Re: [U-Boot] [PATCH v2 16/27] mkimage: Allow a FIT to include an image of any type

Hi Andreas,

On 27 April 2016 at 01:28, Andreas Bießmann andreas.devel@googlemail.com wrote:

Hi Simon,

On 23.02.2016 06:55, Simon Glass wrote:

...

At present FIT images are set up by providing a device tree source file which is a file with a .its extension. We want to support automatically creating this file based on the image supplied to mkimage. This means that even though the final file type is always IH_TYPE_FLATDT, the image inside may be something else.

Signed-off-by: Simon Glass sjg@chromium.org

Changes in v2:

• Always expect the final argument to be the image file

...

diff --git a/tools/mkimage.c b/tools/mkimage.c index b8293f6..2fd1f0b 100644 --- a/tools/mkimage.c +++ b/tools/mkimage.c @@ -112,10 +112,14 @@ static void usage(const char *msg) static void process_args(int argc, char **argv) { char *ptr;

• int type = IH_TYPE_INVALID;
  

• char *datafile = NULL;
  

• int expecting;
  int opt;
  
  

• expecting = IH_TYPE_COUNT;      /* Unknown */
  while ((opt = getopt(argc, argv,
  

•                      "a:A:cC:d:D:e:f:Fk:K:ln:O:rR:sT:vVx")) != -1) {
  

•                      "-a:A:cC:d:D:e:f:Fk:K:ln:O:rR:sT:vVx")) != -1) {
  

I just encountered an error building the latest ToT on my OS X box with this change. Unfortunately the '-' in optstring seems to be an GNU option and is not portable.

...

          switch (opt) {

...

@@ -211,14 +217,31 @@ static void process_args(int argc, char **argv) case 'x': params.xflag++; break;

•         case 1:
  

•                 if (expecting == type || optind == argc) {
  

•                         params.imagefile = optarg;
  

•                         expecting = IH_TYPE_INVALID;
  

•                 }
  

•                 break;
          default:
                  usage("Invalid option");
          }
  }
  
  

• if (optind >= argc)
  

• /*
  

•  * For auto-generated FIT images we need to know the image type to put
  

•  * in the FIT, which is separate from the file's image type (which
  

•  * will always be IH_TYPE_FLATDT in this case).
  

•  */
  

• if (params.type == IH_TYPE_FLATDT) {
  

•         params.fit_image_type = type;
  

•         params.datafile = datafile;
  

• } else if (type != IH_TYPE_INVALID) {
  

•         params.type = type;
  

• }
  

• if (!params.imagefile)
          usage("Missing output filename");
  

• params.imagefile = argv[optind];
  

}

Therefore the imagefile is never set. Do you mind to switch back to the portable version with 'POSIX_CORRECT'?

Hmmm I didn't consider that. I see you have sent a patch (thank you) so let's continue on that thread.

Regards, Simon