Hi Mike,
On Sat, Jan 14, 2012 at 5:42 PM, Mike Frysinger vapier@gentoo.org wrote:
On Saturday 14 January 2012 19:47:24 Simon Glass wrote:
--- a/common/cmd_echo.c +++ b/common/cmd_echo.c @@ -44,8 +44,9 @@ int do_echo(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) } }
- /* Use puts() so that the LCD sees it as a new line */
if (putnl)
- putc('\n');
- puts("\n");
return 0; }
this sounds like a hack for a broken LCD core and so the fix should be somewhere in there
Well it's actually that I don't want to flush after every character (it slows things down a lot) so puts is a good place to look. I suppose we could flush after every newline we see in lcd_putc().
--- a/common/lcd.c +++ b/common/lcd.c
+static char lcd_flush_dcache;
seems like it'd be better as a CONFIG knob
OK, will see if I can do that.
- /*
- * flush_dcache_range() is declared in common.h but it seems that some
- * architectures do not actually implement it. Is there a way to find
- * out whether it exists? For now, ARM is safe.
- */
if those arches don't implement this func, then the failure is on their head. people should feel free to use the cache api we expose in common.h.
I sort-of agree, except that I saw masses of failures. I will look a bit harder.
Regards, Simon
-mike