The DT binding doc says reset-gpios is an optional property but the code
currently bails out if it is omitted.
This is a regression since it breaks previously working device trees.
Fix it by restoring the original documented behaviour.
Fixes: ce03727586 ("mmc: pwrseq_simple: use GPIO descriptors array API")
Tested-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Martin Fuzzey <mfuzzey@parkeon.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
CISTPL_SDIO_STD(0x91) is a known tuple, but sdio_cis don't define it, so
we get the warning below while probing several sdio wifi cards.
Refer to SDIO spec, it's not needed to parse the tuple, so this patch make
it a known one.
[ 4.098980] mmc2: queuing unknown CIS tuple 0x91 (3 bytes)
[ 4.099033] mmc2: new ultra high speed SDR104 SDIO card at address 0001
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
While in sdhci_execute_tuning() the choice whether or not to enable the
tuning is done on the actual timing, in the mmc_sdio_init_uhs_card() the
check is done on the capability of the card.
This difference is causing some issues with some SDIO cards in DDR50
mode where the CDM19 is wrongly issued.
With this patch we modify the check in both
mmc_(sd|sdio)_init_uhs_card() functions to take the proper decision
only according to the actual timing specification.
Cc: stable@vger.kernel.org
Signed-off-by: Carlo Caione <carlo@endlessm.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The SD card specification allows cards to error out a SWITCH command
where the requested function in a group is not supported. The spec
provides for a set of capabilities which indicate which functions are
supported.
In the case of the power limit, requesting an unsupported power level
via the SWITCH command fails, resulting in the power level remaining at
the power-on default of 0.72W, even though the host and card may support
higher powers levels.
This has been seen with SanDisk 8GB cards, which support the default
0.72W and 1.44W (200mA and 400mA) in combination with an iMX6 host,
supporting up to 2.88W (800mA). This currently causes us to try to set
a power limit function value of '3' (2.88W) which the card errors out
on, and thereby causes the power level to remain at 0.72W rather than
the desired 1.44W.
Arrange to limit the selected current limit by the capabilities reported
by the card to avoid the SWITCH command failing. Select the highest
current limit that the host and card combination support.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Fixes: a39ca6ae0a ("mmc: core: Simplify and fix for SD switch processing")
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
A card can be removed while it is runtime suspended.
Do not print an error message.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Ulf 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: Ulf Hansson <ulf.hansson@linaro.org>
Tested-by: Alan Cooper <alcooperx@gmail.com>
Tested-by: Shawn Lin <shawn.lin@rock-chips.com>
there is a time window between __mmc_send_status() and time_afer(),
on some eMMC chip, the timeout_ms is only 10ms, if this thread was
scheduled out during this period, then, even card has already changes
to transfer state by the result of CMD13, this part of code also treat
it to timeout error.
So, need calculate timeout first, then call __mmc_send_status(), if
already timeout and card still in programing state, then treat it to
the real timeout error.
Signed-off-by: Chaotian Jing <chaotian.jing@mediatek.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Now, PM core supports asynchronous suspend/resume mode for devices
during system suspend/resume, and the power state transition of one
device may be completed in separate kernel thread. PM core ensures
all power state transition dependency between devices. This patch
enables MMC/SD/SDIO card and SDIO function devices to suspend/resume
asynchronously. This will take advantage of multicore and improve
system suspend/resume speed. After applying this patch and enabling
all SDIO function's child devices to suspend/resume asynchronously
on ASUS T100TA, the system suspend-to-idle time is reduced from
1645ms to 1108ms, and the system resume time is reduced from 940ms
to 918ms.
Signed-off-by: Zhonghui Fu <zhonghui.fu@linux.intel.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The 'ocr' parameter passed to mmc_set_signal_voltage()
defines the power-on voltage used when power cycling
after a failure to set the voltage. However, in the
case of mmc_sdio_init_card(), the value passed has the
R4_18V_PRESENT flag set which is not valid for power-on
and results in an invalid vdd. Fix by passing the card's
ocr value which does not have the flag.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: stable@vger.kernel.org # v3.13+
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Commit cc4f414c88 ("mmc: mmc: Add driver strength selection")
added driver strength selection for eMMC HS200 and HS400 modes.
That patch also set the driver stength when transitioning through
High Speed mode to HS200/HS400, but driver strength is not defined
for High Speed mode. While the JEDEC specification is not clear
on this point it has been observed to cause problems for some eMMC,
and removing the driver strength setting in this case makes it
consistent with the normal use of High Speed mode.
Signed-off-by: Wenkai Du <wenkai.du@intel.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: stable@vger.kernel.org # v4.2+
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
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: Carlo Caione <carlo@endlessm.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The mmc pm notifiers were recently reworked, but the new
code produces a lot of warnings when CONFIG_PM_SLEEP is disabled:
In file included from ../drivers/mmc/core/sdio_bus.c:27:0:
drivers/mmc/core/core.h:97:13: warning: 'mmc_register_pm_notifier' defined but not used [-Wunused-function]
The obvious solution is to add the 'inline' keyword at the
function definition, as it should be for any function defined
in a header file.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 0e40be7c20e0 ("mmc: core: Refactor code to register the MMC PM notifier")
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The mmc_pwrseq_ops structures are never modified, so declare them as const.
Done with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Ulf 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: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Ulf 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: Ulf 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: Ulf Hansson <ulf.hansson@linaro.org>
MMC_CAP_RUNTIME_RESUME was invented to decrease system PM resume time for
systems that particularly needs this. As the feature has matured let's
make it the default behavior for MMC/SD.
Signed-off-by: Ulf 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: Ulf 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: Ulf Hansson <ulf.hansson@linaro.org>
Though the mmc core driver should/will continue to support the legacy
"enable-sdio-wakeup" property to enable SDIO as the wakeup source, we
need to add support for the new standard property "wakeup-source".
This patch adds support for "wakeup-source" property in addition to the
existing "enable-sdio-wakeup" property.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
mmc_select_hs400() calls __mmc_switch() which checks the switch is
successful using CMD13 (SEND_STATUS). The problem is that it does that
using the timing settings of the previous mode. That is prone to error,
especially when switching from HS to HS400 because the timing parameters
for HS mode are tighter than the timing parameters for HS400 mode.
In the case when CMD13 polling is used (i.e. not MMC_CAP_WAIT_WHILE_BUSY)
with the switch command, it must be assumed that using different modes on
the card and host must work.
However in the case when CMD13 polling is not used
(i.e. MMC_CAP_WAIT_WHILE_BUSY) mmc_select_hs400() can be made more
reliable by setting the host to the correct timing before sending CMD13.
This patch does that.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: <stable@vger.kernel.org> # 4.2+
Tested-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Move the mmc_switch_status() function in preparation for calling it
in mmc_select_hs400().
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: <stable@vger.kernel.org> # 4.2+
Tested-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
mmc_select_hs400() begins with the card and host in HS200 mode.
Therefore, any commands sent to the card should use HS200 timing.
It is incorrect to set the host to High Speed (HS) timing before
sending the switch command. Doing so is unreliable because
the timing parameters for HS mode are tighter than the timing
parameters for HS200 mode. Thus the HS timings should be set
only after the card has switched mode.
However, it is not unreasonable first to reduce the frequency to
the HS mode frequency, which should make the switch command and
subsequent CMD13 commands more reliable.
This patch does that.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: <stable@vger.kernel.org> # 4.2+
Tested-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Currently mmc_select_hs200() uses __mmc_switch() which checks the
success of the switch to HS200 mode using CMD13 (SEND_STATUS).
The problem is that it does that using the timing settings of legacy
mode. That is prone to error, not least because the timing parameters
for legacy mode are tighter than the timing parameters for HS200 mode.
In the case when CMD13 polling is used (i.e. not MMC_CAP_WAIT_WHILE_BUSY)
with the switch command, it must be assumed that using different modes on
the card and host must work.
However in the case when CMD13 polling is not used
(i.e. MMC_CAP_WAIT_WHILE_BUSY) mmc_select_hs200() can be made more
reliable by setting the host to the correct timing before sending CMD13.
This patch does that.
A complication is that the caller, mmc_select_timing(), will ignore a
switch error (indicated by -EBADMSG), assume the old mode is valid
and continue, so the old timing must be restored in that case.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: <stable@vger.kernel.org> # 4.2+
Tested-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The pwrseq_emmc driver does a eMMC card reset before a system reboot to
allow broken or limited ROM boot-loaders (that don't have an eMMC reset
logic) to be able to read the second stage from the eMMC.
But this has to be called before a system reboot handler and while most
of them use the priority 128, there are other restart handlers (such as
the syscon-reboot one) that use a higher priority. So, use the highest
priority to make sure that the eMMC hw is reset before a system reboot.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Tested-by: Markus Reichl <m.reichl@fivetechno.de>
Tested-by: Anand Moon <linux.amoon@gmail.com>
Reviewed-by: Alim Akhtar <alim.akhtar@samsung.com>
Tested-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The mmc_execute_tuning() has already prepared the opcode,
there is no need to prepare it again at mmc_send_tuning(),
and, there is a BUG of mmc_send_tuning() to determine the opcode
by bus width, assume eMMC was running at HS200, 4bit mode,
then the mmc_send_tuning() will overwrite the opcode from CMD21
to CMD19, then got error.
in addition, extend an argument of "cmd_error" to allow getting
if there was cmd error when tune response.
Signed-off-by: Chaotian Jing <chaotian.jing@mediatek.com>
[Ulf: Rebased patch]
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Sometime only need set MMC_CAP_HW_RESET for one of MMC hosts,
So set it in device tree is better.
Signed-off-by: Chaotian Jing <chaotian.jing@mediatek.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
There's little sense in releasing the host on mmc_add_card() error
immediately after reclaiming it, so reclaim the host only in case
of success.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Ulf 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: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Ulf 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: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Ulf 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: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
As SD Specifications Part1 Physical Layer Specification Version
3.01 says, CMD19 tuning is available for unlocked cards in transfer
state of 1.8V signaling mode. The small difference between v3.00
and 3.01 spec means that CMD19 tuning is also available for DDR50
mode.
Signed-off-by: Weijun Yang <york.yang@csr.com>
Signed-off-by: Barry Song <Baohua.Song@csr.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
This patch add MMC_TIMING_UHS_SDR12 and MMC_TIMING_UHS_SDR25
for mmc_ios_show to show the ios->timing if mmc card runs under
these two modes.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@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: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Add a helper function to check if an opcode is a sd-io-rw-* opcode.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The simple power sequence provider sets a value for multiple GPIOs in one
go so it is better to use the API already provided by the GPIO descriptor
API instead of open coding the same logic.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
This patch add ios->drv_type for mmc_ios_show to show the
card's driver type.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The gpiod_get() function expands to gpiod_get_index() with index 0
so it's better to use it since is easier to read and more concise.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
As there are no users of the __mmc_switch() API, except for the mmc core
itself, let's convert it from an exported function into an internal.
Signed-off-by: Ulf 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: Ulf Hansson <ulf.hansson@linaro.org>
Acked-by: Kishon Vijay Abraham I <kishon@ti.com>
Suppose that we got a data crc error, and it triggers the mmc_reset.
mmc_reset will call mmc_send_status to see if HW reset was supported.
before issue CMD13, it will do retune, and if EMMC was in HS400 mode,
it will reduce frequency to 52Mhz firstly, then results in card init
was doing at 52Mhz.
The mmc_send_status was originally only done for mmc_test, should drop
it. And, rename the "eMMC hardware reset" to "Reset test", as we would
also be able to use the test for SD-cards.
Signed-off-by: Chaotian Jing <chaotian.jing@mediatek.com>
Suggested-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: bd11e8bd03 ("mmc: core: Flag re-tuning is needed on CRC errors")
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
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: Chaotian Jing <chaotian.jing@mediatek.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: bd11e8bd03 ("mmc: core: Flag re-tuning is needed on CRC errors")
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
When CONFIG_GPIOLIB is unset, its stubs will return -ENOSYS. That means
when the mmc core parses DT for CD/WP GPIOs via mmc_of_parse(), -ENOSYS
becomes propagated to the caller. Typically this means that the mmc host
driver fails to probe.
As the CD/WP GPIOs are already treated as optional, let's extend that to
cover the case when CONFIG_GPIOLIB is unset.
Reported-by: Michal Simek <michal.simek@xilinx.com>
Fixes: 16b23787fc ("mmc: sdhci-of-arasan: Call OF parsing for MMC")
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Tested-by: Michal Simek <michal.simek@xilinx.com>
Acked-by: Venu Byravarasu <vbyravarasu@nvidia.com>
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: Jialing Fu <jlfu@marvell.com>
Tested-by: Shawn Lin <shawn.lin@rock-chips.com>
Fixes: 2220eedfd7 ("mmc: fix async request mechanism ....")
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf 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: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Use more compact of_property_read_bool() calls instead of the
of_find_property() calls.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Ulf 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: Robin van der Gracht <robin@protonic.nl>
Signed-off-by: Ulf 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: David Jander <david@protonic.nl>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Martin Fuzzey