diff mbox

[08/13] ARM: dts: armada-385-linksys: use binary unit prefixes

Message ID 20170521124903.11050-9-ralph.sennhauser@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

Ralph Sennhauser May 21, 2017, 12:48 p.m. UTC
Use IEEE 1541-2002 unit prefixes for sizes.

Signed-off-by: Ralph Sennhauser <ralph.sennhauser@gmail.com>
---
 arch/arm/boot/dts/armada-385-linksys.dtsi | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

Comments

Andrew Lunn May 22, 2017, 3 p.m. UTC | #1
On Sun, May 21, 2017 at 02:48:57PM +0200, Ralph Sennhauser wrote:
> Use IEEE 1541-2002 unit prefixes for sizes.

Does ePAPR recommend this?

I think this looks ugly, so i would not do it.

  Andrew
Ralph Sennhauser May 22, 2017, 6:24 p.m. UTC | #2
On Mon, 22 May 2017 17:00:01 +0200
Andrew Lunn <andrew@lunn.ch> wrote:

> On Sun, May 21, 2017 at 02:48:57PM +0200, Ralph Sennhauser wrote:
> > Use IEEE 1541-2002 unit prefixes for sizes.  
> 
> Does ePAPR recommend this?

Not directly, but the paper (v1.1) itself is using MiB as well which
could be seen as a recommendation.

SI and NIST to my knowledge explicitly state to not use SI prefixes for
anything but powers of 10. Also as far as I'm aware any relevant
standard body adopted / endorsed these.

> 
> I think this looks ugly, so i would not do it.

Needs getting used to indeed. I for myself I can say I've gotten over
it by now. Grepping the Linux tree shows quite a few other examples of
MiB so this one wont feel lonely either.

Ralph
Ralph Sennhauser May 23, 2017, 8:33 a.m. UTC | #3
On Mon, 22 May 2017 20:24:16 +0200
Ralph Sennhauser <ralph.sennhauser@gmail.com> wrote:

> On Mon, 22 May 2017 17:00:01 +0200
> Andrew Lunn <andrew@lunn.ch> wrote:
> 
> > On Sun, May 21, 2017 at 02:48:57PM +0200, Ralph Sennhauser wrote:  
> > > Use IEEE 1541-2002 unit prefixes for sizes.    
> > 
> > Does ePAPR recommend this?  
> 
> Not directly, but the paper (v1.1) itself is using MiB as well which
> could be seen as a recommendation.

Found the pdf again on the web, so here a link:
http://elinux.org/images/c/cf/Power_ePAPR_APPROVED_v1.1.pdf

Funny thing it's using KB(instead of KiB) and MiB. So it's
inconsistent, also the unit with SI prefix would be kB and not KB. So
I'd say ePAPR doesn't care / the authors didn't pay any attention to
this.

Ralph
Andrew Lunn May 23, 2017, 12:02 p.m. UTC | #4
On Tue, May 23, 2017 at 10:33:02AM +0200, Ralph Sennhauser wrote:
> On Mon, 22 May 2017 20:24:16 +0200
> Ralph Sennhauser <ralph.sennhauser@gmail.com> wrote:
> 
> > On Mon, 22 May 2017 17:00:01 +0200
> > Andrew Lunn <andrew@lunn.ch> wrote:
> > 
> > > On Sun, May 21, 2017 at 02:48:57PM +0200, Ralph Sennhauser wrote:  
> > > > Use IEEE 1541-2002 unit prefixes for sizes.    
> > > 
> > > Does ePAPR recommend this?  
> > 
> > Not directly, but the paper (v1.1) itself is using MiB as well which
> > could be seen as a recommendation.
> 
> Found the pdf again on the web, so here a link:
> http://elinux.org/images/c/cf/Power_ePAPR_APPROVED_v1.1.pdf
> 
> Funny thing it's using KB(instead of KiB) and MiB. So it's
> inconsistent, also the unit with SI prefix would be kB and not KB. So
> I'd say ePAPR doesn't care / the authors didn't pay any attention to
> this.

Hi Ralph

Thanks for checking into this.

My preference, and it is only a preference, is to not do this. Keep
with MB, KB.

   Andrew
diff mbox

Patch

diff --git a/arch/arm/boot/dts/armada-385-linksys.dtsi b/arch/arm/boot/dts/armada-385-linksys.dtsi
index f1d6b77..4049eaf 100644
--- a/arch/arm/boot/dts/armada-385-linksys.dtsi
+++ b/arch/arm/boot/dts/armada-385-linksys.dtsi
@@ -52,7 +52,7 @@ 
 
 	memory {
 		device_type = "memory";
-		reg = <0x00000000 0x20000000>; /* 512 MB */
+		reg = <0x00000000 0x20000000>; /* 512 MiB */
 	};
 
 	soc {
@@ -180,50 +180,50 @@ 
 
 	partition@0 {
 		label = "u-boot";
-		reg = <0x0000000 0x200000>;  /* 2MB */
+		reg = <0x0000000 0x200000>;  /* 2MiB */
 		read-only;
 	};
 
 	partition@100000 {
 		label = "u_env";
-		reg = <0x200000 0x40000>;    /* 256KB */
+		reg = <0x200000 0x40000>;    /* 256KiB */
 	};
 
 	partition@140000 {
 		label = "s_env";
-		reg = <0x240000 0x40000>;    /* 256KB */
+		reg = <0x240000 0x40000>;    /* 256KiB */
 	};
 
 	partition@900000 {
 		label = "devinfo";
-		reg = <0x900000 0x100000>;   /* 1MB */
+		reg = <0x900000 0x100000>;   /* 1MiB */
 		read-only;
 	};
 
 	/* kernel1 overlaps with rootfs1 by design */
 	partition@a00000 {
 		label = "kernel1";
-		reg = <0xa00000 0x2800000>;  /* 40MB */
+		reg = <0xa00000 0x2800000>;  /* 40MiB */
 	};
 
 	partition@1000000 {
 		label = "rootfs1";
-		reg = <0x1000000 0x2200000>;  /* 34MB */
+		reg = <0x1000000 0x2200000>;  /* 34MiB */
 	};
 
 	/* kernel2 overlaps with rootfs2 by design */
 	partition@3200000 {
 		label = "kernel2";
-		reg = <0x3200000 0x2800000>; /* 40MB */
+		reg = <0x3200000 0x2800000>; /* 40MiB */
 	};
 
 	partition@3800000 {
 		label = "rootfs2";
-		reg = <0x3800000 0x2200000>; /* 34MB */
+		reg = <0x3800000 0x2200000>; /* 34MiB */
 	};
 
 	/*
-	 * 38MB, last MB is for the BBT, not writable
+	 * 38MiB, last MiB is for the BBT, not writable
 	 */
 	partition@5a00000 {
 		label = "syscfg";
@@ -238,7 +238,7 @@ 
 	 */
 	partition@180000 {
 		label = "unused_area";
-		reg = <0x280000 0x680000>;   /* 6.5MB */
+		reg = <0x280000 0x680000>;   /* 6.5MiB */
 	};
 };