pci: extend MSI interrupt vector number to 32 bits

Also increase MSI message data size to 32 bits according to PCIe spec.

Remove 0xff check for MSI interrupts because it is potentially valid
interrupt vector number. Reject 0xff only for legacy pin interrupts.

- MSI-X supports up to 2048 interrupts per device that do not fit to
`uint8`.

- Non-x86 systems may use separate interrupt vector ranges for
hard-wired interrupts and MSI interrupts so `uint8` is not enough to
represent all of them.

Change-Id: Iaf9ffb197ec23db0f97ffe3ea756d28d7bfc8705
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7433
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>

bus & drivers: drop PCI_x86

Change-Id: I494deaf24a4793a5e0fe9fa46ecdce32f65e616a
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6226
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
Reviewed-by: Fredrik Holmqvist <fredrik.holmqvist@gmail.com>

hda: add missing product entries for newer devices

should fix #17972
update some product entries for intel which requires the skylake quirk
remove product entries for intel which don't require snoop

Change-Id: I08f2bd640fe9bef4405abc9b2eaa83070b9adce9
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5719
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

drivers/audio: Adjust physical memory mapping after recent VM changes.

Fixes more KDLs.

kernel/vm: Remove default kernel read/write flags

`fix_protection` will not apply `B_KERNEL_READ_AREA` and
`B_KERNEL_WRITE_AREA` by default.

Kernel drivers that directly call `create_area` or `create_area_etc`
and do not pass any protection flags have been updated to
apply `B_KERNEL_READ_AREA | B_KERNEL_WRITE_AREA` instead.

Bug: #17751
Change-Id: I43e7ee6b5396e0309cdcff750e28262942c6d01c
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5330
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: enable power on init

Change-Id: I96b6084a40270aa8007b916a35e54b1fc3b26b39
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4977
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: detect some audio devices per device ID

as it's possible they use the audio class_sub instead of hd_audio.
* also power up the audio function, this seems required on Jasper Lake.
* tested on Acer Swift 1 SF114-34.

Change-Id: I603018b1b973e7884aa99a58be1c24e568f2d5ae
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4928
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>

hda: remove invalid product entries for intel

A few invalid ones were introduced in 2071dc2

Change-Id: Icbe5bb4aafd9d6e03ec4175c10eccdb2e65465d7
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4874
Reviewed-by: Alex von Gluck IV <kallisti5@unixzen.com>
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>

hda:sync intel ids list from freebsd

Change-Id: I8a9610a10709de1d54f315d4868b7b4001c06a0f
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4500
Reviewed-by: Alex von Gluck IV <kallisti5@unixzen.com>

hda/intel: disable MISCBDCGE on chip init

* Disabled Misc Backbone Dynamic Clock Gating during init.
* The alsa and linux kernel drivers both do this on
these skylake HD Audio chipsets.
First iteration:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6639484dda
Later iteration (all skylake and broxton):
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/pci/hda/hda_intel.c#n555

Change-Id: Id914fb5541f7020b9edb4fa635fcfffde9847663
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4556
Reviewed-by: Alex von Gluck IV <kallisti5@unixzen.com>

hda: disable 64bits dma addressing for some devices

* creative, not supported
* amd/ati/nvidia, supports 40/48bits, disabled for simplicity
* should help with #16491

Change-Id: I736d17b7dc5d4798688f9bd2741705362837b082
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3680
Reviewed-by: Jérôme Duval <jerome.duval@gmail.com>

hda: Remove interrupt hanlder before freeing interrupt vector.

Freeing an interrupt vector that still has handlers is not allowed.

Change-Id: If2f256e5288c7a61b7e23bdb202f259f5b67b7ae
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3112
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Use B_32_BIT_CONTIGUOUS when controller is not 64 bit.

This ensures that the allocated physical memory uses 32 bit addresses.

Also avoid writing to the 64 bit address registers when the controller
is not 64 bit.

Change-Id: I58008d68bcd9579903534f824db2660f73e39160
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3065
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Style cleanup only, no functional change.

Change-Id: Id36bf67a51369099093ae24412fa681f88c4c8d0
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3064
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Reset streams before (re-)setting DMA memory addresses.

The spec says that changing the DMA memory addresses of a stream should
really only be done after a reset and before the first run. Stopping the
stream is not enough, as the controller is allowed to cache the
addresses even when the stream is paused.

Note that this is applied to both, the initial configuration of buffers
and to the case of dynamically changing stream parameters. In the initial
case this should always be a no-op as the controller was previously
globally reset.

In the dynamic reconfiguration case this fixes random memory corruption
due to the controller still accessing and modifying physical memory that
is released by the driver. With some controllers, such corruption was
easily triggered by changing the input and output sampling rates for the
HDA device in the media preferences.

The return value of the reset call is intentionally not checked to allow
for it to partially or fully fail. At least for HDA under QEMU, the
stream reset fails to start, while the initial stream stop still works
and prevents the memory corruption.

Change-Id: I7c6e6547ee08889a2990de26d5eed5e73b2ca13c
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3063
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Make interrupt setup more conservative.

Don't unconditionally enable PCI interrupt generation as this only
applies when MSIs are not in use. Delay the evaluation and setting of
the interrupt disable bit until after it is clear whether or not MSIs
are to be used.

Initially force interrupt generation off as early as possible. Depending
on the initial state of the controller, some spurious interrupts might
come up and we really don't want them to occur before everything is set
up in the right way.

Change-Id: Ifb12f5eef1cbf8ecfa3f352564139125d8ad2169
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3062
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Remove doubled and early put of gPCIx86Module.

This is already properly done in uninit_driver.

Putting the module in device instead of driver uninit meant that on a
device reopen the gPCIx86Module would always be gone and therefore MSIs
could not be used anymore even if they previously were.

Change-Id: Ib3bddefbe6bcb02fa1c97f3c843cecaf31c80e3e
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3061
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Unconfigure the MSI as well.

Just disabling it will leak the vector reservation and makes it
impossible to configure an MSI on the device again.

Also removes the check for the gPCIx86Module as having an MSI configured
really implies its presence.

Change-Id: I9f0e3e63c988af515d23ad93d86918cde9add97a
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3060
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Disable MSIs when installing the interrupt handler fails.

This was previously only done if reset failed.

Change-Id: I30ebe1d302087816bccfe67e3d236c9446425aea
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3059
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Wait for streams to actually start/stop.

In the start case this adds debug output in case a start fails.

For stops this makes sure the stream DMA engine has actually stopped
operating, which is a precondition for some other operations.

Change-Id: I0e2731513febd0e0cd4e1f0ef89cf3d7312a47e1
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3058
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: Factor out waiting for bits to be set/cleared.

It is a recurring pattern to set some bits, then wait for them to be
read as set or the reverse. For resets this is done once in both
directions, so factoring this out removes quite a bit of similar code.

No functional change intended.

Change-Id: I2ae4b6cce490c33e1a027328f01fbffea67c5180
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3057
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

drivers: Fix 64 bit BAR checks and remove redundant masking.

Fix address type mask to use the right bits via PCI_address_type. The
previous 0xc value applies to the upper of the two address type and
the prefetchable bit instead of the two adddress type bits. This would
cause prefetchable 64 bit BARs to be misidentified as not being 64 bit.

Remove address masking as this is already done on the field in pci_info.

Use uint64 instead of phys_addr_t explicitly as on pure 32 bit platforms
this would otherwise result in a shift greater than type size.

Note that the sizes should theoretically be adjusted as well, but in all
of these cases a size needing the upper 32 bits doesn't make much sense.

Change-Id: Idd0d6e9a6cc18656b43cafcac720b132fc1f2ad3
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3024
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

HDA: Fix mapping 64-bit PCI addresses.

Some small cleanup.

Change-Id: I07f6cfeedc9f650740e1be7ab998b4fc3e0fd4d2
Reviewed-on: https://review.haiku-os.org/c/haiku/+/2875
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

hda: disable TCSEL on ATI/AMD for non-hmdi

also disable MSI with C-MEDIA 0x5011 (reference alsa)

Change-Id: I51e0b8f046df13a23a0b39998e4155a2ec40f5f2
Reviewed-on: https://review.haiku-os.org/c/haiku/+/2290
Reviewed-by: Jérôme Duval <jerome.duval@gmail.com>

hda: sync snoop quirk intel pci ids list based on alsa.

Change-Id: I08d8e8aa05d03ce49cbf3937eccdb109b00e83f2
Reviewed-on: https://review.haiku-os.org/c/haiku/+/2288
Reviewed-by: Jérôme Duval <jerome.duval@gmail.com>

hda: enable pci snooping on AMD non-hdmi.

disable snooping on ATI non-hdmi.

Change-Id: Ic62e8c7ae376a1e26646f88f308242eef3aa119e
Reviewed-on: https://review.haiku-os.org/c/haiku/+/2287
Reviewed-by: Jérôme Duval <jerome.duval@gmail.com>

hda: add necessary quirk for SiS Azalia Audio Controller

Fixes #15499.

Change-Id: I8bc6df5770f902b6b43e54d80bcfee4afc42f0bd
Reviewed-on: https://review.haiku-os.org/c/haiku/+/1981
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

PVS V595: use before null-check

Change-Id: I1e74b376d77686c08778882b6f2c4c67a51b6eb3
Reviewed-on: https://review.haiku-os.org/c/1614
Reviewed-by: Stephan Aßmus <superstippi@gmx.de>

hda: wait between set and unset of the GLOBAL_CONTROL_RESET bit.

* the spec requires 521us, 1000us is generally used.
* also only uninit streams, corb and rirb when a codec reset is in progress (seen in zircon).
linux instead clears HDAC_STATE_STATUS.

HDA: VirtualBox no longer needs QUIRK_NO_CORBRP_RESET_ACK.

HDA: VMware's emulated controllers have HDA_QUIRK_NO_CORBRP_RESET_ACK.

On my VMware Player installation, the device ID is "1977", but presumably
VMware will have only one HDA codebase, and so we just enable the fix on
all devices.

This fixes initialization of the HDA driver on VMware, and means that
it is now possible to get sound output on VMware without using the
OpenSound driver package.

hda: fix 64-bit warnings.

hda: remove a few ids from the quirks snoop list for intel.

should fix #14231. misread from the alsa quirk list.

hda: sync snoop quirk pci ids list based on alsa.

hda: sync snoop quirk pci ids list based on alsa.

hda: Create HDA_QUIRK_NO_CORBRP_RESET_ACK and use it.

This is instead of just checking for NVIDIA devices. Also add an entry
for Intel 0x2668 (VirtualBox's HDA controller). Gets HDA a bit further
in VirtualBox, but an acquire_sem() times out later on.

hda: cleanup tracing.

* Make sure all traces are prefixed with the driver name (for easier
grepping)
* Remove some useless traces
* 80 column limit fixes

Style fix: line length

Relaxed CORB read pointer reset checking for nVidia HDA

* At least some nVidia HDA controllers do not acknowledge the CORB
read pointer reset, which is required by the spec (cf. HDA 1.0a
ch. 3.3.21). For these controllers, do not fail when acknowledge
is missing.

* Makes hda driver work for nVidia HDA controllers again, thus fixing
the regression pointed out in #10212.

Use uncached memory areas if snooping is not possible

* Some chipsets do not support bus-snooping for the HDA data
transfers. So, when trying to enable snooping, verify that the
setting was accepted.

* When snooping is not possible, set memory areas for audio buffers,
buffer descriptor lists and command ring-buffers as uncached to
ensure memory coherency with the HDA controller.

* This fixes KDLs during booting on some systems where snooping
is not supported (seen on two systems with AMD chipsets, maybe
applies to others as well).
After setting up the buffer descriptor list for a stream, it was a
gamble whether these writes would be visible to the HDA controller
when subsequently enabling the stream. On some boots it was unlucky
and then the HDA controller's DMA engine would go haywire and start
overwriting random kernel memory with audio input data. The
scheduler was usually the first victim and caused page fault KDLs
due to corrupted data structures.

Preserve reserved bits in registers

* Do read-modify-write cycle to preserve bits marked as RsvdP in the
following registers: WAKEEN, CORBWP, CORBCTL, CORBSIZE, RIRBWP,
RINTCNT, RIRBCTL, RIRBSIZE.

Use correct sequence for resetting CORB read pointer

* Spec requires software to wait for hardware to acknowledge the
reset by waiting for the bit to become set and then manually
unset it and again wait for it to become unset.
(cf. HDA spec 1.0a ch. 3.3.21)

* Also, do read-modify-write to protect preserved bits

Read back CORB/RIRB control register to ensure DMA has stopped

* Spec is somewhat unclear about whether that is really necessary,
but let's play safe.

hda: add snoop quirk for a few other pci ids.

* based on alsa.

hda: Add Baytrail HDA controller

Signed-off-by: Jérôme Duval <jerome.duval@gmail.com>

hdaudio: enable MSI support

* disabled for nVidia and Creative Technology devices.

Enable PCI memory access and interrupt

hda: fixes KDL on HDMI, some digital quirks

* avoid crashing in case of lack of playback or record stream
* set format on digital output widgets
* accept digital output on the output path

Added HDA quirks for enabling snoop.

* Enabled snoop for all Intel hardware as found in the Linux driver.
* This fixes #8949.

Disabled HDA MSI for now again.

* At least on my hardware, audio becomes a bit flaky (ie. sometimes it would
just stop doing anything at all).

Minor cleanup, no functional change.

Cleanup, no functional change intended.

MSI interrupts were not disabled on error.

Reset controller::active_codec on uninit.

* The controller structure is static, so if you closed the device,
and the driver failed to initialize correctly (ie. finds no codec)
on the second try, it would have crashed accessing an already freed
codec.
* Not sure why it fails to detect any codecs on second open, yet,
though.

Implemented MSI support.

* Does not seem to change anything, though (I'm also a victim of bug
#8949)
* Minor cleanup.

Fixed 64 bit issue for the buffer descriptor base.

spaces -> tab


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@41388 a95241bf-73f2-0310-859d-f6bbb57e9c96

* If an unknown type is encountered, return immediately. This fixes CID 1504.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38408 a95241bf-73f2-0310-859d-f6bbb57e9c96

some verbs use a 16 bits payload


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37114 a95241bf-73f2-0310-859d-f6bbb57e9c96

* add some helper functions to read/write verbs.
* try at support realtek alc888. alsa uses this init sequence.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37113 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Changed physical_entry::{address,size} to phys_{addr,size}_t and changed
map_physical_memory()'s physicalAddress parameter type from void* to
phys_addr_t. This breaks source compatibility, but -- as long as
phys_{addr,size}_t remain 32 bit wide -- keeps binary compatibility with
BeOS.
* Adjusted all code using the affected interfaces (Oh what fun!). Added a few
TODOs in places where the wrong types (e.g. void* for physical addresses
are used). Looks like quite a few drivers aren't 64 bit safe and others
will break with PAE.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36960 a95241bf-73f2-0310-859d-f6bbb57e9c96

* We now mute internal speakers when using headphones. We don't distinguish between headphones and mic jacks though.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35088 a95241bf-73f2-0310-859d-f6bbb57e9c96

* warn_count is unused.
* the last valid index should be written in HDAC_STREAM_LAST_VALID instead of the fragment count.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34730 a95241bf-73f2-0310-859d-f6bbb57e9c96

Try to detect whether the stream's DMA position is broken and switch to using
the LPIB, if it is.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34684 a95241bf-73f2-0310-859d-f6bbb57e9c96

* stream_handle_interrupt():
- Round the DMA position for the buffer cycle computation. Apparently some
chipsets trigger the interrupt before the position has been updated.
- Don't just assume that stream->buffer_length frames have been processed
at that time. Use the exact stream position at that time. This makes the
performance time computation more precise and immune to the interrupt
being delayed.
* Init hda_stream::frames_count.

Audio skips on I/O seem to be gone for me, now. Not obviously motivated skips
still happen.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34671 a95241bf-73f2-0310-859d-f6bbb57e9c96

* stream_handle_interrupt(): Fixed a misunderstanding of the SDnLPIB (stream
descriptor n link position in buffer) registers. They contain "the number
of bytes that have been received off the link", which is not to be confused
with the number of bytes that have been transferred by the DMA engine.
The interrupt is triggered when the last byte of the buffer has been fetched
by the DMA engine, at which point the stream's LPIB is still somewhere in
the last buffer. So the interrupt handler could compute the wrong buffer
index, which would lead to the multi audio add-on filling the wrong
(currently being transferred) buffer, resulting in noisy sound. Now we use
the DMA position. Should fix #4072.
* Also removed the not (always) working hack-around for the "wrong" buffer
positions in the interrupt handler.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34664 a95241bf-73f2-0310-859d-f6bbb57e9c96

* buffer_ready_sem was released with B_DO_NOT_RESCHEDULE, but without
returning B_INVOKE_SCHEDULER from the interrupt handler, causing
latencies up to a full quantum for the multi audio output thread. This
change improves audio clicks quite a bit on my machine. Though they still
happen from time to time and particulary on FS activity.
* Automatic whitespace cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34633 a95241bf-73f2-0310-859d-f6bbb57e9c96

* cleanup
* use gpio count for gpio quirks


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34355 a95241bf-73f2-0310-859d-f6bbb57e9c96

* added vref and unsol responses definitions
* untested unsolicited response support
* added quirk support for vref and gpio
* vref are now enabled for all inputs, and gpio for some Apple Macs
* replaced dprintf with TRACE and ERROR macros


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34352 a95241bf-73f2-0310-859d-f6bbb57e9c96

* prints the stream fifo size


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33121 a95241bf-73f2-0310-859d-f6bbb57e9c96

try to align the offset to be more compliant


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33047 a95241bf-73f2-0310-859d-f6bbb57e9c96

prints out stream status errors


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33046 a95241bf-73f2-0310-859d-f6bbb57e9c96

try at workaround early playback interrupts on non intel.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33011 a95241bf-73f2-0310-859d-f6bbb57e9c96

the flag means no snoop


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32924 a95241bf-73f2-0310-859d-f6bbb57e9c96

we now print a warning message in the log when the interrupt comes in the second half of a buffer: this should mean the interrupt comes earlier than expected.
Please report such warnings, they can help to investigate timing issues with some hda codecs.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32589 a95241bf-73f2-0310-859d-f6bbb57e9c96

* now uses only one buffer_ready semaphore, this way the buffer exchange is done asap, useful when rates/formats are different between playback and record.
* checked what offsets my hardware my hardware really had: it affected only playback, and was 192 for 16 bits and 256 for 20/24 bits. With these values, playback and record are crystal clear.
As I can't find any references for such offset values anywhere, sorry it's not supposed to work out of the box on all hardware. Maybe we could adjust the offset at runtime.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31875 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fix by Cyan: Do the offset calculation taking the rate base into account.
Fixes using the HDA driver with frame rates based on 44100Hz.
* Automatic white space cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30704 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Use a default buffer frame count that is adapted to the sample rate
of the audio group. This is supposed to keep the latency about the same
regardless of sample rate and lessen the requirements on the system
performance when using higher sample rates. Currently the multi-audio
addon uses the highest available rate.
* Added TODO about the highest sample rate seemingly being forgotten in one
place.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29421 a95241bf-73f2-0310-859d-f6bbb57e9c96

* compute offset based on vendor (32 for non intel) and sample rate


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29327 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Stream interrupts seem to arrive too early on most HDA chips:
we adjust buffer descriptors to take this into account. It defaults to one sample, but it should depend also on the sample rate or the chip vendor.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29322 a95241bf-73f2-0310-859d-f6bbb57e9c96

Do not enable interrupts on codec status change, since we do not handle those. This fixes the lockup on the EeePC, and follows the description in the HDA specs.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29172 a95241bf-73f2-0310-859d-f6bbb57e9c96

use uint32 as temp variable, noticed by Marcus. Thanks!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28914 a95241bf-73f2-0310-859d-f6bbb57e9c96

try to complete snooping enablement


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28906 a95241bf-73f2-0310-859d-f6bbb57e9c96

* reading CORB_READ_POS_RESET as zero fails on some chips (ie nForce).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28849 a95241bf-73f2-0310-859d-f6bbb57e9c96

* try to enable EAPD on capable pin widgets
* added a pin capabilities attribute instead of input and output pin attributes
* added ATI and nVidia vendor ids definitions
* uses "mic in" and "line in" when pin colors are undefined


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28839 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch from tqh, modified by myself: enable PCIe snooping through vendor specific registers (ATI and nVidia).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28838 a95241bf-73f2-0310-859d-f6bbb57e9c96

* uses the default output stream on all audio outputs (and all output paths)
* enable trace to debug mixer controls build


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28699 a95241bf-73f2-0310-859d-f6bbb57e9c96

added a mixer interface


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28677 a95241bf-73f2-0310-859d-f6bbb57e9c96

* added pin capabilities and power support constants
* added a record stream, assigned it to input widgets: no recorded sound yet, though buffer cycling is ok


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26953 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed a ton of wrong usages of B_CONTIGUOUS + B_FULL_LOCK.
* The use of B_{READ|WRITE}_AREA throughout the drivers is surely alarming.
Defining these flags means that *every user* application can access these
buffers read/write, it becomes visible in userspace like any other memory
(just shared among all apps). I would like to ask each driver maintainer
to see if that is really wished here. If you only need one app to be able
to access it, cloning the area would be more appropriate.
* I came across the use of B_ANY_KERNEL_BLOCK_ADDRESS a number of times. This
is almost completely useless for most usages, as it tries to align the
virtual to a multiple of the size of the area. It just makes the allocation
more likely to fail. Please only use where appropriate, and please review
your code.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26858 a95241bf-73f2-0310-859d-f6bbb57e9c96

* added an ALIGN macro for easing readability and use it for 128 alignments
* use HDAC_STREAM_POSITION register value to check the current buffer cycle in interrupt handler
* added B_FULL_LOCK flags for area allocation, not sure it's handled but at least more correct
* buffer descriptors now use a low and high address fields
* applied a byte mask on format
* enabled PCI bus mastering if not yet done
* the PCI space register TCSEL (Traffic Class Select Register, which sets PCI express QOS) is now reset to TC0 (clear 0-2 bits) as needed on some boards like mine to
ensure good playback

Playback is finally working correctly here on ICH8 HDA!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26847 a95241bf-73f2-0310-859d-f6bbb57e9c96

fix alignment


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26446 a95241bf-73f2-0310-859d-f6bbb57e9c96

* some methods to access stream registers were defined with wrong types
* use HDAC_BIDIR_STREAM_OFFSET and HDAC_OUTPUT_STREAM_OFFSET when applicable
* use a PAGE_ALIGN macro


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26330 a95241bf-73f2-0310-859d-f6bbb57e9c96

moved stateStatus read/write out of the interrupt handler
added some error verbose


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26282 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rewrote controller reset - it now performs a full reset, and does no longer
assume the driver to be in reset when started.
* Major cleanup of the register access: now hda_controller and hda_stream both
have member functions for this, the OREGx() and REGx() macros are gone.
* Made the register names and definitions more descriptive - the short names
of the specs are still mentioned in the comments.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25352 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rewrote widget output path discovery, it's now very similar to what the
FreeBSD driver is doing.
* hda_codec_new_audio_group() did not free the audio group's widgets on
failure.
* No longer create the input stream for now.
* Reworked multiaudio-support to work regardless if there is only a playback
or record stream.
* With these changes, I hear nothing on my laptop anymore (before there was
noise), but on another system, I can finally hear something that sounds very
much like the sinus wave the multi_audio_test application produces; the
sound quality is pretty bad though (lots of periodical noise and glitches).
* Made B_MULTI_GET_DESCRIPTION safe to be called from userland.
* Minor other cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24383 a95241bf-73f2-0310-859d-f6bbb57e9c96

A bit more meaningful constant names.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24183 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Shuffled the functions a bit around to separate private from public functions.
* Renamed some functions, added comments.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24182 a95241bf-73f2-0310-859d-f6bbb57e9c96

Accidently read one response less than needed since reworking it.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24181 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed .c to .cpp, and fixed all warnings.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24178 a95241bf-73f2-0310-859d-f6bbb57e9c96

hda: cleanup tracing.

* Make sure all traces are prefixed with the driver name (for easier
grepping)
* Remove some useless traces
* 80 column limit fixes

Style fix: line length

Relaxed CORB read pointer reset checking for nVidia HDA

* At least some nVidia HDA controllers do not acknowledge the CORB
read pointer reset, which is required by the spec (cf. HDA 1.0a
ch. 3.3.21). For these controllers, do not fail when acknowledge
is missing.

* Makes hda driver work for nVidia HDA controllers again, thus fixing
the regression pointed out in #10212.

Use uncached memory areas if snooping is not possible

* Some chipsets do not support bus-snooping for the HDA data
transfers. So, when trying to enable snooping, verify that the
setting was accepted.

* When snooping is not possible, set memory areas for audio buffers,
buffer descriptor lists and command ring-buffers as uncached to
ensure memory coherency with the HDA controller.

* This fixes KDLs during booting on some systems where snooping
is not supported (seen on two systems with AMD chipsets, maybe
applies to others as well).
After setting up the buffer descriptor list for a stream, it was a
gamble whether these writes would be visible to the HDA controller
when subsequently enabling the stream. On some boots it was unlucky
and then the HDA controller's DMA engine would go haywire and start
overwriting random kernel memory with audio input data. The
scheduler was usually the first victim and caused page fault KDLs
due to corrupted data structures.

Preserve reserved bits in registers

* Do read-modify-write cycle to preserve bits marked as RsvdP in the
following registers: WAKEEN, CORBWP, CORBCTL, CORBSIZE, RIRBWP,
RINTCNT, RIRBCTL, RIRBSIZE.

Use correct sequence for resetting CORB read pointer

* Spec requires software to wait for hardware to acknowledge the
reset by waiting for the bit to become set and then manually
unset it and again wait for it to become unset.
(cf. HDA spec 1.0a ch. 3.3.21)

* Also, do read-modify-write to protect preserved bits

Read back CORB/RIRB control register to ensure DMA has stopped

* Spec is somewhat unclear about whether that is really necessary,
but let's play safe.

hda: add snoop quirk for a few other pci ids.

* based on alsa.

hda: Add Baytrail HDA controller

Signed-off-by: Jérôme Duval <jerome.duval@gmail.com>

hdaudio: enable MSI support

* disabled for nVidia and Creative Technology devices.

Enable PCI memory access and interrupt

hda: fixes KDL on HDMI, some digital quirks

* avoid crashing in case of lack of playback or record stream
* set format on digital output widgets
* accept digital output on the output path

Added HDA quirks for enabling snoop.

* Enabled snoop for all Intel hardware as found in the Linux driver.
* This fixes #8949.

Disabled HDA MSI for now again.

* At least on my hardware, audio becomes a bit flaky (ie. sometimes it would
just stop doing anything at all).

Minor cleanup, no functional change.

Cleanup, no functional change intended.

MSI interrupts were not disabled on error.

Reset controller::active_codec on uninit.

* The controller structure is static, so if you closed the device,
and the driver failed to initialize correctly (ie. finds no codec)
on the second try, it would have crashed accessing an already freed
codec.
* Not sure why it fails to detect any codecs on second open, yet,
though.

Implemented MSI support.

* Does not seem to change anything, though (I'm also a victim of bug
#8949)
* Minor cleanup.

Fixed 64 bit issue for the buffer descriptor base.

spaces -> tab


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@41388 a95241bf-73f2-0310-859d-f6bbb57e9c96

* If an unknown type is encountered, return immediately. This fixes CID 1504.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38408 a95241bf-73f2-0310-859d-f6bbb57e9c96

some verbs use a 16 bits payload


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37114 a95241bf-73f2-0310-859d-f6bbb57e9c96

* add some helper functions to read/write verbs.
* try at support realtek alc888. alsa uses this init sequence.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37113 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Changed physical_entry::{address,size} to phys_{addr,size}_t and changed
map_physical_memory()'s physicalAddress parameter type from void* to
phys_addr_t. This breaks source compatibility, but -- as long as
phys_{addr,size}_t remain 32 bit wide -- keeps binary compatibility with
BeOS.
* Adjusted all code using the affected interfaces (Oh what fun!). Added a few
TODOs in places where the wrong types (e.g. void* for physical addresses
are used). Looks like quite a few drivers aren't 64 bit safe and others
will break with PAE.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36960 a95241bf-73f2-0310-859d-f6bbb57e9c96

* We now mute internal speakers when using headphones. We don't distinguish between headphones and mic jacks though.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35088 a95241bf-73f2-0310-859d-f6bbb57e9c96

* warn_count is unused.
* the last valid index should be written in HDAC_STREAM_LAST_VALID instead of the fragment count.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34730 a95241bf-73f2-0310-859d-f6bbb57e9c96

Try to detect whether the stream's DMA position is broken and switch to using
the LPIB, if it is.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34684 a95241bf-73f2-0310-859d-f6bbb57e9c96

* stream_handle_interrupt():
- Round the DMA position for the buffer cycle computation. Apparently some
chipsets trigger the interrupt before the position has been updated.
- Don't just assume that stream->buffer_length frames have been processed
at that time. Use the exact stream position at that time. This makes the
performance time computation more precise and immune to the interrupt
being delayed.
* Init hda_stream::frames_count.

Audio skips on I/O seem to be gone for me, now. Not obviously motivated skips
still happen.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34671 a95241bf-73f2-0310-859d-f6bbb57e9c96

* stream_handle_interrupt(): Fixed a misunderstanding of the SDnLPIB (stream
descriptor n link position in buffer) registers. They contain "the number
of bytes that have been received off the link", which is not to be confused
with the number of bytes that have been transferred by the DMA engine.
The interrupt is triggered when the last byte of the buffer has been fetched
by the DMA engine, at which point the stream's LPIB is still somewhere in
the last buffer. So the interrupt handler could compute the wrong buffer
index, which would lead to the multi audio add-on filling the wrong
(currently being transferred) buffer, resulting in noisy sound. Now we use
the DMA position. Should fix #4072.
* Also removed the not (always) working hack-around for the "wrong" buffer
positions in the interrupt handler.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34664 a95241bf-73f2-0310-859d-f6bbb57e9c96

* buffer_ready_sem was released with B_DO_NOT_RESCHEDULE, but without
returning B_INVOKE_SCHEDULER from the interrupt handler, causing
latencies up to a full quantum for the multi audio output thread. This
change improves audio clicks quite a bit on my machine. Though they still
happen from time to time and particulary on FS activity.
* Automatic whitespace cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34633 a95241bf-73f2-0310-859d-f6bbb57e9c96

* cleanup
* use gpio count for gpio quirks


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34355 a95241bf-73f2-0310-859d-f6bbb57e9c96

* added vref and unsol responses definitions
* untested unsolicited response support
* added quirk support for vref and gpio
* vref are now enabled for all inputs, and gpio for some Apple Macs
* replaced dprintf with TRACE and ERROR macros


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34352 a95241bf-73f2-0310-859d-f6bbb57e9c96

* prints the stream fifo size


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33121 a95241bf-73f2-0310-859d-f6bbb57e9c96

try to align the offset to be more compliant


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33047 a95241bf-73f2-0310-859d-f6bbb57e9c96

prints out stream status errors


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33046 a95241bf-73f2-0310-859d-f6bbb57e9c96

try at workaround early playback interrupts on non intel.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33011 a95241bf-73f2-0310-859d-f6bbb57e9c96

the flag means no snoop


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32924 a95241bf-73f2-0310-859d-f6bbb57e9c96

we now print a warning message in the log when the interrupt comes in the second half of a buffer: this should mean the interrupt comes earlier than expected.
Please report such warnings, they can help to investigate timing issues with some hda codecs.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32589 a95241bf-73f2-0310-859d-f6bbb57e9c96

* now uses only one buffer_ready semaphore, this way the buffer exchange is done asap, useful when rates/formats are different between playback and record.
* checked what offsets my hardware my hardware really had: it affected only playback, and was 192 for 16 bits and 256 for 20/24 bits. With these values, playback and record are crystal clear.
As I can't find any references for such offset values anywhere, sorry it's not supposed to work out of the box on all hardware. Maybe we could adjust the offset at runtime.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31875 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fix by Cyan: Do the offset calculation taking the rate base into account.
Fixes using the HDA driver with frame rates based on 44100Hz.
* Automatic white space cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30704 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Use a default buffer frame count that is adapted to the sample rate
of the audio group. This is supposed to keep the latency about the same
regardless of sample rate and lessen the requirements on the system
performance when using higher sample rates. Currently the multi-audio
addon uses the highest available rate.
* Added TODO about the highest sample rate seemingly being forgotten in one
place.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29421 a95241bf-73f2-0310-859d-f6bbb57e9c96

* compute offset based on vendor (32 for non intel) and sample rate


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29327 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Stream interrupts seem to arrive too early on most HDA chips:
we adjust buffer descriptors to take this into account. It defaults to one sample, but it should depend also on the sample rate or the chip vendor.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29322 a95241bf-73f2-0310-859d-f6bbb57e9c96

Do not enable interrupts on codec status change, since we do not handle those. This fixes the lockup on the EeePC, and follows the description in the HDA specs.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29172 a95241bf-73f2-0310-859d-f6bbb57e9c96

use uint32 as temp variable, noticed by Marcus. Thanks!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28914 a95241bf-73f2-0310-859d-f6bbb57e9c96

try to complete snooping enablement


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28906 a95241bf-73f2-0310-859d-f6bbb57e9c96

* reading CORB_READ_POS_RESET as zero fails on some chips (ie nForce).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28849 a95241bf-73f2-0310-859d-f6bbb57e9c96

* try to enable EAPD on capable pin widgets
* added a pin capabilities attribute instead of input and output pin attributes
* added ATI and nVidia vendor ids definitions
* uses "mic in" and "line in" when pin colors are undefined


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28839 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch from tqh, modified by myself: enable PCIe snooping through vendor specific registers (ATI and nVidia).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28838 a95241bf-73f2-0310-859d-f6bbb57e9c96

* uses the default output stream on all audio outputs (and all output paths)
* enable trace to debug mixer controls build


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28699 a95241bf-73f2-0310-859d-f6bbb57e9c96

added a mixer interface


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28677 a95241bf-73f2-0310-859d-f6bbb57e9c96

* added pin capabilities and power support constants
* added a record stream, assigned it to input widgets: no recorded sound yet, though buffer cycling is ok


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26953 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed a ton of wrong usages of B_CONTIGUOUS + B_FULL_LOCK.
* The use of B_{READ|WRITE}_AREA throughout the drivers is surely alarming.
Defining these flags means that *every user* application can access these
buffers read/write, it becomes visible in userspace like any other memory
(just shared among all apps). I would like to ask each driver maintainer
to see if that is really wished here. If you only need one app to be able
to access it, cloning the area would be more appropriate.
* I came across the use of B_ANY_KERNEL_BLOCK_ADDRESS a number of times. This
is almost completely useless for most usages, as it tries to align the
virtual to a multiple of the size of the area. It just makes the allocation
more likely to fail. Please only use where appropriate, and please review
your code.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26858 a95241bf-73f2-0310-859d-f6bbb57e9c96

* added an ALIGN macro for easing readability and use it for 128 alignments
* use HDAC_STREAM_POSITION register value to check the current buffer cycle in interrupt handler
* added B_FULL_LOCK flags for area allocation, not sure it's handled but at least more correct
* buffer descriptors now use a low and high address fields
* applied a byte mask on format
* enabled PCI bus mastering if not yet done
* the PCI space register TCSEL (Traffic Class Select Register, which sets PCI express QOS) is now reset to TC0 (clear 0-2 bits) as needed on some boards like mine to
ensure good playback

Playback is finally working correctly here on ICH8 HDA!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26847 a95241bf-73f2-0310-859d-f6bbb57e9c96

fix alignment


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26446 a95241bf-73f2-0310-859d-f6bbb57e9c96

* some methods to access stream registers were defined with wrong types
* use HDAC_BIDIR_STREAM_OFFSET and HDAC_OUTPUT_STREAM_OFFSET when applicable
* use a PAGE_ALIGN macro


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26330 a95241bf-73f2-0310-859d-f6bbb57e9c96

moved stateStatus read/write out of the interrupt handler
added some error verbose


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26282 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rewrote controller reset - it now performs a full reset, and does no longer
assume the driver to be in reset when started.
* Major cleanup of the register access: now hda_controller and hda_stream both
have member functions for this, the OREGx() and REGx() macros are gone.
* Made the register names and definitions more descriptive - the short names
of the specs are still mentioned in the comments.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25352 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rewrote widget output path discovery, it's now very similar to what the
FreeBSD driver is doing.
* hda_codec_new_audio_group() did not free the audio group's widgets on
failure.
* No longer create the input stream for now.
* Reworked multiaudio-support to work regardless if there is only a playback
or record stream.
* With these changes, I hear nothing on my laptop anymore (before there was
noise), but on another system, I can finally hear something that sounds very
much like the sinus wave the multi_audio_test application produces; the
sound quality is pretty bad though (lots of periodical noise and glitches).
* Made B_MULTI_GET_DESCRIPTION safe to be called from userland.
* Minor other cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24383 a95241bf-73f2-0310-859d-f6bbb57e9c96

A bit more meaningful constant names.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24183 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Shuffled the functions a bit around to separate private from public functions.
* Renamed some functions, added comments.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24182 a95241bf-73f2-0310-859d-f6bbb57e9c96

Accidently read one response less than needed since reworking it.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24181 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed .c to .cpp, and fixed all warnings.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24178 a95241bf-73f2-0310-859d-f6bbb57e9c96