Dear Macpaul Lin,
In message <1303797876-28548-1-git-send-email-macp...@andestech.com> you wrote:
> ftahbc020s.h provides basic definitions of this controller
> to help a SoC which use this AHB Controller could
> do scalable software settings in lowlevel_init.S.
>
> Signed-off-by: Macpaul Lin <macp...@andestech.com>
> ---
> Changes for v2:
> - Add __ASSEMBLY__ protecton to register offset for supporting lowlevel_init
> Changes for v3:
> - Patch: no change. Changed a mail server to resend this.
...
> +/* this section is used by lowlevel_init.S */
> +#define FTAHBC020S_SLAVE_BSR_0 0x00 /* Slave n Base/Size
> Reg */
> +#define FTAHBC020S_SLAVE_BSR_1 0x04
> +#define FTAHBC020S_SLAVE_BSR_2 0x08
> +#define FTAHBC020S_SLAVE_BSR_3 0x0C
> +#define FTAHBC020S_SLAVE_BSR_4 0x10
> +#define FTAHBC020S_SLAVE_BSR_5 0x14
> +#define FTAHBC020S_SLAVE_BSR_6 0x18
> +#define FTAHBC020S_SLAVE_BSR_7 0x1C
> +#define FTAHBC020S_SLAVE_BSR_8 0x20
> +#define FTAHBC020S_SLAVE_BSR_9 0x24
> +#define FTAHBC020S_SLAVE_BSR_10 0x28
See previous comment: I think this should be done using asm-offsets
instead.
> +#define FTAHBC020S_SLAVE_BSR_BASE(x) (((x) & 0xFFF) << 20)
> +#define FTAHBC020S_SLAVE_BSR_SIZE(x) (((x) & 0xF) << 16)
> +
> +#define FTAHBC020S_SLAVE_BSR_SIZE_1M 0x0
> +#define FTAHBC020S_SLAVE_BSR_SIZE_2M 0x1
> +#define FTAHBC020S_SLAVE_BSR_SIZE_4M 0x2
> +#define FTAHBC020S_SLAVE_BSR_SIZE_8M 0x3
> +#define FTAHBC020S_SLAVE_BSR_SIZE_16M 0x4
> +#define FTAHBC020S_SLAVE_BSR_SIZE_32M 0x5
> +#define FTAHBC020S_SLAVE_BSR_SIZE_64M 0x6
> +#define FTAHBC020S_SLAVE_BSR_SIZE_128M 0x7
> +#define FTAHBC020S_SLAVE_BSR_SIZE_256M 0x8
> +#define FTAHBC020S_SLAVE_BSR_SIZE_512M 0x9
> +#define FTAHBC020S_SLAVE_BSR_SIZE_1G 0xA
> +#define FTAHBC020S_SLAVE_BSR_SIZE_2G 0xB
I recommend to use a generic macro here, as recommended for the other
patch.
> +/*
> + * FTAHBC020S_PCR - Priority Control Register
> + */
> +#define FTAHBC020S_PCR_PLEVEL_15 (1 << 15)
> +#define FTAHBC020S_PCR_PLEVEL_14 (1 << 14)
> +#define FTAHBC020S_PCR_PLEVEL_13 (1 << 13)
> +#define FTAHBC020S_PCR_PLEVEL_12 (1 << 12)
> +#define FTAHBC020S_PCR_PLEVEL_11 (1 << 11)
> +#define FTAHBC020S_PCR_PLEVEL_10 (1 << 10)
> +#define FTAHBC020S_PCR_PLEVEL_09 (1 << 9)
> +#define FTAHBC020S_PCR_PLEVEL_08 (1 << 8)
> +#define FTAHBC020S_PCR_PLEVEL_07 (1 << 7)
> +#define FTAHBC020S_PCR_PLEVEL_06 (1 << 6)
> +#define FTAHBC020S_PCR_PLEVEL_05 (1 << 5)
> +#define FTAHBC020S_PCR_PLEVEL_04 (1 << 4)
> +#define FTAHBC020S_PCR_PLEVEL_03 (1 << 3)
> +#define FTAHBC020S_PCR_PLEVEL_02 (1 << 2)
> +#define FTAHBC020S_PCR_PLEVEL_01 (1 << 1)
Ditto here. Why flooding the code with tons of (mostly) unused
defines?
Use:
#define FTAHBC020S_PCR_PLEVEL(x) (1 << (x))
Best regards,
Wolfgang Denk
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: w...@denx.de
Vor allem kein Gedanke! Nichts ist kompromittierender als ein Gedan-
ke! - Friedrich Wilhelm Nietzsche _Der Fall Wagner_ (1888)
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot
