Message ID | 1432128909-25272-5-git-send-email-hdegoede@redhat.com |
---|---|
State | Accepted |
Delegated to: | Hans de Goede |
Headers | show |
On Wed, 2015-05-20 at 15:35 +0200, Hans de Goede wrote: > The A1 SRAM Base differs between sun9i and the others, update sunxi-common.h > to deal with this, so that we do not set the initial stack pointer to point > to the BROM. > > This removes the need for the weird undocumented register write I took from > the allwiner u-boot sources and which needed #ifdef-ery in start.S as it > needed to be done really early on. (typo: "allwinner") Removes in the sense that it simply no longer appears in the series, I think? (As opposed to a removal hunk being missing from this or a subsequent patch) Assuming that is the case: > Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hi, On 21-05-15 21:07, Ian Campbell wrote: > On Wed, 2015-05-20 at 15:35 +0200, Hans de Goede wrote: >> The A1 SRAM Base differs between sun9i and the others, update sunxi-common.h >> to deal with this, so that we do not set the initial stack pointer to point >> to the BROM. >> >> This removes the need for the weird undocumented register write I took from >> the allwiner u-boot sources and which needed #ifdef-ery in start.S as it >> needed to be done really early on. > > (typo: "allwinner") > > Removes in the sense that it simply no longer appears in the series, I > think? (As opposed to a removal hunk being missing from this or a > subsequent patch) Correct, remove as in simply no longer appears in the series. Regsrds, Hans > > Assuming that is the case: > >> Signed-off-by: Hans de Goede <hdegoede@redhat.com> > > Acked-by: Ian Campbell <ijc@hellion.org.uk> >
On Fri, 2015-05-22 at 00:27 +0200, Hans de Goede wrote: > Hi, > > On 21-05-15 21:07, Ian Campbell wrote: > > On Wed, 2015-05-20 at 15:35 +0200, Hans de Goede wrote: > >> The A1 SRAM Base differs between sun9i and the others, update sunxi-common.h > >> to deal with this, so that we do not set the initial stack pointer to point > >> to the BROM. > >> > >> This removes the need for the weird undocumented register write I took from > >> the allwiner u-boot sources and which needed #ifdef-ery in start.S as it > >> needed to be done really early on. > > > > (typo: "allwinner") > > > > Removes in the sense that it simply no longer appears in the series, I > > think? (As opposed to a removal hunk being missing from this or a > > subsequent patch) > > Correct, remove as in simply no longer appears in the series. Perhaps s/removes/avoids/ and s/write I took/write I previously took/ to clarify? Ack regardless. > > Regsrds, > > Hans > > > > > Assuming that is the case: > > > >> Signed-off-by: Hans de Goede <hdegoede@redhat.com> > > > > Acked-by: Ian Campbell <ijc@hellion.org.uk> > > >
Hi, On 22-05-15 08:32, Ian Campbell wrote: > On Fri, 2015-05-22 at 00:27 +0200, Hans de Goede wrote: >> Hi, >> >> On 21-05-15 21:07, Ian Campbell wrote: >>> On Wed, 2015-05-20 at 15:35 +0200, Hans de Goede wrote: >>>> The A1 SRAM Base differs between sun9i and the others, update sunxi-common.h >>>> to deal with this, so that we do not set the initial stack pointer to point >>>> to the BROM. >>>> >>>> This removes the need for the weird undocumented register write I took from >>>> the allwiner u-boot sources and which needed #ifdef-ery in start.S as it >>>> needed to be done really early on. >>> >>> (typo: "allwinner") >>> >>> Removes in the sense that it simply no longer appears in the series, I >>> think? (As opposed to a removal hunk being missing from this or a >>> subsequent patch) >> >> Correct, remove as in simply no longer appears in the series. > > Perhaps s/removes/avoids/ and s/write I took/write I previously took/ to > clarify? Updated the commit msg with the above suggestion. Thanks & Regards, Hans
diff --git a/include/configs/sunxi-common.h b/include/configs/sunxi-common.h index 2b90681..76f42f5 100644 --- a/include/configs/sunxi-common.h +++ b/include/configs/sunxi-common.h @@ -97,8 +97,20 @@ #define CONFIG_SPL_BSS_MAX_SIZE 0x00080000 /* 512 KiB */ #define CONFIG_SYS_SPL_MALLOC_SIZE 0x00080000 /* 512 KiB */ +#ifdef CONFIG_MACH_SUN9I +/* + * The A80's A1 sram starts at 0x00010000 rather then at 0x00000000 and is + * slightly bigger. Note that it is possible to map the first 32 KiB of the + * A1 at 0x00000000 like with older SoCs by writing 0x16aa0001 to the + * undocumented 0x008000e0 SYS_CTRL register. Where the 16aa is a key and + * the 1 actually activates the mapping of the first 32 KiB to 0x00000000. + */ +#define CONFIG_SYS_INIT_RAM_ADDR 0x10000 +#define CONFIG_SYS_INIT_RAM_SIZE 0x0a000 /* 40 KiB */ +#else #define CONFIG_SYS_INIT_RAM_ADDR 0x0 #define CONFIG_SYS_INIT_RAM_SIZE 0x8000 /* 32 KiB */ +#endif #define CONFIG_SYS_INIT_SP_OFFSET \ (CONFIG_SYS_INIT_RAM_SIZE - GENERATED_GBL_DATA_SIZE)
The A1 SRAM Base differs between sun9i and the others, update sunxi-common.h to deal with this, so that we do not set the initial stack pointer to point to the BROM. This removes the need for the weird undocumented register write I took from the allwiner u-boot sources and which needed #ifdef-ery in start.S as it needed to be done really early on. Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- include/configs/sunxi-common.h | 12 ++++++++++++ 1 file changed, 12 insertions(+)