| Message ID | 1393393865-11608-9-git-send-email-yamada.m@jp.panasonic.com |
|---|---|
| State | Deferred |
| Delegated to: | Tom Rini |
| Headers | show |
On Wed, Feb 26, 2014 at 02:51:05PM +0900, Masahiro Yamada wrote: > Seeing arch/arm/cpu/armv8/config.mk, -mstrict-align option > was added to both PLATFORM_NO_UNALIGNED and PLATFORM_CPPFLAGS. > > $(PLATFORM_CPPFLAGS) is given to all compiled objects. > Adding the same option to also PLATFORM_NO_UNALIGNED is > redundant. > > For example, common/hush.o was compiled with double -mstrict-align. > You can comfirm it by checking common/.hush.o.cmd file. > > Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> > Cc: David Feng <fenghua@phytium.com.cn> > Cc: Albert ARIBAUD <albert.u.boot@aribaud.net> > Cc: Tom Rini <trini@ti.com> With the v3 series about unaligned bits this patch should no longer be needed.
Hi Tom, On Wed, 26 Feb 2014 08:55:43 -0500, Tom Rini <trini@ti.com> wrote: > On Wed, Feb 26, 2014 at 02:51:05PM +0900, Masahiro Yamada wrote: > > > Seeing arch/arm/cpu/armv8/config.mk, -mstrict-align option > > was added to both PLATFORM_NO_UNALIGNED and PLATFORM_CPPFLAGS. > > > > $(PLATFORM_CPPFLAGS) is given to all compiled objects. > > Adding the same option to also PLATFORM_NO_UNALIGNED is > > redundant. > > > > For example, common/hush.o was compiled with double -mstrict-align. > > You can comfirm it by checking common/.hush.o.cmd file. > > > > Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> > > Cc: David Feng <fenghua@phytium.com.cn> > > Cc: Albert ARIBAUD <albert.u.boot@aribaud.net> > > Cc: Tom Rini <trini@ti.com> > > With the v3 series about unaligned bits this patch should no longer be > needed. Hmm... This patch is about not repeating -mstrict-align twice, which is a bug unrelated with switching to -mno-unaligned-access. Besides, -mstrict-align means "Don't assume that unaligned accesses are handled", which I think is for aarch64 what -mno-unaligned-access is for arm(32), so it *should* be specified. Amicalement,
On Wed, Feb 26, 2014 at 10:58:10PM +0100, Albert ARIBAUD wrote: > Hi Tom, > > On Wed, 26 Feb 2014 08:55:43 -0500, Tom Rini <trini@ti.com> wrote: > > > On Wed, Feb 26, 2014 at 02:51:05PM +0900, Masahiro Yamada wrote: > > > > > Seeing arch/arm/cpu/armv8/config.mk, -mstrict-align option > > > was added to both PLATFORM_NO_UNALIGNED and PLATFORM_CPPFLAGS. > > > > > > $(PLATFORM_CPPFLAGS) is given to all compiled objects. > > > Adding the same option to also PLATFORM_NO_UNALIGNED is > > > redundant. > > > > > > For example, common/hush.o was compiled with double -mstrict-align. > > > You can comfirm it by checking common/.hush.o.cmd file. > > > > > > Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> > > > Cc: David Feng <fenghua@phytium.com.cn> > > > Cc: Albert ARIBAUD <albert.u.boot@aribaud.net> > > > Cc: Tom Rini <trini@ti.com> > > > > With the v3 series about unaligned bits this patch should no longer be > > needed. > > Hmm... This patch is about not repeating -mstrict-align twice, which is > a bug unrelated with switching to -mno-unaligned-access. > > Besides, -mstrict-align means "Don't assume that unaligned accesses are > handled", which I think is for aarch64 what -mno-unaligned-access is > for arm(32), so it *should* be specified. Yes, but my patch cleaned up the places that set PLATFORM_NO_UNALIGNED so this duplication was dropped.
Hi Tom, On Wed, 26 Feb 2014 17:03:08 -0500, Tom Rini <trini@ti.com> wrote: > On Wed, Feb 26, 2014 at 10:58:10PM +0100, Albert ARIBAUD wrote: > > Hi Tom, > > > > On Wed, 26 Feb 2014 08:55:43 -0500, Tom Rini <trini@ti.com> wrote: > > > > > On Wed, Feb 26, 2014 at 02:51:05PM +0900, Masahiro Yamada wrote: > > > > > > > Seeing arch/arm/cpu/armv8/config.mk, -mstrict-align option > > > > was added to both PLATFORM_NO_UNALIGNED and PLATFORM_CPPFLAGS. > > > > > > > > $(PLATFORM_CPPFLAGS) is given to all compiled objects. > > > > Adding the same option to also PLATFORM_NO_UNALIGNED is > > > > redundant. > > > > > > > > For example, common/hush.o was compiled with double -mstrict-align. > > > > You can comfirm it by checking common/.hush.o.cmd file. > > > > > > > > Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> > > > > Cc: David Feng <fenghua@phytium.com.cn> > > > > Cc: Albert ARIBAUD <albert.u.boot@aribaud.net> > > > > Cc: Tom Rini <trini@ti.com> > > > > > > With the v3 series about unaligned bits this patch should no longer be > > > needed. > > > > Hmm... This patch is about not repeating -mstrict-align twice, which is > > a bug unrelated with switching to -mno-unaligned-access. > > > > Besides, -mstrict-align means "Don't assume that unaligned accesses are > > handled", which I think is for aarch64 what -mno-unaligned-access is > > for arm(32), so it *should* be specified. > > Yes, but my patch cleaned up the places that set PLATFORM_NO_UNALIGNED > so this duplication was dropped. Oh, ok. My bad, then. Amicalement,
diff --git a/arch/arm/cpu/armv8/config.mk b/arch/arm/cpu/armv8/config.mk index 027a68c..7e59b4d 100644 --- a/arch/arm/cpu/armv8/config.mk +++ b/arch/arm/cpu/armv8/config.mk @@ -8,7 +8,6 @@ PLATFORM_RELFLAGS += -fno-common -ffixed-x18 # SEE README.arm-unaligned-accesses PF_NO_UNALIGNED := $(call cc-option, -mstrict-align) -PLATFORM_NO_UNALIGNED := $(PF_NO_UNALIGNED) PF_CPPFLAGS_ARMV8 := $(call cc-option, -march=armv8-a) PLATFORM_CPPFLAGS += $(PF_CPPFLAGS_ARMV8)
Seeing arch/arm/cpu/armv8/config.mk, -mstrict-align option was added to both PLATFORM_NO_UNALIGNED and PLATFORM_CPPFLAGS. $(PLATFORM_CPPFLAGS) is given to all compiled objects. Adding the same option to also PLATFORM_NO_UNALIGNED is redundant. For example, common/hush.o was compiled with double -mstrict-align. You can comfirm it by checking common/.hush.o.cmd file. Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Cc: David Feng <fenghua@phytium.com.cn> Cc: Albert ARIBAUD <albert.u.boot@aribaud.net> Cc: Tom Rini <trini@ti.com> --- arch/arm/cpu/armv8/config.mk | 1 - 1 file changed, 1 deletion(-)