[U-Boot-Users] mii_send error or at least strange
I had a look to mii_send code in u-boot/cpu/mpc8xx/fec.c . It supports both FEC1 and FEC2 but anyway it seems to write always register mii-data of FEC1, look here: /* send command to phy using mii, wait for result */ static uint mii_send(uint mii_cmd) { uint mii_reply; volatile fec_t *ep; int cnt; ep = &(((immap_t *)CFG_IMMR)->im_cpm.cp_fec); ep->fec_mii_data = mii_cmd; /* command to phy */ /* wait for mii complete */ cnt = 0; while (!(ep->fec_ievent & FEC_ENET_MII)) { if (++cnt > 1000) { printf("mii_send STUCK!\n"); break; } } mii_reply = ep->fec_mii_data; /* result from phy */ ep->fec_ievent = FEC_ENET_MII; /* clear MII complete */ #if 0 printf("%s[%d] %s: sent=0x%8.8x, reply=0x%8.8x\n", __FILE__,__LINE__,__FUNCTION__,mii_cmd,mii_reply); #endif return (mii_reply & 0xffff); /* data read from phy */ }
Antonio,
I'm not sure about the 8xx PowerQUICCs, but other Freescale CPUs with multiple ethernet ports support MII data plane on all ports but only have one MDIO port (control plane for MII). Maybe that's why the code is this way?
regards, Ben
On Wed, 2006-03-15 at 22:31 +0100, Antonio Di Bacco wrote:
I had a look to mii_send code in u-boot/cpu/mpc8xx/fec.c . It supports both FEC1 and FEC2 but anyway it seems to write always register mii-data of FEC1, look here:
/* send command to phy using mii, wait for result */ static uint mii_send(uint mii_cmd) { uint mii_reply; volatile fec_t *ep; int cnt;
ep = &(((immap_t *)CFG_IMMR)->im_cpm.cp_fec);
ep->fec_mii_data = mii_cmd; /* command to phy */
/* wait for mii complete */ cnt = 0; while (!(ep->fec_ievent & FEC_ENET_MII)) { if (++cnt > 1000) { printf("mii_send STUCK!\n"); break; } } mii_reply = ep->fec_mii_data; /* result from phy */ ep->fec_ievent = FEC_ENET_MII; /* clear MII complete */ #if 0 printf("%s[%d] %s: sent=0x%8.8x, reply=0x%8.8x\n", __FILE__,__LINE__,__FUNCTION__,mii_cmd,mii_reply); #endif return (mii_reply & 0xffff); /* data read from phy */ }
This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid%110944&bid$1720&dat%... _______________________________________________ U-Boot-Users mailing list U-Boot-Users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/u-boot-users
That's a good answer.
Thank you.
But why do I receive "mii_send STUCK" for example when tftp tries to use FEC2?
Bye.
On Wednesday 15 March 2006 22:38, Ben Warren wrote:
Antonio,
I'm not sure about the 8xx PowerQUICCs, but other Freescale CPUs with multiple ethernet ports support MII data plane on all ports but only have one MDIO port (control plane for MII). Maybe that's why the code is this way?
regards, Ben
On Wed, 2006-03-15 at 22:31 +0100, Antonio Di Bacco wrote:
I had a look to mii_send code in u-boot/cpu/mpc8xx/fec.c . It supports both FEC1 and FEC2 but anyway it seems to write always register mii-data of FEC1, look here:
/* send command to phy using mii, wait for result */ static uint mii_send(uint mii_cmd) { uint mii_reply; volatile fec_t *ep; int cnt;
ep = &(((immap_t *)CFG_IMMR)->im_cpm.cp_fec);
ep->fec_mii_data = mii_cmd; /* command to phy */
/* wait for mii complete */ cnt = 0; while (!(ep->fec_ievent & FEC_ENET_MII)) { if (++cnt > 1000) { printf("mii_send STUCK!\n"); break; } } mii_reply = ep->fec_mii_data; /* result from phy */ ep->fec_ievent = FEC_ENET_MII; /* clear MII complete */ #if 0 printf("%s[%d] %s: sent=0x%8.8x, reply=0x%8.8x\n", __FILE__,__LINE__,__FUNCTION__,mii_cmd,mii_reply); #endif return (mii_reply & 0xffff); /* data read from phy */ }
This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid%110944&bid$1720&dat%... _______________________________________________ U-Boot-Users mailing list U-Boot-Users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/u-boot-users
On Mar 15, 2006, at 4:38 PM, Ben Warren wrote:
I'm not sure about the 8xx PowerQUICCs, but other Freescale CPUs with multiple ethernet ports support MII data plane on all ports but only have one MDIO port (control plane for MII). Maybe that's why the code is this way?
That's exactly why.
-- Dan
participants (3)
-
Antonio Di Bacco -
Ben Warren -
Dan Malek