| Message ID | 20190711191813.19550-3-simon.k.r.goldschmidt@gmail.com |
|---|---|
| State | Superseded, archived |
| Delegated to: | Simon Goldschmidt |
| Headers | show |
| Series | arm: socfpga: gen5 SPL bss cleanup | expand |
On 7/11/19 9:18 PM, Simon Goldschmidt wrote: > The socfpga gen5 SPL manually zeroed bss in board_init_f(). Now that the > DDR driver does not use bss any more, bss is not used before board_init_r() > and we can remove this hack. > > Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> > --- > > arch/arm/mach-socfpga/spl_gen5.c | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/arch/arm/mach-socfpga/spl_gen5.c b/arch/arm/mach-socfpga/spl_gen5.c > index 87b76b47de..47e63709ad 100644 > --- a/arch/arm/mach-socfpga/spl_gen5.c > +++ b/arch/arm/mach-socfpga/spl_gen5.c > @@ -79,8 +79,6 @@ void board_init_f(ulong dummy) > writel(SYSMGR_ECC_OCRAM_DERR | SYSMGR_ECC_OCRAM_EN, > &sysmgr_regs->eccgrp_ocram); > > - memset(__bss_start, 0, __bss_end - __bss_start); > - > socfpga_sdram_remap_zero(); > socfpga_pl310_clear(); So who will zero out the BSS ? [...]
On Fri, Jul 12, 2019 at 7:15 AM Marek Vasut <marex@denx.de> wrote: > > On 7/11/19 9:18 PM, Simon Goldschmidt wrote: > > The socfpga gen5 SPL manually zeroed bss in board_init_f(). Now that the > > DDR driver does not use bss any more, bss is not used before board_init_r() > > and we can remove this hack. > > > > Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> > > --- > > > > arch/arm/mach-socfpga/spl_gen5.c | 2 -- > > 1 file changed, 2 deletions(-) > > > > diff --git a/arch/arm/mach-socfpga/spl_gen5.c b/arch/arm/mach-socfpga/spl_gen5.c > > index 87b76b47de..47e63709ad 100644 > > --- a/arch/arm/mach-socfpga/spl_gen5.c > > +++ b/arch/arm/mach-socfpga/spl_gen5.c > > @@ -79,8 +79,6 @@ void board_init_f(ulong dummy) > > writel(SYSMGR_ECC_OCRAM_DERR | SYSMGR_ECC_OCRAM_EN, > > &sysmgr_regs->eccgrp_ocram); > > > > - memset(__bss_start, 0, __bss_end - __bss_start); > > - > > socfpga_sdram_remap_zero(); > > socfpga_pl310_clear(); > > So who will zero out the BSS ? BSS is zeroed by crt0.S, but after board_init_f(), before board_init_r(). Socfpga just had this double-zeroing because it invalidly used BSS in board_init_f(). Some weeks ago on this list, we've had a discussion whether it would be good to generally allow such early usage of BSS, possibly via a config option (socfpga is not the only platform affected). The outcome was negative and so I've started this patch. Regards, Simon > > [...]
On 7/12/19 7:29 AM, Simon Goldschmidt wrote: > On Fri, Jul 12, 2019 at 7:15 AM Marek Vasut <marex@denx.de> wrote: >> >> On 7/11/19 9:18 PM, Simon Goldschmidt wrote: >>> The socfpga gen5 SPL manually zeroed bss in board_init_f(). Now that the >>> DDR driver does not use bss any more, bss is not used before board_init_r() >>> and we can remove this hack. >>> >>> Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> >>> --- >>> >>> arch/arm/mach-socfpga/spl_gen5.c | 2 -- >>> 1 file changed, 2 deletions(-) >>> >>> diff --git a/arch/arm/mach-socfpga/spl_gen5.c b/arch/arm/mach-socfpga/spl_gen5.c >>> index 87b76b47de..47e63709ad 100644 >>> --- a/arch/arm/mach-socfpga/spl_gen5.c >>> +++ b/arch/arm/mach-socfpga/spl_gen5.c >>> @@ -79,8 +79,6 @@ void board_init_f(ulong dummy) >>> writel(SYSMGR_ECC_OCRAM_DERR | SYSMGR_ECC_OCRAM_EN, >>> &sysmgr_regs->eccgrp_ocram); >>> >>> - memset(__bss_start, 0, __bss_end - __bss_start); >>> - >>> socfpga_sdram_remap_zero(); >>> socfpga_pl310_clear(); >> >> So who will zero out the BSS ? > > BSS is zeroed by crt0.S, but after board_init_f(), before board_init_r(). > Socfpga just had this double-zeroing because it invalidly used BSS in > board_init_f(). Can you add that to the commit message and resend just this patch ? I applied the other one. > Some weeks ago on this list, we've had a discussion whether it would be good to > generally allow such early usage of BSS, possibly via a config option (socfpga > is not the only platform affected). The outcome was negative and so I've started > this patch.
On Fri, Jul 12, 2019 at 12:29 PM Marek Vasut <marex@denx.de> wrote: > > On 7/12/19 7:29 AM, Simon Goldschmidt wrote: > > On Fri, Jul 12, 2019 at 7:15 AM Marek Vasut <marex@denx.de> wrote: > >> > >> On 7/11/19 9:18 PM, Simon Goldschmidt wrote: > >>> The socfpga gen5 SPL manually zeroed bss in board_init_f(). Now that the > >>> DDR driver does not use bss any more, bss is not used before board_init_r() > >>> and we can remove this hack. > >>> > >>> Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> > >>> --- > >>> > >>> arch/arm/mach-socfpga/spl_gen5.c | 2 -- > >>> 1 file changed, 2 deletions(-) > >>> > >>> diff --git a/arch/arm/mach-socfpga/spl_gen5.c b/arch/arm/mach-socfpga/spl_gen5.c > >>> index 87b76b47de..47e63709ad 100644 > >>> --- a/arch/arm/mach-socfpga/spl_gen5.c > >>> +++ b/arch/arm/mach-socfpga/spl_gen5.c > >>> @@ -79,8 +79,6 @@ void board_init_f(ulong dummy) > >>> writel(SYSMGR_ECC_OCRAM_DERR | SYSMGR_ECC_OCRAM_EN, > >>> &sysmgr_regs->eccgrp_ocram); > >>> > >>> - memset(__bss_start, 0, __bss_end - __bss_start); > >>> - > >>> socfpga_sdram_remap_zero(); > >>> socfpga_pl310_clear(); > >> > >> So who will zero out the BSS ? > > > > BSS is zeroed by crt0.S, but after board_init_f(), before board_init_r(). > > Socfpga just had this double-zeroing because it invalidly used BSS in > > board_init_f(). > > Can you add that to the commit message and resend just this patch ? I > applied the other one. Sure. > > > Some weeks ago on this list, we've had a discussion whether it would be good to > > generally allow such early usage of BSS, possibly via a config option (socfpga > > is not the only platform affected). The outcome was negative and so I've started > > this patch.
diff --git a/arch/arm/mach-socfpga/spl_gen5.c b/arch/arm/mach-socfpga/spl_gen5.c index 87b76b47de..47e63709ad 100644 --- a/arch/arm/mach-socfpga/spl_gen5.c +++ b/arch/arm/mach-socfpga/spl_gen5.c @@ -79,8 +79,6 @@ void board_init_f(ulong dummy) writel(SYSMGR_ECC_OCRAM_DERR | SYSMGR_ECC_OCRAM_EN, &sysmgr_regs->eccgrp_ocram); - memset(__bss_start, 0, __bss_end - __bss_start); - socfpga_sdram_remap_zero(); socfpga_pl310_clear();
The socfpga gen5 SPL manually zeroed bss in board_init_f(). Now that the DDR driver does not use bss any more, bss is not used before board_init_r() and we can remove this hack. Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> --- arch/arm/mach-socfpga/spl_gen5.c | 2 -- 1 file changed, 2 deletions(-)