From: Vijaya Krishna Nivarthi <vijaya.nivarthi@oss.qualcomm.com>

The maximum size for a DMA data descriptor is 64KB-1 because the size
field in HW is 16 bits wide. For this reason, transfers fail at 64KB
and beyond.

Lower max_dma_len to 60KB so larger transfers are split into multiple
DMA blocks and do not hit the failing 64KB boundary. 60KB is chosen as
a safe round number below the 64KB-1 hardware limit while satisfying
alignment requirements.

Tested on x1e80100 (Hamoa) with SPI-NOR flash (/dev/mtd0):

Without patch:
  dd if=/dev/mtd0 of=/tmp/spi_dump.bin bs=32768 count=2  # works
  dd if=/dev/mtd0 of=/tmp/spi_dump.bin bs=65536 count=1  # fails

With patch:
  dd if=/dev/mtd0 of=/tmp/spi_dump.bin bs=65536 count=1  # works

Fixes: b5762d95607e ("spi: spi-qcom-qspi: Add DMA mode support")
Cc: stable@vger.kernel.org
Signed-off-by: Vijaya Krishna Nivarthi <vijaya.nivarthi@oss.qualcomm.com>
---
 drivers/spi/spi-qcom-qspi.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/spi/spi-qcom-qspi.c b/drivers/spi/spi-qcom-qspi.c
index caf55a6f70b3..e9fdefa9d65d 100644
--- a/drivers/spi/spi-qcom-qspi.c
+++ b/drivers/spi/spi-qcom-qspi.c
@@ -775,7 +775,8 @@ static int qcom_qspi_probe(struct platform_device *pdev)
 		return dev_err_probe(dev, ret, "could not set DMA mask\n");
 
 	host->max_speed_hz = 300000000;
-	host->max_dma_len = 65536; /* as per HPG */
+	/* as per HPG, it is 64KB, limit to 60KB to avoid boundary condition failures */
+	host->max_dma_len = 0xf000;
 	host->dma_alignment = QSPI_ALIGN_REQ;
 	host->num_chipselect = QSPI_NUM_CS;
 	host->bus_num = -1;
-- 
2.34.1

On Wed, 22 Jul 2026 14:53:58 +0530, Vijaya Krishna Nivarthi wrote:
> spi: qcom-qspi: Correct max DMA length to avoid 64K boundary failure

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-7.2

Thanks!

[1/1] spi: qcom-qspi: Correct max DMA length to avoid 64K boundary failure
      https://git.kernel.org/broonie/spi/c/90ef2f2961c2

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark