http://blog.gmane.org/gmane.linux.toybox hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.linux.toybox/439 <pre>Would you consider adding the busybox showkey, but with a user set time including 0 for immediate, rather than the default 10 seconds? We wish to add the ability of checking a &lt;ctrl&gt;, &lt;alt&gt;, etc. key in the init and control the response depending on the keycode for function, rather than expect them to wait 10 seconds each time. thanks scsijon </pre> scsijon 2013-05-15T01:04:33 http://comments.gmane.org/gmane.linux.toybox/432 <pre>Hi, Attached is a pretty simple implementation of lsub. It doesn't implement any of the flags/options, but does the raw output. I believe this is essentially the same functionality as what is available from busybox. Regards, Andre _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Andre Renaud 2013-05-14T05:56:00 http://comments.gmane.org/gmane.linux.toybox/429 <pre>http://landley.net/hg/toybox/rev/898 Design issue for today: replace if/else staircase with loop over a table. The existing option parsing in ifconfig_main() is a giant if/else staircase with repeated function calls. It makes multiple calls to strcmp() with mostly the same arguments, the first several of which call set_flags() (again with mostly the same arguments). The set_flags() function calls an ioctl to read an interface's flags, switch off one set of bits and switch on another set of bits, and then call a second ioctl to write those new flags to the interface. The existing logic calling set_flags() does a strcmp("string"), and then a second strcmp("-string") which mostly does the same thing but with the set/clear arguments reversed. So, new code: 1) Remove redundant curly bracket scope and move local variable declarations to the start of function. (Don't reindent everything, just the bits we're changing now. Leave the rest indented an extra two spaces so the diff doesn't lo</pre> Rob Landley 2013-05-13T02:20:44 http://comments.gmane.org/gmane.linux.toybox/428 <pre>I'm impressed: -rwxr-xr-x 1 143640 May 11 23:16 toybox -rwxr-xr-x 1 252748 May 10 21:32 toybox-895 -rwxr-xr-x 1 166499 May 11 23:16 toybox_unstripped (Coincidentally, this is the output of ls -lgo with coreutils; toybox appears to have no way to do a full ls with neither group nor owner...but I'd expect that one could deal with that via pipes.) No config changes, and the list of applets is constant: for i in ./toybox*; do $i |md5sum ; done 58671088549e63bde77622705dc677af - 58671088549e63bde77622705dc677af - 58671088549e63bde77622705dc677af - So a change of less than 30 lines nearly halved size. But gcc is warning about line 21 of lib/getmountlist.c: for (mtlist = 0; me = getmntent(fp); mtlist = mt) { Output of make: scripts/make.sh Make generated/config.h from .config. Extract configuration information from toys/*.c files... Generate headers from toys/*/*.c... generated/newtoys.h generated/globals.h generated/help.h Extract help text from Config.in. Library probe... Compile toybox... lib/getmountlis</pre> idunham-N9AOi2cAC9ZBDgjK7y7TUQ< at >public.gmane.org 2013-05-12T06:58:12 http://comments.gmane.org/gmane.linux.toybox/426 <pre>Hi, Stilling getting familiar with the toybox infrastructure. Here's a simple addition to begin with. To discuss: 1. for 'groups username' FSF's groups prints the username and a colon at the start of the line. The LSB does not specifically describes that. Should we stay compatible with GNU coreutils or not? 2. FSF's groups accepts more than one username, which is not conform the LSB's description, which states one optional user argument. 3. id is in posix/ and groups is from the LSB. If we decide to ignore coreutils, groups could move to lsb/ and call id -Gn [user]. Regards, Ivo _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Ivo van Poorten 2013-05-11T10:25:26 http://comments.gmane.org/gmane.linux.toybox/425 <pre>Hey Rob, I have a case, where setuid is required in nested case. nested case: lets say there are two commands in toybox called __acmd__ and __bcmd__. __acmd__ doesn't require setuid bit, whereas __bcmd__ does require setuid bit. Take a case where __acmd__ will run/launch __bcmd__. Now since __acmd__ doesn't need TOYFLAG_STAYROOT, this sets the euid to be of uid in toy_init() function. When __bcmd__ is run from __acmd__, this would have lost the setuid bit and hence fails to run due to this. Pls do give your opinion on the same. regards, Ashwini _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Ashwini Sharma 2013-05-10T08:29:46 http://comments.gmane.org/gmane.linux.toybox/420 <pre>Hi Rob, I think we should include toynet.h in toys.h, before we include #include "generated/newtoys.h" #include "generated/globals.h" so that whenever someone has a variable in GLOBALS ( ) whose type is declared in networking headers, like a __struct sockaddr_in __ will get the type, else he will have to keep the variable outside the GLOBALS macro. May be this can be included conditionaly, as done for CONFIG_HELP, or straight away. regards, Ashwini _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Ashwini Sharma 2013-05-03T06:05:13 http://comments.gmane.org/gmane.linux.toybox/418 <pre>Hello, the build of toybox was broken if CONFIG_TOYBOX_HELP was enabled but CONFIG_HELP was disabled (and help.h was not yet generated). The attached patch fixes it. Felix _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Felix Janda 2013-04-30T19:13:11 http://comments.gmane.org/gmane.linux.toybox/417 <pre>Sorry about that, version skew in the repository format. Should be working again now. Rob </pre> Rob Landley 2013-04-30T15:59:50 http://comments.gmane.org/gmane.linux.toybox/410 <pre>Hello all, I have a question about treating long(--) option in toybox. There are two style options - short(-) and long(--) - in linux system. ( http://unix.stackexchange.com/questions/21852/single-dashes-for-single-character-options-but-double-dashes-for-words ) I know toybox supports only short option, so toybox cannot treat '--help' option. Is there special reason for not supporting long option? Thanks. Kyungwan Han. _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Kyungwan Han 2013-04-29T06:17:46 http://comments.gmane.org/gmane.linux.toybox/409 <pre>Hi Rob, Are the long options handled completely at present? I think the long options are not handled if they don't have a corresponding short-option. Whats your opinion and do you have any design or plan to get this feature in. -Ashwini _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Ashwini Sharma 2013-04-29T05:05:51 http://comments.gmane.org/gmane.linux.toybox/406 <pre>Hi, find -mtime didn't work as expected in the first place and I've broken it even more with the cleanups. The attached patch should fix it (and simplify the relevant code a bit). Felix _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Felix Janda 2013-04-26T21:27:43 http://comments.gmane.org/gmane.linux.toybox/403 <pre>Hi Rob, Can you change the getmountlist to accept the file path from user, as to in which file he is interested. As of now you bind it to read only /proc/mounts, giving the option will make it usable for /etc/fstab or /etc/mtab regards, Ashwini _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Ashwini Sharma 2013-04-26T09:45:27 http://comments.gmane.org/gmane.linux.toybox/401 <pre>http://landley.net/hg/toybox/rev/878 The diff adds xsocket(), cleanups up some free() logic (free is defined to accept NULL and treat it as a NOP, and we don't need to set a local variable on the way out of a function), get_device_info() can return void and handle its own error_exit(), remove a half-dozen redundant strncpy() of the same value into the same field that's still there, and move the llist_traverse() call into show_ifconfig(). http://landley.net/hg/toybox/rev/879 This is basically inlining three functions in display_ifconfig(), and take advantage of the opportunities for further cleanup revealed by this. 1) Inline print_hw_addr() at the one callsite (in display_ifconfig()), and perform obvious simplifications around it: replace the function's variable names to the arguments fed to the function. Move printing the "HWaddr" label inside the ARPHRD_ETHER test (since that's the only type we support the alternative is potentially printing a label with nothing after it), and c</pre> Rob Landley 2013-04-25T06:00:14 http://comments.gmane.org/gmane.linux.toybox/400 <pre>Hi Rob, Calling xabspath() function with "exact=1", makes sure that the last path component exists. I think the decision making statement fd = openat(dirfd, s, 0); if (fd == -1 &amp;&amp; (exact || todo || errno != ENOENT)) goto error; will jump to error:, when openat fails, though the 's' is the last component in the path. Failing to open the last component with errno != ENOENT doesn't mean the component doesn't exist. The check should be modified to see if the fail happens at last component or otherwise. Do provide your inputs on the same. regards, Ashwini _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Ashwini Sharma 2013-04-25T05:12:20 http://comments.gmane.org/gmane.linux.toybox/399 <pre>nbd_client needs &lt;netinet/in.h&gt; for IPPROTO_TCP on musl; netdb.h includes it already on glibc, but POSIX only specifies it for netinet/in.h. (It just got annoying to see make allyesconfig; make CC=musl-gcc fail on something this trivial.) Isaac Dunham _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> idunham-N9AOi2cAC9ZBDgjK7y7TUQ< at >public.gmane.org 2013-04-24T23:11:41 http://comments.gmane.org/gmane.linux.toybox/393 <pre>Hello, The kernel maps all occurences of '-' in module names to '_', so that if you load a module such as snd-pcm-oss.ko or phc-k8.ko, it is presented by lsmod and /proc/modules as module snd_pcm_oss or phc_k8. Modinfo from module-init-tools makes up for that by reversing the logic and treating '-' and '_' as equivalent, while toybox takes only the string passed. This can result in missing results that should be returned, especially in commands like this: for m in `lsmod |cut -f 1 -d ' '`; do /sbin/modinfo $m; done (ie, anything that relies on lsmod's output to get a name for modinfo). As far as I am aware, there are no modules that use both an underscore and a hyphen in the filename. Another difference is that standard modinfo is documented to accept a filename as a parameter (modinfo /path/to/module.ko). Thanks, Isaac Dunham </pre> Isaac Dunham 2013-04-22T22:40:28 http://comments.gmane.org/gmane.linux.toybox/387 <pre>Hello, I did make allyesconfig; make (to test ifconfig) and get this: Compile toybox... toys/pending/ifconfig.c: In function 'is_host_unix': toys/pending/ifconfig.c:121: error: invalid application of 'sizeof' to incomplete type 'struct sockaddr_un' toys/pending/ifconfig.c:124: error: dereferencing pointer to incomplete type toys/pending/ifconfig.c:124: error: dereferencing pointer to incomplete type toys/pending/ifconfig.c: In function 'address_to_name': toys/pending/ifconfig.c:273: error: dereferencing pointer to incomplete type toys/pending/ifconfig.c:273: error: dereferencing pointer to incomplete type This struct is specified by POSIX to be in &lt;sys/un.h&gt;, so here's the fix. Compile tested on glibc and musl (though musl also misses IPPROTO_TCP in nbd_client.c) </pre> Isaac Dunham 2013-04-21T23:56:03 http://comments.gmane.org/gmane.linux.toybox/383 <pre>Helping my friend Adrienne pack out her apartment today, but it's still more free time than a weekend with the nieces and nephews. Commit 869 (http://landley.net/hg/toybox/rev/869): Just moving code around, no actual code changes. Reorder functions to eliminate the need for function prototypes, and put the command_main() function at the end of the file. (That way it's easy to spot.) In general, C code ordered to avoid function prototypes reads from the bottom up. The main function is at the end, and each other function is after the code it uses. Commit 870 (http://landley.net/hg/toybox/rev/870): Looking at the code for a rough edge, I immediately spot this: -struct iface_list { -... - unsigned long long receive_bytes; //total bytes received - unsigned long long receive_packets; //total packets received - unsigned long receive_errors; //bad packets received - unsigned long receive_drop; //no space in linux buffers - unsigned long receive_fifo; //receiver fifo overrun - un</pre> Rob Landley 2013-04-21T03:40:12 http://comments.gmane.org/gmane.linux.toybox/379 <pre>Hello, some more find cleanup in an attached patch. The main change was to make some code in build_filter_list() less repetitive using a suitable struct and a loop. The command line parsing is however now less efficient since after an argument has been tested for equality with different strings like "-mtime" and the corresponding type (CHECK_MTIME) has been saved, the saved integer has again to be checked for equality with stuff to find out what further command line parsing is necessary. Apart from this I moved around some variables and got rid of the unnecessary index "i" in build_filter_list(). Felix BTW: POSIX-1.2008 TC1 is out _______________________________________________ Toybox mailing list Toybox-oU9gvf+ajcRUPo+8YfT7LV6hYfS7NtTn&lt; at &gt;public.gmane.org http://lists.landley.net/listinfo.cgi/toybox-landley.net </pre> Felix Janda 2013-04-20T18:39:10 http://comments.gmane.org/gmane.linux.toybox/378 <pre>Hello, attached are two patches cleaning up stat a bit. The first changes from 4 to 2 space indentation, the second is more interesting. First of all it extends stat to correctly handle format strings. (Previously it considered only the second char of the format string.) This made it possible to remove duplication between print_stat_format() and stat_main(). The option processing now uses the toybox infrastructure better and by sorting the options in an appropriate way stat_main() could be simplified greatly. get_access_str() is now no longer broken. Apart from this there are various other small changes. The functionality of get_access_str() is also implemented in the ls toy (likely much better). Should something like it be in the library? Actually since stat and ls are both frontends to the stat system call more is duplicated. I also wonder whether toybox will ever have SELinux support... My local man page does even not document the -Z option to stat. Felix _____________________________________________</pre> Felix Janda 2013-04-19T22:28:27 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.linux.toybox