Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Hi Mike, On Wednesday 13 May 2009 04:41:29 Detlev Zundel wrote: I never ran across this outside of linux arm where people seriously repeat the statement over and over that a whole initramfs with an initial userspace and a pivot_root is plausible for nfs root whilst patches of a few lines float around doing the same in the kernel. i never said that a completely populated initramfs was needed. it can be accomplished pretty easily with very little code. Sure, but still, how big is such an initramfs, do you have any numbers here? And how many lines of C code does it take to set the MAC in the kernel? Is this really worth it? that is the problem of whoever needs this functionality. Ah. Now this is exactly what reminded me of the ARM Linux ML in the first place :( Seeing that I do not learn anything new here, I'll stop. Cheers Detlev -- We can forgive a man for making a useful thing as long as he does not admire it. The only excuse for making a useless thing is that one admires it intensely. --- Oscar Wilde -- DENX Software Engineering GmbH, MD: Wolfgang Denk Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: d...@denx.de ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Hi Scott, On Tue, May 12, 2009 at 10:48:28AM +0200, Detlev Zundel wrote: Can you please explain to me, why you think it to be reasonable to demand providing an initramfs in the order of 100s of k to set an attribute of a hardware device which has its own driver? Kind of defeats any space-usage argument against putting device trees in ARM instead... Good point. Apart from being constantly repeated, I do not understand this reasoning at all. My (old-school) belief was that an operating system deals with abstracting the hardware thus userspace does not need to (nor should) know too many hw details. There's more to the operating system than the kernel. Sure - but how does this affect the question at hand? Cheers Detlev -- Practice random senselessness and act kind of beautiful. -- DENX Software Engineering GmbH, MD: Wolfgang Denk Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: d...@denx.de ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Hi Mike, I never ran across this outside of linux arm where people seriously repeat the statement over and over that a whole initramfs with an initial userspace and a pivot_root is plausible for nfs root whilst patches of a few lines float around doing the same in the kernel. i never said that a completely populated initramfs was needed. it can be accomplished pretty easily with very little code. Sure, but still, how big is such an initramfs, do you have any numbers here? And how many lines of C code does it take to set the MAC in the kernel? Is this really worth it? Cheers Detlev -- Basically, Barnes Noble separates things by how old they are -- current stuff is Fiction, stuff from 20 years ago is Literature, stuff from 100 years ago is Classics, stuff from 400 years ago is Shakespeare [..] and stuff from 2000 years ago is History. -- James Kibo Parry in kibo-1207032212000...@10.0.1.2 -- DENX Software Engineering GmbH, MD: Wolfgang Denk Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: d...@denx.de ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Wednesday 13 May 2009 04:41:29 Detlev Zundel wrote: I never ran across this outside of linux arm where people seriously repeat the statement over and over that a whole initramfs with an initial userspace and a pivot_root is plausible for nfs root whilst patches of a few lines float around doing the same in the kernel. i never said that a completely populated initramfs was needed. it can be accomplished pretty easily with very little code. Sure, but still, how big is such an initramfs, do you have any numbers here? And how many lines of C code does it take to set the MAC in the kernel? Is this really worth it? that is the problem of whoever needs this functionality. -mike ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Hi, how do set a mac for NFS Rootfs? use initramfs don't you think it's overkill to use a initramfs just for set a mac address?? no, i think it's perfectly reasonable. and considering you have no other option here that'll get merged ... Can you please explain to me, why you think it to be reasonable to demand providing an initramfs in the order of 100s of k to set an attribute of a hardware device which has its own driver? Apart from being constantly repeated, I do not understand this reasoning at all. My (old-school) belief was that an operating system deals with abstracting the hardware thus userspace does not need to (nor should) know too many hw details. Knowing that there is not a clear distinction line, I still fail to see why a mac address of a network interface should be handled by userspace. Can someone enlighten me here? Thanks Detlev -- Ftpd never switches uid and euid, it uses setfsuid(2) instead. The main reason is that uid switching has been exploited in several breakins, but the sheer ugliness of uid switching counts too. -- pure-ftpd(8) -- DENX Software Engineering GmbH, MD: Wolfgang Denk Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: d...@denx.de ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Tuesday 12 May 2009 04:48:28 Detlev Zundel wrote: how do set a mac for NFS Rootfs? use initramfs don't you think it's overkill to use a initramfs just for set a mac address?? no, i think it's perfectly reasonable. and considering you have no other option here that'll get merged ... Can you please explain to me, why you think it to be reasonable to demand providing an initramfs in the order of 100s of k to set an attribute of a hardware device which has its own driver? Apart from being constantly repeated, I do not understand this reasoning at all. My (old-school) belief was that an operating system deals with abstracting the hardware thus userspace does not need to (nor should) know too many hw details. Knowing that there is not a clear distinction line, I still fail to see why a mac address of a network interface should be handled by userspace. Can someone enlighten me here? no one said it must be done in userspace, that was just one method for doing it. read the FAQ for other possibilities. -mike ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Hi Mike, On Tuesday 12 May 2009 04:48:28 Detlev Zundel wrote: how do set a mac for NFS Rootfs? use initramfs don't you think it's overkill to use a initramfs just for set a mac address?? no, i think it's perfectly reasonable. and considering you have no other option here that'll get merged ... Can you please explain to me, why you think it to be reasonable to demand providing an initramfs in the order of 100s of k to set an attribute of a hardware device which has its own driver? Apart from being constantly repeated, I do not understand this reasoning at all. My (old-school) belief was that an operating system deals with abstracting the hardware thus userspace does not need to (nor should) know too many hw details. Knowing that there is not a clear distinction line, I still fail to see why a mac address of a network interface should be handled by userspace. Can someone enlighten me here? no one said it must be done in userspace, that was just one method for doing it. read the FAQ for other possibilities. No you lost me completely. The question cited above was whether you find it plausible to use initramfs - and thus userspace - to set a mac address. You answered that you find this plausible. This is what I do not understand. I never ran across this outside of linux arm where people seriously repeat the statement over and over that a whole initramfs with an initial userspace and a pivot_root is plausible for nfs root whilst patches of a few lines float around doing the same in the kernel. Now proof by repetition is not my preferred form of understanding a problem, so I was asking for an explanation from someone arguing along the same lines. Cheers Detlev -- We can forgive a man for making a useful thing as long as he does not admire it. The only excuse for making a useless thing is that one admires it intensely. --- Oscar Wilde -- DENX Software Engineering GmbH, MD: Wolfgang Denk Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: d...@denx.de ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Tue, 12 May 2009, Detlev Zundel wrote: I never ran across this outside of linux arm where people seriously repeat the statement over and over that a whole initramfs with an initial userspace and a pivot_root is plausible for nfs root whilst patches of a few lines float around doing the same in the kernel. Now proof by repetition is not my preferred form of understanding a problem, so I was asking for an explanation from someone arguing along the same lines. The wiki page that explains the problem and provides the suggested solutions does not repeat that same mantra. It mentions the initramfs method as one way out of several - I don't think it mentions that solution as particularly better in any way. Wolfgang also repeated (most of) them here just a few mails ago in this thread. -- / daniel.haxx.se ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Tue, May 12, 2009 at 10:48:28AM +0200, Detlev Zundel wrote: Can you please explain to me, why you think it to be reasonable to demand providing an initramfs in the order of 100s of k to set an attribute of a hardware device which has its own driver? Kind of defeats any space-usage argument against putting device trees in ARM instead... Apart from being constantly repeated, I do not understand this reasoning at all. My (old-school) belief was that an operating system deals with abstracting the hardware thus userspace does not need to (nor should) know too many hw details. There's more to the operating system than the kernel. -Scott ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Tue, May 12, 2009 at 02:04:46AM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote: i dont know why you're emphasizing mainline. if it's your board and if you implement the change correctly, submit it to mainline. because I already known RMK answer NACK So go to RMK and say you need device trees because you got NACKed in u-boot. :-) -Scott ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Tuesday 12 May 2009 10:18:25 Detlev Zundel wrote: Hi Mike, On Tuesday 12 May 2009 04:48:28 Detlev Zundel wrote: how do set a mac for NFS Rootfs? use initramfs don't you think it's overkill to use a initramfs just for set a mac address?? no, i think it's perfectly reasonable. and considering you have no other option here that'll get merged ... Can you please explain to me, why you think it to be reasonable to demand providing an initramfs in the order of 100s of k to set an attribute of a hardware device which has its own driver? Apart from being constantly repeated, I do not understand this reasoning at all. My (old-school) belief was that an operating system deals with abstracting the hardware thus userspace does not need to (nor should) know too many hw details. Knowing that there is not a clear distinction line, I still fail to see why a mac address of a network interface should be handled by userspace. Can someone enlighten me here? no one said it must be done in userspace, that was just one method for doing it. read the FAQ for other possibilities. No you lost me completely. The question cited above was whether you find it plausible to use initramfs - and thus userspace - to set a mac address. You answered that you find this plausible. This is what I do not understand. I never ran across this outside of linux arm where people seriously repeat the statement over and over that a whole initramfs with an initial userspace and a pivot_root is plausible for nfs root whilst patches of a few lines float around doing the same in the kernel. i never said that a completely populated initramfs was needed. it can be accomplished pretty easily with very little code. -mike ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
[U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
init it at the eth_initialize this will allow to add later a hook to update the enetadd hw storage when it's updated in env or when the env is saved Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD plagn...@jcrosoft.com Cc: Ben Warren biggerbadder...@gmail.com --- include/net.h |2 ++ net/eth.c |4 2 files changed, 6 insertions(+), 0 deletions(-) diff --git a/include/net.h b/include/net.h index 5a1d36e..2ed6263 100644 --- a/include/net.h +++ b/include/net.h @@ -93,6 +93,7 @@ enum eth_state_t { }; struct eth_device { + int num; char name[NAMESIZE]; unsigned char enetaddr[6]; int iobase; @@ -105,6 +106,7 @@ struct eth_device { #ifdef CONFIG_MCAST_TFTP int (*mcast) (struct eth_device*, u32 ip, u8 set); #endif + int (*set_hw_enetaddr) (struct eth_device*, bd_t*); struct eth_device *next; void *priv; }; diff --git a/net/eth.c b/net/eth.c index c6fa5b9..99bb6c2 100644 --- a/net/eth.c +++ b/net/eth.c @@ -217,6 +217,8 @@ int eth_initialize(bd_t *bis) if (eth_number) puts (, ); + dev-num = eth_number; + printf(%s, dev-name); if (ethprime strcmp (dev-name, ethprime) == 0) { @@ -240,6 +242,8 @@ int eth_initialize(bd_t *bis) } memcpy(dev-enetaddr, env_enetaddr, 6); + if(dev-set_hw_enetaddr) + dev-set_hw_enetaddr(dev, bis); } eth_number++; -- 1.6.1.3 ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Monday 11 May 2009 02:39:26 Jean-Christophe PLAGNIOL-VILLARD wrote: this will allow to add later a hook to update the enetadd hw storage when it's updated in env or when the env is saved there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. -mike signature.asc Description: This is a digitally signed message part. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On 03:48 Mon 11 May , Mike Frysinger wrote: On Monday 11 May 2009 02:39:26 Jean-Christophe PLAGNIOL-VILLARD wrote: this will allow to add later a hook to update the enetadd hw storage when it's updated in env or when the env is saved there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. Best Regards, J. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511120809.ga1...@game.jcrosoft.org you wrote: there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. That's why I have to NAK this. The problem should be fixed in the Linux kernel. I know that RMK does not alow for things like a MAC ATAG or similar, so the best you can do is hope for quick approach of device tree based drivers. Until then, you probably have to resort to out-of-tree (Linux) solutions like everybody else who hit this problem before. 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: w...@denx.de Nobody trips over mountains. It is the small pebble that causes you to stumble. Pass all the pebbles in your path and you will find you have crossed the mountain. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On 15:26 Mon 11 May , Wolfgang Denk wrote: Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511120809.ga1...@game.jcrosoft.org you wrote: there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr Best Regards, J. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511142454.gb1...@game.jcrosoft.org you wrote: Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr Yes, I perfectly understood this. You did not have to explain it again. Did you understand what I wrote, too? 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: w...@denx.de Minds are like parachutes - they only function when open. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Monday 11 May 2009 10:24:54 Jean-Christophe PLAGNIOL-VILLARD wrote: On 15:26 Mon 11 May , Wolfgang Denk wrote: Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511120809.ga1...@game.jcrosoft.org you wrote: there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr then we agree you're doing it wrong and this patch is not needed/wanted there is this FAQ, but i think we can add more info to it: http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux -mike signature.asc Description: This is a digitally signed message part. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On 18:01 Mon 11 May , Wolfgang Denk wrote: Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511142454.gb1...@game.jcrosoft.org you wrote: Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr Yes, I perfectly understood this. You did not have to explain it again. So no I do not understand why you nack it Best Regards, J. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On 12:24 Mon 11 May , Mike Frysinger wrote: On Monday 11 May 2009 10:24:54 Jean-Christophe PLAGNIOL-VILLARD wrote: On 15:26 Mon 11 May , Wolfgang Denk wrote: Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511120809.ga1...@game.jcrosoft.org you wrote: there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr then we agree you're doing it wrong and this patch is not needed/wanted there is this FAQ, but i think we can add more info to it: http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux It does not work ether how do you deal with multiple nic? how do set a mac for NFS Rootfs? how do you boot a *mainline* kernel? Best Regards, J. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Monday 11 May 2009 12:37:28 Jean-Christophe PLAGNIOL-VILLARD wrote: On 12:24 Mon 11 May , Mike Frysinger wrote: On Monday 11 May 2009 10:24:54 Jean-Christophe PLAGNIOL-VILLARD wrote: On 15:26 Mon 11 May , Wolfgang Denk wrote: In message Jean-Christophe wrote: there is no need for that on purpose. the network layer will recheck the environment every time it gets initialized, and it gets initialized with every network step. this is already documented in doc/README.enetaddr. Unfortunatly we need it to set the enetaddr in the register of the soc evenif there is no ethernet transfer as for the macb on at91 or the dm9000. This how the kernel pass the mac address for the bootloader. You mean, this is how U-Boot is supposed to pass the MAC address to the Kernel, right? Well, strictly speaking this is (1) in direct violation of U-Boot's design principles which say NOT to initialize things it does not use itself, and (2) a bug in the Linux kernel that should be fixed there. In this case it may be just writing to a register without further initialization of the hardware, but if we allow for that you can bet that tomorrow someone will try and add code that requires more extensive initialization of the hardware for the same purpose, and he will argue that his code must go in because there is preexisting code that does the same. this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr then we agree you're doing it wrong and this patch is not needed/wanted there is this FAQ, but i think we can add more info to it: http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux It does not work ether your solution blatantly ignores the issues raised in the FAQ. you cant push a change without clearing addressing those. how do you deal with multiple nic? the ATAG/FDT tree handles most arches. for the rest you have two major issues to address: - how do i pass it to the next OS ? * have the OS read the u-boot environment * have the OS read the command line - how does the OS set it ? * add a board hook to set it * set it via user space and normal methods (`ifconfig hw`) how do set a mac for NFS Rootfs? use initramfs how do you boot a *mainline* kernel? i dont know why you're emphasizing mainline. if it's your board and if you implement the change correctly, submit it to mainline. -mike signature.asc Description: This is a digitally signed message part. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Dear Jean-Christophe PLAGNIOL-VILLARD, In message 20090511163728.gb32...@game.jcrosoft.org you wrote: then we agree you're doing it wrong and this patch is not needed/wanted there is this FAQ, but i think we can add more info to it: http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux It does not work ether What is it in this sentence? You start a discussion here which is off topic on the U-Boot list. It deals with a problem that is caused by the ARM Linuxkernel design. It's a very old problem, that has been discussed many times before, and it's unlikely that we willsolve it here. If you want to solve it, then please explain it to RMK. how do you deal with multiple nic? Other (more reasonable architectures) use a device tree (current PowerPC), a binary data structure (old PowerPC), or command line arguments. how do set a mac for NFS Rootfs? Other (more reasonable architectures) use a device tree (current PowerPC), a binary data structure (old PowerPC), or command line arguments. how do you boot a *mainline* kernel? Other (more reasonable architectures) use a device tree (current PowerPC), a binary data structure (old PowerPC), or command line arguments. The fact that mainline ARM has serious deficiencies is another story. Or can you explain why the ARM kernel cannot load and uncompress a ramdisk image from NOR flash, why we instead have to load it to RAM first, adding a completely unnecessary copy operation? It's just one of the issues with the ARM Linux kernel that need to be fixed *there*. But, as mentioned, this is off topic here. 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: w...@denx.de The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of doubts. - Bertrand Russell ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
Dear Mike, in message 200905111324.58673.vap...@gentoo.org you wrote: ive filled it out some more ... i'm assuming this is the best place for such information Thanks. also, i dont really know this wiki, so i cheated with the FDT wiki link and used the full path since i was unable to guess at how to make it work relatively. See http://www.denx.de/wiki/TWiki/TextFormattingRules next time you have such an issue. I fixed it. 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: w...@denx.de Everyting looks interesting until you do it. Then you find it's just another job. - Terry Pratchett, _Moving Pictures_ ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Monday 11 May 2009 12:24:31 Mike Frysinger wrote: there is this FAQ, but i think we can add more info to it: http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux ive filled it out some more ... i'm assuming this is the best place for such information also, i dont really know this wiki, so i cheated with the FDT wiki link and used the full path since i was unable to guess at how to make it work relatively. -mike signature.asc Description: This is a digitally signed message part. ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
Re: [U-Boot] [PATCH 1/4] net: extend the netdev to have a common way to set the hw mac address
On Monday 11 May 2009 20:04:46 Jean-Christophe PLAGNIOL-VILLARD wrote: this exactly what is done only write mac hw register without init the eth stack before we do need to init the mac and detect the phy if we want to store now we have to call the set_hw_enetaddr then we agree you're doing it wrong and this patch is not needed/wanted there is this FAQ, but i think we can add more info to it: http://www.denx.de/wiki/view/DULG/EthernetDoesNotWorkInLinux It does not work ether your solution blatantly ignores the issues raised in the FAQ. you cant push a change without clearing addressing those. how do you deal with multiple nic? the ATAG/FDT tree handles most arches. both not possible for the rest you have two major issues to address: - how do i pass it to the next OS ? * have the OS read the u-boot environment * have the OS read the command line - how does the OS set it ? * add a board hook to set it * set it via user space and normal methods (`ifconfig hw`) how do set a mac for NFS Rootfs? use initramfs don't you think it's overkill to use a initramfs just for set a mac address?? no, i think it's perfectly reasonable. and considering you have no other option here that'll get merged ... -mike ___ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot