Add missing kerneldoc entry for sdio_set_host_pm_flags()'s 'flag'
function argument.

Fixes the following W=1 kernel build warning:

 kernel/drivers/mmc/core/sdio_io.c:722: warning: Function parameter or member 'flags' not described in 'sdio_set_host_pm_flags'

Cc: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Link: https://lore.kernel.org/r/20200701124702.908713-3-lee.jones@linaro.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Not all source files which include quirks.h make use of the all of
the available fixup information.  When this happens the compiler
complains that some constant variables are defined by never used.
We can fix this by telling the compiler that this intentional by
simply marking them as __maybe_unused.

Fixes the following W=1 kernel build warnings:

 In file included from drivers/mmc/core/sdio.c:22:
 drivers/mmc/core/quirks.h:105:31: warning: ‘mmc_ext_csd_fixups’ defined but not used [-Wunused-const-variable=]
 105 | static const struct mmc_fixup mmc_ext_csd_fixups[] = {
 | ^~~~~~~~~~~~~~~~~~
 drivers/mmc/core/quirks.h:17:31: warning: ‘mmc_blk_fixups’ defined but not used [-Wunused-const-variable=]
 17 | static const struct mmc_fixup mmc_blk_fixups[] = {
 | ^~~~~~~~~~~~~~
 In file included from drivers/mmc/core/mmc.c:25:
 drivers/mmc/core/quirks.h:123:31: warning: ‘sdio_fixup_methods’ defined but not used [-Wunused-const-variable=]
 123 | static const struct mmc_fixup sdio_fixup_methods[] = {
 | ^~~~~~~~~~~~~~~~~~
 drivers/mmc/core/quirks.h:17:31: warning: ‘mmc_blk_fixups’ defined but not used [-Wunused-const-variable=]
 17 | static const struct mmc_fixup mmc_blk_fixups[] = {
 | ^~~~~~~~~~~~~~

Cc: Andrei Warkentin <andreiw@motorola.com>
Cc: Martin Mares <mj@ucw.cz>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Link: https://lore.kernel.org/r/20200701124702.908713-2-lee.jones@linaro.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Fix sparse build warning:

drivers/mmc/host/sdhci-msm.c:1888:6: warning:
 symbol 'sdhci_msm_dump_vendor_regs' was not declared. Should it be static?
Signed-off-by: Hulk Robot <hulkci@huawei.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/20200702020347.77214-1-weiyongjun1@huawei.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

If this controller doesn't issue interrupt for response of command,
this driver lacks to call dma_unmap_sg(). So, fix the issue by
using end() of struct tmio_mmc_dma_ops.
Reported-by: Takeshi Saito <takeshi.saito.xv@renesas.com>
Fixes: 2a68ea78 ("mmc: renesas-sdhi: add support for R-Car Gen3 SDHI DMAC")
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/1590044466-28372-4-git-send-email-yoshihiro.shimoda.uh@renesas.comTested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>

To add end() operation in the future, clean the code of
renesas_sdhi_internal_dmac_complete_tasklet_fn(). No behavior change.
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/1590044466-28372-3-git-send-email-yoshihiro.shimoda.uh@renesas.comTested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Related drivers like renesas_sdhi_internal_dmac are possible
to lack dma unmaping in error cases (for example response timeout).

Since tmio_mmc_finish_request() will be always called in any case,
to fix the issue, add end operation into struct tmio_mmc_dma_ops and
call the operation in tmio_mmc_finish_request() to call dma_ummap API
by the related drivers correctly.
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/1590044466-28372-2-git-send-email-yoshihiro.shimoda.uh@renesas.comTested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Define appropriate macro names for consistency with other macros.
Signed-off-by: Pali Rohár <pali@kernel.org>
Link: https://lore.kernel.org/r/20200629072144.24351-1-pali@kernel.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

On qcom SD host controllers voltage switching be done after the HW
is ready for it. The HW informs its readiness through power irq.
The voltage switching should happen only then.

Use the internal voltage switching and then control the voltage
switching using power irq.

IO-bus supply of eMMC would be kept always-on. So set the load
for this supply to configure it in LPM when eMMC is suspend state
and in HPM when eMMC is active.
Co-developed-by: Asutosh Das <asutoshd@codeaurora.org>
Signed-off-by: Asutosh Das <asutoshd@codeaurora.org>
Co-developed-by: Vijay Viswanath <vviswana@codeaurora.org>
Signed-off-by: Vijay Viswanath <vviswana@codeaurora.org>
Co-developed-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org>
Signed-off-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org>
Link: https://lore.kernel.org/r/1592919288-1020-4-git-send-email-vbadigan@codeaurora.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Set the default power mode, MMC_POWER_UNDEFINED, in mmc_alloc_host() rather
than in mmc_start_host(). This enables host drivers to make use of the
initial state during ->probe().

Signed-off-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org> Link:
https://lore.kernel.org/r/1592919288-1020-3-git-send-email-vbadigan@codeaurora.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

If vendor platform drivers are controlling whole logic of voltage
switching, then sdhci driver no need control vqmmc regulator.
So skip enabling/disable vqmmc from SDHC driver.
Signed-off-by: Vijay Viswanath <vviswana@codeaurora.org>
Signed-off-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/1592919288-1020-2-git-send-email-vbadigan@codeaurora.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

HS400/HS200/eMMC HS doesn't have Preset Value register.
Hence, sdhci_set_ios function overrides the value set by fmw to
SDHCI_CTRL_DRV_TYPE_B.
This patch sets drv_type to MMC_SET_DRIVER_TYPE_A
so that host_control2 register gets updated with the required
strength value.
Signed-off-by: Akshu Agrawal <akshu.agrawal@amd.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/20200623133347.4598-1-akshu.agrawal@amd.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

USDHC of i.MX has internal IC debug register, which record the IC
logical status. So dump these logical status in error condition,
this can help analyzing issue.
Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/1592985830-13038-1-git-send-email-haibo.chen@nxp.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

In current code logic, when work in SDR12/SDR25 mode, the final clock
rate is incorrect, just the legancy 400KHz, because the
card->sw_caps.sd3_bus_mode do not has the flag SD_MODE_UHS_SDR12 or
SD_MODE_UHS_SDR25. Besides, SDIO_SPEED_SDR12 is actually value 0, and
every mode need to config the timing and clock rate, so remove the
‘if’ operator.
Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Tested-by: Pali Rohár <pali@kernel.org>
Link: https://lore.kernel.org/r/1592813959-5914-1-git-send-email-haibo.chen@nxp.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Add support for writing new clkbuf_sel property for the J721e 4 bit IP.
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Link: https://lore.kernel.org/r/20200619125801.9530-7-faiz_abbas@ti.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

For speed modes where DLL is not enabled write to TXDLYCLK to enable
transmitter delay chain mode.
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Link: https://lore.kernel.org/r/20200619125801.9530-6-faiz_abbas@ti.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

The clock > CLOCK_TOO_SLOW_HZ condition gating phy configuration
is only required because dll should not be enabled at too low a
clock frequency or too low timing. Make sure that this condition
only gates dll enablement.
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Link: https://lore.kernel.org/r/20200619125801.9530-5-faiz_abbas@ti.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Add Support for AM65x PG2.0. Use the SoC bus framework to fixup
the driver data and skip do DLL calibration if the revision is 1.0
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Link: https://lore.kernel.org/r/20200619125801.9530-4-faiz_abbas@ti.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Not all controllers need calibration for the PHY DLL. Add a DLL_CALIB flag
to indicate the same.
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Link: https://lore.kernel.org/r/20200619125801.9530-3-faiz_abbas@ti.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Add documentation for the clock buffer select phy property in the
am654x and j721e 4 bit IP
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Link: https://lore.kernel.org/r/20200619125801.9530-2-faiz_abbas@ti.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Convert the pwrseq binding to DT schema format using json-schema.

At the same time, fix a couple of issues with the examples discovered by
the validation tool -- missing ";"
Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Link: https://lore.kernel.org/r/20200622164431.3dbc8c5a@xhacker.debianSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

"K" stands for "kelvin".

While at it, make the spacing before units consistent.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20200618080321.16678-1-geert+renesas@glider.beReviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Fix typo: "trigered" --> "triggered"
Signed-off-by: Flavio Suligoi <f.suligoi@asem.it>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/20200617151938.30217-1-f.suligoi@asem.itSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

This patch adds datactrl_mask_sdio for sdmmc revisions.
sdmmc revisions used same bit of previous ST variant.
Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
Link: https://lore.kernel.org/r/20200611132839.4515-1-ludovic.barre@st.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

These functions do not check the return value of devm_clk_register():
  - sdhci_arasan_register_sdcardclk()
  - sdhci_arasan_register_sampleclk()

Therefore, add the missed checks to fix them.

Fixes: c390f211 ("mmc: sdhci-of-arasan: Add ability to export card clock")
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20200608162226.3259186-1-hslester96@gmail.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Commit 6b5eda36 ("sdio: put active devices into 1-bit mode during
suspend") disabled 4-bit mode during system suspend. After this patch,
commit 7310ece8 ("mmc: implement SD-combo (IO+mem) support") used
new sdio_enable_4bit_bus() instead of sdio_enable_wide() to support
SD-combo cards, also for card resume. However, no corresponding support
added during suspend. That is not correct. Let's fix it.
Signed-off-by: Yue Hu <huyue2@yulong.com>
Link: https://lore.kernel.org/r/20200609081431.6376-1-zbestahu@gmail.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

If the card type is SD combo(MMC_TYPE_SD_COMBO) and the memory part does
not support wider bus(SD_SCR_BUS_WIDTH_4), nothing will be done except
return 0. However, we should check whether IO part support wider bus or
not. We should use available IO ability if supported.

In addition, there's a duplicated check to MMC_CAP_4_BIT_DATA since
sdio_enable_wide() will include that check. And we can also save one
call site to sdio_enable_wide() after this change.
Signed-off-by: Yue Hu <huyue2@yulong.com>
Link: https://lore.kernel.org/r/20200608103009.5000-1-zbestahu@gmail.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

We should return any possible error returned by mmc_io_rw_direct()
rather than only -EIO in sdio_disable_func() failure path.
Signed-off-by: Yue Hu <huyue2@yulong.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20200604100102.13572-1-zbestahu@gmail.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

This patch adds comments about NVQUIRKS HAS_PADCALIB and NEEDS_PAD_CONTROL.
Signed-off-by: Sowjanya Komatineni <skomatineni@nvidia.com>
Link: https://lore.kernel.org/r/1591326240-28928-1-git-send-email-skomatineni@nvidia.comSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Add interconnect bandwidth scaling supported strings for qcom-sdhci
controller.
Signed-off-by: Pradeep P V K <ppvk@codeaurora.org>
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1591691846-7578-3-git-send-email-ppvk@codeaurora.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

Interconnect bandwidth scaling support is now added as a
part of OPP. So, make sure interconnect driver is ready
before handling interconnect scaling.
Signed-off-by: Pradeep P V K <ppvk@codeaurora.org>
Reviewed-by: Sibi Sankar <sibis@codeaurora.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Link: https://lore.kernel.org/r/1591691846-7578-2-git-send-email-ppvk@codeaurora.orgSigned-off-by: Ulf Hansson <ulf.hansson@linaro.org>

The approach to allow userspace ~5s to consume the uevent, which is
triggered when a new card is inserted/initialized, currently requires the
mmc host to support system wakeup.

This is unnecessary limiting, especially for an mmc host that relies on a
GPIO IRQ for card detect. More precisely, the mmc host may not support
system wakeup for its corresponding struct device, while the GPIO IRQ still
could be configured as a wakeup IRQ via enable_irq_wake().

To support all various cases, let's simply drop the need for the wakeup
support. Instead let's always register a wakeup source and activate it for
all card detect IRQs by calling __pm_wakeup_event().
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Link: https://lore.kernel.org/r/20200529102341.12529-1-ulf.hansson@linaro.org

Pull Xtensa fixes from Max Filippov:

 - fix __sync_fetch_and_{and,or}_4 declarations to avoid build warning

 - update *pos in cpuinfo_op.next to avoid runtime warning

 - use for_each_set_bit in xtensa_pmu_irq_handler instead of open-coding
   it

* tag 'xtensa-20200712' of git://github.com/jcmvbkbc/linux-xtensa:
  xtensa: simplify xtensa_pmu_irq_handler
  xtensa: update *pos in cpuinfo_op.next
  xtensa: fix __sync_fetch_and_{and,or}_4 declarations

Pull io_uring fixes from Jens Axboe:
 "Two late fixes again:

   - Fix missing msg_name assignment in certain cases (Pavel)

   - Correct a previous fix for full coverage (Pavel)"

* tag 'io_uring-5.8-2020-07-12' of git://git.kernel.dk/linux-block:
  io_uring: fix not initialised work->flags
  io_uring: fix missing msg_name assignment

Pull btrfs fixes from David Sterba:
 "Two refcounting fixes and one prepartory patch for upcoming splice
  cleanup:

   - fix double put of block group with nodatacow

   - fix missing block group put when remounting with discard=async

   - explicitly set splice callback (no functional change), to ease
     integrating splice cleanup patches"

* tag 'for-5.8-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: wire up iter_file_splice_write
  btrfs: fix double put of block group with nocow
  btrfs: discard: add missing put when grabbing block group from unused list

59960b9d ("io_uring: fix lazy work init") tried to fix missing
io_req_init_async(), but left out work.flags and hash. Do it earlier.

Fixes: 7cdaf587 ("io_uring: avoid whole io_wq_work copy for requests completed inline")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Ensure to set msg.msg_name for the async portion of send/recvmsg,
as the header copy will copy to/from it.

Cc: stable@vger.kernel.org # v5.5+
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Pull RISC-V fixes from Palmer Dabbelt:
 "I have a few KGDB-related fixes. They're mostly fixes for build
  warnings, but there's also:

   - Support for the qSupported and qXfer packets, which are necessary
     to pass around GDB XML information which we need for the RISC-V GDB
     port to fully function.

   - Users can now select STRICT_KERNEL_RWX instead of forcing it on"

* tag 'riscv-for-linus-5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: Avoid kgdb.h including gdb_xml.h to solve unused-const-variable warning
  kgdb: Move the extern declaration kgdb_has_hit_break() to generic kgdb.h
  riscv: Fix "no previous prototype" compile warning in kgdb.c file
  riscv: enable the Kconfig prompt of STRICT_KERNEL_RWX
  kgdb: enable arch to support XML packet.

Pull SCSI fixes from James Bottomley:
 "Five small fixes, four in driver and one in the SCSI Parallel
  transport, which fixes an incredibly old bug so I suspect no-one has
  actually used the functionality it fixes"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: dh: Add Fujitsu device to devinfo and dh lists
  scsi: mpt3sas: Fix error returns in BRM_status_show
  scsi: mpt3sas: Fix unlock imbalance
  scsi: iscsi: Change iSCSI workqueue max_active back to 1
  scsi: scsi_transport_spi: Fix function pointer check

Pull xen fix from Juergen Gross:
 "Just one fix of a recent patch (double free in an error path)"

* tag 'for-linus-5.8b-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/xenbus: Fix a double free in xenbus_map_ring_pv()