http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132303 <pre>Hi Rob, On Fri, May 25, 2012 at 3:47 PM, Rob Herring &lt;robherring2&lt; at &gt;gmail.com&gt; wrote: It applies now. Thanks, -Joe </pre> Joe Hershberger 2012-05-25T21:44:35 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132302 <pre>U-Boot port is based on sources forked from GRUB-0.97 by Sun in 2004, which can be found here: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/grub/grub-0.97/stage2/zfs-include/zfs.h Released by Sun for GRUB under the license: * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. GRUB official releases include ZFS in version: ftp://alpha.gnu.org/gnu/grub/grub-1.99~rc1.tar.gz And patched against GRUB Bazaar repository for ashift fixes (4KB HDDs) more conveniently found at github: https://github.com/pendor/grub-zfs/commit/e7b6ef3ac3b9685ac4c394c897b1d4221b7381f1 Signed-off-by: Jorgen Lundman &lt;lundman&lt; at &gt;lundman.net&gt; --- v4: * Add doc/README.zfs documentation v3: * add missing patch revision history (this text) * Submitted as single patch per Wolfgang Denk instructions v2: * Keep Makefi</pre> Jorgen Lundman 2012-05-25T02:11:46 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132301 <pre>Hi Michael, On Fri, May 25, 2012 at 2:54 PM, Michael Walle &lt;michael&lt; at &gt;walle.cc&gt; wrote: OK. Since it is specific to your board support anyway, that seems fine. It does, yes. There is an inherent discrepancy, though, given that one is used and the other is not. I have no problem adding the accessor if there are clients for it to justify its code space. Not a problem at all. -Joe </pre> Joe Hershberger 2012-05-25T20:50:00 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132300 <pre>From: Rob Herring &lt;rob.herring&lt; at &gt;calxeda.com&gt; Add a new command "sysboot" which parses syslinux menu files and boots using kernel and initrd specified by menu files. The operation is similar to "pxe boot" except local files on ext2 or fat filesystem are parsed. Signed-off-by: Rob Herring &lt;rob.herring&lt; at &gt;calxeda.com&gt; --- v2: - rebase to current mainline - fix compile if ext2 or fat is disabled common/cmd_pxe.c | 141 ++++++++++++++++++++++++++++++++++++++++++++++++++---- include/common.h | 6 +++ 2 files changed, 139 insertions(+), 8 deletions(-) diff --git a/common/cmd_pxe.c b/common/cmd_pxe.c index 0fc863b..2b1115d 100644 --- a/common/cmd_pxe.c +++ b/common/cmd_pxe.c &lt; at &gt;&lt; at &gt; -132,6 +132,51 &lt; at &gt;&lt; at &gt; static int get_bootfile_path(const char *file_path, char *bootfile_path, return 1; } +static int (*do_getfile)(char *file_path, char *file_addr); + +static int do_get_tftp(char *file_path, char *file_addr) +{ +char *tftp_argv[] = {"tftp", NULL, NULL, NULL}; + +tftp_argv[1] = file_addr; +tftp_argv[2] = file_path;</pre> Rob Herring 2012-05-25T20:47:39 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132299 <pre>From: Rob Herring &lt;rob.herring&lt; at &gt;calxeda.com&gt; Include files outside of a menu were not getting included and parsed. Signed-off-by: Rob Herring &lt;rob.herring&lt; at &gt;calxeda.com&gt; --- v2: - ensure include file load address is aligned common/cmd_pxe.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/common/cmd_pxe.c b/common/cmd_pxe.c index b3c1f67..5949b24 100644 --- a/common/cmd_pxe.c +++ b/common/cmd_pxe.c &lt; at &gt;&lt; at &gt; -1110,6 +1110,11 &lt; at &gt;&lt; at &gt; static int parse_pxefile_top(char *p, struct pxe_menu *cfg, int nest_level) break; +case T_INCLUDE: +err = handle_include(&amp;p, b + ALIGN(strlen(b), 4), cfg, +nest_level + 1); +break; + case T_PROMPT: err = parse_integer(&amp;p, &amp;cfg-&gt;prompt); break; </pre> Rob Herring 2012-05-25T20:43:16 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132298 <pre>Since run_command() and run_command_list() are important and a little confusing, add some basic tests to check that the behaviour is correct. Note: I am not sure that this should be committed, nor where it should go in the source tree. Comments welcome. To run the unit tests use the ut_cmd command available in sandbox: make sandbox_config make ./u-boot -c ut_cmd (To test both hush and built-in parsers, you need to manually change CONFIG_SYS_HUSH_PARSER in include/configs/sandbox.h and build/run again) Signed-off-by: Simon Glass &lt;sjg&lt; at &gt;chromium.org&gt; --- Changes in v3: - Add some unit tests for run_command() and run_command_list() Changes in v4: - Add some unit tests for run_command() and run_command_list() - Fix code nits in unit test and test/Makefile Makefile | 1 + test/Makefile | 45 +++++++++++++++++++++++++++ test/command_ut.c | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 133 insertions(+), 0 deletions(-) create mode 100644 test/Makefile create </pre> Simon Glass 2012-05-25T20:22:03 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132297 <pre>This new function runs a list of commands separated by semicolon or newline. We move this out of cmd_source so that it can be used by other code. The PXE code also uses the new function. Suggested-by: Michael Walle &lt;michael&lt; at &gt;walle.cc&gt; Signed-off-by: Simon Glass &lt;sjg&lt; at &gt;chromium.org&gt; --- Changes in v3: - Added a few more comments on the need for malloc() - Change PXE's printf() to a debug() - Fix bug in built-in run_command() - Fix commit message to mention ; and \n - Move run_command_list() comment to header file Changes in v4: - Remove length param from run_command_list() call in PXE common/cmd_pxe.c | 20 ++---------- common/cmd_source.c | 49 +---------------------------- common/main.c | 85 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/common.h | 13 ++++++++ 4 files changed, 102 insertions(+), 65 deletions(-) diff --git a/common/cmd_pxe.c b/common/cmd_pxe.c index b3c1f67..55d1c05 100644 --- a/common/cmd_pxe.c +++ b/common/cmd_pxe.c &lt; at &gt;&lt; at &gt; -515,33 +515,19 &lt; at &gt;&lt; at &gt; static void lab</pre> Simon Glass 2012-05-25T20:22:01 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132296 <pre>Any environment variable can hold commands to be executed by the 'run' command. The environment variables preboot, bootcmd and menucmd have special code for triggering execution in certain circumstances. We adjust these calls to use run_command_list() instead of run_command(). This change permits these variables to have embedded newlines so that they work the same as the 'source' command. Signed-off-by: Simon Glass &lt;sjg&lt; at &gt;chromium.org&gt; --- Changes in v2: - Update commit message to be less confusing Changes in v3: - Remove unneeded len parameter in calls to run_command_list common/main.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/common/main.c b/common/main.c index d96ba0a..9405922 100644 --- a/common/main.c +++ b/common/main.c &lt; at &gt;&lt; at &gt; -334,7 +334,7 &lt; at &gt;&lt; at &gt; void main_loop (void) int prev = disable_ctrlc(1);/* disable Control C checking */ # endif -run_command(p, 0); +run_command_list(p, -1, 0); # ifdef CONFIG_AUTOBOOT_KEYED disable_ctrlc(prev);/* restore Control C</pre> Simon Glass 2012-05-25T20:22:02 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132295 <pre>Am Freitag 25 Mai 2012, 20:50:09 schrieb Joe Hershberger: [..snip..] Hi Joe, thanks for the review. I think i'll drop the dynamic "ethernet variable name" support entirely. Two reasons: - it isn't really dynamic, eg. the function always evaluates the name to "ethaddr" - back then when i wrote the support for the board the name was "eth1addr" and mike suggested to use the by index function for the getter and a new function for the setter. But i guess its not worth the hassle ;) btw, imho your solution introduces a discrepancy between the setter and getter. i hope it doesnt bother you if i don't adapt your solution. But thanks for the work. </pre> Michael Walle 2012-05-25T19:54:51 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132294 <pre> Hi Thierry, On 25.05.2012 19:40, Thierry Reding wrote: As an aside: A pulse wont be enough in this specific case as a pull-down on the carrier would lead to the nRST going low thus resetting the baseboard peripherals. Plus there is a status LED on the Tamonten Evaluation carrier that would then indicate the red "system not up". Cheers, Kai </pre> Kai Poggensee 2012-05-25T18:56:09 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132293 <pre>Hi Rob, On Wed, May 23, 2012 at 6:02 PM, Joe Hershberger &lt;joe.hershberger&lt; at &gt;gmail.com&gt; wrote: Any luck figuring out why the patch at http://patchwork.ozlabs.org/patch/149285/ will not apply? Perhaps just rebase (since that works for you) and resend. Thanks, -Joe </pre> Joe Hershberger 2012-05-25T18:56:56 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132292 <pre>Hi Michael, On Thu, May 17, 2012 at 3:43 PM, Michael Walle &lt;michael&lt; at &gt;walle.cc&gt; wrote: How about something like this instead... -----------8&lt;-----------8&lt;-----------8&lt;-----------8&lt;----------- diff --git a/include/net.h b/include/net.h index a092f29..5bc4b0d 100644 --- a/include/net.h +++ b/include/net.h &lt; at &gt;&lt; at &gt; -600,6 +600,22 &lt; at &gt;&lt; at &gt; static inline int is_broadcast_ether_addr(const u8 *addr) } /* + * Create a variable name for the ethernet address + * + * &lt; at &gt;param base_name Input string constant (typically "eth") prepended to name + * &lt; at &gt;param index Index to be added to the name if not 0 + * &lt; at &gt;param ethaddr_var Output string variable name for MAC address + */ +static inline void eth_enetaddr_var_by_index(const char *base_name, int index, +char *ethaddr_var) +{ +if (index) +sprintf(ethaddr_var, "%s%daddr", base_name, index); +else +sprintf(ethaddr_var, "%saddr", base_name); +} + +/* * is_valid_ether_addr - Determine if the given Ethernet address is valid * &lt; at &gt;addr: Pointer to a six-byte array containing the Eth</pre> Joe Hershberger 2012-05-25T18:50:09 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132291 <pre> This isn't true. But I'll still have to respin because of the MAINTAINERS entry and the other points that you brought up, so let me know if I should add a patch for Seaboard. Thierry _______________________________________________ U-Boot mailing list U-Boot&lt; at &gt;lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot </pre> Thierry Reding 2012-05-25T17:59:30 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132290 <pre> Acked-by: Joe Hershberger &lt;joe.hershberger&lt; at &gt;gmail.com&gt; </pre> Joe Hershberger 2012-05-25T17:52:18 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132289 <pre> I believe that even a pulse would be enough for the undelying mechanisms to take effect, but I'll have to check with our hardware engineers to be sure. You're probably right, though, that keeping it requested until control is handed over to the kernel may be safer. I'll update the patch. Thierry _______________________________________________ U-Boot mailing list U-Boot&lt; at &gt;lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot </pre> Thierry Reding 2012-05-25T17:40:03 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132288 <pre> I just noticed that I forgot to add an entry to MAINTAINERS for the TEC. Furthermore I'll need to rebase the patches onto the Tegra branch due to the major restructuring of the configuration files. Since I'll respin the series anyway, I could add a patch to make the corresponding change to Seaboard. I can't really comment on this because I've never used the Seaboard. Maybe Simon (CC'ed) can comment on this. If we decide to completely remove the SPI support then the above patch becomes obsolete. Thierry _______________________________________________ U-Boot mailing list U-Boot&lt; at &gt;lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot </pre> Thierry Reding 2012-05-25T17:31:34 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132287 <pre>Hi Thanks for the answer! Wonderfull! It's a nice coincidence that you are writing from a Samsung address. Because the platform I'm trying to get u-boot working on is the GT-9100G. (part of the Samsung Galaxy S2 series) The GT-9100G is the only Galaxy S2 with an OMAP4430 on it, and after soldering me a debugging cabel for getting access to the UART [1] , I'm now tinkering around on making u-boot start on the device. I must say, of all the smartphones I received during the past few years, this one makes to me the best impression from the point of electrical engineering. Already the choice of the SoC (OMAP) which is my preferred chip because it's so well documented, but then also the choice of the other parts for which mostly datasheets are available. The more I analyze this specific hardware, the more I'm starting to love it. I owe the engineers who have designed this board a beer I guess :-) Maybe, if you are working within Samsung you could try to make the basic support for this device in u-boot? Thi</pre> David Lanzendörfer 2012-05-25T16:59:57 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132286 <pre> FYI, in the prior version, I'm seeing some issues with reading sub-directories on ext4 filesystem. When doing an ls some directories are present, but have a size of 0 and can't be listed. In Linux, they are fine. Rob </pre> Rob Herring 2012-05-25T16:57:41 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132285 <pre> That define is no longer used. I'd recommend explicitly either making your bootcmd define the entire command-line you want (by setting bootargs), or loading a script to do so. Aside from the comments I've already made, the series, Acked-by: Stephen Warren &lt;swarren&lt; at &gt;wwwdotorg.org&gt; </pre> Stephen Warren 2012-05-25T16:31:09 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132284 <pre> Do you really mean to free the GPIO here? While gpio_free() does not do this at present, it seems perfectly reasonable for someone to modify gpio_free() so that instead of leaving the HW in some random state when free, it actively reprograms the pin to be an input instead, since that's the most conflict-free setting when the pin is actively unused. </pre> Stephen Warren 2012-05-25T16:27:02 http://permalink.gmane.org/gmane.comp.boot-loaders.u-boot/132283 <pre> I assert that all early GPIO initialization should be reworked to use this new gpio_early_init(). In other words, now that there's a single generic gpio_early_init(), we should remove the following code right after it: #ifdef CONFIG_SPI_UART_SWITCH gpio_early_init_uart(); #else gpio_config_uart(); #endif and any boards that need to do any of that (which is just Seaboard I think) should do that from an implementation of the new generic gpio_early_init() in their board file. That way, common code won't have to know about crap like the SPI/UART muxing on Seaboard. That said, this need not be part of this patch or even this series, so I don't mean to nak this patch with these comments. And actually, I'd be all for ripping out all the SPI support on Seaboard and all related code anyway... </pre> Stephen Warren 2012-05-25T16:24:50 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.boot-loaders.u-boot