 | Inside the ioctl you're in user context to a process. When a error
|
| 176 | occurs you return a negated errno (see
|
| 177 | ``include/uapi/asm-generic/errno-base.h``,
|
| 178 | ``include/uapi/asm-generic/errno.h`` and ``include/linux/errno.h``),
|
| 179 | otherwise you return 0.
|
| 180 |
|
| 181 | After you slept you should check if a signal occurred: the Unix/Linux
|
| 182 | way of handling signals is to temporarily exit the system call with the
|
| 183 | ``-ERESTARTSYS`` error. The system call entry code will switch back to
|
| 184 | user context, process the signal handler and then your system call will
|
| 185 | be restarted (unless the user disabled that). So you should be prepared
|
| 186 | to process the restart, e.g. if you're in the middle of manipulating
|
| 187 | some data structure.
|
| 188 |
|
| 189 | ::
|
| 190 |
|
| 191 | if (signal_pending(current))
|
| 192 | return -ERESTARTSYS;
|
| 193 |
|
| 194 |
|
| 195 | If you're doing longer computations: first think userspace. If you
|
| 196 | **really** want to do it in kernel you should regularly check if you need
|
| 197 | to give up the CPU (remember there is cooperative multitasking per CPU).
|
| 198 | Idiom::
|
| 199 |
|
| 200 | cond_resched(); /* Will sleep */
|
| 201 |
|
| 202 |
|
| 203 | A short note on interface design: the UNIX system call motto is "Provide
|
| 204 | mechanism not policy".
|
| 205 |
|
| 206 | Recipes for Deadlock
|
| 207 | ====================
|
| 208 |
|
| 209 | You cannot call any routines which may sleep, unless:
|
| 210 |
|
| 211 | - You are in user context.
|
| 212 |
|
| 213 | - You do not own any spinlocks.
|
| 214 |
|
| 215 | - You have interrupts enabled (actually, Andi Kleen says that the
|
| 216 | scheduling code will enable them for you, but that's probably not
|
| 217 | what you wanted).
|
| 218 |
|
| 219 | Note that some functions may sleep implicitly: common ones are the user
|
| 220 | space access functions (\*_user) and memory allocation functions
|
| 221 | without ``GFP_ATOMIC``.
|
| 222 |
|
| 223 | You should always compile your kernel ``CONFIG_DEBUG_ATOMIC_SLEEP`` on,
|
| 224 | and it will warn you if you break these rules. If you **do** break the
|
| 225 | rules, you will eventually lock up your box.
|
| 226 |
|
| 227 | Really.
|
| 228 |
|
| 229 | Common Routines
|
| 230 | ===============
|
| 231 |
|
| 232 | :c:func:`printk()`
|
| 233 | ------------------
|
| 234 |
|
| 235 | Defined in ``include/linux/printk.h``
|
| 236 |
|
| 237 | :c:func:`printk()` feeds kernel messages to the console, dmesg, and
|
| 238 | the syslog daemon. It is useful for debugging and reporting errors, and
|
| 239 | can be used inside interrupt context, but use with caution: a machine
|
| 240 | which has its console flooded with printk messages is unusable. It uses
|
| 241 | a format string mostly compatible with ANSI C printf, and C string
|
| 242 | concatenation to give it a first "priority" argument::
|
| 243 |
|
| 244 | printk(KERN_INFO "i = %u\n", i);
|
| 245 |
|
| 246 |
|
| 247 | See ``include/linux/kern_levels.h``; for other ``KERN_`` values; these are
|
| 248 | interpreted by syslog as the level. Special case: for printing an IP
|
| 249 | address use::
|
| 250 |
|
| 251 | __be32 ipaddress;
|
| 252 | printk(KERN_INFO "my ip: %pI4\n", &ipaddress);
|
| 253 |
|
| 254 |
|
| 255 | :c:func:`printk()` internally uses a 1K buffer and does not catch
|
| 256 | overruns. Make sure that will be enough.
|
| 257 |
|
| 258 | .. note::
|
| 259 |
|
| 260 | You will know when you are a real kernel hacker when you start
|
| 261 | typoing printf as printk in your user programs :)
|
| 262 |
|
| 263 | .. note::
|
| 264 |
|
| 265 | Another sidenote: the original Unix Version 6 sources had a comment
|
| 266 | on top of its printf function: "Printf should not be used for
|
| 267 | chit-chat". You should follow that advice.
|
| 268 |
|
| 269 | :c:func:`copy_to_user()` / :c:func:`copy_from_user()` / :c:func:`get_user()` / :c:func:`put_user()`
|
| 270 | ---------------------------------------------------------------------------------------------------
|
| 271 |
|
| 272 | Defined in ``include/linux/uaccess.h`` / ``asm/uaccess.h``
|
| 273 |
|
| 274 | **[SLEEPS]**
|
| 275 |
|
| 276 | :c:func:`put_user()` and :c:func:`get_user()` are used to get
|
| 277 | and put single values (such as an int, char, or long) from and to
|
| 278 | userspace. A pointer into userspace should never be simply dereferenced:
|
| 279 | data should be copied using these routines. Both return ``-EFAULT`` or
|
| 280 | 0.
|
| 281 |
|
| 282 | :c:func:`copy_to_user()` and :c:func:`copy_from_user()` are
|
| 283 | more general: they copy an arbitrary amount of data to and from
|
| 284 | userspace.
|
| 285 |
|
| 286 | .. warning::
|
| 287 |
|
| 288 | Unlike :c:func:`put_user()` and :c:func:`get_user()`, they
|
| 289 | return the amount of uncopied data (ie. 0 still means success).
|
| 290 |
|
| 291 | [Yes, this moronic interface makes me cringe. The flamewar comes up
|
| 292 | every year or so. --RR.]
|
| 293 |
|
| 294 | The functions may sleep implicitly. This should never be called outside
|
| 295 | user context (it makes no sense), with interrupts disabled, or a
|
| 296 | spinlock held.
|
| 297 |
|
| 298 | :c:func:`kmalloc()`/:c:func:`kfree()`
|
| 299 | -------------------------------------
|
| 300 |
|
| 301 | Defined in ``include/linux/slab.h``
|
| 302 |
|
| 303 | **[MAY SLEEP: SEE BELOW]**
|
| 304 |
|
| 305 | These routines are used to dynamically request pointer-aligned chunks of
|
| 306 | memory, like malloc and free do in userspace, but
|
| 307 | :c:func:`kmalloc()` takes an extra flag word. Important values:
|
| 308 |
|
| 309 | ``GFP_KERNEL``
|
| 310 | May sleep and swap to free memory. Only allowed in user context, but
|
| 311 | is the most reliable way to allocate memory.
|
| 312 |
|
| 313 | ``GFP_ATOMIC``
|
| 314 | Don't sleep. Less reliable than ``GFP_KERNEL``, but may be called
|
| 315 | from interrupt context. You should **really** have a good
|
| 316 | out-of-memory error-handling strategy.
|
| 317 |
|
| 318 | ``GFP_DMA``
|
| 319 | Allocate ISA DMA lower than 16MB. If you don't know what that is you
|
| 320 | don't need it. Very unreliable.
|
| 321 |
|
| 322 | If you see a sleeping function called from invalid context warning
|
| 323 | message, then maybe you called a sleeping allocation function from
|
| 324 | interrupt context without ``GFP_ATOMIC``. You should really fix that.
|
| 325 | Run, don't walk.
|
| 326 |
|
| 327 | If you are allocating at least ``PAGE_SIZE`` (``asm/page.h`` or
|
| 328 | ``asm/page_types.h``) bytes, consider using :c:func:`__get_free_pages()`
|
| 329 | (``include/linux/gfp.h``). It takes an order argument (0 for page sized,
|
| 330 | 1 for double page, 2 for four pages etc.) and the same memory priority
|
| 331 | flag word as above.
|
| 332 |
|
| 333 | If you are allocating more than a page worth of bytes you can use
|
| 334 | :c:func:`vmalloc()`. It'll allocate virtual memory in the kernel
|
| 335 | map. This block is not contiguous in physical memory, but the MMU makes
|
| 336 | it look like it is for you (so it'll only look contiguous to the CPUs,
|
| 337 | not to external device drivers). If you really need large physically
|
| 338 | contiguous memory for some weird device, you have a problem: it is
|
| 339 | poorly supported in Linux because after some time memory fragmentation
|
| 340 | in a running kernel makes it hard. The best way is to allocate the block
|
| 341 | early in the boot process via the :c:func:`alloc_bootmem()`
|
| 342 | routine.
|
| 343 |
|
| 344 | Before inventing your own cache of often-used objects consider using a
|
| 345 | slab cache in ``include/linux/slab.h``
|
| 346 |
|
| 347 | :c:func:`current()`
|
| 348 | -------------------
|
| 349 |
|
| 350 | Defined in ``include/asm/current.h``
|
| 351 |
|
| 352 | This global variable (really a macro) contains a pointer to the current
|
| 353 | task structure, so is only valid in user context. For example, when a
|
| 354 | process makes a system call, this will point to the task structure of
|
| 355 | the calling process. It is **not NULL** in interrupt context.
|
| 356 |
|
| 357 | :c:func:`mdelay()`/:c:func:`udelay()`
|
| 358 | -------------------------------------
|
| 359 |
|
| 360 | Defined in ``include/asm/delay.h`` / ``include/linux/delay.h``
|
| 361 |
|
| 362 | The :c:func:`udelay()` and :c:func:`ndelay()` functions can be
|
| 363 | used for small pauses. Do not use large values with them as you risk
|
| 364 | overflow - the helper function :c:func:`mdelay()` is useful here, or
|
| 365 | consider :c:func:`msleep()`.
|
| 366 |
|
| 367 | :c:func:`cpu_to_be32()`/:c:func:`be32_to_cpu()`/:c:func:`cpu_to_le32()`/:c:func:`le32_to_cpu()`
|
| 368 | -----------------------------------------------------------------------------------------------
|
| 369 |
|
| 370 | Defined in ``include/asm/byteorder.h``
|
| 371 |
|
| 372 | The :c:func:`cpu_to_be32()` family (where the "32" can be replaced
|
| 373 | by 64 or 16, and the "be" can be replaced by "le") are the general way
|
| 374 | to do endian conversions in the kernel: they return the converted value.
|
| 375 | All variations supply the reverse as well:
|
| 376 | :c:func:`be32_to_cpu()`, etc.
|
| 377 |
|
| 378 | There are two major variations of these functions: the pointer
|
| 379 | variation, such as :c:func:`cpu_to_be32p()`, which take a pointer
|
| 380 | to the given type, and return the converted value. The other variation
|
| 381 | is the "in-situ" family, such as :c:func:`cpu_to_be32s()`, which
|
| 382 | convert value referred to by the pointer, and return void.
|
| 383 |
|
| 384 | :c:func:`local_irq_save()`/:c:func:`local_irq_restore()`
|
| 385 | --------------------------------------------------------
|
| 386 |
|
| 387 | Defined in ``include/linux/irqflags.h``
|
| 388 |
|
| 389 | These routines disable hard interrupts on the local CPU, and restore
|
| 390 | them. They are reentrant; saving the previous state in their one
|
| 391 | ``unsigned long flags`` argument. If you know that interrupts are
|
| 392 | enabled, you can simply use :c:func:`local_irq_disable()` and
|
| 393 | :c:func:`local_irq_enable()`.
|
| 394 |
|
| 395 | .. _local_bh_disable:
|
| 396 |
|
| 397 | :c:func:`local_bh_disable()`/:c:func:`local_bh_enable()`
|
| 398 | --------------------------------------------------------
|
| 399 |
|
| 400 | Defined in ``include/linux/bottom_half.h``
|
| 401 |
|
| 402 |
|
| 403 | These routines disable soft interrupts on the local CPU, and restore
|
| 404 | them. They are reentrant; if soft interrupts were disabled before, they
|
| 405 | will still be disabled after this pair of functions has been called.
|
| 406 | They prevent softirqs and tasklets from running on the current CPU.
|
| 407 |
|
| 408 | :c:func:`smp_processor_id()`
|
| 409 | ----------------------------
|
| 410 |
|
| 411 | Defined in ``include/linux/smp.h``
|
| 412 |
|
| 413 | :c:func:`get_cpu()` disables preemption (so you won't suddenly get
|
| 414 | moved to another CPU) and returns the current processor number, between
|
| 415 | 0 and ``NR_CPUS``. Note that the CPU numbers are not necessarily
|
| 416 | continuous. You return it again with :c:func:`put_cpu()` when you
|
| 417 | are done.
|
| 418 |
|
| 419 | If you know you cannot be preempted by another task (ie. you are in
|
| 420 | interrupt context, or have preemption disabled) you can use
|
| 421 | smp_processor_id().
|
| 422 |
|
| 423 | ``__init``/``__exit``/``__initdata``
|
| 424 | ------------------------------------
|
| 425 |
|
| 426 | Defined in ``include/linux/init.h``
|
| 427 |
|
| 428 | After boot, the kernel frees up a special section; functions marked with
|
| 429 | ``__init`` and data structures marked with ``__initdata`` are dropped
|
| 430 | after boot is complete: similarly modules discard this memory after
|
| 431 | initialization. ``__exit`` is used to declare a function which is only
|
| 432 | required on exit: the function will be dropped if this file is not
|
| 433 | compiled as a module. See the header file for use. Note that it makes no
|
| 434 | sense for a function marked with ``__init`` to be exported to modules
|
| 435 | with :c:func:`EXPORT_SYMBOL()` or :c:func:`EXPORT_SYMBOL_GPL()`- this
|
| 436 | will break.
|
| 437 |
|
| 438 | :c:func:`__initcall()`/:c:func:`module_init()`
|
| 439 | ----------------------------------------------
|
| 440 |
|
| 441 | Defined in ``include/linux/init.h`` / ``include/linux/module.h``
|
| 442 |
|
| 443 | Many parts of the kernel are well served as a module
|
| 444 | (dynamically-loadable parts of the kernel). Using the
|
| 445 | :c:func:`module_init()` and :c:func:`module_exit()` macros it
|
| 446 | is easy to write code without #ifdefs which can operate both as a module
|
| 447 | or built into the kernel.
|
| 448 |
|
| 449 | The :c:func:`module_init()` macro defines which function is to be
|
| 450 | called at module insertion time (if the file is compiled as a module),
|
| 451 | or at boot time: if the file is not compiled as a module the
|
| 452 | :c:func:`module_init()` macro becomes equivalent to
|
| 453 | :c:func:`__initcall()`, which through linker magic ensures that
|
| 454 | the function is called on boot.
|
| 455 |
|
| 456 | The function can return a negative error number to cause module loading
|
| 457 | to fail (unfortunately, this has no effect if the module is compiled
|
| 458 | into the kernel). This function is called in user context with
|
| 459 | interrupts enabled, so it can sleep.
|
| 460 |
|
| 461 | :c:func:`module_exit()`
|
| 462 | -----------------------
|
| 463 |
|
| 464 |
|
| 465 | Defined in ``include/linux/module.h``
|
| 466 |
|
| 467 | This macro defines the function to be called at module removal time (or
|
| 468 | never, in the case of the file compiled into the kernel). It will only
|
| 469 | be called if the module usage count has reached zero. This function can
|
| 470 | also sleep, but cannot fail: everything must be cleaned up by the time
|
| 471 | it returns.
|
| 472 |
|
| 473 | Note that this macro is optional: if it is not present, your module will
|
| 474 | not be removable (except for 'rmmod -f').
|
| 475 |
|
| 476 | :c:func:`try_module_get()`/:c:func:`module_put()`
|
| 477 | -------------------------------------------------
|
| 478 |
|
| 479 | Defined in ``include/linux/module.h``
|
| 480 |
|
| 481 | These manipulate the module usage count, to protect against removal (a
|
| 482 | module also can't be removed if another module uses one of its exported
|
| 483 | symbols: see below). Before calling into module code, you should call
|
| 484 | :c:func:`try_module_get()` on that module: if it fails, then the
|
| 485 | module is being removed and you should act as if it wasn't there.
|
| 486 | Otherwise, you can safely enter the module, and call
|
| 487 | :c:func:`module_put()` when you're finished.
|
| 488 |
|
| 489 | Most registerable structures have an owner field, such as in the
|
| 490 | :c:type:`struct file_operations <file_operations>` structure.
|
| 491 | Set this field to the macro ``THIS_MODULE``.
|
| 492 |
|
| 493 | Wait Queues ``include/linux/wait.h``
|
| 494 | ====================================
|
| 495 |
|
| 496 | **[SLEEPS]**
|
| 497 |
|
| 498 | A wait queue is used to wait for someone to wake you up when a certain
|
| 499 | condition is true. They must be used carefully to ensure there is no
|
| 500 | race condition. You declare a :c:type:`wait_queue_head_t`, and then processes
|
| 501 | which want to wait for that condition declare a :c:type:`wait_queue_entry_t`
|
| 502 | referring to themselves, and place that in the queue.
|
| 503 |
|
| 504 | Declaring
|
| 505 | ---------
|
| 506 |
|
| 507 | You declare a ``wait_queue_head_t`` using the
|
| 508 | :c:func:`DECLARE_WAIT_QUEUE_HEAD()` macro, or using the
|
| 509 | :c:func:`init_waitqueue_head()` routine in your initialization
|
| 510 | code.
|
| 511 |
|
| 512 | Queuing
|
| 513 | -------
|
| 514 |
|
| 515 | Placing yourself in the waitqueue is fairly complex, because you must
|
| 516 | put yourself in the queue before checking the condition. There is a
|
| 517 | macro to do this: :c:func:`wait_event_interruptible()`
|
| 518 | (``include/linux/wait.h``) The first argument is the wait queue head, and
|
| 519 | the second is an expression which is evaluated; the macro returns 0 when
|
| 520 | this expression is true, or ``-ERESTARTSYS`` if a signal is received. The
|
| 521 | :c:func:`wait_event()` version ignores signals.
|
| 522 |
|
| 523 | Waking Up Queued Tasks
|
| 524 | ----------------------
|
| 525 |
|
| 526 | Call :c:func:`wake_up()` (``include/linux/wait.h``);, which will wake
|
| 527 | up every process in the queue. The exception is if one has
|
| 528 | ``TASK_EXCLUSIVE`` set, in which case the remainder of the queue will
|
| 529 | not be woken. There are other variants of this basic function available
|
| 530 | in the same header.
|
| 531 |
|
| 532 | Atomic Operations
|
| 533 | =================
|
| 534 |
|
| 535 | Certain operations are guaranteed atomic on all platforms. The first
|
| 536 | class of operations work on :c:type:`atomic_t` (``include/asm/atomic.h``);
|
| 537 | this contains a signed integer (at least 32 bits long), and you must use
|
| 538 | these functions to manipulate or read :c:type:`atomic_t` variables.
|
| 539 | :c:func:`atomic_read()` and :c:func:`atomic_set()` get and set
|
| 540 | the counter, :c:func:`atomic_add()`, :c:func:`atomic_sub()`,
|
| 541 | :c:func:`atomic_inc()`, :c:func:`atomic_dec()`, and
|
| 542 | :c:func:`atomic_dec_and_test()` (returns true if it was
|
| 543 | decremented to zero).
|
| 544 |
|
| 545 | Yes. It returns true (i.e. != 0) if the atomic variable is zero.
|
| 546 |
|
| 547 | Note that these functions are slower than normal arithmetic, and so
|
| 548 | should not be used unnecessarily.
|
| 549 |
|
| 550 | The second class of atomic operations is atomic bit operations on an
|
| 551 | ``unsigned long``, defined in ``include/linux/bitops.h``. These
|
| 552 | operations generally take a pointer to the bit pattern, and a bit
|
| 553 | number: 0 is the least significant bit. :c:func:`set_bit()`,
|
| 554 | :c:func:`clear_bit()` and :c:func:`change_bit()` set, clear,
|
| 555 | and flip the given bit. :c:func:`test_and_set_bit()`,
|
| 556 | :c:func:`test_and_clear_bit()` and
|
| 557 | :c:func:`test_and_change_bit()` do the same thing, except return
|
| 558 | true if the bit was previously set; these are particularly useful for
|
| 559 | atomically setting flags.
|
| 560 |
|
| 561 | It is possible to call these operations with bit indices greater than
|
| 562 | ``BITS_PER_LONG``. The resulting behavior is strange on big-endian
|
| 563 | platforms though so it is a good idea not to do this.
|
| 564 |
|
| 565 | Symbols
|
| 566 | =======
|
| 567 |
|
| 568 | Within the kernel proper, the normal linking rules apply (ie. unless a
|
| 569 | symbol is declared to be file scope with the ``static`` keyword, it can
|
| 570 | be used anywhere in the kernel). However, for modules, a special
|
| 571 | exported symbol table is kept which limits the entry points to the
|
| 572 | kernel proper. Modules can also export symbols.
|
| 573 |
|
| 574 | :c:func:`EXPORT_SYMBOL()`
|
| 575 | -------------------------
|
| 576 |
|
| 577 | Defined in ``include/linux/export.h``
|
| 578 |
|
| 579 | This is the classic method of exporting a symbol: dynamically loaded
|
| 580 | modules will be able to use the symbol as normal.
|
| 581 |
|
| 582 | :c:func:`EXPORT_SYMBOL_GPL()`
|
| 583 | -----------------------------
|
| 584 |
|
| 585 | Defined in ``include/linux/export.h``
|
| 586 |
|
| 587 | Similar to :c:func:`EXPORT_SYMBOL()` except that the symbols
|
| 588 | exported by :c:func:`EXPORT_SYMBOL_GPL()` can only be seen by
|
| 589 | modules with a :c:func:`MODULE_LICENSE()` that specifies a GPL
|
| 590 | compatible license. It implies that the function is considered an
|
| 591 | internal implementation issue, and not really an interface. Some
|
| 592 | maintainers and developers may however require EXPORT_SYMBOL_GPL()
|
| 593 | when adding any new APIs or functionality.
|
| 594 |
|
| 595 | Routines and Conventions
|
| 596 | ========================
|
| 597 |
|
| 598 | Double-linked lists ``include/linux/list.h``
|
| 599 | --------------------------------------------
|
| 600 |
|
| 601 | There used to be three sets of linked-list routines in the kernel
|
| 602 | headers, but this one is the winner. If you don't have some particular
|
| 603 | pressing need for a single list, it's a good choice.
|
| 604 |
|
| 605 | In particular, :c:func:`list_for_each_entry()` is useful.
|
| 606 |
|
| 607 | Return Conventions
|
| 608 | ------------------
|
| 609 |
|
| 610 | For code called in user context, it's very common to defy C convention,
|
| 611 | and return 0 for success, and a negative error number (eg. ``-EFAULT``) for
|
| 612 | failure. This can be unintuitive at first, but it's fairly widespread in
|
| 613 | the kernel.
|
| 614 |
|
| 615 | Using :c:func:`ERR_PTR()` (``include/linux/err.h``) to encode a
|
| 616 | negative error number into a pointer, and :c:func:`IS_ERR()` and
|
| 617 | :c:func:`PTR_ERR()` to get it back out again: avoids a separate
|
| 618 | pointer parameter for the error number. Icky, but in a good way.
|
| 619 |
|
| 620 | Breaking Compilation
|
| 621 | --------------------
|
| 622 |
|
| 623 | Linus and the other developers sometimes change function or structure
|
| 624 | names in development kernels; this is not done just to keep everyone on
|
| 625 | their toes: it reflects a fundamental change (eg. can no longer be
|
| 626 | called with interrupts on, or does extra checks, or doesn't do checks
|
| 627 | which were caught before). Usually this is accompanied by a fairly
|
| 628 | complete note to the linux-kernel mailing list; search the archive.
|
| 629 | Simply doing a global replace on the file usually makes things **worse**.
|
| 630 |
|
| 631 | Initializing structure members
|
| 632 | ------------------------------
|
| 633 |
|
| 634 | The preferred method of initializing structures is to use designated
|
| 635 | initialisers, as defined by ISO C99, eg::
|
| 636 |
|
| 637 | static struct block_device_operations opt_fops = {
|
| 638 | .open = opt_open,
|
| 639 | .release = opt_release,
|
| 640 | .ioctl = opt_ioctl,
|
| 641 | .check_media_change = opt_media_change,
|
| 642 | };
|
| 643 |
|
| 644 |
|
| 645 | This makes it easy to grep for, and makes it clear which structure
|
| 646 | fields are set. You should do this because it looks cool.
|
| 647 |
|
| 648 | GNU Extensions
|
| 649 | --------------
|
| 650 |
|
| 651 | GNU Extensions are explicitly allowed in the Linux kernel. Note that
|
| 652 | some of the more complex ones are not very well supported, due to lack
|
| 653 | of general use, but the following are considered standard (see the GCC
|
| 654 | info page section "C Extensions" for more details - Yes, really the info
|
| 655 | page, the man page is only a short summary of the stuff in info).
|
| 656 |
|
| 657 | - Inline functions
|
| 658 |
|
| 659 | - Statement expressions (ie. the ({ and }) constructs).
|
| 660 |
|
| 661 | - Declaring attributes of a function / variable / type
|
| 662 | (__attribute__)
|
| 663 |
|
| 664 | - typeof
|
| 665 |
|
| 666 | - Zero length arrays
|
| 667 |
|
| 668 | - Macro varargs
|
| 669 |
|
| 670 | - Arithmetic on void pointers
|
| 671 |
|
| 672 | - Non-Constant initializers
|
| 673 |
|
| 674 | - Assembler Instructions (not outside arch/ and include/asm/)
|
| 675 |
|
| 676 | - Function names as strings (__func__).
|
| 677 |
|
| 678 | - __builtin_constant_p()
|
| 679 |
|
| 680 | Be wary when using long long in the kernel, the code gcc generates for
|
| 681 | it is horrible and worse: division and multiplication does not work on
|
| 682 | i386 because the GCC runtime functions for it are missing from the
|
| 683 | kernel environment.
|
| 684 |
|
| 685 | C++
|
| 686 | ---
|
| 687 |
|
| 688 | Using C++ in the kernel is usually a bad idea, because the kernel does
|
| 689 | not provide the necessary runtime environment and the include files are
|
| 690 | not tested for it. It is still possible, but not recommended. If you
|
| 691 | really want to do this, forget about exceptions at least.
|
| 692 |
|
| 693 | NUMif
|
| 694 | -----
|
| 695 |
|
| 696 | It is generally considered cleaner to use macros in header files (or at
|
| 697 | the top of .c files) to abstract away functions rather than using \`#if'
|
| 698 | pre-processor statements throughout the source code.
|
| 699 |
|
| 700 | Putting Your Stuff in the Kernel
|
| 701 | ================================
|
| 702 |
|
| 703 | In order to get your stuff into shape for official inclusion, or even to
|
| 704 | make a neat patch, there's administrative work to be done:
|
| 705 |
|
| 706 | - Figure out whose pond you've been pissing in. Look at the top of the
|
| 707 | source files, inside the ``MAINTAINERS`` file, and last of all in the
|
| 708 | ``CREDITS`` file. You should coordinate with this person to make sure
|
| 709 | you're not duplicating effort, or trying something that's already
|
| 710 | been rejected.
|
| 711 |
|
| 712 | Make sure you put your name and EMail address at the top of any files
|
| 713 | you create or mangle significantly. This is the first place people
|
| 714 | will look when they find a bug, or when **they** want to make a change.
|
| 715 |
|
| 716 | - Usually you want a configuration option for your kernel hack. Edit
|
| 717 | ``Kconfig`` in the appropriate directory. The Config language is
|
| 718 | simple to use by cut and paste, and there's complete documentation in
|
| 719 | ``Documentation/kbuild/kconfig-language.txt``.
|
| 720 |
|
| 721 | In your description of the option, make sure you address both the
|
| 722 | expert user and the user who knows nothing about your feature.
|
| 723 | Mention incompatibilities and issues here. **Definitely** end your
|
| 724 | description with “if in doubt, say N” (or, occasionally, \`Y'); this
|
| 725 | is for people who have no idea what you are talking about.
|
| 726 |
|
| 727 | - Edit the ``Makefile``: the CONFIG variables are exported here so you
|
| 728 | can usually just add a "obj-$(CONFIG_xxx) += xxx.o" line. The syntax
|
| 729 | is documented in ``Documentation/kbuild/makefiles.txt``.
|
| 730 |
|
| 731 | - Put yourself in ``CREDITS`` if you've done something noteworthy,
|
| 732 | usually beyond a single file (your name should be at the top of the
|
| 733 | source files anyway). ``MAINTAINERS`` means you want to be consulted
|
| 734 | when changes are made to a subsystem, and hear about bugs; it implies
|
| 735 | a more-than-passing commitment to some part of the code.
|
| 736 |
|
| 737 | - Finally, don't forget to read
|
| 738 | ``Documentation/process/submitting-patches.rst`` and possibly
|
| 739 | ``Documentation/process/submitting-drivers.rst``.
|
| 740 |
|
| 741 | Kernel Cantrips
|
| 742 | ===============
|
| 743 |
|
| 744 | Some favorites from browsing the source. Feel free to add to this list.
|
| 745 |
|
| 746 | ``arch/x86/include/asm/delay.h``::
|
| 747 |
|
| 748 | #define ndelay(n) (__builtin_constant_p(n) ? \
|
| 749 | ((n) > 20000 ? __bad_ndelay() : __const_udelay((n) * 5ul)) : \
|
| 750 | __ndelay(n))
|
| 751 |
|
| 752 |
|
| 753 | ``include/linux/fs.h``::
|
| 754 |
|
| 755 | /*
|
| 756 | * Kernel pointers have redundant information, so we can use a
|
| 757 | * scheme where we can return either an error code or a dentry
|
| 758 | * pointer with the same return value.
|
| 759 | *
|
| 760 | * This should be a per-architecture thing, to allow different
|
| 761 | * error and pointer decisions.
|
| 762 | */
|
| 763 | #define ERR_PTR(err) ((void *)((long)(err)))
|
| 764 | #define PTR_ERR(ptr) ((long)(ptr))
|
| 765 | #define IS_ERR(ptr) ((unsigned long)(ptr) > (unsigned long)(-1000))
|
| 766 |
|
| 767 | ``arch/x86/include/asm/uaccess_32.h:``::
|
| 768 |
|
| 769 | #define copy_to_user(to,from,n) \
|
| 770 | (__builtin_constant_p(n) ? \
|
| 771 | __constant_copy_to_user((to),(from),(n)) : \
|
| 772 | __generic_copy_to_user((to),(from),(n)))
|
| 773 |
|
| 774 |
|
| 775 | ``arch/sparc/kernel/head.S:``::
|
| 776 |
|
| 777 | /*
|
| 778 | * Sun people can't spell worth damn. "compatability" indeed.
|
| 779 | * At least we *know* we can't spell, and use a spell-checker.
|
| 780 | */
|
| 781 |
|
| 782 | /* Uh, actually Linus it is I who cannot spell. Too much murky
|
| 783 | * Sparc assembly will do this to ya.
|
| 784 | */
|
| 785 | C_LABEL(cputypvar):
|
| 786 | .asciz "compatibility"
|
| 787 |
|
| 788 | /* Tested on SS-5, SS-10. Probably someone at Sun applied a spell-checker. */
|
| 789 | .align 4
|
| 790 | C_LABEL(cputypvar_sun4m):
|
| 791 | .asciz "compatible"
|
| 792 |
|
| 793 |
|
| 794 | ``arch/sparc/lib/checksum.S:``::
|
| 795 |
|
| 796 | /* Sun, you just can't beat me, you just can't. Stop trying,
|
| 797 | * give up. I'm serious, I am going to kick the living shit
|
| 798 | * out of you, game over, lights out.
|
| 799 | */
|
| 800 |
|
| 801 |
|
| 802 | Thanks
|
| 803 | ======
|
| 804 |
|
| 805 | Thanks to Andi Kleen for the idea, answering my questions, fixing my
|
| 806 | mistakes, filling content, etc. Philipp Rumpf for more spelling and
|
| 807 | clarity fixes, and some excellent non-obvious points. Werner Almesberger
|
| 808 | for giving me a great summary of :c:func:`disable_irq()`, and Jes
|
| 809 | Sorensen and Andrea Arcangeli added caveats. Michael Elizabeth Chastain
|
| 810 | for checking and adding to the Configure section. Telsa Gwynne for
|
| 811 | teaching me DocBook.
|