-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 10/07/2013 09:55 AM, Wolfgang Denk wrote: > Dear Tom, > > In message <20131007121252.GS15917@bill-the-cat> you wrote: >> >>> 331 filename = argv[3]; 332 addr = >>> simple_strtoul(argv[4], NULL, cmdline_base); 333 bytes >>> = simple_strtoul(argv[5], NULL, cmdline_base); 334 if >>> (argc >= 7) 335 pos = simple_strtoul(argv[6], >>> NULL, cmdline_base); 336 else 337 pos = >>> 0; >>> >>> >>> Should we not perform at least minimal error checking, i. e. >>> verify that no garbage arguments have been passed to that >>> function? >> >> Yes, we ought to. If you don't pass fatwrite the right number >> of arguments we get data aborts, for example. > > Well, this is not a problem here, in do_save(): > > ... 325 if (argc < 6 || argc > 7) 326 > return CMD_RET_USAGE; ... > > And are you sure of "fatwrite"? This calls do_fat_fswrite(). and > here I also see some test: > > ... 98 if (argc < 5) 99 return > cmd_usage(cmdtp); ...
Off-by-one error somewhere? 'mmc' 'part' 'ddr-addr' 'filename' will blowup as we use an insane value for size. I suspect we aren't eating 'fatwrite' somewhere along the line. - -- Tom -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJSUsPdAAoJENk4IS6UOR1W6kEP/ijUEElKWLxsfJKLR1y0BbJ6 NaovYZ2Pz2UiOIEbk186hqEnpyaMU+2WlTavs61nufZ7fiteWkHS+536pgg/7GlG pA9Uiq0DzI5RVG0EuaBJ62lt55JV9RWavdYlnGBEcgaD7ESLBTtizGbdKW30vW4S FLfBXvEZVnYzTcIGT0kj8zEOLlg05QN6Asik8sv9GbwjX4wxVKwD0lpLYMmP/hy1 d/H6ql2a682HSAQw/6C+QxK89bSYbFay5+VhmFCZESM97tI1vWYIFC6hQsg8q+jl /1asD1hquvv5Of7O2xIMyvmExPD25mbeflyTYnXoMFb8SL1VKNBfSdljcBPeTe8G TBJMdURCf6zptmR+qUt8fRmWImhYZhalq+oyEImA4uKxu9zSBIBFcILKCJQ9LCXI Hp0oaRaBUZVfD2Dno2g7fYgNwuOkXv/IpK/pWY0F6a2jNXGZooni9TdvkLeqX0sB Bdgk1gz2EDeXIxvG/H4vTTWF5lVDh7Y5NHnqPW+HJ0fhaSqqxsklaXIU4np7EjIB hfmwhi25fJZrGlUVgDG6YTkjTE8FOKfCaYVlFtIbrGZo14U27JucUBbLOeVxtEU0 VcdVcJx/pzi9377DbpWd8dF6XryMAMR3/de58ady6/TsoWZ90axIMykz3xE1TgAF cppQkXNHHtao0PDxeWcV =ntyL -----END PGP SIGNATURE----- _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
