In order to clean up the mmc_erase() function and do some optimization
for erase size alignment, factor out the guts of erase size alignment
into mmc_align_erase_size() function.
Signed-off-by: NBaolin Wang <baolin.wang@linaro.org>
Tested-by: NShawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

In cases when the host->max_busy_timeout isn't specified, the calculated
number of maximum discard sectors defaults to UINT_MAX. This may cause a
too long timeout for a discard request.

Avoid this by using a default maximum erase timeout of 60s, used when we
calculate the maximum number of sectors that are allowed to be discarded
per request.

Do note that the minimum number of sectors to be discarded is still at
least one "preferred erase size".
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Reviewed-by: NShawn Lin <shawn.lin@rock-chips.com>

When mmc host HW supports busy signalling (using R1B as response), don't
use the host->max_busy_timeout as the limitation when deciding the max
discard sectors, which we inform the generic BLOCK layer about. Instead,
let's use at least one preferred erase size as the max discard sectors.

In cases when the host controller supports HW busy signalling and the
timeout for the erase operation doesn't exceed the max_busy_timeout, we
keep the R1B response, otherwise we prevent the host from doing HW busy
detection by converting to a R1 response.
Signed-off-by: NBaolin Wang <baolin.wang@linaro.org>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Host drivers which needs to valdiate for non-supported MMC
commands and returnn error code for such requests.

To improve and simplify the behaviour, let's invent MMC_CAP2_NO_MMC
which these host drivers can set to tell the mmc core to skip sending MMC
commands during card initialization.
Signed-off-by: NShawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

There are host drivers which needs to valdiate for non-supported SD
commands and returnn error code for such requests.

To improve and simplify the behaviour, let's invent MMC_CAP2_NO_SD
which these host drivers can set to tell the mmc core to skip sending SD
commands during card initialization.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

If available, eMMC stack uses HC_ERASE_GRP_SIZE as the preferred erase
size. As some high capacity eMMC (64MB) reports this size to 512kB, the
discard operations end up taking very long time.

Improve the behaviour by instead calculating the preferred erase size
based on the eMMC size. In this way the discard operations becomes faster.
Signed-off-by: NGwendal Grignou <gwendal@chromium.org>
[Ulf: Updated changelog and improved comment in code]
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Controllers use data strobe line to latch data from devices
under hs400 mode, but not for cmd line. So since emmc 5.1, JEDEC
introduces enhanced strobe mode for latching cmd response from
emmc devices to host controllers. This new feature is optional,
so it depends both on device's cap and host's cap to decide
whether to use it or not.
Signed-off-by: NShawn Lin <shawn.lin@rock-chips.com>
Reviewed-by: NJaehoon Chung <jh80.chung@samsung.com>
Tested-by: NDouglas Anderson <dianders@chromium.org>
Tested-by: NJaehoon Chung <jh80.chung@samsung.com>
Acked-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

008GE0 Toshiba mmc in some Intel Baytrail tablets responds to
MMC_SEND_EXT_CSD in 450-600ms.

This patch will...

() Increase the long read time quirk timeout from 300ms to 600ms. Original
   author of that quirk says 300ms was only a guess and that the number
   may need to be raised in the future.

() Add this specific MMC to the quirk
Signed-off-by: NMatt Gumbel <matthew.k.gumbel@intel.com>
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

The eMMC HW reset may be implemented either via the host ops ->hw_reset()
callback or through DT and the eMMC pwrseq. Additionally some eMMC cards
don't support HW reset.

To allow a reset to be done for the different combinations of mmc hosts
and eMMC/MMC cards, let's implement a fallback via trying a regular power
cycle. This improves the mmc block layer retry mechanism of failing I/O
requests.
Signed-off-by: NGwendal Grignou <gwendal@chromium.org>
[Ulf: Rewrote changelog]
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

This patch provides some tracepoints for the lifecycle of a mmc request
from starting to completion to help with performance analysis of MMC
subsystem.
Signed-off-by: NBaolin Wang <baolin.wang@linaro.org>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Print the error code when the tuning command fails.  This allows the
reason for the failure to be reported, which aids debugging.
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Improve mmc_of_parse_voltage()'s return values so that drivers can tell
whether a voltage-range specification was present, and whether it has
been successfully parsed, or there was an error while parsing.

We return a negative errno when parsing fails, zero if no voltage-range
specification is present, or one if a voltage-range specification is
successfully parsed.

No users need modifying as no users check the return value.
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Each time a driver such as sdhci-esdhc-imx is probed, we get a info
printk complaining that the DT voltage-ranges property has not been
specified.

However, the DT binding specifically says that the voltage-ranges
property is optional.  That means we should not be complaining that
DT hasn't specified this property: by indicating that it's optional,
it is valid not to have the property in DT.

Silence the warning if the property is missing.
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

The bus width is sometimes the actual bus width, and sometimes indices
to different arrays encoding the bus width. In my debugging case "2"
could mean 8-bit as well as 4-bit, which was extremly confusing. Let's
use the human-readable actual bus width in all places.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

In linux/mmc/host.h, mmc_card_is_removable() is already defined.
There is no reason that it doesn't use.
Signed-off-by: NJaehoon Chung <jh80.chung@samsung.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

The mmc workqueue is an ordered workqueue, allowing only one work to
execute per given time. As this workqueue is used for card detection, the
conseqeunce is that cards will be detected one by one waiting for each
other.

Moreover, most of the time spent during card initialization is waiting for
the card's internal firmware to be ready. From a CPU perspective this
typically means waiting for a completion variable to be kicked via an
IRQ-handler or waiting for a sleep timer to finish.

This behaviour of detecting/initializing cards is sub-optimal, especially
for SOCs having several controllers/cards.

Let's convert to use the system_freezable_wq for the mmc detect works.
This enables several works to be executed simultaneously and thus also
cards to be detected like so.

Tests on UX500, which holds two eMMC cards and an SD-card (actually also
an SDIO card, currently not detected), shows a significant improved
behaviour due to this change.

Before this change, both the eMMC cards waited for the SD card to be
initialized as its detect work entered the workqueue first. In some cases,
depending on the characteristic of the SD-card, they got delayed 1-1.5 s.

Additionally for the second eMMC, it needed to wait for the first eMMC to
be initialized which added another 120-190 ms.

Converting to the system_freezable_wq, removed these delays and made both
the eMMC cards available far earlier in the boot sequence.

Selecting the system_freezable_wq, in favour of for example the system_wq,
is because we need card detection mechanism to be disabled once userspace
are frozen during system PM. Currently the mmc core deal with this via PM
notifiers, but following patches may utilize the behaviour of the
system_freezable_wq, to simplify the use of the PM notifiers.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Tested-by: NAlan Cooper <alcooperx@gmail.com>
Tested-by: NShawn Lin <shawn.lin@rock-chips.com>

This patch introduce a new MMC_CAP2_NO_SDIO cap used to tell the mmc
core to not send SDIO specific commands.
Signed-off-by: NCarlo Caione <carlo@endlessm.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Turn the informative message about no vmmc/vqmmc regulator found in
debug one. There is no need to indicate that something optional is
missing. Moreover, it can bring confusion, people who doesn't know
it is optional may consider these messages as warnings or errors.
Signed-off-by: NLudovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

_mmc_detect_card_removed() validates that the card is removable, but when
being called via the bus_ops ->detect() callbacks, the validation is
redundant as it's already done in mmc_rescan().

Move the validation of a removable card to the mmc_detect_card_removed()
API, which is where it's applicable, to allow the blk error recovery path
to get the response a bit earlier.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Instead of checking for "#ifdef" directly in the code, let's invent a pair
of mmc core functions to deal with register/unregister the MMC PM notifier
block. Implement stubs for these functions when CONFIG_PM_SLEEP is unset,
as in that case the PM notifiers isn't used.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

As mmc_claim_host() invokes pm_runtime_get_sync() for the mmc host device,
it's important that the host is kept claimed for *all* accesses to it via
the host_ops callbacks.

In mmc_rescan(), the ->card_event() and the ->get_cd() callback are being
invoked without claiming the host, let's fix this.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

The ->card_event() callback may be called when re-scan is disabled and for
non-removable cards, which both cases are unnecessary.

Instead let's move the call later in mmc_rescan() where these constraints
have been validated.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

This adds logic to the MMC core to set VQMMC.  This is expected to be
called by MMC drivers like dw_mmc as part of (or instead of) their
start_signal_voltage_switch() callback.

A few notes:

* When setting the signal voltage to 3.3V we do our best to make VQMMC
  and VMMC match.  It's been reported that this makes some old cards
  happy since they were tested back in the day before UHS when VQMMC
  and VMMC were provided by the same regulator.  A nice side effect of
  this is that we don't end up on the hairy edge of VQMMC (2.7V),
  which some EEs claim is a little too close to the minimum for
  comfort.
  This is done in two steps. At first we try to find a VQMMC within
  a 0.3V tolerance of VMMC and if this is not supported by the
  supplying regulator we try to find a suitable voltage within the
  whole 2.7V-3.6V area of the spec.

* The two step approach is currently necessary, as the used
  regulator_set_voltage_triplet(min, target, max) uses a simple
  implementation that just tries two basic steps:
	regulator_set_voltage(target, max);
	regulator_set_voltage(min, target);
  So with only one step with 2.7-3.6V borders, if a suitable voltage
  is a bit below VMMC, we would directly get the lowest 2.7V
  which some boards (like Rockchips) don't like at all.

* When setting the signal voltage to 1.8V or 1.2V we aim for that
  specific voltage instead of picking the lowest one in the range.

* We very purposely don't print errors in mmc_regulator_set_vqmmc().
  There are cases where the MMC core will try several different
  voltages and we don't want to pollute the logs.
Signed-off-by: NDouglas Anderson <dianders@chromium.org>
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

We will shortly need the calculation of an ocr-bit to the actual
voltage in a second place too, so move it from mmc_regulator_set_ocr
to a common function mmc_ocrbitnum_to_vdd to make that possible.
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

MMC_CLKGATE was once invented to save power by gating the bus clock at
request inactivity. At that time it served its purpose. The modern way to
deal with power saving for these scenarios, is by using runtime PM.

Nowadays, several host drivers have deployed runtime PM, but for those
that haven't and which still cares power saving at request inactivity,
it's certainly time to deploy runtime PM as it has been around for several
years now.

To simplify code to mmc core and thus decrease maintenance efforts, this
patch removes all code related to MMC_CLKGATE.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Some sdio wifi chips will not work properly if we try to start new
sdio-rw requests while the device is signalling that it is busy.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

As mmc_claim_host() invokes pm_runtime_get_sync() for the mmc host device,
it's important that the host is kept claimed for *all* accesses to it via
the host_ops callbacks.

In some code paths for SDIO, particularly related to the PM support,
mmc_power_off|up() is invoked without keeping the host claimed. Let's fix
these.

Moreover, mmc_start|stop_host() also invokes mmc_power_off|up() without
claiming the host, let's fix these as well.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Acked-by: NKishon Vijay Abraham I <kishon@ti.com>

When get a CRC error, start the mmc_retune, it will issue CMD19/CMD21
to do tune, assume there were 10 clock phase need to try, phase 0 to
phase 6 is ok, phase 7 to phase 9 is NG, we try it from 0 to 9, so
the last CMD19/CMD21 will get CRC error, host->need_retune was set and
cause mmc_retune was called, then dead loop of mmc_retune
Signed-off-by: NChaotian Jing <chaotian.jing@mediatek.com>
Acked-by: NAdrian Hunter <adrian.hunter@intel.com>
Fixes: bd11e8bd ("mmc: core: Flag re-tuning is needed on CRC errors")
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

The following panic is captured in ker3.14, but the issue still exists
in latest kernel.
---------------------------------------------------------------------
[   20.738217] c0 3136 (Compiler) Unable to handle kernel NULL pointer dereference
at virtual address 00000578
......
[   20.738499] c0 3136 (Compiler) PC is at _raw_spin_lock_irqsave+0x24/0x60
[   20.738527] c0 3136 (Compiler) LR is at _raw_spin_lock_irqsave+0x20/0x60
[   20.740134] c0 3136 (Compiler) Call trace:
[   20.740165] c0 3136 (Compiler) [<ffffffc0008ee900>] _raw_spin_lock_irqsave+0x24/0x60
[   20.740200] c0 3136 (Compiler) [<ffffffc0000dd024>] __wake_up+0x1c/0x54
[   20.740230] c0 3136 (Compiler) [<ffffffc000639414>] mmc_wait_data_done+0x28/0x34
[   20.740262] c0 3136 (Compiler) [<ffffffc0006391a0>] mmc_request_done+0xa4/0x220
[   20.740314] c0 3136 (Compiler) [<ffffffc000656894>] sdhci_tasklet_finish+0xac/0x264
[   20.740352] c0 3136 (Compiler) [<ffffffc0000a2b58>] tasklet_action+0xa0/0x158
[   20.740382] c0 3136 (Compiler) [<ffffffc0000a2078>] __do_softirq+0x10c/0x2e4
[   20.740411] c0 3136 (Compiler) [<ffffffc0000a24bc>] irq_exit+0x8c/0xc0
[   20.740439] c0 3136 (Compiler) [<ffffffc00008489c>] handle_IRQ+0x48/0xac
[   20.740469] c0 3136 (Compiler) [<ffffffc000081428>] gic_handle_irq+0x38/0x7c
----------------------------------------------------------------------
Because in SMP, "mrq" has race condition between below two paths:
path1: CPU0: <tasklet context>
  static void mmc_wait_data_done(struct mmc_request *mrq)
  {
     mrq->host->context_info.is_done_rcv = true;
     //
     // If CPU0 has just finished "is_done_rcv = true" in path1, and at
     // this moment, IRQ or ICache line missing happens in CPU0.
     // What happens in CPU1 (path2)?
     //
     // If the mmcqd thread in CPU1(path2) hasn't entered to sleep mode:
     // path2 would have chance to break from wait_event_interruptible
     // in mmc_wait_for_data_req_done and continue to run for next
     // mmc_request (mmc_blk_rw_rq_prep).
     //
     // Within mmc_blk_rq_prep, mrq is cleared to 0.
     // If below line still gets host from "mrq" as the result of
     // compiler, the panic happens as we traced.
     wake_up_interruptible(&mrq->host->context_info.wait);
  }

path2: CPU1: <The mmcqd thread runs mmc_queue_thread>
  static int mmc_wait_for_data_req_done(...
  {
     ...
     while (1) {
           wait_event_interruptible(context_info->wait,
                   (context_info->is_done_rcv ||
                    context_info->is_new_req));
     	   static void mmc_blk_rw_rq_prep(...
           {
           ...
           memset(brq, 0, sizeof(struct mmc_blk_request));

This issue happens very coincidentally; however adding mdelay(1) in
mmc_wait_data_done as below could duplicate it easily.

   static void mmc_wait_data_done(struct mmc_request *mrq)
   {
     mrq->host->context_info.is_done_rcv = true;
+    mdelay(1);
     wake_up_interruptible(&mrq->host->context_info.wait);
    }

At runtime, IRQ or ICache line missing may just happen at the same place
of the mdelay(1).

This patch gets the mmc_context_info at the beginning of function, it can
avoid this race condition.
Signed-off-by: NJialing Fu <jlfu@marvell.com>
Tested-by: NShawn Lin <shawn.lin@rock-chips.com>
Fixes: 2220eedf ("mmc: fix async request mechanism ....")
Signed-off-by: NShawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

For some mass production of kingston eMMCs which adopt Phison's
firmware will meet an unrecoverable data conrruption occasionally
if performing trim due to a firmware bug confirmed by vendor. We
found it on Intel-C3230RK platform. So we add fixup of broken trim
for it.
Signed-off-by: NShawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

When requesting a trim for several bytes, everything up to the next
erase-group is erased. This causes data corruption.
Signed-off-by: NRobin van der Gracht <robin@protonic.nl>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

In the (not so unlikely) case that the mmc controller timeout budget is
enough for exactly one erase-group, the simplification of allowing one
sector has an enormous performance penalty. We optimize this special case
by introducing a flag that prohibits erase-group boundary crossing, so
that we can allow trimming more than one sector at a time.
Signed-off-by: NDavid Jander <david@protonic.nl>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Make a new function out of common code used for drive
strength selection.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

IO state variable drv_type could be set during card
initialization. Consequently, it must be reset to the
default value when setting the initial state.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Since the regulator used for the SDMMC IO voltage is not expected to
draw a lot of current, most systems will probably use an inexpensive
LDO for it.  LDO regulators apparently have the feature that they
don't actively drive the voltage down--they wait for other components
in the system to drag the voltage down.  Thus they will transition
faster under heavy loads and slower under light loads.

During an SDMMC voltage change from 3.3V to 1.8V, we are almost
certainly under a light load.  To be specific:
* The regulator is hooked through pulls to CMD0-3 and DAT.  Probably
  the CMD pulls are something like 47K and the DAT is something like
  10K.
* The card is supposed to be driving DAT0-3 low during voltage change
  which will draw _some_ current, but not a lot.
* The regulator is also provided to the SDMMC host controller, but the
  SDMMC host controller is in open drain mode during the voltage
  change and so shouldn't be drawing much current.

In order to keep the SDMMC host working properly (or for noise
reasons), there might also be a capacitor attached to the SDMMC IO
regulator.  This also will have the effect of slowing down transitions
of the regulator, especially under light loads.

From experimental evidence, we've seen the voltage change fail if the
card doesn't detect that the voltage fell to less than about 2.3V when
we turn on the clock.  On one device (that admittedly had a 47K CMD
pullup instead of a 10K CMD pullup) we saw that the voltage was just
about 2.3V after 5ms and thus the voltage change would sometimes fail.
Doubling the delay gave margin and made the voltage change work 100%
of the time, despite the slightly weaker CMD pull.

At the moment submitting this as an RFC patch since my problem _could_
be fixed by increasing the pull strength (or using a smaller
capacitor).  However being a little bit more lenient to strange
hardware could also be a good thing.
Signed-off-by: NDoug Anderson <dianders@chromium.org>
Acked-by: NMark Brown <broonie@kernel.org>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Check the error code for EOPNOTSUPP and do not print
reset warning in that case.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

CRC errors could possibly be alleviated by
re-tuning so flag re-tuning needed in those cases.
Note this has no effect if re-tuning has not been
enabled.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Hold re-tuning during bkops to prevent
it from conflicting with the busy state.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

Hold re-tuning during erase commands to prevent
it from conflicting with the sequence of commands.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>

At the start of each request, re-tune if needed and
then hold off re-tuning again until the request is done.

Note that though there is one function that starts
requests (mmc_start_request) there are two that wait for
the request to be done (mmc_wait_for_req_done and
mmc_wait_for_data_req_done).  Also note that
mmc_wait_for_data_req_done can return even when the
request is not done (which allows the block driver
to prepare a newly arrived request while still
waiting for the previous request).

This patch ensures re-tuning is held for the duration
of a request.  Subsequent patches will also hold
re-tuning at other times when it might cause a
conflict.

In addition, possibly a command is failing because
re-tuning is needed. Use mmc_retune_recheck() to check
re-tuning. At that point re-tuning is held, at least by
the request, so mmc_retune_recheck() flags host->retune_now
if the hold count is 1.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>