linux-kernel - [PATCH v2 18/19] mtd: spinand: winbond: Add octal support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20250403-winbond-6-14-rc1-octal-v2-18-7846bd88fe83@bootlin.com>
Date: Thu, 03 Apr 2025 11:19:28 +0200
From: Miquel Raynal <miquel.raynal@...tlin.com>
To: Richard Weinberger <richard@....at>, 
 Vignesh Raghavendra <vigneshr@...com>, Santhosh Kumar K <s-k6@...com>
Cc: Tudor Ambarus <tudor.ambarus@...aro.org>, 
 Pratyush Yadav <pratyush@...nel.org>, Michael Walle <michael@...le.cc>, 
 Thomas Petazzoni <thomas.petazzoni@...tlin.com>, 
 Steam Lin <stlin2@...bond.com>, linux-mtd@...ts.infradead.org, 
 linux-kernel@...r.kernel.org, Miquel Raynal <miquel.raynal@...tlin.com>
Subject: [PATCH v2 18/19] mtd: spinand: winbond: Add octal support

Describe the octal "read from cache" (1S_1S_8S, 1S_8S_8S and
1S_1D_8D_OP) and "program load" (1S_1S_8S and 1S_8S_8S) operations
supported by the W35N01JW chip.

There is no observable improvement between the two octal SDR modes,
which anyway offered a great read speed improvement compared to single
mode.

flash_speed test results ran on the TI AM62A7 LP SK using a 25MHz SPI bus:

* Single mode:
eraseblock  read speed is  2346 KiB/s
page        read speed is  2342 KiB/s
eraseblock write speed is  2028 KiB/s
page       write speed is  2028 KiB/s

* Octal modes:
eraseblock  read speed is 10801 KiB/s
page        read speed is 10711 KiB/s
eraseblock write speed is  7335 KiB/s
page       write speed is  7293 KiB/s

However, testing on the cadence qspi controller driver did not allow
mixed modes (where SDR and DTR cycles are mixed in the same operation),
thus 1S-1D-8D mode could not be tested and is provided for reference (it
is anyway skipped on platforms not supporting this). It did not allow
easy "update cache" tests either, so these macros are untested and given
as-is.

Reviewed-by: Tudor Ambarus <tudor.ambarus@...aro.org>
Signed-off-by: Miquel Raynal <miquel.raynal@...tlin.com>
---
 drivers/mtd/nand/spi/winbond.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/mtd/nand/spi/winbond.c b/drivers/mtd/nand/spi/winbond.c
index 9b0369a077efadac8bd6be632f15b822d42a7227..c263c9b5affe79a14466bc0e2ef2da53ecfd48cf 100644
--- a/drivers/mtd/nand/spi/winbond.c
+++ b/drivers/mtd/nand/spi/winbond.c
@@ -24,13 +24,19 @@
  */
 
 static SPINAND_OP_VARIANTS(read_cache_octal_variants,
+		SPINAND_PAGE_READ_FROM_CACHE_1S_1D_8D_OP(0, 2, NULL, 0, 105 * HZ_PER_MHZ),
+		SPINAND_PAGE_READ_FROM_CACHE_1S_8S_8S_OP(0, 16, NULL, 0, 86 * HZ_PER_MHZ),
+		SPINAND_PAGE_READ_FROM_CACHE_1S_1S_8S_OP(0, 1, NULL, 0, 133 * HZ_PER_MHZ),
 		SPINAND_PAGE_READ_FROM_CACHE_FAST_1S_1S_1S_OP(0, 1, NULL, 0),
 		SPINAND_PAGE_READ_FROM_CACHE_1S_1S_1S_OP(0, 1, NULL, 0));
 
 static SPINAND_OP_VARIANTS(write_cache_octal_variants,
+		SPINAND_PROG_LOAD_1S_8S_8S_OP(true, 0, NULL, 0),
+		SPINAND_PROG_LOAD_1S_1S_8S_OP(0, NULL, 0),
 		SPINAND_PROG_LOAD_1S_1S_1S_OP(true, 0, NULL, 0));
 
 static SPINAND_OP_VARIANTS(update_cache_octal_variants,
+		SPINAND_PROG_LOAD_1S_8S_8S_OP(false, 0, NULL, 0),
 		SPINAND_PROG_LOAD_1S_1S_1S_OP(false, 0, NULL, 0));
 
 static SPINAND_OP_VARIANTS(read_cache_dual_quad_dtr_variants,

-- 
2.48.1