From: Simon Goldschmidt Date: Wed, 11 Sep 2019 22:24:20 +0200 Subject: [U-Boot] Issue in u-boot; TFTP error: trying to overwrite reserved memory... In-Reply-To: References: <4e43f516-5452-d8f0-cd09-82e2bb7d9dda@gmail.com> Message-ID: List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: u-boot@lists.denx.de On 11.09.19 21:33, Moses Christopher wrote: > > On Wed, 11 Sep, 2019, 9:12 PM Simon Goldschmidt, > > wrote: > > > > On 11.09.19 20:59, Moses Christopher wrote: > > > > On Wed, 11 Sep, 2019, 4:43 PM Simon Goldschmidt, > > > > >> wrote: > > > >     On Wed, Sep 11, 2019 at 11:44 AM Moses Christopher > >      > >> > >     wrote: > >      > > >      > > >      > > >      > On Wed, 11 Sep, 2019, 10:32 AM Simon Goldschmidt, > >      > >      >> wrote: > >      >> > >      >> On Mon, Sep 9, 2019 at 11:29 AM Moses Christopher > >      >> > >      >> wrote: > >      >> > > >      >> > Hi Simon, > >      >> > > >      >> > > >      >> > Thanks for the prompt reply. > >      >> > > >      >> > On Fri, 6 Sep, 2019, 8:13 AM Simon Goldschmidt, > >      > >      >> wrote: > >      >> > > >      >> > Hi, > >      >> > > >      >> > On Thu, Sep 5, 2019 at 4:14 PM Moses Christopher > >      >> > > >      >> wrote: > >      >> > > Hello together, > >      >> > > > >      >> > > I was trying to build u-boot and spl for the arm > target and > >     tried to boot via usb-ethernet. > >      >> > > I found an issue with one of the commit made in the > early 2019, > >      >> > > http://patchwork.ozlabs.org/patch/1024795/ > >      >> > > > >      >> > > When using this CONFIG_LMB the max_size or the > >     lmb_get_free_size(&lmb, load_addr); returns 0, no matter what. > >      >> > > And it triggers the following error, > >      >> > > TFTP error: trying to overwrite reserved memory... > >      >> > > I did a quick fix by adding #undef CONFIG_LMB in the > file, > >     net/tftp.c > >      >> > > So, I would like to know why this doesn’t work as it was > >     working before applying this patch ? > >      >> > > >      >> > Can you add "#define DEBUG" as the first line in > 'lib/lmb.c'? That > >      >> > should give you debug > >      >> > output when lmb is used. > >      >> > > >      >> > > >      >> > I did add DEBUG macro to lmb.c but the function having the > >     debug messages isn't getting called. I suppose it was from > fs/fs.c > >      >> > >      >> Right, tftp.c is missing the call to that funcftion. > Could you > >     add the > >      >> call to 'lmb_dump_all(&lmb);' > >      >> right below 'lmb_init_and_reserve()' in tftp.c? > >      >> > >      >> That should give you the output required. And while at > it, tell us > >      >> what 'load_addr' is set to > >      >> (by adding a printf in tftp.c, too). > >      >> > >      >> Thanks, > >      >> Simon > >      > > >      > > >      > Thanks for your patience and time. > >      > > >      > Please find the log below, > >      > > >      > > >      > > >      > DHCP client bound to address 172.17.0.2 (1285 ms) > >      > > >      > Using usb_ether device > >      > > >      > TFTP from server 172.17.0.1; our IP address is 172.17.0.2 > >      > > >      > Filename 'u-boot.img'. > >      > > >      > lmb_dump_all: > >      > > >      >     memory.cnt             = 0x0 > >      > > >      >     memory.size            = 0xx > >      > > >      > > >      > > >      >     reserved.cnt           = 0x0 > >      > > >      >     reserved.size          = 0xx > >      > > >      > load_addr: 0x82000000 > >      > > >      > > >      > > >      > TFTP error: trying to overwrite reserved memory... > >      > > >      > Problem booting with BOOTP > >      > > >      > > >      > > >      > In my u-boot it shows the DRAM size properly as 256MiB > >      > > >      > So, do I need to configure my RAM size in SPL stage as > well, such > >     that SPL is aware of the memory size ? > > > >     Ehrm, are you doing this from SPL? > > > > > > Yes, SPL loads u-boot onto the DRAM from Network, when the NAND > is empty. > > > > By flashing the same MLO and u-boot binaries onto the NAND, and then > > doing *dhcp MLO *from u-boot prompt gives the following log, > > > > Using usb_ether device > > TFTP from server 172.17.0.1; our IP address is 172.17.0.2 > > Filename 'MLO'. > > *lmb_dump_all:** > > memory.cnt = 0x1 > > memory.size = 0x0 > > memory.reg[0x0].base = 0x80000000 > > .size = 0x10000000 > > > > reserved.cnt = 0x1 > > reserved.size = 0x0 > > reserved.reg[0x0].base = 0x8df2ab98 > > .size = 0x20d5468* > > load_addr: 0x80200000 > > OK, so that works for a start ;-) > > > > > > > > >     You need the RAM size for 'lmb_init_and_reserve()' to read, yes. > >     Otherwise it can't know > >     where to safely allocate things. > > > >     Regards, > >     Simon > > > > > > Would you think, SPL should also do the same or is this valid > only for > > u-boot ? > > Yes, SPL should do the same. > > > > > Because, during SPL stage, the RAM would be mostly free anyway, > so would > > you think, we can just use this lmb check only for u-boot but not > for SPL ? > > No. SPL still needs this check. On some platforms, SPL runs with BSS, > stack or heap in RAM when executing this, so this is still an attack > vector that should be prevented. > > Even if SPL does not use DRAM (only SRAM), we still have to check that > we only write into valid DRAM addresses. E.g. a file larger than the > DRAM would write to whatever comes next in the address range... > > > Ah, that really makes sense. Sorry, if I'm asking too many basic questions. > > Which platform are you running this on, does SPL use a devicetree? In > any case, you need your SPL to provide the valid RAM range. > > > I'm using TI's AM335x based custom board, yes the SPL has the minimal > devicetree embedded. > > If I understand it correctly, I've to configure my RAM information > properly such that the SPL is able to get the RAM size, reserved size, > free size, etc.., Yes. I would have expected that to come from the devicetree though... I'd appreciate if you could come back here with a solution what exactly was missing to get it running, if only for the archives that other users find the result. Regards, Simon > > Thanks again for you patience and time. > > Regards, > Moses Christopher > > > Regards, > Simon > > > > > Regards, > > Moses Christopher > > > >      > > > > >      > > >      >> > >      >> > > >      >> > FYI, > >      >> > I'm trying to load SPL and uboot on RAM, using USB-ETH. > Also > >     the environment is not stored separately, neither the device > tree. > >      >> > > >      >> > > >      >> > The lmb code works by getting the RAM size, adding reserved > >     areas and then only > >      >> > allowing allocations in non-reserved areay. However, > the RAM > >     size is > >      >> > not fully used > >      >> > depending on some config options and/or environment > variables. > >     There's possibly > >      >> > something wrong in your configuration around that. > >      >> > > >      >> > > >      >> > Because, earlier to this patch, net/tftp.c isn't actually > >     checking for the reserved memory regions and is able to > download the > >     files properly on the RAM and it works. I know, that's not a good > >     approach, hence you've made the necessary changes to correct it. > >      >> > > >      >> > Could you kindly provide me some information, where I > can read > >     more about the reserved memory regions and how exactly some > region > >     is treated as reserved region ? > >      >> > > >      >> > Also, it'd be great if you could provide some information > >     related to the configuration of Reserved and free addresses > of RAM. > >      >> > > >      >> > Thank you for your patience and time. > >      >> > > >      >> > Regards, > >      >> > Simon > >      >> > > >      >> > > > >      >> > > Best regards, > >      >> > > Moses Christopher > >      >> > > >      >> > Best regards, > >      >> > Moses Christopher > > >