I am going to name Mr. Copy and Mr. Paste as the culprits, your honor.
Since they were acting on my direct orders, I ask mercy from the court.
I definitely know better and will try to do better in the future. :)
--Z
On Mon, 2009-11-16 at 15:54 -0700, David Brownell wrote:
> Newly introduced failure mode I saw:
>
> Info : JTAG tap: omap5912.dsp tap/device found: 0xffffff81 (mfg: 0x7c0, part:
> 0xffff, ver: 0xf)
> Warn : JTAG tap: omap5912.dsp UNEXPECTED: 0xffffff81 (mfg: 0x7c0, part:
> 0xffff, ver: 0xf)
> Error: JTAG tap: omap5912.dsp expected 1 of 1: 0x03df1d81 (mfg: 0x6c0, part:
> 0x3df1, ver: 0x0)
> Info : JTAG tap: omap5912.arm tap/device found: 0x0692602f (mfg: 0x017, part:
> 0x6926, ver: 0x0)
> Warn : TAP omap5912.unknown does not have IDCODE
>
> Hmm, that "Warn" should at worst be "Info" -- different issue.
>
> This patch was fingered by "git bisect". The appended tweak
> resolves this regression. Never assume "char" is unsigned!!
>
> - Dave
>
> --- a/src/helper/binarybuffer.h
> +++ b/src/helper/binarybuffer.h
> @@ -41,7 +41,8 @@
> static inline void buf_set_u32(void *_buffer,
> unsigned first, unsigned num, uint32_t value)
> {
> - char *buffer = (char *)_buffer;
> + uint8_t *buffer = (uint8_t *)_buffer;
> +
> if ((num == 32) && (first == 0)) {
> buffer[3] = (value >> 24) & 0xff;
> buffer[2] = (value >> 16) & 0xff;
> @@ -69,7 +70,8 @@ static inline void buf_set_u32(void *_buffer,
> static inline uint32_t buf_get_u32(const void *_buffer,
> unsigned first, unsigned num)
> {
> - char *buffer = (char *)_buffer;
> + uint8_t *buffer = (uint8_t *)_buffer;
> +
> if ((num == 32) && (first == 0)) {
> return (((uint32_t)buffer[3]) << 24) |
> (((uint32_t)buffer[2]) << 16) |
_______________________________________________
Openocd-development mailing list
Openocd-development@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/openocd-development
