92de28b751
Instead of reverting whole commit it's enough to just revert a single line change. It seems the real problem with the regressing commit was a bump of read chunk size. Switching back to 256 B chunks is enough to fix the problem/regression. Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
42 lines
1.4 KiB
Diff
42 lines
1.4 KiB
Diff
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <rafal@milecki.pl>
|
|
Date: Thu, 11 Oct 2018 09:07:31 +0200
|
|
Subject: [PATCH FIX] spi: bcm-qspi: switch back to reading flash using smaller
|
|
chunks
|
|
MIME-Version: 1.0
|
|
Content-Type: text/plain; charset=UTF-8
|
|
Content-Transfer-Encoding: 8bit
|
|
|
|
Fixing/optimizing bcm_qspi_bspi_read() performance introduced two
|
|
changes:
|
|
1) It added a loop to read all requested data using multiple BSPI ops.
|
|
2) It bumped max size of a single BSPI block request from 256 to 512 B.
|
|
|
|
The later change resulted in occasional BSPI timeouts causing a
|
|
regression.
|
|
|
|
For some unknown reason hardware doesn't always handle reads as expected
|
|
when using 512 B chunks. In such cases it may happen that BSPI returns
|
|
amount of requested bytes without the last 1-3 ones. It provides the
|
|
remaining bytes later but doesn't raise an interrupt until another LR
|
|
start.
|
|
|
|
Switching back to 256 B reads fixes that problem and regression.
|
|
|
|
Fixes: 345309fa7c0c ("spi: bcm-qspi: Fix bcm_qspi_bspi_read() performance")
|
|
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
|
|
Cc: stable@vger.kernel.org # 4.11+
|
|
---
|
|
drivers/spi/spi-bcm-qspi.c | 2 +-
|
|
1 file changed, 1 insertion(+), 1 deletion(-)
|
|
|
|
--- a/drivers/spi/spi-bcm-qspi.c
|
|
+++ b/drivers/spi/spi-bcm-qspi.c
|
|
@@ -88,7 +88,7 @@
|
|
#define BSPI_BPP_MODE_SELECT_MASK BIT(8)
|
|
#define BSPI_BPP_ADDR_SELECT_MASK BIT(16)
|
|
|
|
-#define BSPI_READ_LENGTH 512
|
|
+#define BSPI_READ_LENGTH 256
|
|
|
|
/* MSPI register offsets */
|
|
#define MSPI_SPCR0_LSB 0x000
|