Hi, On 12/16/25 16:31, Patrice Chotard wrote:
When WATCHDOG_RESET() was replaced by schedule() in commit 29caf9305b6f ("cyclic: Use schedule() instead of WATCHDOG_RESET()") we not only reset the watchdog but also call the cyclic infrastructure which takes time and has impact on read accesses performances.Move schedule() from _stm32_qspi_read_fifo() to _stm32_qspi_poll() and call schedule() only every 1MB chunk of data. Test performed by reading 64MB on sNOR on stm32mp157c-ev1 board: before after ratio Read : 201 KB/s 520KB/s +258% Signed-off-by: Patrice Chotard <[email protected]> --- drivers/spi/stm32_qspi.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/spi/stm32_qspi.c b/drivers/spi/stm32_qspi.c index 3216ec8010e..1778f8c1458 100644 --- a/drivers/spi/stm32_qspi.c +++ b/drivers/spi/stm32_qspi.c @@ -165,7 +165,6 @@ static int _stm32_qspi_wait_cmd(struct stm32_qspi_priv *priv, static void _stm32_qspi_read_fifo(u8 *val, void __iomem *addr) { *val = readb(addr); - schedule(); }static void _stm32_qspi_write_fifo(u8 *val, void __iomem *addr)@@ -200,6 +199,9 @@ static int _stm32_qspi_poll(struct stm32_qspi_priv *priv, }fifo(buf++, &priv->regs->dr);+ + if (!(len % SZ_1M)) + schedule(); }return 0;
Reviewed-by: Patrick Delaunay <[email protected]> Thanks Patrick
