All of lore.kernel.org
 help / color / mirror / Atom feed
* [U-Boot] [PATCH] arm: mvebu: a38x: Weed out floating point use
@ 2016-04-30 12:45 Marek Vasut
  2016-05-19 23:46 ` Marek Vasut
  2016-05-20  9:04 ` Stefan Roese
  0 siblings, 2 replies; 3+ messages in thread
From: Marek Vasut @ 2016-04-30 12:45 UTC (permalink / raw)
  To: u-boot

For reason unknown, recently, the DDR init code writers are really fond
of hiding some small floating point operating deep in their creations.
This patch removes one from the Marvell A38x code.

Instead of returning size of chip as float from ddr3_get_device_size()
in GiB units, return it as int in MiB units. Since this would interfere
with the huge switch code in ddr3_calc_mem_cs_size(), rework the code
to match the change.

Before this patch, the cs_mem_size variable could have these values:
 ( { 16, 32 } x { 8, 16 } x { 0.01, 0.5, 1, 2, 4, 8 } ) / 8 =
   { 0.000000, 0.001250, 0.002500, 0.005000, 0.062500, 0.125000,
     0.250000, 0.500000, 1.000000, 2.000000, 4.000000, }
The switch code checked for a subset of the resulting RAM sizes, which
is in range 128 MiB ... 2048 MiB.

With this patch, the cs_mem_size variable can have these values:
 ( { 16, 32 } x { 8, 16 } x { 0, 512, 1024, 2048, 4096, 8192 } ) / 8 =
   { 0, 64, 128, 256, 512, 1024, 2048, 4096 }
To retain previous behavior, filter out 0 MiB (invalid size), 64 MiB
and 4096 MiB options.

Removing the floating point stuff also saves 1.5k from text segment:
  clearfog       :  spl/u-boot-spl:all -1592  spl/u-boot-spl:text -1592

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dirk Eibach <dirk.eibach@gdsys.cc>
Cc: Stefan Roese <sr@denx.de>
---
 drivers/ddr/marvell/a38x/ddr3_init.c | 29 ++++++++++-------------------
 1 file changed, 10 insertions(+), 19 deletions(-)

diff --git a/drivers/ddr/marvell/a38x/ddr3_init.c b/drivers/ddr/marvell/a38x/ddr3_init.c
index ee05f57..55baad4 100644
--- a/drivers/ddr/marvell/a38x/ddr3_init.c
+++ b/drivers/ddr/marvell/a38x/ddr3_init.c
@@ -678,7 +678,7 @@ u32 ddr3_get_device_width(u32 cs)
 	return (device_width == 0) ? 8 : 16;
 }
 
-float ddr3_get_device_size(u32 cs)
+static int ddr3_get_device_size(u32 cs)
 {
 	u32 device_size_low, device_size_high, device_size;
 	u32 data, cs_low_offset, cs_high_offset;
@@ -695,15 +695,15 @@ float ddr3_get_device_size(u32 cs)
 
 	switch (device_size) {
 	case 0:
-		return 2;
+		return 2048;
 	case 2:
-		return 0.5;
+		return 512;
 	case 3:
-		return 1;
+		return 1024;
 	case 4:
-		return 4;
+		return 4096;
 	case 5:
-		return 8;
+		return 8192;
 	case 1:
 	default:
 		DEBUG_INIT_C("Error: Wrong device size of Cs: ", cs, 1);
@@ -711,13 +711,13 @@ float ddr3_get_device_size(u32 cs)
 		 * Small value will give wrong emem size in
 		 * ddr3_calc_mem_cs_size
 		 */
-		return 0.01;
+		return 0;
 	}
 }
 
 int ddr3_calc_mem_cs_size(u32 cs, u32 *cs_size)
 {
-	float cs_mem_size;
+	int cs_mem_size;
 
 	/* Calculate in GiB */
 	cs_mem_size = ((ddr3_get_bus_width() / ddr3_get_device_width(cs)) *
@@ -731,21 +731,12 @@ int ddr3_calc_mem_cs_size(u32 cs, u32 *cs_size)
 	 */
 	cs_mem_size *= DDR_CONTROLLER_BUS_WIDTH_MULTIPLIER;
 
-	if (cs_mem_size == 0.125) {
-		*cs_size = 128 << 20;
-	} else if (cs_mem_size == 0.25) {
-		*cs_size = 256 << 20;
-	} else if (cs_mem_size == 0.5) {
-		*cs_size = 512 << 20;
-	} else if (cs_mem_size == 1) {
-		*cs_size = 1 << 30;
-	} else if (cs_mem_size == 2) {
-		*cs_size = 2 << 30;
-	} else {
+	if (!cs_mem_size || (cs_mem_size == 64) || (cs_mem_size == 4096)) {
 		DEBUG_INIT_C("Error: Wrong Memory size of Cs: ", cs, 1);
 		return MV_BAD_VALUE;
 	}
 
+	*cs_size = cs_mem_size << 20;
 	return MV_OK;
 }
 
-- 
2.7.0

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* [U-Boot] [PATCH] arm: mvebu: a38x: Weed out floating point use
  2016-04-30 12:45 [U-Boot] [PATCH] arm: mvebu: a38x: Weed out floating point use Marek Vasut
@ 2016-05-19 23:46 ` Marek Vasut
  2016-05-20  9:04 ` Stefan Roese
  1 sibling, 0 replies; 3+ messages in thread
From: Marek Vasut @ 2016-05-19 23:46 UTC (permalink / raw)
  To: u-boot

On 04/30/2016 02:45 PM, Marek Vasut wrote:
> For reason unknown, recently, the DDR init code writers are really fond
> of hiding some small floating point operating deep in their creations.
> This patch removes one from the Marvell A38x code.
> 
> Instead of returning size of chip as float from ddr3_get_device_size()
> in GiB units, return it as int in MiB units. Since this would interfere
> with the huge switch code in ddr3_calc_mem_cs_size(), rework the code
> to match the change.
> 
> Before this patch, the cs_mem_size variable could have these values:
>  ( { 16, 32 } x { 8, 16 } x { 0.01, 0.5, 1, 2, 4, 8 } ) / 8 =
>    { 0.000000, 0.001250, 0.002500, 0.005000, 0.062500, 0.125000,
>      0.250000, 0.500000, 1.000000, 2.000000, 4.000000, }
> The switch code checked for a subset of the resulting RAM sizes, which
> is in range 128 MiB ... 2048 MiB.
> 
> With this patch, the cs_mem_size variable can have these values:
>  ( { 16, 32 } x { 8, 16 } x { 0, 512, 1024, 2048, 4096, 8192 } ) / 8 =
>    { 0, 64, 128, 256, 512, 1024, 2048, 4096 }
> To retain previous behavior, filter out 0 MiB (invalid size), 64 MiB
> and 4096 MiB options.
> 
> Removing the floating point stuff also saves 1.5k from text segment:
>   clearfog       :  spl/u-boot-spl:all -1592  spl/u-boot-spl:text -1592
> 
> Signed-off-by: Marek Vasut <marex@denx.de>
> Cc: Dirk Eibach <dirk.eibach@gdsys.cc>
> Cc: Stefan Roese <sr@denx.de>

Bump?

> ---
>  drivers/ddr/marvell/a38x/ddr3_init.c | 29 ++++++++++-------------------
>  1 file changed, 10 insertions(+), 19 deletions(-)
> 
> diff --git a/drivers/ddr/marvell/a38x/ddr3_init.c b/drivers/ddr/marvell/a38x/ddr3_init.c
> index ee05f57..55baad4 100644
> --- a/drivers/ddr/marvell/a38x/ddr3_init.c
> +++ b/drivers/ddr/marvell/a38x/ddr3_init.c
> @@ -678,7 +678,7 @@ u32 ddr3_get_device_width(u32 cs)
>  	return (device_width == 0) ? 8 : 16;
>  }
>  
> -float ddr3_get_device_size(u32 cs)
> +static int ddr3_get_device_size(u32 cs)
>  {
>  	u32 device_size_low, device_size_high, device_size;
>  	u32 data, cs_low_offset, cs_high_offset;
> @@ -695,15 +695,15 @@ float ddr3_get_device_size(u32 cs)
>  
>  	switch (device_size) {
>  	case 0:
> -		return 2;
> +		return 2048;
>  	case 2:
> -		return 0.5;
> +		return 512;
>  	case 3:
> -		return 1;
> +		return 1024;
>  	case 4:
> -		return 4;
> +		return 4096;
>  	case 5:
> -		return 8;
> +		return 8192;
>  	case 1:
>  	default:
>  		DEBUG_INIT_C("Error: Wrong device size of Cs: ", cs, 1);
> @@ -711,13 +711,13 @@ float ddr3_get_device_size(u32 cs)
>  		 * Small value will give wrong emem size in
>  		 * ddr3_calc_mem_cs_size
>  		 */
> -		return 0.01;
> +		return 0;
>  	}
>  }
>  
>  int ddr3_calc_mem_cs_size(u32 cs, u32 *cs_size)
>  {
> -	float cs_mem_size;
> +	int cs_mem_size;
>  
>  	/* Calculate in GiB */
>  	cs_mem_size = ((ddr3_get_bus_width() / ddr3_get_device_width(cs)) *
> @@ -731,21 +731,12 @@ int ddr3_calc_mem_cs_size(u32 cs, u32 *cs_size)
>  	 */
>  	cs_mem_size *= DDR_CONTROLLER_BUS_WIDTH_MULTIPLIER;
>  
> -	if (cs_mem_size == 0.125) {
> -		*cs_size = 128 << 20;
> -	} else if (cs_mem_size == 0.25) {
> -		*cs_size = 256 << 20;
> -	} else if (cs_mem_size == 0.5) {
> -		*cs_size = 512 << 20;
> -	} else if (cs_mem_size == 1) {
> -		*cs_size = 1 << 30;
> -	} else if (cs_mem_size == 2) {
> -		*cs_size = 2 << 30;
> -	} else {
> +	if (!cs_mem_size || (cs_mem_size == 64) || (cs_mem_size == 4096)) {
>  		DEBUG_INIT_C("Error: Wrong Memory size of Cs: ", cs, 1);
>  		return MV_BAD_VALUE;
>  	}
>  
> +	*cs_size = cs_mem_size << 20;
>  	return MV_OK;
>  }
>  
> 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [U-Boot] [PATCH] arm: mvebu: a38x: Weed out floating point use
  2016-04-30 12:45 [U-Boot] [PATCH] arm: mvebu: a38x: Weed out floating point use Marek Vasut
  2016-05-19 23:46 ` Marek Vasut
@ 2016-05-20  9:04 ` Stefan Roese
  1 sibling, 0 replies; 3+ messages in thread
From: Stefan Roese @ 2016-05-20  9:04 UTC (permalink / raw)
  To: u-boot

On 30.04.2016 14:45, Marek Vasut wrote:
> For reason unknown, recently, the DDR init code writers are really fond
> of hiding some small floating point operating deep in their creations.
> This patch removes one from the Marvell A38x code.
>
> Instead of returning size of chip as float from ddr3_get_device_size()
> in GiB units, return it as int in MiB units. Since this would interfere
> with the huge switch code in ddr3_calc_mem_cs_size(), rework the code
> to match the change.
>
> Before this patch, the cs_mem_size variable could have these values:
>   ( { 16, 32 } x { 8, 16 } x { 0.01, 0.5, 1, 2, 4, 8 } ) / 8 =
>     { 0.000000, 0.001250, 0.002500, 0.005000, 0.062500, 0.125000,
>       0.250000, 0.500000, 1.000000, 2.000000, 4.000000, }
> The switch code checked for a subset of the resulting RAM sizes, which
> is in range 128 MiB ... 2048 MiB.
>
> With this patch, the cs_mem_size variable can have these values:
>   ( { 16, 32 } x { 8, 16 } x { 0, 512, 1024, 2048, 4096, 8192 } ) / 8 =
>     { 0, 64, 128, 256, 512, 1024, 2048, 4096 }
> To retain previous behavior, filter out 0 MiB (invalid size), 64 MiB
> and 4096 MiB options.
>
> Removing the floating point stuff also saves 1.5k from text segment:
>    clearfog       :  spl/u-boot-spl:all -1592  spl/u-boot-spl:text -1592
>
> Signed-off-by: Marek Vasut <marex@denx.de>
> Cc: Dirk Eibach <dirk.eibach@gdsys.cc>
> Cc: Stefan Roese <sr@denx.de>

Marek, thanks for working on this.

Applied to u-boot-marvell/master

Thanks,
Stefan

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2016-05-20  9:04 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-04-30 12:45 [U-Boot] [PATCH] arm: mvebu: a38x: Weed out floating point use Marek Vasut
2016-05-19 23:46 ` Marek Vasut
2016-05-20  9:04 ` Stefan Roese

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.