[PATCH 3/3] spi: tegra210-quad: Process small PIO transfers in hard IRQ context

[Vishwaroop A]

On heavily loaded systems, workqueue scheduling delays can exceed
transfer timeouts even for high-priority queues, causing false timeouts
for latency-sensitive devices like TPM despite hardware completing in
microseconds.

Process small PIO transfers (≤256 bytes) directly in hard IRQ context
instead of deferring to workqueue. This reduces completion latency
from 1000ms+ to microseconds and matches the pattern used by other
SPI drivers.

The 256-byte threshold (FIFO depth) ensures small transfers for devices
like TPMs use the fast path, while larger transfers continue using
workqueue.

Signed-off-by: Vishwaroop A <va@xxxxxxxxxx>
---
 drivers/spi/spi-tegra210-quad.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/spi/spi-tegra210-quad.c b/drivers/spi/spi-tegra210-quad.c
index 6148267a51cd..435e14d80bfa 100644
--- a/drivers/spi/spi-tegra210-quad.c
+++ b/drivers/spi/spi-tegra210-quad.c
@@ -1666,6 +1666,15 @@ static irqreturn_t tegra_qspi_isr(int irq, void *context_data)
 
 	spin_unlock(&tqspi->lock);
 
+	/*
+	 * For small PIO transfers (e.g., TPM), process directly in hard IRQ
+	 * context unless there was a FIFO error. Error recovery calls
+	 * device_reset() which can sleep, so must be deferred to workqueue.
+	 */
+	if (!tqspi->is_curr_dma_xfer && tqspi->curr_dma_words <= QSPI_FIFO_DEPTH &&
+	    !tqspi->tx_status && !tqspi->rx_status)
+		return handle_cpu_based_xfer(tqspi);
+
 	queue_work(tqspi->wq, &tqspi->irq_work);
 
 	return IRQ_HANDLED;
-- 
2.17.1