Date: Tue, 25 Apr 2006 17:31:44 GMT From: Warner Losh <imp@FreeBSD.org> To: Perforce Change Reviews <perforce@freebsd.org> Subject: PERFORCE change 96073 for review Message-ID: <200604251731.k3PHViBv049807@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=96073 Change 96073 by imp@imp_hammer on 2006/04/25 17:31:12 OK. It appears we have to wait for at least 23 SLCK ticks before we can try to get the status again. This corresponds to ~70us. I don't see anything on the datasheet that would suggest why this is necessary, but weird shit happens when I don't do this :-(. 22 or 21 also seem to work mostly , but 20 seems to never work. We may need to bump this value up with experience. Affected files ... .. //depot/projects/arm/src/sys/boot/arm/at91/libat91/spi_flash.c#7 edit Differences ... ==== //depot/projects/arm/src/sys/boot/arm/at91/libat91/spi_flash.c#7 (text+ko) ==== @@ -35,15 +35,11 @@ static void Delay(void) { -#if 0 unsigned later; - later = (AT91C_BASE_ST->ST_CRTR + 10) & AT91C_ST_CRTV; + later = (AT91C_BASE_ST->ST_CRTR + 23) & AT91C_ST_CRTV; while (later != AT91C_BASE_ST->ST_CRTR) continue; -#else - printf("V--------"); -#endif } /* @@ -306,8 +302,9 @@ value = pSPI->SPI_RDR; value = pSPI->SPI_SR; + // Increment real time counter every SLCK + AT91C_BASE_ST->ST_RTMR = 1; + if (((initStatus = GetFlashStatus()) & 0xFC) != 0xBC) printf(" Unexpected SPI flash status: 0x%x\r\n", initStatus); - // Increment real time counter every SLCK - AT91C_BASE_ST->ST_RTMR = 1; }
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200604251731.k3PHViBv049807>