From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sourav Poddar Date: Mon, 6 Jan 2014 15:51:50 +0530 Subject: [U-Boot] [PATCH v6 08/12] sf: Discover read dummy_cycles In-Reply-To: References: <1388847888-19474-1-git-send-email-jaganna@xilinx.com> <4c399e81-548a-4121-a938-6df85186f5b4@DB9EHSMHS029.ehs.local> <52CA557F.3070402@ti.com> Message-ID: <52CA83BE.7030102@ti.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Monday 06 January 2014 03:48 PM, Jagan Teki wrote: > Hi Sourav, > > On Mon, Jan 6, 2014 at 12:34 PM, Sourav Poddar wrote: >> On Saturday 04 January 2014 08:34 PM, Jagannadha Sutradharudu Teki wrote: >>> Discovered the read dummy_cycles based on the configured >>> read command. >>> >>> Signed-off-by: Jagannadha Sutradharudu Teki >>> --- >>> drivers/mtd/spi/sf_internal.h | 2 ++ >>> drivers/mtd/spi/sf_ops.c | 10 ++++++---- >>> drivers/mtd/spi/sf_probe.c | 12 ++++++++++++ >>> include/spi_flash.h | 2 ++ >>> 4 files changed, 22 insertions(+), 4 deletions(-) >>> >>> diff --git a/drivers/mtd/spi/sf_internal.h b/drivers/mtd/spi/sf_internal.h >>> index 7be0292..a9f5a81 100644 >>> --- a/drivers/mtd/spi/sf_internal.h >>> +++ b/drivers/mtd/spi/sf_internal.h >>> @@ -10,6 +10,8 @@ >>> #ifndef _SF_INTERNAL_H_ >>> #define _SF_INTERNAL_H_ >>> >>> +#define SPI_FLASH_3B_ADDR_LEN 3 >>> +#define SPI_FLASH_CMD_LEN (1 + SPI_FLASH_3B_ADDR_LEN) >>> #define SPI_FLASH_16MB_BOUN 0x1000000 >>> >>> /* CFI Manufacture ID's */ >>> diff --git a/drivers/mtd/spi/sf_ops.c b/drivers/mtd/spi/sf_ops.c >>> index 827f719..dda75b1 100644 >>> --- a/drivers/mtd/spi/sf_ops.c >>> +++ b/drivers/mtd/spi/sf_ops.c >>> @@ -9,6 +9,7 @@ >>> */ >>> >>> #include >>> +#include >>> #include >>> #include >>> #include >>> @@ -216,7 +217,7 @@ int spi_flash_write_common(struct spi_flash *flash, >>> const u8 *cmd, >>> int spi_flash_cmd_erase_ops(struct spi_flash *flash, u32 offset, size_t >>> len) >>> { >>> u32 erase_size; >>> - u8 cmd[4]; >>> + u8 cmd[SPI_FLASH_CMD_LEN]; >>> int ret = -1; >>> >>> erase_size = flash->erase_size; >>> @@ -255,7 +256,7 @@ int spi_flash_cmd_write_ops(struct spi_flash *flash, >>> u32 offset, >>> { >>> unsigned long byte_addr, page_size; >>> size_t chunk_len, actual; >>> - u8 cmd[4]; >>> + u8 cmd[SPI_FLASH_CMD_LEN]; >>> int ret = -1; >>> >>> page_size = flash->page_size; >>> @@ -317,7 +318,7 @@ int spi_flash_read_common(struct spi_flash *flash, >>> const u8 *cmd, >>> int spi_flash_cmd_read_ops(struct spi_flash *flash, u32 offset, >>> size_t len, void *data) >>> { >>> - u8 cmd[5], bank_sel = 0; >>> + u8 *cmd, bank_sel = 0; >>> u32 remain_len, read_len; >>> int ret = -1; >>> >>> @@ -335,8 +336,9 @@ int spi_flash_cmd_read_ops(struct spi_flash *flash, >>> u32 offset, >>> return 0; >>> } >>> >>> + cmd = malloc(SPI_FLASH_CMD_LEN + flash->dummy_cycles); >>> + memset(cmd, 0, SPI_FLASH_CMD_LEN + flash->dummy_cycles); >>> cmd[0] = flash->read_cmd; >>> - cmd[4] = 0x00; >>> >>> while (len) { >>> #ifdef CONFIG_SPI_FLASH_BAR >>> diff --git a/drivers/mtd/spi/sf_probe.c b/drivers/mtd/spi/sf_probe.c >>> index a049e72..b070adc 100644 >>> --- a/drivers/mtd/spi/sf_probe.c >>> +++ b/drivers/mtd/spi/sf_probe.c >>> @@ -140,6 +140,18 @@ static struct spi_flash >>> *spi_flash_validate_params(struct spi_slave *spi, >>> } >>> } >>> >>> + /* Read dummy cycles */ >>> + switch (flash->read_cmd) { >>> + case CMD_READ_QUAD_IO_FAST: >>> + flash->dummy_cycles = 2; >>> + break; >>> + case CMD_READ_ARRAY_SLOW: >>> + flash->dummy_cycles = 0; >>> + break; >>> + default: >>> + flash->dummy_cycles = 1; >>> + } >>> + >> what about dummy cycles for dual i/o(0xbb), it has 4 clock bit dummy >> cycles(macronix), so >> by your code you keep it to 1(8 bit) dummy cycle. ? > Yes- Actually i miss used the names here. > These are dummy_byte_count > > Ex: Except fast reads (array, dual and quad fast) > dummy_byte_count = dummy_cycles * no.of lines > dual i/o (4 dummy_cycles) - 4 * 2 = 1 dummy_byte > quad i/o (4 dummy_cycles) - 4 * 4 = 2 dummy_byte > is this calculation documented in any of the flash sheet? > fast, dual and quad fast have 8 dummy_cycles means 1 byte dummy the > reason why I am not including > no.of lines here is irrespective of fast(1 line), dual(2 lines) and > quad(4 lines) only data can travel among the difference lines > but the cmd,inst, dummy can travel only on single line. this can be > opposite in i/o commands. > > Hope you understand - I just said based on my knowledge. > I will change the dummy_cycles into dummy_bytes - nevertheless everything same. >