Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
On 7/11/2010 10:45 PM, Prafulla Wadaskar wrote:
-Original Message-
From: u-boot-boun...@lists.denx.de
[mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud
Sent: Sunday, July 11, 2010 1:32 PM
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other
than kirkwood
Rename all references to kirkwood in mv_egiga symbols
throughout the whole codebase.
Signed-off-by: Albert Aribaudalbert.arib...@free.fr
---
arch/arm/cpu/arm926ejs/kirkwood/cpu.c |4 +-
arch/arm/include/asm/arch-kirkwood/kirkwood.h |5 +
drivers/net/Makefile |2 +-
drivers/net/mv_egiga.c| 322
+-
drivers/net/mv_egiga.h| 466
include/configs/guruplug.h|4 +-
include/configs/km_arm.h |4 +-
include/configs/mv88f6281gtw_ge.h |4 +-
include/configs/openrd_base.h |4 +-
include/configs/rd6281a.h |4 +-
include/configs/sheevaplug.h |4 +-
include/netdev.h |2 +-
12 files changed, 418 insertions(+), 407 deletions(-)
diff --git a/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
b/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
index 6fc3902..786ffc6 100644
--- a/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
+++ b/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
@@ -378,10 +378,10 @@ int arch_misc_init(void)
}
#endif /* CONFIG_ARCH_MISC_INIT */
-#ifdef CONFIG_KIRKWOOD_EGIGA
+#ifdef CONFIG_MV_EGIGA
int cpu_eth_init(bd_t *bis)
{
-kirkwood_egiga_initialize(bis);
+mv_egiga_initialize(bis);
return 0;
}
#endif
diff --git a/arch/arm/include/asm/arch-kirkwood/kirkwood.h
b/arch/arm/include/asm/arch-kirkwood/kirkwood.h
index 2470efb..9200605 100644
--- a/arch/arm/include/asm/arch-kirkwood/kirkwood.h
+++ b/arch/arm/include/asm/arch-kirkwood/kirkwood.h
@@ -60,6 +60,11 @@
#define KW_EGIGA0_BASE (KW_REGISTER(0x72000))
#define KW_EGIGA1_BASE (KW_REGISTER(0x76000))
+#if defined (CONFIG_MV_EGIGA)
I think you don't need ifdef here
+#define MV_EGIGA0_BASE KW_EGIGA0_BASE
+#define MV_EGIGA1_BASE KW_EGIGA1_BASE
+#endif
+
#if defined (CONFIG_KW88F6281)
#includeasm/arch/kw88f6281.h
#elif defined (CONFIG_KW88F6192)
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 1599c26..04f8de0 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -52,7 +52,7 @@ COBJS-$(CONFIG_MACB) += macb.o
COBJS-$(CONFIG_MCFFEC) += mcffec.o mcfmii.o
COBJS-$(CONFIG_MPC5xxx_FEC) += mpc5xxx_fec.o
COBJS-$(CONFIG_MPC512x_FEC) += mpc512x_fec.o
-COBJS-$(CONFIG_KIRKWOOD_EGIGA) += mv_egiga.o
+COBJS-$(CONFIG_MV_EGIGA) += mv_egiga.o
COBJS-$(CONFIG_NATSEMI) += natsemi.o
COBJS-$(CONFIG_DRIVER_NE2000) += ne2000.o ne2000_base.o
COBJS-$(CONFIG_DRIVER_AX88796L) += ax88796.o ne2000_base.o
diff --git a/drivers/net/mv_egiga.c b/drivers/net/mv_egiga.c
index b7ecc9d..4fecf1e 100644
--- a/drivers/net/mv_egiga.c
+++ b/drivers/net/mv_egiga.c
@@ -35,13 +35,17 @@
#includeasm/errno.h
#includeasm/types.h
#includeasm/byteorder.h
+
+#if defined (CONFIG_KIRKWOOD)
#includeasm/arch/kirkwood.h
+#endif
+
#include mv_egiga.h
DECLARE_GLOBAL_DATA_PTR;
-#define KIRKWOOD_PHY_ADR_REQUEST 0xee
-#define KWGBE_SMI_REG (((struct kwgbe_registers
*)KW_EGIGA0_BASE)-smi)
+#define MV_PHY_ADR_REQUEST 0xee
+#define MV_EGIGA_SMI_REG (((struct mv_egiga_registers
*)MV_EGIGA0_BASE)-smi)
/*
* smi_reg_read - miiphy_read callback function.
@@ -51,16 +55,16 @@ DECLARE_GLOBAL_DATA_PTR;
static int smi_reg_read(char *devname, u8 phy_adr, u8
reg_ofs, u16 * data)
{
struct eth_device *dev = eth_get_dev_by_name(devname);
-struct kwgbe_device *dkwgbe = to_dkwgbe(dev);
-struct kwgbe_registers *regs = dkwgbe-regs;
+struct mv_egiga_device *dmvegiga = to_mv_egiga(dev);
+struct mv_egiga_registers *regs = dmvegiga-regs;
I suggest to keep name as mvgbe here instead of mv_egiga, 3 additional chars,
increases overall code size
huh? The name is consistent with the rest of his work, and *if* the
code really increases in size, I can't imagine that 3 chars really
matters...
u32 smi_reg;
u32 timeout;
/* Phyadr read request */
-if (phy_adr == KIRKWOOD_PHY_ADR_REQUEST
-reg_ofs == KIRKWOOD_PHY_ADR_REQUEST) {
+if (phy_adr == MV_PHY_ADR_REQUEST
+reg_ofs == MV_PHY_ADR_REQUEST) {
/* */
-*data = (u16) (KWGBEREG_RD(regs-phyadr) PHYADR_MASK);
+*data = (u16) (MV_EGIGA_REG_RD(regs-phyadr)
Same here, pls use MVGBEREG instead of MV_EGIGA_REG, 4 extra chars
Same comment as above. Clarity is king.
Pls just replace kw/KW by mv/MV to avoide indentation issues
I
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
-Original Message- From: Ben Warren [mailto:biggerbadder...@gmail.com] Sent: Monday, July 12, 2010 11:54 AM To: Prafulla Wadaskar Cc: Albert Aribaud; u-boot@lists.denx.de Subject: Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood On 7/11/2010 10:45 PM, Prafulla Wadaskar wrote: -Original Message- From: u-boot-boun...@lists.denx.de [mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud Sent: Sunday, July 11, 2010 1:32 PM To: u-boot@lists.denx.de Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood ...snip... struct eth_device *dev = eth_get_dev_by_name(devname); - struct kwgbe_device *dkwgbe = to_dkwgbe(dev); - struct kwgbe_registers *regs = dkwgbe-regs; + struct mv_egiga_device *dmvegiga = to_mv_egiga(dev); + struct mv_egiga_registers *regs = dmvegiga-regs; I suggest to keep name as mvgbe here instead of mv_egiga, 3 additional chars, increases overall code size huh? The name is consistent with the rest of his work, and *if* the code really increases in size, I can't imagine that 3 chars really matters... That's true. But if we can do it why to avoid it? again it helps to keep same indentation (keeping them below 80char size) Regards.. Prafulla . . ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
Le 12/07/2010 08:53, Prafulla Wadaskar a écrit : - struct kwgbe_device *dkwgbe = to_dkwgbe(dev); - struct kwgbe_registers *regs = dkwgbe-regs; + struct mv_egiga_device *dmvegiga = to_mv_egiga(dev); + struct mv_egiga_registers *regs = dmvegiga-regs; I suggest to keep name as mvgbe here instead of mv_egiga, 3 additional chars, increases overall code size huh? The name is consistent with the rest of his work, and *if* the code really increases in size, I can't imagine that 3 chars really matters... That's true. But if we can do it why to avoid it? again it helps to keep same indentation (keeping them below 80char size) I don't think I changed indentation here, and the issue is about line lengths, right? Initially I chose mv egiga because the file names used egiga while the code used gbe, and I wanted clarity, so I decided to keep only one of egiga and gbe. Now which one I should keep is not really important to me, and a Google search for marvell egiga vs marvell gbe indicates gbe appears much more frequently, so someone looking into this will probably know GbE more than egiga. I suggest that: - I switch the file names from mv_egiga to mvgbe (to be consistent with Prafulla's comment on mv_sata becoming mvsata), and - I replace mv_egiga/MV_EGIGA symbols with mvgbe/MVGBE. That will retain (as much) clarity and uniformity (as egiga does), which is what I think Ben is looking for, and it'll keep name length at a minimum, which should satisfy Prafulla. Ben, Prafulla (and others as well, of course), do you agree? Amicalement, -- Albert. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
Le 12/07/2010 07:49, Prafulla Wadaskar a écrit : -Original Message- From: u-boot-boun...@lists.denx.de [mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud Sent: Sunday, July 11, 2010 1:32 PM To: u-boot@lists.denx.de Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood Rename all references to kirkwood in mv_egiga symbols throughout the whole codebase. Signed-off-by: Albert Aribaudalbert.arib...@free.fr --- Pls break this patch as- arch/arm/cpu/arm926ejs/kirkwood/cpu.c |4 +- arch/arm/include/asm/arch-kirkwood/kirkwood.h |5 + Kirkwood SOC specific drivers/net/Makefile |2 +- drivers/net/mv_egiga.c| 322 +- drivers/net/mv_egiga.h| 466 include/netdev.h |2 +- Net: driver specific for net repo include/configs/guruplug.h|4 +- include/configs/km_arm.h |4 +- include/configs/mv88f6281gtw_ge.h |4 +- include/configs/openrd_base.h |4 +- include/configs/rd6281a.h |4 +- include/configs/sheevaplug.h |4 +- Board updates In the same order, so that it is easier to pick them separately by net and Marvell repos. Regards.. Prafulla . . If I split this patch, it will not build correctly any more. Amicalement, -- Albert. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
Hi Albert, On Mon, Jul 12, 2010 at 1:53 AM, Albert ARIBAUD albert.arib...@free.frwrote: Le 12/07/2010 08:53, Prafulla Wadaskar a écrit : - struct kwgbe_device *dkwgbe = to_dkwgbe(dev); - struct kwgbe_registers *regs = dkwgbe-regs; + struct mv_egiga_device *dmvegiga = to_mv_egiga(dev); + struct mv_egiga_registers *regs = dmvegiga-regs; I suggest to keep name as mvgbe here instead of mv_egiga, 3 additional chars, increases overall code size huh? The name is consistent with the rest of his work, and *if* the code really increases in size, I can't imagine that 3 chars really matters... That's true. But if we can do it why to avoid it? again it helps to keep same indentation (keeping them below 80char size) I don't think I changed indentation here, and the issue is about line lengths, right? Initially I chose mv egiga because the file names used egiga while the code used gbe, and I wanted clarity, so I decided to keep only one of egiga and gbe. Now which one I should keep is not really important to me, and a Google search for marvell egiga vs marvell gbe indicates gbe appears much more frequently, so someone looking into this will probably know GbE more than egiga. I suggest that: - I switch the file names from mv_egiga to mvgbe (to be consistent with Prafulla's comment on mv_sata becoming mvsata), and - I replace mv_egiga/MV_EGIGA symbols with mvgbe/MVGBE. That will retain (as much) clarity and uniformity (as egiga does), which is what I think Ben is looking for, and it'll keep name length at a minimum, which should satisfy Prafulla. Ben, Prafulla (and others as well, of course), do you agree? OK with me. Amicalement, -- Albert. regards, Ben ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
Hi Albert, On Mon, Jul 12, 2010 at 1:56 AM, Albert ARIBAUD albert.arib...@free.frwrote: Le 12/07/2010 07:49, Prafulla Wadaskar a écrit : -Original Message- From: u-boot-boun...@lists.denx.de [mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud Sent: Sunday, July 11, 2010 1:32 PM To: u-boot@lists.denx.de Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood Rename all references to kirkwood in mv_egiga symbols throughout the whole codebase. Signed-off-by: Albert Aribaudalbert.arib...@free.fr --- Pls break this patch as- arch/arm/cpu/arm926ejs/kirkwood/cpu.c |4 +- arch/arm/include/asm/arch-kirkwood/kirkwood.h |5 + Kirkwood SOC specific drivers/net/Makefile |2 +- drivers/net/mv_egiga.c| 322 +- drivers/net/mv_egiga.h| 466 include/netdev.h |2 +- Net: driver specific for net repo include/configs/guruplug.h|4 +- include/configs/km_arm.h |4 +- include/configs/mv88f6281gtw_ge.h |4 +- include/configs/openrd_base.h |4 +- include/configs/rd6281a.h |4 +- include/configs/sheevaplug.h |4 +- Board updates In the same order, so that it is easier to pick them separately by net and Marvell repos. Regards.. Prafulla . . If I split this patch, it will not build correctly any more. Usually, these should be split up. In this case, the dependencies are significant, so just leave it as-is. If Prafulla ACKs, I'll pull the whole group into the net repo. Amicalement, -- Albert. regards, Ben ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
Le 12/07/2010 16:19, Ben Warren a écrit : I suggest that: - I switch the file names from mv_egiga to mvgbe (to be consistent with Prafulla's comment on mv_sata becoming mvsata), and - I replace mv_egiga/MV_EGIGA symbols with mvgbe/MVGBE. That will retain (as much) clarity and uniformity (as egiga does), which is what I think Ben is looking for, and it'll keep name length at a minimum, which should satisfy Prafulla. Ben, Prafulla (and others as well, of course), do you agree? OK with me. Thanks Ben. I assume Prafulla will agree too. I'll regenerate a patch set with all changes requested by Prafulla, keeping the symbol-change in a single patch as discussed elsewhere in this thread. Note that the last patch of the series (edminiv2 support) will only build above the CONFIG_SKIP_LOCAL_MAC_RANDOMIZATION patch that I posted separately. Amicalement, -- Albert. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
Le 12/07/2010 07:45, Prafulla Wadaskar a écrit : --- a/arch/arm/include/asm/arch-kirkwood/kirkwood.h +++ b/arch/arm/include/asm/arch-kirkwood/kirkwood.h +#if defined (CONFIG_MV_EGIGA) I think you don't need ifdef here +#define MV_EGIGA0_BASE KW_EGIGA0_BASE +#define MV_EGIGA1_BASE KW_EGIGA1_BASE +#endif Actually the ifdef is intended to make sure these symbols do not get defined uselessly for kirkwood boards that don't have Ethernet support. Granted, their being defined without being used is no big deal. However the whole thing about the egiga driver supporting one or two ports could be streamlined (e.g., one might define KIRKWOOD_EGIGAn_BASE while not enabling port n in CONFIG_KIRKWOOD_EGIGA_PORTS) and integrate the suggestion about the PHYs that occurred some time ago: http://lists.denx.de/pipermail/u-boot/2010-June/073138.html So I'll leave that one to a later patch set. Amicalement, -- Albert. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
-Original Message- From: Ben Warren [mailto:biggerbadder...@gmail.com] Sent: Monday, July 12, 2010 7:54 PM To: Albert ARIBAUD Cc: Prafulla Wadaskar; u-boot@lists.denx.de Subject: Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood Hi Albert, On Mon, Jul 12, 2010 at 1:56 AM, Albert ARIBAUD albert.arib...@free.fr wrote: Le 12/07/2010 07:49, Prafulla Wadaskar a écrit : -Original Message- From: u-boot-boun...@lists.denx.de [mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud Sent: Sunday, July 11, 2010 1:32 PM To: u-boot@lists.denx.de Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood Rename all references to kirkwood in mv_egiga symbols throughout the whole codebase. Signed-off-by: Albert Aribaudalbert.arib...@free.fr --- Pls break this patch as- arch/arm/cpu/arm926ejs/kirkwood/cpu.c |4 +- arch/arm/include/asm/arch-kirkwood/kirkwood.h |5 + Kirkwood SOC specific drivers/net/Makefile |2 +- drivers/net/mv_egiga.c| 322 +- drivers/net/mv_egiga.h| 466 include/netdev.h |2 +- Net: driver specific for net repo include/configs/guruplug.h|4 +- include/configs/km_arm.h |4 +- include/configs/mv88f6281gtw_ge.h |4 +- include/configs/openrd_base.h |4 +- include/configs/rd6281a.h |4 +- include/configs/sheevaplug.h |4 +- Board updates In the same order, so that it is easier to pick them separately by net and Marvell repos. Regards.. Prafulla . . If I split this patch, it will not build correctly any more. Usually, these should be split up. In this case, the dependencies are significant, so just leave it as-is. If Prafulla ACKs, I'll pull the whole group into the net repo. Ack Hi Ben, its good if you pull entire patch series. Regards.. Prafulla . . ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
-Original Message-
From: u-boot-boun...@lists.denx.de
[mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud
Sent: Sunday, July 11, 2010 1:32 PM
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other
than kirkwood
Rename all references to kirkwood in mv_egiga symbols
throughout the whole codebase.
Signed-off-by: Albert Aribaud albert.arib...@free.fr
---
arch/arm/cpu/arm926ejs/kirkwood/cpu.c |4 +-
arch/arm/include/asm/arch-kirkwood/kirkwood.h |5 +
drivers/net/Makefile |2 +-
drivers/net/mv_egiga.c| 322
+-
drivers/net/mv_egiga.h| 466
include/configs/guruplug.h|4 +-
include/configs/km_arm.h |4 +-
include/configs/mv88f6281gtw_ge.h |4 +-
include/configs/openrd_base.h |4 +-
include/configs/rd6281a.h |4 +-
include/configs/sheevaplug.h |4 +-
include/netdev.h |2 +-
12 files changed, 418 insertions(+), 407 deletions(-)
diff --git a/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
b/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
index 6fc3902..786ffc6 100644
--- a/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
+++ b/arch/arm/cpu/arm926ejs/kirkwood/cpu.c
@@ -378,10 +378,10 @@ int arch_misc_init(void)
}
#endif /* CONFIG_ARCH_MISC_INIT */
-#ifdef CONFIG_KIRKWOOD_EGIGA
+#ifdef CONFIG_MV_EGIGA
int cpu_eth_init(bd_t *bis)
{
- kirkwood_egiga_initialize(bis);
+ mv_egiga_initialize(bis);
return 0;
}
#endif
diff --git a/arch/arm/include/asm/arch-kirkwood/kirkwood.h
b/arch/arm/include/asm/arch-kirkwood/kirkwood.h
index 2470efb..9200605 100644
--- a/arch/arm/include/asm/arch-kirkwood/kirkwood.h
+++ b/arch/arm/include/asm/arch-kirkwood/kirkwood.h
@@ -60,6 +60,11 @@
#define KW_EGIGA0_BASE (KW_REGISTER(0x72000))
#define KW_EGIGA1_BASE (KW_REGISTER(0x76000))
+#if defined (CONFIG_MV_EGIGA)
I think you don't need ifdef here
+#define MV_EGIGA0_BASE KW_EGIGA0_BASE
+#define MV_EGIGA1_BASE KW_EGIGA1_BASE
+#endif
+
#if defined (CONFIG_KW88F6281)
#include asm/arch/kw88f6281.h
#elif defined (CONFIG_KW88F6192)
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 1599c26..04f8de0 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -52,7 +52,7 @@ COBJS-$(CONFIG_MACB) += macb.o
COBJS-$(CONFIG_MCFFEC) += mcffec.o mcfmii.o
COBJS-$(CONFIG_MPC5xxx_FEC) += mpc5xxx_fec.o
COBJS-$(CONFIG_MPC512x_FEC) += mpc512x_fec.o
-COBJS-$(CONFIG_KIRKWOOD_EGIGA) += mv_egiga.o
+COBJS-$(CONFIG_MV_EGIGA) += mv_egiga.o
COBJS-$(CONFIG_NATSEMI) += natsemi.o
COBJS-$(CONFIG_DRIVER_NE2000) += ne2000.o ne2000_base.o
COBJS-$(CONFIG_DRIVER_AX88796L) += ax88796.o ne2000_base.o
diff --git a/drivers/net/mv_egiga.c b/drivers/net/mv_egiga.c
index b7ecc9d..4fecf1e 100644
--- a/drivers/net/mv_egiga.c
+++ b/drivers/net/mv_egiga.c
@@ -35,13 +35,17 @@
#include asm/errno.h
#include asm/types.h
#include asm/byteorder.h
+
+#if defined (CONFIG_KIRKWOOD)
#include asm/arch/kirkwood.h
+#endif
+
#include mv_egiga.h
DECLARE_GLOBAL_DATA_PTR;
-#define KIRKWOOD_PHY_ADR_REQUEST 0xee
-#define KWGBE_SMI_REG (((struct kwgbe_registers
*)KW_EGIGA0_BASE)-smi)
+#define MV_PHY_ADR_REQUEST 0xee
+#define MV_EGIGA_SMI_REG (((struct mv_egiga_registers
*)MV_EGIGA0_BASE)-smi)
/*
* smi_reg_read - miiphy_read callback function.
@@ -51,16 +55,16 @@ DECLARE_GLOBAL_DATA_PTR;
static int smi_reg_read(char *devname, u8 phy_adr, u8
reg_ofs, u16 * data)
{
struct eth_device *dev = eth_get_dev_by_name(devname);
- struct kwgbe_device *dkwgbe = to_dkwgbe(dev);
- struct kwgbe_registers *regs = dkwgbe-regs;
+ struct mv_egiga_device *dmvegiga = to_mv_egiga(dev);
+ struct mv_egiga_registers *regs = dmvegiga-regs;
I suggest to keep name as mvgbe here instead of mv_egiga, 3 additional chars,
increases overall code size
u32 smi_reg;
u32 timeout;
/* Phyadr read request */
- if (phy_adr == KIRKWOOD_PHY_ADR_REQUEST
- reg_ofs == KIRKWOOD_PHY_ADR_REQUEST) {
+ if (phy_adr == MV_PHY_ADR_REQUEST
+ reg_ofs == MV_PHY_ADR_REQUEST) {
/* */
- *data = (u16) (KWGBEREG_RD(regs-phyadr) PHYADR_MASK);
+ *data = (u16) (MV_EGIGA_REG_RD(regs-phyadr)
Same here, pls use MVGBEREG instead of MV_EGIGA_REG, 4 extra chars
Pls just replace kw/KW by mv/MV to avoide indentation issues
I think these are the only changes in this patch (i.e. find and replace), are
there any other?
Regards..
Prafulla . .
___
U-Boot mailing list
U-Boot@lists.denx.de
Re: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood
-Original Message- From: u-boot-boun...@lists.denx.de [mailto:u-boot-boun...@lists.denx.de] On Behalf Of Albert Aribaud Sent: Sunday, July 11, 2010 1:32 PM To: u-boot@lists.denx.de Subject: [U-Boot] [PATCH V3 2/5] mv_egiga: support SoCs other than kirkwood Rename all references to kirkwood in mv_egiga symbols throughout the whole codebase. Signed-off-by: Albert Aribaud albert.arib...@free.fr --- Pls break this patch as- arch/arm/cpu/arm926ejs/kirkwood/cpu.c |4 +- arch/arm/include/asm/arch-kirkwood/kirkwood.h |5 + Kirkwood SOC specific drivers/net/Makefile |2 +- drivers/net/mv_egiga.c| 322 +- drivers/net/mv_egiga.h| 466 include/netdev.h |2 +- Net: driver specific for net repo include/configs/guruplug.h|4 +- include/configs/km_arm.h |4 +- include/configs/mv88f6281gtw_ge.h |4 +- include/configs/openrd_base.h |4 +- include/configs/rd6281a.h |4 +- include/configs/sheevaplug.h |4 +- Board updates In the same order, so that it is easier to pick them separately by net and Marvell repos. Regards.. Prafulla . . ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
