CVS log for src/sys/dev/i2c/i2c.c

Up to [cvs.NetBSD.org] / src / sys / dev / i2c

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

iic(4): Use config_detach_children to simplify.

i2c(9): Nix smbus intr API.

It was introduced in 2007 for some Xbox thing which was removed in
2011.  The API and the threads it spawned have been sitting around
idly for over a decade serving no purpose -- sometimes causing kernel
lock spinouts in the event of panic.

Add ic_tag_private to obviate need for future ABI changes.  Not
currently used, but we can privately allocate memory in iic_tag_init
for the purpose later if need be without changing ABI.

XXX kernel revbump -- changes struct i2c_controller

If there is a compat string list, parenthetically print the first one
in iic_print_direct().  (From thorpej-i2c-spi-conf2 branch.)

Allocate data buffer instead of using the stack.

Split i2c_subr.c into a separate module rather than including it in
the iic module.  There are valid configurations where i2c_subr code
can be both built-in and part of a loaded module (eg, piixpm is in
the kernel, but the iic module is loaded later).  This causes the
in-kernel linker to detect a duplicate symbol.

driver(9): devsw_detach never fails.  Make it return void.

Prune a whole lotta dead branches as a result of this.  (Some logic
calling this is also wrong for other reasons; devsw_detach is final
-- you should never have any reason to decide to roll it back.  To be
cleaned up in subsequent commits...)

XXX kernel ABI change to devsw_detach signature requires bump

remove double "with" in comments and usage text. Also fix one typo.

Change the devhandle_from_*() functions to also take a "super handle",
from which the newly created handle will inherit it's implementation.
The root implementation for a new handle type is used if an invalid
"super handle" is passed.

Replace devhandle_invalidate(), which invalidates a devhandle, with
devhandle_invalid(), which returns an invalid devhandle.

Minimal hack to pass child devices a devhandle created from the OF
or ACPI "cookie".  Temporary measure until the i2c autoconfiguration
overhaul is merged (which fixes this in a more generic way).

Move the i2c enumeration helpers to their own header file to detangle
it from reliance on "iic" being configured into the kernel.

iic_devslot_remove(): remove an errant "return false;" (the result of
a paste-o).

Re-factor the code in sandpoint and sparc64 that enumerates a static
table of i2c device entries into something sharable.

Remove unnecessary references to i2c_attach_args::ia_prop and mark the ones
that remain as XXX.

In iic_devslot_reserve(), don't set a pointer to NULL before dereferencing
it in a KASSERT().

Use designated initialiers in iic_attach().

In iic_close(), mark the "sc" variable as __diagused.

Port over the changes from thorpej-i2c-spi-conf to thorpej-i2c-spi-conf2,
which is based on a newer HEAD revision.

Merge thorpej-cfargs2.

Adapt to CFARGS().

fix proplib deprecation

Correct a comment.

Remove "count" from the device enumeration callback arguments.

Check for errors from iic_acquire_bus() in iic_ioctl_exec().

- Set D_MCLOSE in the iic_cdevsw so that we get a d_close call for
  each close so that the module ref counting works properly.
- Rearrange things a little to avoid holding the iic_mtx a long as
  previously done.

Rather than allocating 8KB (!!) of space per i2c bus for a sparsely
populated array of child devices, use a sorted list instead, optimized
a bit for the common usage pattern.

iic_print_direct(): In the "not configured" case, parenthetically print
the first element of the compat list, if we got a compat list.

Stop using a prop_array of child device data to perform direct
configuration of i2c devices.  Instead, introduce and use a new
device call "i2c-enumerate-devices" that can be implemented by
devhandle providers that have I2C bindings.  This device call
gathers up the data into i2c_attach_args and invoked a callback
(back to the i2c layer) to attach the device.  The I2C controller
device supplies its devhandle (or the appropriate devhandle for
the I2C bus) when attaching the i2c bus instance using CFARG_DEVHANDLE.

Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
  actually needed.
- Don't be explicit about what interface attribute is attaching if
  the device only has one.  (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
  situations, making is visibly easier to see when indirect config is
  in play, and allowing for future change in semantics.  (As of now,
  this is just a wrapper around config_match(), but that is an
  implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance.  This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

config_match() -> config_probe().  It's a little over-loaded here because
we're mixing direct and indirect in the same place, but that will be
addresed at a later time.

CFARG_SUBMATCH -> CFARG_SEARCH for the indirect configuration uses.

Sync with HEAD.

Give config_attach() the tagged variadic argument treatment and
mechanically convert all call sites.

"iic" only has a single interface attribute, so no need to be explicit.

Audit CFARG_IATTR in config_found() calls, and remove it in situations
where the interface attribute is not ambiguous.

Mechanical conversion of config_found_sm_loc() -> config_found().
CFARG_IATTR usage needs to be audited.

In "rescan" routines, always pass locators and the interface attribute
straight through to config_search().  Also, for devices that carry only
one interface attribute, no need to do an ifattr_match(), because
rescan_with_cfdata() will have already validated that the parent is
eligible, which includes an interface attribute check.

The proliferation if config_search_*() and config_found_*() combinations
is a little absurd, so begin to tidy this up:

- Introduce a new cfarg_t enumerated type, that defines the types of
  tag-value variadic arguments that can be passed to the various
  config_*() functions (CFARG_SUBMATCH, CFARG_IATTR, and CFARG_LOCATORS,
  for now, plus a CFARG_EOL sentinel).
- Collapse config_search_*() into config_search() that takes these
  variadic arguments.
- Convert all call sites of config_search_*() to the new signature.
  Noticed several incorrect usages along the way, which will be
  audited in a future commit.

Add support for ACPI-based I2C mux attachment.

Change the device_compatible_match() function to only perform the match.
Introduce a device_compatible_lookup() function to return an entry based
on the same matching criteria (a'la of_search_compatible()).

Update iic_compatible_match() to reflect the above change, and introduce
iic_compatible_lookup().  This pattern is less awkward to use.

whitelist -> permitlist

i2c-indirect-config == false  -->  i2c-no-indirect-config == true

Update for proplib(3) API changes.

Merge changes from current as of 20200406

No need to use I2C_F_POLL here.

Cleanup i2c bus acquire / release, centralizing all of the logic into
iic_acquire_bus() / iic_release_bus().  "acquire" and "release" hooks
no longer need to be provided by back-end controller drivers (only if
they need special handling, e.g. powering on the i2c controller).
This results in the removal of a bunch of rendundant code from each
back-end controller driver.

Assert that we are not in hard interrupt context in iic_acquire_bus(),
iic_exec(), and iic_release_bus().

Pull up following revision(s) (requested by mlelstev in ticket #552):

	sys/dev/i2c/i2c.c: revision 1.70

Revert previous. Indirect matches are not wanted on platforms that
use external configuration data (FDT or OF).

Revert previous. Indirect matches are not wanted on platforms that
use external configuration data (FDT or OF).

Sync with HEAD

touch match_result only on success.

Merge the [pgoyette-compat] branch

Sync with HEAD, resolve a few conflicts

try to detach on shutdown; for now the only practical reason is to test that
the driver detach paths work

Sync with HEAD

Remove the i2c "size" locator and corresponding property / attach arg.
It was a hack meant only for EEPROMs, which have another way to specify
size in the config directive ("flag" paramter), as well as a better way
to detect size based on "compatible" string.

Change device_compatible_match() and iic_compatible_match() to return
the weighted match value and take an optional compatible-entry pointer,
rather than the other way around.

Sync with HEAD

iic_use_direct_match(): when iic_compatible_match() does not find a
comptible entry, set match quality to 0. Otherwise callers might use
random stack garbage.

- Rename iic_compat_match() to iic_compatible_match() and change it
  to use the new device_compatible_match() routine.  A pointer to
  the matching device_compatible_entry is returned if a match is
  found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
  direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
  simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
  all of the logic for direct-config matching.  If it returns true,
  the driver returns the match result (which may be 0).  If it returns
  false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
  lower-indexed "compatible" device property are more-specific matches,
  and return a better match quality accordingly.

Changes / enhancements to i2c indirect device auto-configuration:

— iic_search() chooses a “probe strategy� based on the
  "i2c-indirect-probe-strategy� property on the “iic� instance.
  Valid values are "smbus-quick-write�, "smbus-receive-byte�, and
  “none�.  If no value is specified, the default is "smbus-quick-write�.

— If the "i2c-indirect-device-whitelist� exists on the “iic� instance,
  iic_search() will first check the driver name in the cfdata_t against
  this list, and only allow the match/probe to move forward if the
  cfdata_t driver name is in the list.  This is primarily to accommodate
  the Intel integrated memory controller neutered-i2c-thing.

— If the cfdata_t specifies a wildcard address, each address of the i2c
  bus will be consulted.  If the cfdata_t contains a nailed-down address,
  then we limit the bus scan to that specific address.

— We explicitly skip reserved / special i2c addresses, such as the
  General-Call address, etc.

— We introduce the notion of a “match quality� for i2c drivers.  From
  lowest-quality to highest-quality: matched by plausible address only,
  matched by plausible address and poking at the bus to see if the
  device looks reasonable, matched by direct-config “compatible� string,
  matched by direct-config “driver name� string.

— If the “match quality� is merely “plausible address only�, then
  iic_search() will use the probe strategy selected above to see if
  a device responds to that address.

Back out unintended commits.

Adjust come defaults:
- Initial gain -> 16x
- auto_gain -> true

Sync with HEAD

Fix a problem reported by jmcneill@ where by a system with multuple i2c
busses would end up with "ghost" device instances on the second bus.  This
issue was previously masked on ARM systems by the empty-child-devices
array issue fixed recently (that effectively blocked all indirect config
of i2c busses on those systems).

To fix this problem, we require that indirectly-configured devices have
to fully specify their parent spec and address, e.g.:

foo* at iic0 addr 0x55

NOT

foo* at iic? addr ?

or even:

foo* at iic? addr 0x55

This is needed because of how indirect configuration works... attach
directives in the kernel config file are enumerated, calling the bus's
search routine, which in the case of i2c, enumerates all i2c addresses
and calls the match routine for each address.  Because we can't always
reliably probe for i2c devices, we ended up with erroneous matches.

Direct configuration of i2c is still allowed to use wildcarded parent specs
and locators.

For direct-config devices, pass the whole dictionary to the child as
ia_prop, so that chil drivers can look up device-specific properties.
Needed for the upcoming HID over I2C support, proposed on tech-kern@
on Dec, 1.

update from HEAD

Kill some more extern struct cfdriver declarations.

Down with externs in .c!

Sync with HEAD

remove checks for failure after memory allocation calls that cannot fail:

  kmem_alloc() with KM_SLEEP
  kmem_zalloc() with KM_SLEEP
  percpu_alloc()
  pserialize_create()
  psref_class_create()

all of these paths include an assertion that the allocation has not failed,
so callers should not assert that again.

Revise previous.  Rather than explicitly including <sys/localcount.h>
in all the places where {b,c}devsw is initialized, just include it
from <sys/conf.h>.  This avoids an include-sequence dependancy.

Add DEVSW_MODULE_INIT to existing device-driver modules, so that they
willl have a localcount defined and thus be permitted to load.  Without
a localcount, loading the module will return EINVAL.

XXX the dtrace and drm stuff might need to be fed back upstream?

Sync with HEAD

Sync with HEAD

Use one more value from "locators.h", also make sure 'size' is initialized.

Use values from "locators.h" when constructing locators array to be sure
of correct index;

Sync with HEAD

Create separate modules for i2c_bitbang and bpf_filter so these files
can be included in kernels which need them without also duplicating
them in other modules.  Removes the duplicate symbols I found which
prevented loading i2c and bpf modules after having fixed PR 45125.

Sync with HEAD (as of 26th Dec)

allow child devices to be passed in attach args instead of device dictionary

When loaded as a non-built-in module, make sure we attach the cdevsw.
Without this, the i2c bus works but userland programs (such as
i2cscan(8)) cannot open /dev/iic*.

While we're here, add a ref-count to make sure that the device doesn't
get detached while it is open.

Sync with HEAD

Pull up following revision(s) (requested by maxv in ticket #761):
	sys/arch/sgimips/dev/scn.c: revision 1.8
	sys/arch/sgimips/mace/macekbc.c: revision 1.8
	sys/dev/hpc/hpcfb.c: revision 1.59
	sys/dev/i2c/i2c.c: revision 1.47
	sys/dev/ic/mfi.c: revision 1.57
	sys/dev/if_ndis/if_ndis_pci.c: revision 1.21
	sys/dev/sysmon/sysmon_power.c: revisions 1.50-1.52
	sys/dev/usb/umass_isdata.c: revision 1.31
fix double free, found by Brainy.
--
Free cmd on error if we allocated it.
Found by Brainy, reported by maxv@.
--
don't leak rl (but there are other leaks), found by Brainy.
--
Free ped if we can't hand it to the power daemon.
Found by Brainy, reported by maxv@.
--
don't forget to free the dictionary.
--
another missing free dict.
--
fix leak, found by Brainy.
--
don't malloc a tiny, fixed size buffer to scribble into, then not use it
and never free it either
found by Brainy
--
malloc() -> kmem_alloc() for private data, also kmem_free() them if we
don't finish attaching for whatever reason
found by Brainy

Add required dependency on i2cexec

Sync with HEAD

dedup code.

Free cmd on error if we allocated it.

Found by Brainy, reported by maxv@.

Pull up following revision(s) (requested by jakllsch in ticket #444):
	sys/dev/i2c/i2c.c: revision 1.46
More correctly handle I²C exec ioctls with both a command and write data
phase. Previously the data phase of ioctl writes was sending uninitialized
kernel stack to the addressed device.

More correctly handle I²C exec ioctls with both a command and write data
phase. Previously the data phase of ioctl writes was sending uninitialized
kernel stack to the addressed device.

Let kernel config override I2C_MAX_ADDR

Rebase to HEAD as of a few days ago.

Rebase.

Add d_discard to all struct cdevsw instances I could find.

All have been set to "nodiscard"; some should get a real implementation.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

sync with head

Change (mostly mechanically) every cdevsw/bdevsw I can find to use
designated initializers.

I have not built every extant kernel so I have probably broken at
least one build; however I've also found and fixed some wrong
cdevsw/bdevsw entries so even if so I think we come out ahead.

Add a property "i2c-indirect-config" to allow/disallow i2c indirect config.
If missing, "allowed" is assumed.  Set it to false for OFW machines.
Patch from martin@.  Removes erroneous i2c probe messages that appeared with
wildcard support.

Remove unused variable

sync with head

Allow i2c addr wildcard matching. Use with care!

Added two locators for I2C devices: "intr" and "intrbase".

They are required to allow the TPS65950 companion chip to both handle its
interrupts on the system, and allow its extra GPIO pins to be used as
interrupts as well.

XXX This change is not adequate because intr_establish() is a MD interface,
    so it will have to be re-designed before merging to the main tree.

resync with head

Also set ia.ia_size from size when doing direct config.

sync with head

sync to latest -current.

Don't try to match a device if there's already a device attached at
the specified address.

remove spurious warning about duplicate devices when using info from MD
code ( like OpenFirmware )
ok jmcneill

iic_ioctl_exec: if the i2c controller's "exec" function uses -1 instead of
an error code to report failure, convert it to EIO.

fix build for kernels with i2c host controller drivers but no iic(4)
driver present

iic_ioctl_exec: initialize cmd before using it

add support for detaching iic(4)

I2C_IOCTL_EXEC: only allow i2c write ops if the device was opened for writing

Install dev/i2c/i2c_io.h and implement the API in the iic(4) driver.
Obsolete the I2C_SCAN option as this can now be done from userland.

Consistent use of 'return'.

Build iic(4) and gpioiic(4) as modules.

ii2c can be built as module.

Prevent multiple attachments at the same i2c address.

OK jmcneill@

add rescan support

Sync with HEAD.

sync with head

Do initialise i2c_attach_args new members, for indirect configuration
too.

Generic (device property based) framework for optional direct configuration
of i2c devices, as discussed on tech-kern.

sync with head.

Sync with HEAD.

Change the i2c bus scan routine to use the write_quick protocol for most
addresses, rather than trying to read a byte of data from the device.

Some devices don't like to be read from (certain clock devices are, I'm
told, write-only!), while others expect to be asked only for words (or
pairs of bytes).

While here, skip a bunch of i2c addresses that can't (or at the very
least, shouldn't) have any slave devices.

This is the only use in NetBSD of the quick_read/quick_write protocol,
and it remains disabled by default.  I've updated all the generic i2c
drivers to handle the quick_* protocols, but several port-specific
drivers have not been updated since I'm in no position to verify that
the changes work.  Assistance from sandpoint, arm/xscale, evbarm/gumstix,
mips/alchemy, and macppc would be greatly appreciated.

Sync with HEAD.

Sync with HEAD.

Sync with HEAD.

When scanning i2c bus for responding devices, skip over the multicast
i2c Alert Response Address.  Skipping this won't hurt (the address is
allegedly reserved), and it might avoid the lock-ups that have been
seen by others.

Sync with HEAD.

Merge in changes between wrstuden-revivesa-base-2 and
wrstuden-revivesa-base-3.

Provide an option I2C_SCAN to enable sanning i2c bus for responses.  This
is disabled by default, preserving current behavior.

OK garbled@

Sync with HEAD.

sync with head.

sync with head.

device_t/softc split and other related cosmetic changes.

use aprint_*_dev and device_xname

sync with head

sync with HEAD

Sync with HEAD.

Sync with head.

Sync with HEAD

sync with head.

use __KERNEL_RCSID()

Merge jmcneill-pm branch.

Rename pnp(9) -> pmf(9), as requested by many.

Register with PM framework. i2c is by definition dumb so nothing to
do...

Sync with HEAD.

sync with head.

Sync with HEAD

From jmcneill-pm: don't probe for I2C device. It has limited usefulness
and hangs a number of machines.

sync with HEAD

Sync with head.

Disable i2c scan, it's dangerous.

Sync with HEAD.

sync with head.

Sync with HEAD.

For SMBus, add the ability to enumerate devices on the bus.
This does NOT identify the devices, merely indicates the
presence of devices at certain addresses.  Tested on ichsmb
and nfsmb - other SMBus devices will need to ensure the
proper bus type is set. (I2C_TYPE_SMBUS)

From Nicolas Joly, via Paul Goyette, in PR#36744.

Sync with head.

Merge some of the less invasive changes from the vmlocking branch:

- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements

- Pass the error number and residual count to biodone(), and let it handle
  setting error indicators. Prepare to eliminate B_ERROR.
- Add a flag argument to brelse() to be set into the buf's flags, instead
  of doing it directly. Typically used to set B_INVAL.
- Add a "struct cpu_info *" argument to kthread_create(), to be used to
  create bound threads. Change "bool mpsafe" to "int flags".
- Allow exit of LWPs in the IDL state when (l != curlwp).
- More locking fixes & conversion to the new API.

Nuke the deferred kthread creation stuff, as it's no longer needed.
Pointed out by thorpej@.

- Add two new arguments to kthread_create1: pri_t pri, bool mpsafe.
- Fork kthreads off proc0 as new LWPs, not new processes.

sync with head.

Sync with HEAD.

Missing LIST_INIT. Pointed out by uebayasi@

Remove debug printf accidentally added in last commit.

Allow i2c child devices to register interrupt handlers to the smbus host
controller. Both interrupt context and process context type handlers are
supported.

sync with head.

sync with head.

Sync with head.

__unused removal on arguments; approved by core.

sync with head

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

sync with head

sync with head

Merge from HEAD.

remove the "name" element from the i2cbus attach args, that's what
interface attributes are good for

sync with head.

Sync with head.

sync with head.

sync with head.

Merge 2006-03-31 NetBSD-current into the "peter-altq" branch.

Use device_private().

Merge 2006-03-28 NetBSD-current into the "peter-altq" branch.

Implement a handful of missing smbus protocol functions. Proposed on
tech-kern: http://mail-index.netbsd.org/tech-kern/2006/03/16/0029.html

merge ktrace-lwp.

Sync with HEAD. Here we go again...

s/locdesc_t/int/g

Fix the sync with head I botched.

Sync with HEAD.

a round of autoconf cleanup:
-convert submatch() style functions (passed to config_search() or
 config_found_sm()) to the locator passing variants
-pass interface attributes in some cases
-make submatch() functions look uniformly as far as possible
-avoid macros which just hide cfdata members, and reduce dependencies
 on "locators.h"

Sync with HEAD

file i2c.c was added on branch ktrace-lwp on 2004-08-03 10:46:05 +0000

Fix typo in comment, reported by Alexander Yurchenko grange at rt mipt ru.

New generic I2C framework.  Supports bit-bang and "intelligent" I2C
interface controllers (of varying intelligence levels).

Contributed by Wasabi Systems, Inc.  Primarily written by Steve Woodford,
with some modification by me.