On Thursday 12 July 2012 11:25:13 Simon Glass wrote:
> +void lcd_sync(void)
> +{
> + /*
> + * 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.
> + */
> +#ifdef CONFIG_ARM
> + int line_length;
> +
> + if (lcd_flush_dcache)
> + flush_dcache_range((u32)lcd_base,
> + (u32)(lcd_base + lcd_get_size(&line_length)));
> +#endif
> +}hmm, i think this should be more fine grained > /* Clear the last one */ > memset (CONSOLE_ROW_LAST, COLOR_MASK(lcd_color_bg), CONSOLE_ROW_SIZE); > + lcd_sync(); > } this requires only updating one row and not the entire display right ? so we should only flush that one line. seems like other parts suffer similar over- flushing behavior. > +/** > + * Set whether we need to flush the dcache when changing the LCD image. > + * This defaults to off. i wonder if this default is correct. perhaps it should default to 1 since setting up custom memory maps to disable cache is fairly specially. -mike
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
