Takashi Iwai

tiwai@suse.de

November 17, 2005 0.3.6 This document describes how to write an ALSA (Advanced Linux Sound Architecture) driver. Copyright (c) 2002-2005 Takashi Iwai tiwai@suse.de This document is free; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This document is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA This document describes how to write an ALSA (Advanced Linux Sound Architecture) driver. The document focuses mainly on the PCI soundcard. In the case of other device types, the API might be different, too. However, at least the ALSA kernel API is consistent, and therefore it would be still a bit help for writing them. The target of this document is ones who already have enough skill of C language and have the basic knowledge of linux kernel programming. This document doesn't explain the general topics of linux kernel codes and doesn't cover the detail of implementation of each low-level driver. It describes only how is the standard way to write a PCI sound driver on ALSA. If you are already familiar with the older ALSA ver.0.5.x, you can check the drivers such as es1938.c or maestro3.c which have also almost the same code-base in the ALSA 0.5.x tree, so you can compare the differences. This document is still a draft version. Any feedbacks and corrections, please!!

The ALSA drivers are provided in the two ways. One is the trees provided as a tarball or via cvs from the ALSA's ftp site, and another is the 2.6 (or later) Linux kernel tree. To synchronize both, the ALSA driver tree is split into two different trees: alsa-kernel and alsa-driver. The former contains purely the source codes for the Linux 2.6 (or later) tree. This tree is designed only for compilation on 2.6 or later environment. The latter, alsa-driver, contains many subtle files for compiling the ALSA driver on the outside of Linux kernel like configure script, the wrapper functions for older, 2.2 and 2.4 kernels, to adapt the latest kernel API, and additional drivers which are still in development or in tests. The drivers in alsa-driver tree will be moved to alsa-kernel (eventually 2.6 kernel tree) once when they are finished and confirmed to work fine. The file tree structure of ALSA driver is depicted below. Both alsa-kernel and alsa-driver have almost the same file structure, except for core directory. It's named as acore in alsa-driver tree. sound /core /oss /seq /oss /instr /ioctl32 /include /drivers /mpu401 /opl3 /i2c /l3 /synth /emux /pci /(cards) /isa /(cards) /arm /ppc /sparc /usb /pcmcia /(cards) /oss

This directory contains the middle layer, that is, the heart of ALSA drivers. In this directory, the native ALSA modules are stored. The sub-directories contain different modules and are dependent upon the kernel config.

The codes for PCM and mixer OSS emulation modules are stored in this directory. The rawmidi OSS emulation is included in the ALSA rawmidi code since it's quite small. The sequencer code is stored in core/seq/oss directory (see below).

This directory contains the 32bit-ioctl wrappers for 64bit architectures such like x86-64, ppc64 and sparc64. For 32bit and alpha architectures, these are not compiled.

This and its sub-directories are for the ALSA sequencer. This directory contains the sequencer core and primary sequencer modules such like snd-seq-midi, snd-seq-virmidi, etc. They are compiled only when CONFIG_SND_SEQUENCER is set in the kernel config.

This contains the OSS sequencer emulation codes.

This directory contains the modules for the sequencer instrument layer.

This is the place for the public header files of ALSA drivers, which are to be exported to the user-space, or included by several files at different directories. Basically, the private header files should not be placed in this directory, but you may still find files there, due to historical reason :)

This directory contains the codes shared among different drivers on the different architectures. They are hence supposed not to be architecture-specific. For example, the dummy pcm driver and the serial MIDI driver are found in this directory. In the sub-directories, there are the codes for components which are independent from bus and cpu architectures.

The MPU401 and MPU401-UART modules are stored here.

The OPL3 and OPL4 FM-synth stuff is found here.

This contains the ALSA i2c components. Although there is a standard i2c layer on Linux, ALSA has its own i2c codes for some cards, because the soundcard needs only a simple operation and the standard i2c API is too complicated for such a purpose.

This is a sub-directory for ARM L3 i2c.

This contains the synth middle-level modules. So far, there is only Emu8000/Emu10k1 synth driver under synth/emux sub-directory.

This and its sub-directories hold the top-level card modules for PCI soundcards and the codes specific to the PCI BUS. The drivers compiled from a single file is stored directly on pci directory, while the drivers with several source files are stored on its own sub-directory (e.g. emu10k1, ice1712).

This and its sub-directories hold the top-level card modules for ISA soundcards.

These are for the top-level card modules which are specific to each given architecture.

This contains the USB-audio driver. On the latest version, the USB MIDI driver is integrated together with usb-audio driver.

The PCMCIA, especially PCCard drivers will go here. CardBus drivers will be on pci directory, because its API is identical with the standard PCI cards.

The OSS/Lite source files are stored here on Linux 2.6 (or later) tree. (In the ALSA driver tarball, it's empty, of course :)

The minimum flow of PCI soundcard is like the following: define the PCI ID table (see the section PCI Entries ). create probe() callback. create remove() callback. create pci_driver table which contains the three pointers above. create init() function just calling pci_register_driver() to register the pci_driver table defined above. create exit() function to call pci_unregister_driver() function.

The code example is shown below. Some parts are kept unimplemented at this moment but will be filled in the succeeding sections. The numbers in comment lines of snd_mychip_probe() function are the markers. #include #include #include #include #include /* module parameters (see "Module Parameters") */ static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX; static char *id[SNDRV_CARDS] = SNDRV_DEFAULT_STR; static int enable[SNDRV_CARDS] = SNDRV_DEFAULT_ENABLE_PNP; /* definition of the chip-specific record */ struct mychip { struct snd_card *card; // rest of implementation will be in the section // "PCI Resource Managements" }; /* chip-specific destructor * (see "PCI Resource Managements") */ static int snd_mychip_free(struct mychip *chip) { .... // will be implemented later... } /* component-destructor * (see "Management of Cards and Components") */ static int snd_mychip_dev_free(struct snd_device *device) { return snd_mychip_free(device->device_data); } /* chip-specific constructor * (see "Management of Cards and Components") */ static int __devinit snd_mychip_create(struct snd_card *card, struct pci_dev *pci, struct mychip **rchip) { struct mychip *chip; int err; static struct snd_device_ops ops = { .dev_free = snd_mychip_dev_free, }; *rchip = NULL; // check PCI availability here // (see "PCI Resource Managements") .... /* allocate a chip-specific data with zero filled */ chip = kzalloc(sizeof(*chip), GFP_KERNEL); if (chip == NULL) return -ENOMEM; chip->card = card; // rest of initialization here; will be implemented // later, see "PCI Resource Managements" .... if ((err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops)) < 0) { snd_mychip_free(chip); return err; } snd_card_set_dev(card, &pci->dev); *rchip = chip; return 0; } /* constructor -- see "Constructor" sub-section */ static int __devinit snd_mychip_probe(struct pci_dev *pci, const struct pci_device_id *pci_id) { static int dev; struct snd_card *card; struct mychip *chip; int err; /* (1) */ if (dev >= SNDRV_CARDS) return -ENODEV; if (!enable[dev]) { dev++; return -ENOENT; } /* (2) */ card = snd_card_new(index[dev], id[dev], THIS_MODULE, 0); if (card == NULL) return -ENOMEM; /* (3) */ if ((err = snd_mychip_create(card, pci, &chip)) < 0) { snd_card_free(card); return err; } /* (4) */ strcpy(card->driver, "My Chip"); strcpy(card->shortname, "My Own Chip 123"); sprintf(card->longname, "%s at 0x%lx irq %i", card->shortname, chip->ioport, chip->irq); /* (5) */ .... // implemented later /* (6) */ if ((err = snd_card_register(card)) < 0) { snd_card_free(card); return err; } /* (7) */ pci_set_drvdata(pci, card); dev++; return 0; } /* destructor -- see "Destructor" sub-section */ static void __devexit snd_mychip_remove(struct pci_dev *pci) { snd_card_free(pci_get_drvdata(pci)); pci_set_drvdata(pci, NULL); } ]]>

The real constructor of PCI drivers is probe callback. The probe callback and other component-constructors which are called from probe callback should be defined with __devinit prefix. You cannot use __init prefix for them, because any PCI device could be a hotplug device. In the probe callback, the following scheme is often used.

= SNDRV_CARDS) return -ENODEV; if (!enable[dev]) { dev++; return -ENOENT; } ]]> where enable[dev] is the module option. At each time probe callback is called, check the availability of the device. If not available, simply increment the device index and returns. dev will be incremented also later (step 7).

The detail will be explained in the section Management of Cards and Components.

In this part, the PCI resources are allocated. The detail will be explained in the section PCI Resource Managements.

driver, "My Chip"); strcpy(card->shortname, "My Own Chip 123"); sprintf(card->longname, "%s at 0x%lx irq %i", card->shortname, chip->ioport, chip->irq); ]]> The driver field holds the minimal ID string of the chip. This is referred by alsa-lib's configurator, so keep it simple but unique. Even the same driver can have different driver IDs to distinguish the functionality of each chip type. The shortname field is a string shown as more verbose name. The longname field contains the information which is shown in /proc/asound/cards.

Here you define the basic components such as PCM, mixer (e.g. AC97), MIDI (e.g. MPU-401), and other interfaces. Also, if you want a proc file, define it here, too.

Will be explained in the section Management of Cards and Components, too.

In the above, the card record is stored. This pointer is referred in the remove callback and power-management callbacks, too.

The destructor, remove callback, simply releases the card instance. Then the ALSA middle layer will release all the attached components automatically. It would be typically like the following: The above code assumes that the card pointer is set to the PCI driver data.

For the above example, at least the following include files are necessary. #include #include #include #include #include ]]> where the last one is necessary only when module options are defined in the source file. If the codes are split to several files, the file without module options don't need them. In addition to them, you'll need <linux/interrupt.h> for the interrupt handling, and <asm/io.h> for the i/o access. If you use mdelay() or udelay() functions, you'll need to include <linux/delay.h>, too. The ALSA interfaces like PCM or control API are defined in other header files as <sound/xxx.h>. They have to be included after <sound/core.h>.

For each soundcard, a card record must be allocated. A card record is the headquarters of the soundcard. It manages the list of whole devices (components) on the soundcard, such as PCM, mixers, MIDI, synthesizer, and so on. Also, the card record holds the ID and the name strings of the card, manages the root of proc files, and controls the power-management states and hotplug disconnections. The component list on the card record is used to manage the proper releases of resources at destruction. As mentioned above, to create a card instance, call snd_card_new(). The function takes four arguments, the card-index number, the id string, the module pointer (usually THIS_MODULE), and the size of extra-data space. The last argument is used to allocate card->private_data for the chip-specific data. Note that this data is allocated by snd_card_new().

After the card is created, you can attach the components (devices) to the card instance. On ALSA driver, a component is represented as a struct snd_device object. A component can be a PCM instance, a control interface, a raw MIDI interface, etc. Each of such instances has one component entry. A component can be created via snd_device_new() function. This takes the card pointer, the device-level (SNDRV_DEV_XXX), the data pointer, and the callback pointers (&ops). The device-level defines the type of components and the order of registration and de-registration. For most of components, the device-level is already defined. For a user-defined component, you can use SNDRV_DEV_LOWLEVEL. This function itself doesn't allocate the data space. The data must be allocated manually beforehand, and its pointer is passed as the argument. This pointer is used as the identifier (chip in the above example) for the instance. Each ALSA pre-defined component such as ac97 or pcm calls snd_device_new() inside its constructor. The destructor for each component is defined in the callback pointers. Hence, you don't need to take care of calling a destructor for such a component. If you would like to create your own component, you need to set the destructor function to dev_free callback in ops, so that it can be released automatically via snd_card_free(). The example will be shown later as an implementation of a chip-specific data.

The chip-specific information, e.g. the i/o port address, its resource pointer, or the irq number, is stored in the chip-specific record. In general, there are two ways to allocate the chip record.

As mentioned above, you can pass the extra-data-length to the 4th argument of snd_card_new(), i.e. whether struct mychip is the type of the chip record. In return, the allocated record can be accessed as private_data; ]]> With this method, you don't have to allocate twice. The record is released together with the card instance.

After allocating a card instance via snd_card_new() (with NULL on the 4th arg), call kzalloc(). The chip record should have the field to hold the card pointer at least, Then, set the card pointer in the returned chip instance. card = card; ]]> Next, initialize the fields, and register this chip record as a low-level device with a specified ops, snd_mychip_dev_free() is the device-destructor function, which will call the real destructor. device_data); } ]]> where snd_mychip_free() is the real destructor.

After all components are assigned, register the card instance by calling snd_card_register(). The access to the device files are enabled at this point. That is, before snd_card_register() is called, the components are safely inaccessible from external side. If this call fails, exit the probe function after releasing the card via snd_card_free(). For releasing the card instance, you can call simply snd_card_free(). As already mentioned, all components are released automatically by this call. As further notes, the destructors (both snd_mychip_dev_free and snd_mychip_free) cannot be defined with __devexit prefix, because they may be called from the constructor, too, at the false path. For a device which allows hotplugging, you can use snd_card_free_when_closed. This one will postpone the destruction until all devices are closed.

In this section, we'll finish the chip-specific constructor, destructor and PCI entries. The example code is shown first, below. irq >= 0) free_irq(chip->irq, chip); /* release the i/o ports & memory */ pci_release_regions(chip->pci); /* disable the PCI entry */ pci_disable_device(chip->pci); /* release the data */ kfree(chip); return 0; } /* chip-specific constructor */ static int __devinit snd_mychip_create(struct snd_card *card, struct pci_dev *pci, struct mychip **rchip) { struct mychip *chip; int err; static struct snd_device_ops ops = { .dev_free = snd_mychip_dev_free, }; *rchip = NULL; /* initialize the PCI entry */ if ((err = pci_enable_device(pci)) < 0) return err; /* check PCI availability (28bit DMA) */ if (pci_set_dma_mask(pci, DMA_28BIT_MASK) < 0 || pci_set_consistent_dma_mask(pci, DMA_28BIT_MASK) < 0) { printk(KERN_ERR "error to set 28bit mask DMA\n"); pci_disable_device(pci); return -ENXIO; } chip = kzalloc(sizeof(*chip), GFP_KERNEL); if (chip == NULL) { pci_disable_device(pci); return -ENOMEM; } /* initialize the stuff */ chip->card = card; chip->pci = pci; chip->irq = -1; /* (1) PCI resource allocation */ if ((err = pci_request_regions(pci, "My Chip")) < 0) { kfree(chip); pci_disable_device(pci); return err; } chip->port = pci_resource_start(pci, 0); if (request_irq(pci->irq, snd_mychip_interrupt, IRQF_SHARED, "My Chip", chip)) { printk(KERN_ERR "cannot grab irq %d\n", pci->irq); snd_mychip_free(chip); return -EBUSY; } chip->irq = pci->irq; /* (2) initialization of the chip hardware */ .... // (not implemented in this document) if ((err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops)) < 0) { snd_mychip_free(chip); return err; } snd_card_set_dev(card, &pci->dev); *rchip = chip; return 0; } /* PCI IDs */ static struct pci_device_id snd_mychip_ids[] = { { PCI_VENDOR_ID_FOO, PCI_DEVICE_ID_BAR, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, .... { 0, } }; MODULE_DEVICE_TABLE(pci, snd_mychip_ids); /* pci_driver definition */ static struct pci_driver driver = { .name = "My Own Chip", .id_table = snd_mychip_ids, .probe = snd_mychip_probe, .remove = __devexit_p(snd_mychip_remove), }; /* initialization of the module */ static int __init alsa_card_mychip_init(void) { return pci_register_driver(&driver); } /* clean up the module */ static void __exit alsa_card_mychip_exit(void) { pci_unregister_driver(&driver); } module_init(alsa_card_mychip_init) module_exit(alsa_card_mychip_exit) EXPORT_NO_SYMBOLS; /* for old kernels only */ ]]>

The allocation of PCI resources is done in the probe() function, and usually an extra xxx_create() function is written for this purpose. In the case of PCI devices, you have to call at first pci_enable_device() function before allocating resources. Also, you need to set the proper PCI DMA mask to limit the accessed i/o range. In some cases, you might need to call pci_set_master() function, too. Suppose the 28bit mask, and the code to be added would be like:

The allocation of I/O ports and irqs are done via standard kernel functions. Unlike ALSA ver.0.5.x., there are no helpers for that. And these resources must be released in the destructor function (see below). Also, on ALSA 0.9.x, you don't need to allocate (pseudo-)DMA for PCI like ALSA 0.5.x. Now assume that this PCI device has an I/O port with 8 bytes and an interrupt. Then struct mychip will have the following fields: For an i/o port (and also a memory region), you need to have the resource pointer for the standard resource management. For an irq, you have to keep only the irq number (integer). But you need to initialize this number as -1 before actual allocation, since irq 0 is valid. The port address and its resource pointer can be initialized as null by kzalloc() automatically, so you don't have to take care of resetting them. The allocation of an i/o port is done like this: port = pci_resource_start(pci, 0); ]]> It will reserve the i/o port region of 8 bytes of the given PCI device. The returned value, chip->res_port, is allocated via kmalloc() by request_region(). The pointer must be released via kfree(), but there is some problem regarding this. This issue will be explained more below. The allocation of an interrupt source is done like this: irq, snd_mychip_interrupt, IRQF_DISABLED|IRQF_SHARED, "My Chip", chip)) { printk(KERN_ERR "cannot grab irq %d\n", pci->irq); snd_mychip_free(chip); return -EBUSY; } chip->irq = pci->irq; ]]> where snd_mychip_interrupt() is the interrupt handler defined later. Note that chip->irq should be defined only when request_irq() succeeded. On the PCI bus, the interrupts can be shared. Thus, IRQF_SHARED is given as the interrupt flag of request_irq(). The last argument of request_irq() is the data pointer passed to the interrupt handler. Usually, the chip-specific record is used for that, but you can use what you like, too. I won't define the detail of the interrupt handler at this point, but at least its appearance can be explained now. The interrupt handler looks usually like the following: Now let's write the corresponding destructor for the resources above. The role of destructor is simple: disable the hardware (if already activated) and release the resources. So far, we have no hardware part, so the disabling is not written here. For releasing the resources, check-and-release method is a safer way. For the interrupt, do like this: irq >= 0) free_irq(chip->irq, chip); ]]> Since the irq number can start from 0, you should initialize chip->irq with a negative value (e.g. -1), so that you can check the validity of the irq number as above. When you requested I/O ports or memory regions via pci_request_region() or pci_request_regions() like this example, release the resource(s) using the corresponding function, pci_release_region() or pci_release_regions(). pci); ]]> When you requested manually via request_region() or request_mem_region, you can release it via release_resource(). Suppose that you keep the resource pointer returned from request_region() in chip->res_port, the release procedure looks like below: res_port); ]]> Don't forget to call pci_disable_device() before all finished. And finally, release the chip-specific record. Again, remember that you cannot set __devexit prefix for this destructor. We didn't implement the hardware-disabling part in the above. If you need to do this, please note that the destructor may be called even before the initialization of the chip is completed. It would be better to have a flag to skip the hardware-disabling if the hardware was not initialized yet. When the chip-data is assigned to the card using snd_device_new() with SNDRV_DEV_LOWLELVEL , its destructor is called at the last. That is, it is assured that all other components like PCMs and controls have been already released. You don't have to call stopping PCMs, etc. explicitly, but just stop the hardware in the low-level. The management of a memory-mapped region is almost as same as the management of an i/o port. You'll need three fields like the following: and the allocation would be like below: iobase_phys = pci_resource_start(pci, 0); chip->iobase_virt = ioremap_nocache(chip->iobase_phys, pci_resource_len(pci, 0)); ]]> and the corresponding destructor would be: iobase_virt) iounmap(chip->iobase_virt); .... pci_release_regions(chip->pci); .... } ]]>

At some point, typically after calling snd_device_new(), you need to register the struct device of the chip you're handling for udev and co. ALSA provides a macro for compatibility with older kernels. Simply call like the following: dev); ]]> so that it stores the PCI's device pointer to the card. This will be referred by ALSA core functions later when the devices are registered. In the case of non-PCI, pass the proper device struct pointer of the BUS instead. (In the case of legacy ISA without PnP, you don't have to do anything.)

So far, so good. Let's finish the rest of missing PCI stuffs. At first, we need a pci_device_id table for this chipset. It's a table of PCI vendor/device ID number, and some masks. For example, The first and second fields of pci_device_id struct are the vendor and device IDs. If you have nothing special to filter the matching devices, you can use the rest of fields like above. The last field of pci_device_id struct is a private data for this entry. You can specify any value here, for example, to tell the type of different operations per each device IDs. Such an example is found in intel8x0 driver. The last entry of this list is the terminator. You must specify this all-zero entry. Then, prepare the pci_driver record: The probe and remove functions are what we already defined in the previous sections. The remove should be defined with __devexit_p() macro, so that it's not defined for built-in (and non-hot-pluggable) case. The name field is the name string of this device. Note that you must not use a slash / in this string. And at last, the module entries: Note that these module entries are tagged with __init and __exit prefixes, not __devinit nor __devexit. Oh, one thing was forgotten. If you have no exported symbols, you need to declare it on 2.2 or 2.4 kernels (on 2.6 kernels it's not necessary, though). That's all!

The PCM middle layer of ALSA is quite powerful and it is only necessary for each driver to implement the low-level functions to access its hardware. For accessing to the PCM layer, you need to include <sound/pcm.h> above all. In addition, <sound/pcm_params.h> might be needed if you access to some functions related with hw_param. Each card device can have up to four pcm instances. A pcm instance corresponds to a pcm device file. The limitation of number of instances comes only from the available bit size of the linux's device number. Once when 64bit device number is used, we'll have more available pcm instances. A pcm instance consists of pcm playback and capture streams, and each pcm stream consists of one or more pcm substreams. Some soundcard supports the multiple-playback function. For example, emu10k1 has a PCM playback of 32 stereo substreams. In this case, at each open, a free substream is (usually) automatically chosen and opened. Meanwhile, when only one substream exists and it was already opened, the succeeding open will result in the blocking or the error with EAGAIN according to the file open mode. But you don't have to know the detail in your driver. The PCM middle layer will take all such jobs.

The example code below does not include any hardware access routines but shows only the skeleton, how to build up the PCM interfaces. .... /* hardware definition */ static struct snd_pcm_hardware snd_mychip_playback_hw = { .info = (SNDRV_PCM_INFO_MMAP | SNDRV_PCM_INFO_INTERLEAVED | SNDRV_PCM_INFO_BLOCK_TRANSFER | SNDRV_PCM_INFO_MMAP_VALID), .formats = SNDRV_PCM_FMTBIT_S16_LE, .rates = SNDRV_PCM_RATE_8000_48000, .rate_min = 8000, .rate_max = 48000, .channels_min = 2, .channels_max = 2, .buffer_bytes_max = 32768, .period_bytes_min = 4096, .period_bytes_max = 32768, .periods_min = 1, .periods_max = 1024, }; /* hardware definition */ static struct snd_pcm_hardware snd_mychip_capture_hw = { .info = (SNDRV_PCM_INFO_MMAP | SNDRV_PCM_INFO_INTERLEAVED | SNDRV_PCM_INFO_BLOCK_TRANSFER | SNDRV_PCM_INFO_MMAP_VALID), .formats = SNDRV_PCM_FMTBIT_S16_LE, .rates = SNDRV_PCM_RATE_8000_48000, .rate_min = 8000, .rate_max = 48000, .channels_min = 2, .channels_max = 2, .buffer_bytes_max = 32768, .period_bytes_min = 4096, .period_bytes_max = 32768, .periods_min = 1, .periods_max = 1024, }; /* open callback */ static int snd_mychip_playback_open(struct snd_pcm_substream *substream) { struct mychip *chip = snd_pcm_substream_chip(substream); struct snd_pcm_runtime *runtime = substream->runtime; runtime->hw = snd_mychip_playback_hw; // more hardware-initialization will be done here return 0; } /* close callback */ static int snd_mychip_playback_close(struct snd_pcm_substream *substream) { struct mychip *chip = snd_pcm_substream_chip(substream); // the hardware-specific codes will be here return 0; } /* open callback */ static int snd_mychip_capture_open(struct snd_pcm_substream *substream) { struct mychip *chip = snd_pcm_substream_chip(substream); struct snd_pcm_runtime *runtime = substream->runtime; runtime->hw = snd_mychip_capture_hw; // more hardware-initialization will be done here return 0; } /* close callback */ static int snd_mychip_capture_close(struct snd_pcm_substream *substream) { struct mychip *chip = snd_pcm_substream_chip(substream); // the hardware-specific codes will be here return 0; } /* hw_params callback */ static int snd_mychip_pcm_hw_params(struct snd_pcm_substream *substream, struct snd_pcm_hw_params *hw_params) { return snd_pcm_lib_malloc_pages(substream, params_buffer_bytes(hw_params)); } /* hw_free callback */ static int snd_mychip_pcm_hw_free(struct snd_pcm_substream *substream) { return snd_pcm_lib_free_pages(substream); } /* prepare callback */ static int snd_mychip_pcm_prepare(struct snd_pcm_substream *substream) { struct mychip *chip = snd_pcm_substream_chip(substream); struct snd_pcm_runtime *runtime = substream->runtime; /* set up the hardware with the current configuration * for example... */ mychip_set_sample_format(chip, runtime->format); mychip_set_sample_rate(chip, runtime->rate); mychip_set_channels(chip, runtime->channels); mychip_set_dma_setup(chip, runtime->dma_addr, chip->buffer_size, chip->period_size); return 0; } /* trigger callback */ static int snd_mychip_pcm_trigger(struct snd_pcm_substream *substream, int cmd) { switch (cmd) { case SNDRV_PCM_TRIGGER_START: // do something to start the PCM engine break; case SNDRV_PCM_TRIGGER_STOP: // do something to stop the PCM engine break; default: return -EINVAL; } } /* pointer callback */ static snd_pcm_uframes_t snd_mychip_pcm_pointer(struct snd_pcm_substream *substream) { struct mychip *chip = snd_pcm_substream_chip(substream); unsigned int current_ptr; /* get the current hardware pointer */ current_ptr = mychip_get_hw_pointer(chip); return current_ptr; } /* operators */ static struct snd_pcm_ops snd_mychip_playback_ops = { .open = snd_mychip_playback_open, .close = snd_mychip_playback_close, .ioctl = snd_pcm_lib_ioctl, .hw_params = snd_mychip_pcm_hw_params, .hw_free = snd_mychip_pcm_hw_free, .prepare = snd_mychip_pcm_prepare, .trigger = snd_mychip_pcm_trigger, .pointer = snd_mychip_pcm_pointer, }; /* operators */ static struct snd_pcm_ops snd_mychip_capture_ops = { .open = snd_mychip_capture_open, .close = snd_mychip_capture_close, .ioctl = snd_pcm_lib_ioctl, .hw_params = snd_mychip_pcm_hw_params, .hw_free = snd_mychip_pcm_hw_free, .prepare = snd_mychip_pcm_prepare, .trigger = snd_mychip_pcm_trigger, .pointer = snd_mychip_pcm_pointer, }; /* * definitions of capture are omitted here... */ /* create a pcm device */ static int __devinit snd_mychip_new_pcm(struct mychip *chip) { struct snd_pcm *pcm; int err; if ((err = snd_pcm_new(chip->card, "My Chip", 0, 1, 1, &pcm)) < 0) return err; pcm->private_data = chip; strcpy(pcm->name, "My Chip"); chip->pcm = pcm; /* set operators */ snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, &snd_mychip_playback_ops); snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE, &snd_mychip_capture_ops); /* pre-allocation of buffers */ /* NOTE: this may fail */ snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV, snd_dma_pci_data(chip->pci), 64*1024, 64*1024); return 0; } ]]>

A pcm instance is allocated by snd_pcm_new() function. It would be better to create a constructor for pcm, namely, card, "My Chip", 0, 1, 1, &pcm)) < 0) return err; pcm->private_data = chip; strcpy(pcm->name, "My Chip"); chip->pcm = pcm; .... return 0; } ]]> The snd_pcm_new() function takes the four arguments. The first argument is the card pointer to which this pcm is assigned, and the second is the ID string. The third argument (index, 0 in the above) is the index of this new pcm. It begins from zero. When you will create more than one pcm instances, specify the different numbers in this argument. For example, index = 1 for the second PCM device. The fourth and fifth arguments are the number of substreams for playback and capture, respectively. Here both 1 are given in the above example. When no playback or no capture is available, pass 0 to the corresponding argument. If a chip supports multiple playbacks or captures, you can specify more numbers, but they must be handled properly in open/close, etc. callbacks. When you need to know which substream you are referring to, then it can be obtained from struct snd_pcm_substream data passed to each callback as follows: number; ]]> After the pcm is created, you need to set operators for each pcm stream. The operators are defined typically like this: Each of callbacks is explained in the subsection Operators. After setting the operators, most likely you'd like to pre-allocate the buffer. For the pre-allocation, simply call the following: pci), 64*1024, 64*1024); ]]> It will allocate up to 64kB buffer as default. The details of buffer management will be described in the later section Buffer and Memory Management. Additionally, you can set some extra information for this pcm in pcm->info_flags. The available values are defined as SNDRV_PCM_INFO_XXX in <sound/asound.h>, which is used for the hardware definition (described later). When your soundchip supports only half-duplex, specify like this: info_flags = SNDRV_PCM_INFO_HALF_DUPLEX; ]]>

The destructor for a pcm instance is not always necessary. Since the pcm device will be released by the middle layer code automatically, you don't have to call destructor explicitly. The destructor would be necessary when you created some special records internally and need to release them. In such a case, set the destructor function to pcm->private_free: my_private_pcm_data); // do what you like else .... } static int __devinit snd_mychip_new_pcm(struct mychip *chip) { struct snd_pcm *pcm; .... /* allocate your own data */ chip->my_private_pcm_data = kmalloc(...); /* set the destructor */ pcm->private_data = chip; pcm->private_free = mychip_pcm_free; .... } ]]>

When the PCM substream is opened, a PCM runtime instance is allocated and assigned to the substream. This pointer is accessible via substream->runtime. This runtime pointer holds the various information; it holds the copy of hw_params and sw_params configurations, the buffer pointers, mmap records, spinlocks, etc. Almost everything you need for controlling the PCM can be found there. The definition of runtime instance is found in <sound/pcm.h>. Here is the copy from the file. For the operators (callbacks) of each sound driver, most of these records are supposed to be read-only. Only the PCM middle-layer changes / updates these info. The exceptions are the hardware description (hw), interrupt callbacks (transfer_ack_xxx), DMA buffer information, and the private data. Besides, if you use the standard buffer allocation method via snd_pcm_lib_malloc_pages(), you don't need to set the DMA buffer information by yourself. In the sections below, important records are explained.

The hardware descriptor (struct snd_pcm_hardware) contains the definitions of the fundamental hardware configuration. Above all, you'll need to define this in the open callback. Note that the runtime instance holds the copy of the descriptor, not the pointer to the existing descriptor. That is, in the open callback, you can modify the copied descriptor (runtime->hw) as you need. For example, if the maximum number of channels is 1 only on some chip models, you can still use the same hardware descriptor and change the channels_max later: runtime; ... runtime->hw = snd_mychip_playback_hw; /* common definition */ if (chip->model == VERY_OLD_ONE) runtime->hw.channels_max = 1; ]]> Typically, you'll have a hardware descriptor like below: The info field contains the type and capabilities of this pcm. The bit flags are defined in <sound/asound.h> as SNDRV_PCM_INFO_XXX. Here, at least, you have to specify whether the mmap is supported and which interleaved format is supported. When the mmap is supported, add SNDRV_PCM_INFO_MMAP flag here. When the hardware supports the interleaved or the non-interleaved format, SNDRV_PCM_INFO_INTERLEAVED or SNDRV_PCM_INFO_NONINTERLEAVED flag must be set, respectively. If both are supported, you can set both, too. In the above example, MMAP_VALID and BLOCK_TRANSFER are specified for OSS mmap mode. Usually both are set. Of course, MMAP_VALID is set only if the mmap is really supported. The other possible flags are SNDRV_PCM_INFO_PAUSE and SNDRV_PCM_INFO_RESUME. The PAUSE bit means that the pcm supports the pause operation, while the RESUME bit means that the pcm supports the full suspend/resume operation. If PAUSE flag is set, the trigger callback below must handle the corresponding (pause push/release) commands. The suspend/resume trigger commands can be defined even without RESUME flag. See Power Management section for details. When the PCM substreams can be synchronized (typically, synchronized start/stop of a playback and a capture streams), you can give SNDRV_PCM_INFO_SYNC_START, too. In this case, you'll need to check the linked-list of PCM substreams in the trigger callback. This will be described in the later section. formats field contains the bit-flags of supported formats (SNDRV_PCM_FMTBIT_XXX). If the hardware supports more than one format, give all or'ed bits. In the example above, the signed 16bit little-endian format is specified. rates field contains the bit-flags of supported rates (SNDRV_PCM_RATE_XXX). When the chip supports continuous rates, pass CONTINUOUS bit additionally. The pre-defined rate bits are provided only for typical rates. If your chip supports unconventional rates, you need to add KNOT bit and set up the hardware constraint manually (explained later). rate_min and rate_max define the minimal and maximal sample rate. This should correspond somehow to rates bits. channel_min and channel_max define, as you might already expected, the minimal and maximal number of channels. buffer_bytes_max defines the maximal buffer size in bytes. There is no buffer_bytes_min field, since it can be calculated from the minimal period size and the minimal number of periods. Meanwhile, period_bytes_min and define the minimal and maximal size of the period in bytes. periods_max and periods_min define the maximal and minimal number of periods in the buffer. The period is a term, that corresponds to fragment in the OSS world. The period defines the size at which the PCM interrupt is generated. This size strongly depends on the hardware. Generally, the smaller period size will give you more interrupts, that is, more controls. In the case of capture, this size defines the input latency. On the other hand, the whole buffer size defines the output latency for the playback direction. There is also a field fifo_size. This specifies the size of the hardware FIFO, but it's not used currently in the driver nor in the alsa-lib. So, you can ignore this field.

Ok, let's go back again to the PCM runtime records. The most frequently referred records in the runtime instance are the PCM configurations. The PCM configurations are stored on runtime instance after the application sends hw_params data via alsa-lib. There are many fields copied from hw_params and sw_params structs. For example, format holds the format type chosen by the application. This field contains the enum value SNDRV_PCM_FORMAT_XXX. One thing to be noted is that the configured buffer and period sizes are stored in frames in the runtime In the ALSA world, 1 frame = channels * samples-size. For conversion between frames and bytes, you can use the helper functions, frames_to_bytes() and bytes_to_frames(). period_size); ]]> Also, many software parameters (sw_params) are stored in frames, too. Please check the type of the field. snd_pcm_uframes_t is for the frames as unsigned integer while snd_pcm_sframes_t is for the frames as signed integer.

The DMA buffer is defined by the following four fields, dma_area, dma_addr, dma_bytes and dma_private. The dma_area holds the buffer pointer (the logical address). You can call memcpy from/to this pointer. Meanwhile, dma_addr holds the physical address of the buffer. This field is specified only when the buffer is a linear buffer. dma_bytes holds the size of buffer in bytes. dma_private is used for the ALSA DMA allocator. If you use a standard ALSA function, snd_pcm_lib_malloc_pages(), for allocating the buffer, these fields are set by the ALSA middle layer, and you should not change them by yourself. You can read them but not write them. On the other hand, if you want to allocate the buffer by yourself, you'll need to manage it in hw_params callback. At least, dma_bytes is mandatory. dma_area is necessary when the buffer is mmapped. If your driver doesn't support mmap, this field is not necessary. dma_addr is also not mandatory. You can use dma_private as you like, too.

The running status can be referred via runtime->status. This is the pointer to struct snd_pcm_mmap_status record. For example, you can get the current DMA hardware pointer via runtime->status->hw_ptr. The DMA application pointer can be referred via runtime->control, which points struct snd_pcm_mmap_control record. However, accessing directly to this value is not recommended.

You can allocate a record for the substream and store it in runtime->private_data. Usually, this done in the open callback. Don't mix this with pcm->private_data. The pcm->private_data usually points the chip instance assigned statically at the creation of PCM, while the runtime->private_data points a dynamic data created at the PCM open callback. runtime->private_data = data; .... } ]]> The allocated object must be released in the close callback.

The field transfer_ack_begin and transfer_ack_end are called at the beginning and the end of snd_pcm_period_elapsed(), respectively.

OK, now let me explain the detail of each pcm callback (ops). In general, every callback must return 0 if successful, or a negative number with the error number such as -EINVAL at any error. The callback function takes at least the argument with snd_pcm_substream pointer. For retrieving the chip record from the given substream instance, you can use the following macro. The macro reads substream->private_data, which is a copy of pcm->private_data. You can override the former if you need to assign different data records per PCM substream. For example, cmi8330 driver assigns different private_data for playback and capture directions, because it uses two different codecs (SB- and AD-compatible) for different directions.

This is called when a pcm substream is opened. At least, here you have to initialize the runtime->hw record. Typically, this is done by like this: runtime; runtime->hw = snd_mychip_playback_hw; return 0; } ]]> where snd_mychip_playback_hw is the pre-defined hardware description. You can allocate a private data in this callback, as described in Private Data section. If the hardware configuration needs more constraints, set the hardware constraints here, too. See Constraints for more details.

Obviously, this is called when a pcm substream is closed. Any private instance for a pcm substream allocated in the open callback will be released here. runtime->private_data); .... } ]]>

This is used for any special action to pcm ioctls. But usually you can pass a generic ioctl callback, snd_pcm_lib_ioctl.

This and hw_free callbacks exist only on ALSA 0.9.x. This is called when the hardware parameter (hw_params) is set up by the application, that is, once when the buffer size, the period size, the format, etc. are defined for the pcm substream. Many hardware set-up should be done in this callback, including the allocation of buffers. Parameters to be initialized are retrieved by params_xxx() macros. For allocating a buffer, you can call a helper function, snd_pcm_lib_malloc_pages() is available only when the DMA buffers have been pre-allocated. See the section Buffer Types for more details. Note that this and prepare callbacks may be called multiple times per initialization. For example, the OSS emulation may call these callbacks at each change via its ioctl. Thus, you need to take care not to allocate the same buffers many times, which will lead to memory leak! Calling the helper function above many times is OK. It will release the previous buffer automatically when it was already allocated. Another note is that this callback is non-atomic (schedulable). This is important, because the trigger callback is atomic (non-schedulable). That is, mutex or any schedule-related functions are not available in trigger callback. Please see the subsection Atomicity for details.

This is called to release the resources allocated via hw_params. For example, releasing the buffer via snd_pcm_lib_malloc_pages() is done by calling the following: This function is always called before the close callback is called. Also, the callback may be called multiple times, too. Keep track whether the resource was already released.

This callback is called when the pcm is prepared. You can set the format type, sample rate, etc. here. The difference from hw_params is that the prepare callback will be called at each time snd_pcm_prepare() is called, i.e. when recovered after underruns, etc. Note that this callback became non-atomic since the recent version. You can use schedule-related functions safely in this callback now. In this and the following callbacks, you can refer to the values via the runtime record, substream->runtime. For example, to get the current rate, format or channels, access to runtime->rate, runtime->format or runtime->channels, respectively. The physical address of the allocated buffer is set to runtime->dma_area. The buffer and period sizes are in runtime->buffer_size and runtime->period_size, respectively. Be careful that this callback will be called many times at each set up, too.

This is called when the pcm is started, stopped or paused. Which action is specified in the second argument, SNDRV_PCM_TRIGGER_XXX in <sound/pcm.h>. At least, START and STOP commands must be defined in this callback. When the pcm supports the pause operation (given in info field of the hardware table), PAUSE_PUSE and PAUSE_RELEASE commands must be handled here, too. The former is the command to pause the pcm, and the latter to restart the pcm again. When the pcm supports the suspend/resume operation, regardless of full or partial suspend/resume support, SUSPEND and RESUME commands must be handled, too. These commands are issued when the power-management status is changed. Obviously, the SUSPEND and RESUME do suspend and resume of the pcm substream, and usually, they are identical with STOP and START commands, respectively. See Power Management section for details. As mentioned, this callback is atomic. You cannot call the function going to sleep. The trigger callback should be as minimal as possible, just really triggering the DMA. The other stuff should be initialized hw_params and prepare callbacks properly beforehand.

This callback is called when the PCM middle layer inquires the current hardware position on the buffer. The position must be returned in frames (which was in bytes on ALSA 0.5.x), ranged from 0 to buffer_size - 1. This is called usually from the buffer-update routine in the pcm middle layer, which is invoked when snd_pcm_period_elapsed() is called in the interrupt routine. Then the pcm middle layer updates the position and calculates the available space, and wakes up the sleeping poll threads, etc. This callback is also atomic.

These callbacks are not mandatory, and can be omitted in most cases. These callbacks are used when the hardware buffer cannot be on the normal memory space. Some chips have their own buffer on the hardware which is not mappable. In such a case, you have to transfer the data manually from the memory buffer to the hardware buffer. Or, if the buffer is non-contiguous on both physical and virtual memory spaces, these callbacks must be defined, too. If these two callbacks are defined, copy and set-silence operations are done by them. The detailed will be described in the later section Buffer and Memory Management.

This callback is also not mandatory. This callback is called when the appl_ptr is updated in read or write operations. Some drivers like emu10k1-fx and cs46xx need to track the current appl_ptr for the internal buffer, and this callback is useful only for such a purpose. This callback is atomic.

This callback is also not mandatory. This callback is used mainly for the non-contiguous buffer. The mmap calls this callback to get the page address. Some examples will be explained in the later section Buffer and Memory Management, too.

The rest of pcm stuff is the PCM interrupt handler. The role of PCM interrupt handler in the sound driver is to update the buffer position and to tell the PCM middle layer when the buffer position goes across the prescribed period size. To inform this, call snd_pcm_period_elapsed() function. There are several types of sound chips to generate the interrupts.

This is the most frequently found type: the hardware generates an interrupt at each period boundary. In this case, you can call snd_pcm_period_elapsed() at each interrupt. snd_pcm_period_elapsed() takes the substream pointer as its argument. Thus, you need to keep the substream pointer accessible from the chip instance. For example, define substream field in the chip record to hold the current running substream pointer, and set the pointer value at open callback (and reset at close callback). If you acquire a spinlock in the interrupt handler, and the lock is used in other pcm callbacks, too, then you have to release the lock before calling snd_pcm_period_elapsed(), because snd_pcm_period_elapsed() calls other pcm callbacks inside. A typical coding would be like: lock); .... if (pcm_irq_invoked(chip)) { /* call updater, unlock before it */ spin_unlock(&chip->lock); snd_pcm_period_elapsed(chip->substream); spin_lock(&chip->lock); // acknowledge the interrupt if necessary } .... spin_unlock(&chip->lock); return IRQ_HANDLED; } ]]>

This is the case when the hardware doesn't generate interrupts at the period boundary but do timer-interrupts at the fixed timer rate (e.g. es1968 or ymfpci drivers). In this case, you need to check the current hardware position and accumulates the processed sample length at each interrupt. When the accumulated size overcomes the period size, call snd_pcm_period_elapsed() and reset the accumulator. A typical coding would be like the following. lock); .... if (pcm_irq_invoked(chip)) { unsigned int last_ptr, size; /* get the current hardware pointer (in frames) */ last_ptr = get_hw_ptr(chip); /* calculate the processed frames since the * last update */ if (last_ptr < chip->last_ptr) size = runtime->buffer_size + last_ptr - chip->last_ptr; else size = last_ptr - chip->last_ptr; /* remember the last updated point */ chip->last_ptr = last_ptr; /* accumulate the size */ chip->size += size; /* over the period boundary? */ if (chip->size >= runtime->period_size) { /* reset the accumulator */ chip->size %= runtime->period_size; /* call updater */ spin_unlock(&chip->lock); snd_pcm_period_elapsed(substream); spin_lock(&chip->lock); } // acknowledge the interrupt if necessary } .... spin_unlock(&chip->lock); return IRQ_HANDLED; } ]]>

In both cases, even if more than one period are elapsed, you don't have to call snd_pcm_period_elapsed() many times. Call only once. And the pcm layer will check the current hardware pointer and update to the latest status.

One of the most important (and thus difficult to debug) problem on the kernel programming is the race condition. On linux kernel, usually it's solved via spin-locks or semaphores. In general, if the race condition may happen in the interrupt handler, it's handled as atomic, and you have to use spinlock for protecting the critical session. If it never happens in the interrupt and it may take relatively long time, you should use semaphore. As already seen, some pcm callbacks are atomic and some are not. For example, hw_params callback is non-atomic, while trigger callback is atomic. This means, the latter is called already in a spinlock held by the PCM middle layer. Please take this atomicity into account when you use a spinlock or a semaphore in the callbacks. In the atomic callbacks, you cannot use functions which may call schedule or go to sleep. The semaphore and mutex do sleep, and hence they cannot be used inside the atomic callbacks (e.g. trigger callback). For taking a certain delay in such a callback, please use udelay() or mdelay(). All three atomic callbacks (trigger, pointer, and ack) are called with local interrupts disabled.

If your chip supports unconventional sample rates, or only the limited samples, you need to set a constraint for the condition. For example, in order to restrict the sample rates in the some supported values, use snd_pcm_hw_constraint_list(). You need to call this function in the open callback. runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &constraints_rates); if (err < 0) return err; .... } ]]> There are many different constraints. Look in sound/pcm.h for a complete list. You can even define your own constraint rules. For example, let's suppose my_chip can manage a substream of 1 channel if and only if the format is S16_LE, otherwise it supports any format specified in the snd_pcm_hardware structure (or in any other constraint_list). You can build a rule like this: min < 2) { fmt.bits[0] &= SNDRV_PCM_FMTBIT_S16_LE; return snd_mask_refine(f, &fmt); } return 0; } ]]> Then you need to call this function to add your rule: runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS, hw_rule_channels_by_format, 0, SNDRV_PCM_HW_PARAM_FORMAT, -1); ]]> The rule function is called when an application sets the number of channels. But an application can set the format before the number of channels. Thus you also need to define the inverse rule: bits[0] == SNDRV_PCM_FMTBIT_S16_LE) { ch.min = ch.max = 1; ch.integer = 1; return snd_interval_refine(c, &ch); } return 0; } ]]> ...and in the open callback: runtime, 0, SNDRV_PCM_HW_PARAM_FORMAT, hw_rule_format_by_channels, 0, SNDRV_PCM_HW_PARAM_CHANNELS, -1); ]]> I won't explain more details here, rather I would like to say, Luke, use the source.

The control interface is used widely for many switches, sliders, etc. which are accessed from the user-space. Its most important use is the mixer interface. In other words, on ALSA 0.9.x, all the mixer stuff is implemented on the control kernel API (while there was an independent mixer kernel API on 0.5.x). ALSA has a well-defined AC97 control module. If your chip supports only the AC97 and nothing else, you can skip this section. The control API is defined in <sound/control.h>. Include this file if you add your own controls.

For creating a new control, you need to define the three callbacks: info, get and put. Then, define a struct snd_kcontrol_new record, such as: Most likely the control is created via snd_ctl_new1(), and in such a case, you can add __devinitdata prefix to the definition like above. The iface field specifies the type of the control, SNDRV_CTL_ELEM_IFACE_XXX, which is usually MIXER. Use CARD for global controls that are not logically part of the mixer. If the control is closely associated with some specific device on the sound card, use HWDEP, PCM, RAWMIDI, TIMER, or SEQUENCER, and specify the device number with the device and subdevice fields. The name is the name identifier string. On ALSA 0.9.x, the control name is very important, because its role is classified from its name. There are pre-defined standard control names. The details are described in the subsection Control Names. The index field holds the index number of this control. If there are several different controls with the same name, they can be distinguished by the index number. This is the case when several codecs exist on the card. If the index is zero, you can omit the definition above. The access field contains the access type of this control. Give the combination of bit masks, SNDRV_CTL_ELEM_ACCESS_XXX, there. The detailed will be explained in the subsection Access Flags. The private_value field contains an arbitrary long integer value for this record. When using generic info, get and put callbacks, you can pass a value through this field. If several small numbers are necessary, you can combine them in bitwise. Or, it's possible to give a pointer (casted to unsigned long) of some record to this field, too. The other three are callback functions.

There are some standards for defining the control names. A control is usually defined from the three parts as SOURCE DIRECTION FUNCTION. The first, SOURCE, specifies the source of the control, and is a string such as Master, PCM, CD or Line. There are many pre-defined sources. The second, DIRECTION, is one of the following strings according to the direction of the control: Playback, Capture, Bypass Playback and Bypass Capture. Or, it can be omitted, meaning both playback and capture directions. The third, FUNCTION, is one of the following strings according to the function of the control: Switch, Volume and Route. The example of control names are, thus, Master Capture Switch or PCM Playback Volume. There are some exceptions:

Capture Source, Capture Switch and Capture Volume are used for the global capture (input) source, switch and volume. Similarly, Playback Switch and Playback Volume are used for the global output gain switch and volume.

tone-control switch and volumes are specified like Tone Control - XXX, e.g. Tone Control - Switch, Tone Control - Bass, Tone Control - Center.

3D-control switches and volumes are specified like 3D Control - XXX, e.g. 3D Control - Switch, 3D Control - Center, 3D Control - Space.

Mic-boost switch is set as Mic Boost or Mic Boost (6dB). More precise information can be found in Documentation/sound/alsa/ControlNames.txt.

The access flag is the bit-flags which specifies the access type of the given control. The default access type is SNDRV_CTL_ELEM_ACCESS_READWRITE, which means both read and write are allowed to this control. When the access flag is omitted (i.e. = 0), it is regarded as READWRITE access as default. When the control is read-only, pass SNDRV_CTL_ELEM_ACCESS_READ instead. In this case, you don't have to define put callback. Similarly, when the control is write-only (although it's a rare case), you can use WRITE flag instead, and you don't need get callback. If the control value changes frequently (e.g. the VU meter), VOLATILE flag should be given. This means that the control may be changed without notification. Applications should poll such a control constantly. When the control is inactive, set INACTIVE flag, too. There are LOCK and OWNER flags for changing the write permissions.

The info callback is used to get the detailed information of this control. This must store the values of the given struct snd_ctl_elem_info object. For example, for a boolean control with a single element will be: type = SNDRV_CTL_ELEM_TYPE_BOOLEAN; uinfo->count = 1; uinfo->value.integer.min = 0; uinfo->value.integer.max = 1; return 0; } ]]> The type field specifies the type of the control. There are BOOLEAN, INTEGER, ENUMERATED, BYTES, IEC958 and INTEGER64. The count field specifies the number of elements in this control. For example, a stereo volume would have count = 2. The value field is a union, and the values stored are depending on the type. The boolean and integer are identical. The enumerated type is a bit different from others. You'll need to set the string for the currently given item index. type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; uinfo->count = 1; uinfo->value.enumerated.items = 4; if (uinfo->value.enumerated.item > 3) uinfo->value.enumerated.item = 3; strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]); return 0; } ]]>

This callback is used to read the current value of the control and to return to the user-space. For example, value.integer.value[0] = get_some_value(chip); return 0; } ]]> The value field is depending on the type of control as well as on info callback. For example, the sb driver uses this field to store the register offset, the bit-shift and the bit-mask. The private_value is set like and is retrieved in callbacks like private_value & 0xff; int shift = (kcontrol->private_value >> 16) & 0xff; int mask = (kcontrol->private_value >> 24) & 0xff; .... } ]]> In get callback, you have to fill all the elements if the control has more than one elements, i.e. count > 1. In the example above, we filled only one element (value.integer.value[0]) since it's assumed as count = 1.

This callback is used to write a value from the user-space. For example, current_value != ucontrol->value.integer.value[0]) { change_current_value(chip, ucontrol->value.integer.value[0]); changed = 1; } return changed; } ]]> As seen above, you have to return 1 if the value is changed. If the value is not changed, return 0 instead. If any fatal error happens, return a negative error code as usual. Like get callback, when the control has more than one elements, all elements must be evaluated in this callback, too.

All these three callbacks are basically not atomic.

When everything is ready, finally we can create a new control. For creating a control, there are two functions to be called, snd_ctl_new1() and snd_ctl_add(). In the simplest way, you can do like this: where my_control is the struct snd_kcontrol_new object defined above, and chip is the object pointer to be passed to kcontrol->private_data which can be referred in callbacks. snd_ctl_new1() allocates a new snd_kcontrol instance (that's why the definition of my_control can be with __devinitdata prefix), and snd_ctl_add assigns the given control component to the card.

If you need to change and update a control in the interrupt routine, you can call snd_ctl_notify(). For example, This function takes the card pointer, the event-mask, and the control id pointer for the notification. The event-mask specifies the types of notification, for example, in the above example, the change of control values is notified. The id pointer is the pointer of struct snd_ctl_elem_id to be notified. You can find some examples in es1938.c or es1968.c for hardware volume interrupts.

The ALSA AC97 codec layer is a well-defined one, and you don't have to write many codes to control it. Only low-level control routines are necessary. The AC97 codec API is defined in <sound/ac97_codec.h>.

private_data; .... // read a register value here from the codec return the_register_value; } static void snd_mychip_ac97_write(struct snd_ac97 *ac97, unsigned short reg, unsigned short val) { struct mychip *chip = ac97->private_data; .... // write the given register value to the codec } static int snd_mychip_ac97(struct mychip *chip) { struct snd_ac97_bus *bus; struct snd_ac97_template ac97; int err; static struct snd_ac97_bus_ops ops = { .write = snd_mychip_ac97_write, .read = snd_mychip_ac97_read, }; if ((err = snd_ac97_bus(chip->card, 0, &ops, NULL, &bus)) < 0) return err; memset(&ac97, 0, sizeof(ac97)); ac97.private_data = chip; return snd_ac97_mixer(bus, &ac97, &chip->ac97); } ]]>

For creating an ac97 instance, first call snd_ac97_bus with an ac97_bus_ops_t record with callback functions.