Hi Martin,
On 02/01/2017 22:24, Martin Kaiser wrote:
From: Martin Kaiser martin@kaiser.cx
We can use the same header length calculations for both imximage v1 and v2. This addresses TODO comments about imximage v1 in the current code.
I see, absoulutely a good idea.
With this patch applied, *header_size_ptr in imximage_set_header() will have the correct value for both imximage v1 and v2. This is necessary for people wanting to add proprietary data behind the created imximage.
Signed-off-by: Martin Kaiser martin@kaiser.cx Cc: sbabic@denx.de
I discovered the problem when I tried to use csf_ptr with imximage v1 (as part of a private modification).
*csf_ptr = params->ep + *header_size_ptr - imximage_init_loadsize; = params->ep + sbuf->st_size + imximage_ivt_offset - imximage_init_loadsize; = params->ep + padded data file size + alloc_len - (imximage_init_loadsize - imximage_ivt_offset);
This works only if alloc_len == imximage_init_loadsize - imximage_ivt_offset, not if alloc_len is always 4096.
I remember that this was a "reasonable" value that initially avoided overlapping. But it was not computed.
I rebased my tree to make sure that the patch still applies against current master. I'd really appreciate any feedback, this has been pending since September.
tools/imximage.c | 38 ++++++++++++++++---------------------- 1 file changed, 16 insertions(+), 22 deletions(-)
diff --git a/tools/imximage.c b/tools/imximage.c index 2cd8d88..0c43196 100644 --- a/tools/imximage.c +++ b/tools/imximage.c @@ -300,8 +300,7 @@ static void set_imx_hdr_v1(struct imx_header *imxhdr, uint32_t dcd_len, /* Set magic number */ fhdr_v1->app_code_barker = APP_CODE_BARKER;
- /* TODO: check i.MX image V1 handling, for now use 'old' style */
- hdr_base = entry_point - 4096;
- hdr_base = entry_point - imximage_init_loadsize + flash_offset; fhdr_v1->app_dest_ptr = hdr_base - flash_offset; fhdr_v1->app_code_jump_vector = entry_point;
@@ -833,18 +832,19 @@ static void imximage_set_header(void *ptr, struct stat *sbuf, int ifd, /* Parse dcd configuration file */ dcd_len = parse_cfg_file(imxhdr, params->imagename);
- if (imximage_version == IMXIMAGE_V2) {
- if (imximage_version == IMXIMAGE_V1)
header_size = sizeof(flash_header_v1_t);- else { header_size = sizeof(flash_header_v2_t) + sizeof(boot_data_t); if (!plugin_image) header_size += sizeof(dcd_v2_t); else header_size += MAX_PLUGIN_CODE_SIZE;
if (imximage_init_loadsize < imximage_ivt_offset + header_size)imximage_init_loadsize = imximage_ivt_offset +header_size;
- if (imximage_init_loadsize < imximage_ivt_offset + header_size)
imximage_init_loadsize = imximage_ivt_offset + header_size;- /* Set the imx header */ (*set_imx_hdr)(imxhdr, dcd_len, params->ep, imximage_ivt_offset);
@@ -913,23 +913,21 @@ static int imximage_generate(struct image_tool_params *params, /* Parse dcd configuration file */ parse_cfg_file(&imximage_header, params->imagename);
- /* TODO: check i.MX image V1 handling, for now use 'old' style */
- if (imximage_version == IMXIMAGE_V1) {
alloc_len = 4096;header_size = 4096;- } else {
- if (imximage_version == IMXIMAGE_V1)
header_size = sizeof(imx_header_v1_t);- else { header_size = sizeof(flash_header_v2_t) + sizeof(boot_data_t); if (!plugin_image) header_size += sizeof(dcd_v2_t); else header_size += MAX_PLUGIN_CODE_SIZE;
if (imximage_init_loadsize < imximage_ivt_offset + header_size)imximage_init_loadsize = imximage_ivt_offset +header_size;alloc_len = imximage_init_loadsize - imximage_ivt_offset;
- if (imximage_init_loadsize < imximage_ivt_offset + header_size)
imximage_init_loadsize = imximage_ivt_offset + header_size;- alloc_len = imximage_init_loadsize - imximage_ivt_offset;
- if (alloc_len < header_size) { fprintf(stderr, "%s: header error\n", params->cmdname);
@@ -959,11 +957,7 @@ static int imximage_generate(struct image_tool_params *params,
pad_len = ROUND(sbuf.st_size, 4096) - sbuf.st_size;
- /* TODO: check i.MX image V1 handling, for now use 'old' style */
- if (imximage_version == IMXIMAGE_V1)
return 0;- else
return pad_len;
- return pad_len;
}
IMHO it is ok and it was pending since a lot of time. I merge it.
Best regards, Stefano Babic