[U-Boot] [PATCH] Decreases code size of the nand_spl
From b59f1e5b0bc3684615756c12fd5c5f9fcaa4c812 Mon Sep 17 00:00:00 2001
From: Alex Waterman awaterman@dawning.com Date: Tue, 3 May 2011 15:00:23 -0400 Subject: [PATCH] Decreases code size of the nand_spl
The canyonland boards nand_spl size is just under the maximum 4KByte size. This patch decreases the size of the nand_spl to make a previous commit - commit 65a9db7be0868be91ba81b9b5bf821de82e6d9b0 - fit in the nand_spl.
Signed-off-by: Alex Waterman awaterman@dawning.com --- This patch uses a function pointer declared as a local variable; checkpatch didn't mind but this seems like it could be (stylistically) a very bad idea. Any thoughts?
nand_spl/nand_boot.c | 18 ++++++++++-------- 1 files changed, 10 insertions(+), 8 deletions(-)
diff --git a/nand_spl/nand_boot.c b/nand_spl/nand_boot.c index 4a96878..79048b3 100644 --- a/nand_spl/nand_boot.c +++ b/nand_spl/nand_boot.c @@ -77,6 +77,8 @@ static int nand_command(struct mtd_info *mtd, int block, int page, int offs, u8 { struct nand_chip *this = mtd->priv; int page_addr = page + block * CONFIG_SYS_NAND_PAGE_COUNT; + void (*hwctrl)(struct mtd_info *mtd, int cmd, + unsigned int ctrl) = this->cmd_ctrl;
if (this->dev_ready) while (!this->dev_ready(mtd)) @@ -95,25 +97,25 @@ static int nand_command(struct mtd_info *mtd, int block, int page, int offs, u8 offs >>= 1;
/* Begin command latch cycle */ - this->cmd_ctrl(mtd, cmd, NAND_CTRL_CLE | NAND_CTRL_CHANGE); + hwctrl(mtd, cmd, NAND_CTRL_CLE | NAND_CTRL_CHANGE); /* Set ALE and clear CLE to start address cycle */ /* Column address */ - this->cmd_ctrl(mtd, offs & 0xff, + hwctrl(mtd, offs & 0xff, NAND_CTRL_ALE | NAND_CTRL_CHANGE); /* A[7:0] */ - this->cmd_ctrl(mtd, (offs >> 8) & 0xff, NAND_CTRL_ALE); /* A[11:9] */ + hwctrl(mtd, (offs >> 8) & 0xff, NAND_CTRL_ALE); /* A[11:9] */ /* Row address */ - this->cmd_ctrl(mtd, (page_addr & 0xff), NAND_CTRL_ALE); /* A[19:12] */ - this->cmd_ctrl(mtd, ((page_addr >> 8) & 0xff), + hwctrl(mtd, (page_addr & 0xff), NAND_CTRL_ALE); /* A[19:12] */ + hwctrl(mtd, ((page_addr >> 8) & 0xff), NAND_CTRL_ALE); /* A[27:20] */ #ifdef CONFIG_SYS_NAND_5_ADDR_CYCLE /* One more address cycle for devices > 128MiB */ - this->cmd_ctrl(mtd, (page_addr >> 16) & 0x0f, + hwctrl(mtd, (page_addr >> 16) & 0x0f, NAND_CTRL_ALE); /* A[31:28] */ #endif /* Latch in address */ - this->cmd_ctrl(mtd, NAND_CMD_READSTART, + hwctrl(mtd, NAND_CMD_READSTART, NAND_CTRL_CLE | NAND_CTRL_CHANGE); - this->cmd_ctrl(mtd, NAND_CMD_NONE, NAND_NCE | NAND_CTRL_CHANGE); + hwctrl(mtd, NAND_CMD_NONE, NAND_NCE | NAND_CTRL_CHANGE);
/* * Wait a while for the data to be ready
Hi Alex,
On Wednesday 04 May 2011 15:10:15 Alex Waterman wrote:
From b59f1e5b0bc3684615756c12fd5c5f9fcaa4c812 Mon Sep 17 00:00:00 2001 From: Alex Waterman awaterman@dawning.com Date: Tue, 3 May 2011 15:00:23 -0400 Subject: [PATCH] Decreases code size of the nand_spl
The canyonland boards nand_spl size is just under the maximum 4KByte size. This patch decreases the size of the nand_spl to make a previous commit - commit 65a9db7be0868be91ba81b9b5bf821de82e6d9b0 - fit in the nand_spl.
Signed-off-by: Alex Waterman awaterman@dawning.com
This patch uses a function pointer declared as a local variable; checkpatch didn't mind but this seems like it could be (stylistically) a very bad idea. Any thoughts?
Please see my patches sent a few hours ago:
"nand_spl: nand_boot.c: Init nand_chip.options to 0" "nand_spl: nand_boot.c: Remove CONFIG_SYS_NAND_READ_DELAY"
The 2nd patch fixes the size problem as well. So no need for your patch any more.
But thanks anyway.
BTW: What platform/SoC are you using?
Cheers, Stefan
-- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-0 Fax: (+49)-8142-66989-80 Email: office@denx.de
Stefan,
The 2nd patch fixes the size problem as well. So no need for your patch any more.
Ahh, I saw the first but missed the second I guess.
BTW: What platform/SoC are you using?
I am using a custom board based on the AMCC sequoia board. At some point I plan to try and mainline our port for that board if possible. But I have more work to go on that.
- Alex
On Wed, 4 May 2011 15:47:27 +0200 Stefan Roese sr@denx.de wrote:
Hi Alex,
On Wednesday 04 May 2011 15:10:15 Alex Waterman wrote:
From b59f1e5b0bc3684615756c12fd5c5f9fcaa4c812 Mon Sep 17 00:00:00 2001 From: Alex Waterman awaterman@dawning.com Date: Tue, 3 May 2011 15:00:23 -0400 Subject: [PATCH] Decreases code size of the nand_spl
The canyonland boards nand_spl size is just under the maximum 4KByte size. This patch decreases the size of the nand_spl to make a previous commit - commit 65a9db7be0868be91ba81b9b5bf821de82e6d9b0 - fit in the nand_spl.
Signed-off-by: Alex Waterman awaterman@dawning.com
This patch uses a function pointer declared as a local variable; checkpatch didn't mind but this seems like it could be (stylistically) a very bad idea. Any thoughts?
I don't see what's wrong with a local function pointer.
Please see my patches sent a few hours ago:
"nand_spl: nand_boot.c: Init nand_chip.options to 0" "nand_spl: nand_boot.c: Remove CONFIG_SYS_NAND_READ_DELAY"
The 2nd patch fixes the size problem as well. So no need for your patch any more.
Or we could apply both and save even more space, delaying the next time we run into trouble. :-)
-Scott
On Wed, 4 May 2011 14:24:15 -0400 Alex Waterman awaterman@dawning.com wrote:
Scott,
Or we could apply both and save even more space, delaying the next time we run into trouble. :-)
Since the spl is so limited in space, would it be worth making every function in the spl use a local function pointer instead of lots of dereferences?
It'll only help when the same pointer is called more than once in the function, but I'd be fine with doing it in those cases.
-Scott
This patch decreases the code size of the nand_spl by turning multiple function pointer dereferences in a single function into a single local function pointer.
Signed-off-by: Alex Waterman awaterman@dawning.com Cc: Scott Wood scottwood@freescale.com Cc: Stefan Roese sr@denx.de --- This works on my local board but I haven't tested it on anything else since I do not have access to any other hardware. I did make sure the canyonlands board still builds.
nand_spl/nand_boot.c | 50 ++++++++++++++++++++++++++++---------------------- 1 files changed, 28 insertions(+), 22 deletions(-)
diff --git a/nand_spl/nand_boot.c b/nand_spl/nand_boot.c index 4a96878..dd7f3b2 100644 --- a/nand_spl/nand_boot.c +++ b/nand_spl/nand_boot.c @@ -35,34 +35,37 @@ static int nand_command(struct mtd_info *mtd, int block, int page, int offs, u8 { struct nand_chip *this = mtd->priv; int page_addr = page + block * CONFIG_SYS_NAND_PAGE_COUNT; + void (*cmd_ctrl)(struct mtd_info *mtd, int cmd, + unsigned int ctrl) = this->cmd_ctrl; + int (*dev_ready)(struct mtd_info *mtd) = this->dev_ready;
- if (this->dev_ready) - while (!this->dev_ready(mtd)) + if (dev_ready != NULL) + while (!dev_ready(mtd)) ; else CONFIG_SYS_NAND_READ_DELAY;
/* Begin command latch cycle */ - this->cmd_ctrl(mtd, cmd, NAND_CTRL_CLE | NAND_CTRL_CHANGE); + cmd_ctrl(mtd, cmd, NAND_CTRL_CLE | NAND_CTRL_CHANGE); /* Set ALE and clear CLE to start address cycle */ /* Column address */ - this->cmd_ctrl(mtd, offs, NAND_CTRL_ALE | NAND_CTRL_CHANGE); - this->cmd_ctrl(mtd, page_addr & 0xff, NAND_CTRL_ALE); /* A[16:9] */ - this->cmd_ctrl(mtd, (page_addr >> 8) & 0xff, + cmd_ctrl(mtd, offs, NAND_CTRL_ALE | NAND_CTRL_CHANGE); + cmd_ctrl(mtd, page_addr & 0xff, NAND_CTRL_ALE); /* A[16:9] */ + cmd_ctrl(mtd, (page_addr >> 8) & 0xff, NAND_CTRL_ALE); /* A[24:17] */ #ifdef CONFIG_SYS_NAND_4_ADDR_CYCLE /* One more address cycle for devices > 32MiB */ - this->cmd_ctrl(mtd, (page_addr >> 16) & 0x0f, + cmd_ctrl(mtd, (page_addr >> 16) & 0x0f, NAND_CTRL_ALE); /* A[28:25] */ #endif /* Latch in address */ - this->cmd_ctrl(mtd, NAND_CMD_NONE, NAND_NCE | NAND_CTRL_CHANGE); + cmd_ctrl(mtd, NAND_CMD_NONE, NAND_NCE | NAND_CTRL_CHANGE);
/* * Wait a while for the data to be ready */ - if (this->dev_ready) - while (!this->dev_ready(mtd)) + if (dev_ready != NULL) + while (!dev_ready(mtd)) ; else CONFIG_SYS_NAND_READ_DELAY; @@ -77,9 +80,12 @@ static int nand_command(struct mtd_info *mtd, int block, int page, int offs, u8 { struct nand_chip *this = mtd->priv; int page_addr = page + block * CONFIG_SYS_NAND_PAGE_COUNT; + void (*cmd_ctrl)(struct mtd_info *mtd, int cmd, + unsigned int ctrl) = this->cmd_ctrl; + int (*dev_ready)(struct mtd_info *mtd) = this->dev_ready;
- if (this->dev_ready) - while (!this->dev_ready(mtd)) + if (dev_ready != NULL) + while (!dev_ready(mtd)) ; else CONFIG_SYS_NAND_READ_DELAY; @@ -95,31 +101,31 @@ static int nand_command(struct mtd_info *mtd, int block, int page, int offs, u8 offs >>= 1;
/* Begin command latch cycle */ - this->cmd_ctrl(mtd, cmd, NAND_CTRL_CLE | NAND_CTRL_CHANGE); + cmd_ctrl(mtd, cmd, NAND_CTRL_CLE | NAND_CTRL_CHANGE); /* Set ALE and clear CLE to start address cycle */ /* Column address */ - this->cmd_ctrl(mtd, offs & 0xff, + cmd_ctrl(mtd, offs & 0xff, NAND_CTRL_ALE | NAND_CTRL_CHANGE); /* A[7:0] */ - this->cmd_ctrl(mtd, (offs >> 8) & 0xff, NAND_CTRL_ALE); /* A[11:9] */ + cmd_ctrl(mtd, (offs >> 8) & 0xff, NAND_CTRL_ALE); /* A[11:9] */ /* Row address */ - this->cmd_ctrl(mtd, (page_addr & 0xff), NAND_CTRL_ALE); /* A[19:12] */ - this->cmd_ctrl(mtd, ((page_addr >> 8) & 0xff), + cmd_ctrl(mtd, (page_addr & 0xff), NAND_CTRL_ALE); /* A[19:12] */ + cmd_ctrl(mtd, ((page_addr >> 8) & 0xff), NAND_CTRL_ALE); /* A[27:20] */ #ifdef CONFIG_SYS_NAND_5_ADDR_CYCLE /* One more address cycle for devices > 128MiB */ - this->cmd_ctrl(mtd, (page_addr >> 16) & 0x0f, + cmd_ctrl(mtd, (page_addr >> 16) & 0x0f, NAND_CTRL_ALE); /* A[31:28] */ #endif /* Latch in address */ - this->cmd_ctrl(mtd, NAND_CMD_READSTART, + cmd_ctrl(mtd, NAND_CMD_READSTART, NAND_CTRL_CLE | NAND_CTRL_CHANGE); - this->cmd_ctrl(mtd, NAND_CMD_NONE, NAND_NCE | NAND_CTRL_CHANGE); + cmd_ctrl(mtd, NAND_CMD_NONE, NAND_NCE | NAND_CTRL_CHANGE);
/* * Wait a while for the data to be ready */ - if (this->dev_ready) - while (!this->dev_ready(mtd)) + if (dev_ready != NULL) + while (!dev_ready(mtd)) ; else CONFIG_SYS_NAND_READ_DELAY;
Dear Scott,
In message 4DC1A7F1.7000001@dawning.com Alex Waterman wrote:
This patch decreases the code size of the nand_spl by turning multiple function pointer dereferences in a single function into a single local function pointer.
Signed-off-by: Alex Waterman awaterman@dawning.com Cc: Scott Wood scottwood@freescale.com Cc: Stefan Roese sr@denx.de
This works on my local board but I haven't tested it on anything else since I do not have access to any other hardware. I did make sure the canyonlands board still builds.
nand_spl/nand_boot.c | 50 ++++++++++++++++++++++++++++---------------------- 1 files changed, 28 insertions(+), 22 deletions(-)
Can this patch go in, please? I would like to see the build problems fixed.
Stefan, eventually you want to add your Tested-by: and/or Acked-by: ?
Best regards,
Wolfgang Denk
On Thu, 12 May 2011 23:49:40 +0200 Wolfgang Denk wd@denx.de wrote:
Dear Scott,
In message 4DC1A7F1.7000001@dawning.com Alex Waterman wrote:
This patch decreases the code size of the nand_spl by turning multiple function pointer dereferences in a single function into a single local function pointer.
Signed-off-by: Alex Waterman awaterman@dawning.com Cc: Scott Wood scottwood@freescale.com Cc: Stefan Roese sr@denx.de
This works on my local board but I haven't tested it on anything else since I do not have access to any other hardware. I did make sure the canyonlands board still builds.
nand_spl/nand_boot.c | 50 ++++++++++++++++++++++++++++---------------------- 1 files changed, 28 insertions(+), 22 deletions(-)
Can this patch go in, please? I would like to see the build problems fixed.
Yes, sorry for the delay -- I'll try to get to it tomorrow.
-Scott
Dear Scott,
In message 20110512165759.0692384d@schlenkerla.am.freescale.net you wrote:
Can this patch go in, please? I would like to see the build problems fixed.
Yes, sorry for the delay -- I'll try to get to it tomorrow.
Thanks. Guess you guys are as busy as evryone else ;-)
Best regards,
Wolfgang Denk
On Thursday 12 May 2011 23:49:40 Wolfgang Denk wrote:
This patch decreases the code size of the nand_spl by turning multiple function pointer dereferences in a single function into a single local function pointer.
Signed-off-by: Alex Waterman awaterman@dawning.com Cc: Scott Wood scottwood@freescale.com Cc: Stefan Roese sr@denx.de
This works on my local board but I haven't tested it on anything else since I do not have access to any other hardware. I did make sure the canyonlands board still builds.
nand_spl/nand_boot.c | 50 ++++++++++++++++++++++++++++---------------------- 1 files changed, 28 insertions(+), 22 deletions(-)
Can this patch go in, please? I would like to see the build problems fixed.
Stefan, eventually you want to add your Tested-by: and/or Acked-by: ?
Sure:
Acked-by: Stefan Roese sr@denx.de
Thanks, Stefan
-- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-0 Fax: (+49)-8142-66989-80 Email: office@denx.de
On Wed, May 04, 2011 at 09:10:15AM -0400, Alex Waterman wrote:
From b59f1e5b0bc3684615756c12fd5c5f9fcaa4c812 Mon Sep 17 00:00:00 2001 From: Alex Waterman awaterman@dawning.com Date: Tue, 3 May 2011 15:00:23 -0400 Subject: [PATCH] Decreases code size of the nand_spl
The canyonland boards nand_spl size is just under the maximum 4KByte size. This patch decreases the size of the nand_spl to make a previous commit - commit 65a9db7be0868be91ba81b9b5bf821de82e6d9b0 - fit in the nand_spl.
Signed-off-by: Alex Waterman awaterman@dawning.com
This patch uses a function pointer declared as a local variable; checkpatch didn't mind but this seems like it could be (stylistically) a very bad idea. Any thoughts?
nand_spl/nand_boot.c | 18 ++++++++++-------- 1 files changed, 10 insertions(+), 8 deletions(-)
Applied to u-boot-nand-flash
-Scott
participants (4)
-
Alex Waterman -
Scott Wood -
Stefan Roese -
Wolfgang Denk