Skip to content

L
linux
Switch branch/tag
  1. 28 Nov, 2018 2 commits
    • Lukas Wunner's avatar
      spi: bcm2835: Fix book-keeping of DMA termination · dbc94411
      Lukas Wunner authored 
      If submission of a DMA TX transfer succeeds but submission of the
corresponding RX transfer does not, the BCM2835 SPI driver terminates
the TX transfer but neglects to reset the dma_pending flag to false.

Thus, if the next transfer uses interrupt mode (because it is shorter
than BCM2835_SPI_DMA_MIN_LENGTH) and runs into a timeout,
dmaengine_terminate_all() will be called both for TX (once more) and
for RX (which was never started in the first place).  Fix it.
Signed-off-by: default avatarLukas Wunner <lukas@wunner.de>
Fixes: 3ecd37ed ("spi: bcm2835: enable dma modes for transfers meeting certain conditions")
Cc: stable@vger.kernel.org # v4.2+
Cc: Mathias Duckeck <m.duckeck@kunbus.de>
Cc: Frank Pavlic <f.pavlic@kunbus.de>
Cc: Martin Sperl <kernel@martin.sperl.org>
Cc: Noralf Trønnes <noralf@tronnes.org>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      dbc94411
    • Lukas Wunner's avatar
      spi: bcm2835: Avoid finishing transfer prematurely in IRQ mode · 56c17234
      Lukas Wunner authored 
      The IRQ handler bcm2835_spi_interrupt() first reads as much as possible
from the RX FIFO, then writes as much as possible to the TX FIFO.
Afterwards it decides whether the transfer is finished by checking if
the TX FIFO is empty.

If very few bytes were written to the TX FIFO, they may already have
been transmitted by the time the FIFO's emptiness is checked.  As a
result, the transfer will be declared finished and the chip will be
reset without reading the corresponding received bytes from the RX FIFO.

The odds of this happening increase with a high clock frequency (such
that the TX FIFO drains quickly) and either passing "threadirqs" on the
command line or enabling CONFIG_PREEMPT_RT_BASE (such that the IRQ
handler may be preempted between filling the TX FIFO and checking its
emptiness).

Fix by instead checking whether rx_len has reached zero, which means
that the transfer has been received in full.  This is also more
efficient as it avoids one bus read access per interrupt.  Note that
bcm2835_spi_transfer_one_poll() likewise uses rx_len to determine
whether the transfer has finished.
Signed-off-by: default avatarLukas Wunner <lukas@wunner.de>
Fixes: e34ff011 ("spi: bcm2835: move to the transfer_one driver model")
Cc: stable@vger.kernel.org # v4.1+
Cc: Mathias Duckeck <m.duckeck@kunbus.de>
Cc: Frank Pavlic <f.pavlic@kunbus.de>
Cc: Martin Sperl <kernel@martin.sperl.org>
Cc: Noralf Trønnes <noralf@tronnes.org>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      56c17234
  3. 16 Nov, 2018 1 commit
    • Tony Lindgren's avatar
      spi: omap2-mcspi: Add missing suspend and resume calls · 91b9deef
      Tony Lindgren authored 
      I've been wondering still about omap2-mcspi related suspend and resume
flakeyness and looks like we're missing calls to spi_master_suspend()
and spi_master_resume(). Adding those and using pm_runtime_force_suspend()
and pm_runtime_force_resume() makes things work for suspend and resume
and allows us to stop using noirq suspend and resume.

And while at it, let's use SET_SYSTEM_SLEEP_PM_OPS to simplify things
further.
Signed-off-by: default avatarTony Lindgren <tony@atomide.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      91b9deef
  5. 31 Oct, 2018 1 commit
  7. 24 Oct, 2018 1 commit
  9. 21 Oct, 2018 2 commits
  11. 19 Oct, 2018 5 commits
    • Vignesh R's avatar
      spi: omap2-mcspi: Add slave mode support · 89e8b9cb
      Vignesh R authored 
      Add support to use McSPI controller as SPI slave. In slave mode, DMA TX
completion does not mean entire data has been shifted out as data might
still be stuck in FIFO waiting for master to clock the bus. Therefore,
add an IRQ handler for slave mode to know when entire data in FIFO has
been shifted out.
Signed-off-by: default avatarVignesh R <vigneshr@ti.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      89e8b9cb
    • Vignesh R's avatar
      spi: omap2-mcspi: Set FIFO DMA trigger level to word length · b682cffa
      Vignesh R authored 
      McSPI has 32 byte FIFO in Transmit-Receive mode. Current code tries to
configuration FIFO watermark level for DMA trigger to be GCD of transfer
length and max FIFO size which would mean trigger level may be set to 32
for transmit-receive mode if length is aligned. This does not work in
case of SPI slave mode where FIFO always needs to have data ready
whenever master starts the clock. With DMA trigger size of 32 there will
be a small window during slave TX where DMA is still putting data into
FIFO but master would have started clock for next byte, resulting in
shifting out of stale data. Similarly, on Slave RX side there may be RX
FIFO overflow
Fix this by setting FIFO watermark for DMA trigger to word
length. This means DMA is triggered as soon as FIFO has space for word
length bytes and DMA would make sure FIFO is almost always full
therefore improving FIFO occupancy in both master and slave mode.
Signed-off-by: default avatarVignesh R <vigneshr@ti.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      b682cffa
    • Vignesh R's avatar
      spi: omap2-mcspi: Switch to readl_poll_timeout() · 13d515c7
      Vignesh R authored 
      Use standard readl_poll_timeout() macro for polling on status bits.
Signed-off-by: default avatarVignesh R <vigneshr@ti.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      13d515c7
    • Ludovic Barre's avatar
      spi: spi-mem: add stm32 qspi controller · c530cd1d
      Ludovic Barre authored 
      The qspi controller is a specialized communication interface
targeting single, dual or quad SPI Flash memories (NOR/NAND).

It can operate in any of the following modes:
-indirect mode: all the operations are performed using the quadspi
 registers
-read memory-mapped mode: the external Flash memory is mapped to the
 microcontroller address space and is seen by the system as if it was
 an internal memory

tested on:
-NOR: mx66l51235l
-NAND: MT29F2G01ABAGD
Signed-off-by: default avatarLudovic Barre <ludovic.barre@st.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      c530cd1d
    • Ludovic Barre's avatar
      dt-bindings: spi: add stm32 qspi controller · c3e6845e
      Ludovic Barre authored 
      This patch adds the documentation of device tree bindings
for the STM32 QSPI controller. It is a specialized communication
interface targeting single, dual or quad SPI Flash memories (NOR/NAND).
Signed-off-by: default avatarLudovic Barre <ludovic.barre@st.com>
Reviewed-by: default avatarRob Herring <robh@kernel.org>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      c3e6845e
  13. 17 Oct, 2018 2 commits
  15. 12 Oct, 2018 5 commits
  17. 11 Oct, 2018 16 commits
  19. 10 Oct, 2018 4 commits
  21. 08 Oct, 2018 1 commit
    • Nathan Chancellor's avatar
      spi: spi-ep93xx: Use dma_data_direction for ep93xx_spi_dma_{finish,prepare} · a1108c7b
      Nathan Chancellor authored 
      Clang warns when one enumerated type is implicitly converted to another.

drivers/spi/spi-ep93xx.c:342:62: warning: implicit conversion from
enumeration type 'enum dma_transfer_direction' to different enumeration
type 'enum dma_data_direction' [-Wenum-conversion]
        nents = dma_map_sg(chan->device->dev, sgt->sgl, sgt->nents, dir);
                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
./include/linux/dma-mapping.h:428:58: note: expanded from macro
'dma_map_sg'
#define dma_map_sg(d, s, n, r) dma_map_sg_attrs(d, s, n, r, 0)
                               ~~~~~~~~~~~~~~~~          ^
drivers/spi/spi-ep93xx.c:348:57: warning: implicit conversion from
enumeration type 'enum dma_transfer_direction' to different enumeration
type 'enum dma_data_direction' [-Wenum-conversion]
                dma_unmap_sg(chan->device->dev, sgt->sgl, sgt->nents, dir);
                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
./include/linux/dma-mapping.h:429:62: note: expanded from macro
'dma_unmap_sg'
#define dma_unmap_sg(d, s, n, r) dma_unmap_sg_attrs(d, s, n, r, 0)
                                 ~~~~~~~~~~~~~~~~~~          ^
drivers/spi/spi-ep93xx.c:377:56: warning: implicit conversion from
enumeration type 'enum dma_transfer_direction' to different enumeration
type 'enum dma_data_direction' [-Wenum-conversion]
        dma_unmap_sg(chan->device->dev, sgt->sgl, sgt->nents, dir);
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
./include/linux/dma-mapping.h:429:62: note: expanded from macro
'dma_unmap_sg'
#define dma_unmap_sg(d, s, n, r) dma_unmap_sg_attrs(d, s, n, r, 0)
                                 ~~~~~~~~~~~~~~~~~~          ^
3 warnings generated.

dma_{,un}map_sg expect an enum of type dma_data_direction but this
driver uses dma_transfer_direction for everything. Convert the driver to
use dma_data_direction for these two functions.

There are two places that strictly require an enum of type
dma_transfer_direction: the direction member in struct dma_slave_config
and the direction parameter in dmaengine_prep_slave_sg. To avoid using
an explicit cast, add a simple function, ep93xx_dma_data_to_trans_dir,
to safely map between the two types because they are not 1 to 1 in
meaning.
Signed-off-by: default avatarNathan Chancellor <natechancellor@gmail.com>
Reviewed-by: default avatarNick Desaulniers <ndesaulniers@google.com>
Reviewed-by: default avatarMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      a1108c7b
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7