[U-Boot] Change to miiphy_register() creates warnings
Hi,
In this commit:
5c45a22 mii: miiphy register address width change
the required prototypes for the read and write functions passed to miiphy_register() were changed. However, from what I can tell the callers to miiphy_register() did not all get their functions changes. So there are now a lot of warnings. I doubt they will affect actual operation.
Chandan, do you have time to do a patch for this? Files that use this are listed below, but not all will require changes - just changing the other use in miiphy.h would help a lot.
./arch/arm/cpu/ixp/npe/npe.c ./arch/mips/cpu/mips32/au1x00/au1x00_eth.c ./arch/powerpc/cpu/mpc85xx/ether_fcc.c ./arch/powerpc/cpu/mpc8220/fec.c ./arch/powerpc/cpu/mpc8xx/fec.c ./arch/powerpc/cpu/mpc8260/ether_fcc.c ./board/evb64260/eth.c ./board/prodrive/p3mx/mv_eth.c ./board/gdsys/405ex/io64.c ./board/gdsys/405ep/io.c ./common/miiphyutil.c ./drivers/qe/uec.c ./drivers/net/fsl_mcdmafec.c ./drivers/net/macb.c ./drivers/net/mpc5xxx_fec.c ./drivers/net/davinci_emac.c ./drivers/net/mcffec.c ./drivers/net/enc28j60.c ./drivers/net/smc911x.c ./drivers/net/altera_tse.c ./drivers/net/ep93xx_eth.c ./drivers/net/at91_emac.c ./drivers/net/sh_eth.c ./drivers/net/bfin_mac.c ./drivers/net/mpc512x_fec.c ./drivers/net/xilinx_axi_emac.c ./drivers/net/mvgbe.c ./drivers/net/armada100_fec.c ./drivers/net/designware.c ./drivers/net/4xx_enet.c ./drivers/net/fec_mxc.c ./drivers/net/eepro100.c
Regards, Simon
Dear Mike Frysinger,
In message 201112070114.04179.vapier@gentoo.org you wrote:
odd, i thought that commit was NAK-ed and people were told to convert to the newer mii api. maybe this is a good time to do so ...
No, nobody had the balls to send a clear NAK. It has been discussed that it's not worth applying, and that we should not invest efforts into the old API, but nobody actually NAKed it. So I thought it was harmless to apply.
I'll be OK with reverting it if anybody has any clear opinion.
Best regards,
Wolfgang Denk
Dear Simon Glass,
In message CAPnjgZ0tGPnZOQ7eBRwt5noHHVw_z3KY=QcyH_GSxkFxS2kgEw@mail.gmail.com you wrote:
In this commit:
5c45a22 mii: miiphy register address width change
the required prototypes for the read and write functions passed to miiphy_register() were changed. However, from what I can tell the callers to miiphy_register() did not all get their functions changes. So there are now a lot of warnings. I doubt they will affect actual operation.
Chandan, do you have time to do a patch for this? Files that use this are listed below, but not all will require changes - just changing the other use in miiphy.h would help a lot.
./arch/arm/cpu/ixp/npe/npe.c ./arch/mips/cpu/mips32/au1x00/au1x00_eth.c ./arch/powerpc/cpu/mpc85xx/ether_fcc.c ./arch/powerpc/cpu/mpc8220/fec.c ./arch/powerpc/cpu/mpc8xx/fec.c ./arch/powerpc/cpu/mpc8260/ether_fcc.c ./board/evb64260/eth.c ./board/prodrive/p3mx/mv_eth.c ./board/gdsys/405ex/io64.c ./board/gdsys/405ep/io.c ./common/miiphyutil.c ./drivers/qe/uec.c ./drivers/net/fsl_mcdmafec.c ./drivers/net/macb.c ./drivers/net/mpc5xxx_fec.c ./drivers/net/davinci_emac.c ./drivers/net/mcffec.c ./drivers/net/enc28j60.c ./drivers/net/smc911x.c ./drivers/net/altera_tse.c ./drivers/net/ep93xx_eth.c ./drivers/net/at91_emac.c ./drivers/net/sh_eth.c ./drivers/net/bfin_mac.c ./drivers/net/mpc512x_fec.c ./drivers/net/xilinx_axi_emac.c ./drivers/net/mvgbe.c ./drivers/net/armada100_fec.c ./drivers/net/designware.c ./drivers/net/4xx_enet.c ./drivers/net/fec_mxc.c ./drivers/net/eepro100.c
It makes no sense "fixing" all these drivers just to keep the old API longer alive. I reverted the commit.
Best regards,
Wolfgang Denk
Hi Wolfgang,
On Tue, Dec 6, 2011 at 11:39 PM, Wolfgang Denk wd@denx.de wrote:
Dear Simon Glass,
In message CAPnjgZ0tGPnZOQ7eBRwt5noHHVw_z3KY=QcyH_GSxkFxS2kgEw@mail.gmail.com you wrote:
In this commit:
5c45a22 mii: miiphy register address width change
the required prototypes for the read and write functions passed to miiphy_register() were changed. However, from what I can tell the callers to miiphy_register() did not all get their functions changes. So there are now a lot of warnings. I doubt they will affect actual operation.
Chandan, do you have time to do a patch for this? Files that use this are listed below, but not all will require changes - just changing the other use in miiphy.h would help a lot.
./arch/arm/cpu/ixp/npe/npe.c ./arch/mips/cpu/mips32/au1x00/au1x00_eth.c ./arch/powerpc/cpu/mpc85xx/ether_fcc.c ./arch/powerpc/cpu/mpc8220/fec.c ./arch/powerpc/cpu/mpc8xx/fec.c ./arch/powerpc/cpu/mpc8260/ether_fcc.c ./board/evb64260/eth.c ./board/prodrive/p3mx/mv_eth.c ./board/gdsys/405ex/io64.c ./board/gdsys/405ep/io.c ./common/miiphyutil.c ./drivers/qe/uec.c ./drivers/net/fsl_mcdmafec.c ./drivers/net/macb.c ./drivers/net/mpc5xxx_fec.c ./drivers/net/davinci_emac.c ./drivers/net/mcffec.c ./drivers/net/enc28j60.c ./drivers/net/smc911x.c ./drivers/net/altera_tse.c ./drivers/net/ep93xx_eth.c ./drivers/net/at91_emac.c ./drivers/net/sh_eth.c ./drivers/net/bfin_mac.c ./drivers/net/mpc512x_fec.c ./drivers/net/xilinx_axi_emac.c ./drivers/net/mvgbe.c ./drivers/net/armada100_fec.c ./drivers/net/designware.c ./drivers/net/4xx_enet.c ./drivers/net/fec_mxc.c ./drivers/net/eepro100.c
It makes no sense "fixing" all these drivers just to keep the old API longer alive. I reverted the commit.
Well that will fix the problem, thank you.
Regards, Simon
Best regards,
Wolfgang Denk
-- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de Wait! You have not been prepared! -- Mr. Atoz, "Tomorrow is Yesterday", stardate 3113.2
participants (3)
-
Mike Frysinger -
Simon Glass -
Wolfgang Denk