ARM: OMAP2+: cpu: Make cpu_class_is_omap2 true for all non-omap1 platforms

As omap1 and omap2 will never be compiled together, due to different compiler flags, so we can simply make cpu_class_is_omap2() = true, for all non-omap1 platforms.

Signed-off-by Vaibhav Hiremath <[email protected]>

---
 arch/arm/plat-omap/include/plat/cpu.h |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

Quoted 9 lines · Vaibhav

As omap1 and omap2 will never be compiled together, due to different compiler flags, so we can simply make cpu_class_is_omap2() = true, for all non-omap1 platforms.

Signed-off-by Vaibhav Hiremath <[email protected]>

---
 arch/arm/plat-omap/include/plat/cpu.h |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

Tony,

Ping???

Quoted 11 lines · Hiremath, Vaibhav

Quoted 9 lines · Vaibhav

As omap1 and omap2 will never be compiled together, due to different compiler flags, so we can simply make cpu_class_is_omap2() = true, for all non-omap1 platforms.

Signed-off-by Vaibhav Hiremath <[email protected]>

---
 arch/arm/plat-omap/include/plat/cpu.h |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

Quoted 3 lines

Tony,

Ping???

Thanks applying into devel-am33xx branch.

There are few cleanup patches pending for your review, it would be good if you provide your review comments.

Quoted 3 lines · Hiremath, Vaibhav

There are few cleanup patches pending for your review, it would be good if you provide your review comments.

Hmm care to post the Subject: lines for those? I may have marked something as read accidentally..

Quoted 11 lines · Hiremath, Vaibhav

Quoted 9 lines · Vaibhav

As omap1 and omap2 will never be compiled together, due to different compiler flags, so we can simply make cpu_class_is_omap2() = true, for all non-omap1 platforms.

Signed-off-by Vaibhav Hiremath <[email protected]>

---
 arch/arm/plat-omap/include/plat/cpu.h |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

Quoted 3 lines

Tony,

Ping???

Thanks applying into devel-am33xx branch.

Quoted 12 lines · Tony Lindgren

Quoted 11 lines · Hiremath, Vaibhav

Quoted 9 lines · Vaibhav

As omap1 and omap2 will never be compiled together, due to different compiler flags, so we can simply make cpu_class_is_omap2() = true, for all non-omap1 platforms.

Signed-off-by Vaibhav Hiremath <[email protected]>

---
 arch/arm/plat-omap/include/plat/cpu.h |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

Quoted 5 lines

Quoted 3 lines

Tony,

Ping???

Thanks applying into devel-am33xx branch.

Actually looks like this fails for the dynamic detection case with omap1_defconfig.. Looks like you did not build test your patch with omap1_defconfig? Please do it for every patch!

It seems we need something better though for multi zImage too, we can't really rely on CONFIG_ARCH_OMAP1 or CONFIG_ARCH_OMAP2PLUS here to determine anything.

Anyways, dropping this for now, let's try to come up with something better.

arch/arm/plat-omap/include/plat/cpu.h +1 -0

@@ -42,7 +42,14 @@
 #define OMAP2_DEVICE_TYPE_GP           3
 #define OMAP2_DEVICE_TYPE_BAD          4
 
+#ifdef CONFIG_ARCH_OMAP2PLUS
 int omap_type(void);

Quoted 16 lines · Hiremath, Vaibhav

arch/arm/plat-omap/include/plat/cpu.h +8 -0

@@ -42,7 +42,14 @@
 #define OMAP2_DEVICE_TYPE_GP           3
 #define OMAP2_DEVICE_TYPE_BAD          4
 
+#ifdef CONFIG_ARCH_OMAP2PLUS
 int omap_type(void);
+#else
+inline static int omap_type(void)
+{
+	/* Always return GP, since it is not being used anyway for omap1 */
+	return OMAP2_DEVICE_TYPE_GP;
+}
+#endif
 

arch/arm/plat-omap/include/plat/cpu.h

Quoted 2 lines

arch/arm/plat-omap/include/plat/cpu.h +0 -0

 /*
  * omap_rev bits:

Just to follow-up on this, looks like we need to postpone this for v3.7. While at it, we need to make sure all soc_is_omapxxxx() are false by default.

What you're suggesting above would also depend on some MULTI_ARCH option set if multiple ARM architectures are selected. Above patch would only optimize things when ARCH_OMAP2PLUS alone is selected.

Quoted 17 lines · Tony Lindgren

Quoted 16 lines · Hiremath, Vaibhav

arch/arm/plat-omap/include/plat/cpu.h +8 -0

@@ -42,7 +42,14 @@
 #define OMAP2_DEVICE_TYPE_GP           3
 #define OMAP2_DEVICE_TYPE_BAD          4
 
+#ifdef CONFIG_ARCH_OMAP2PLUS
 int omap_type(void);
+#else
+inline static int omap_type(void)
+{
+	/* Always return GP, since it is not being used anyway for omap1 */
+	return OMAP2_DEVICE_TYPE_GP;
+}
+#endif
 

Quoted 7 lines

arch/arm/plat-omap/include/plat/cpu.h

Quoted 2 lines

arch/arm/plat-omap/include/plat/cpu.h +0 -0

 /*
  * omap_rev bits:

Just to follow-up on this, looks like we need to postpone this for v3.7. While at it, we need to make sure all soc_is_omapxxxx() are false by default.

Just FYI, I had submitted V2 of this patch.

Quoted 4 lines

What you're suggesting above would also depend on some MULTI_ARCH option set if multiple ARM architectures are selected. Above patch would only optimize things when ARCH_OMAP2PLUS alone is selected.

Isn't this only applicable for ARCH_OMAP2PLUS and OMAP1?? And above patch take cares of both, as OMAP1 and 2 will not be build together, since compiler flags itself will be different.

Can you elaborate on how things will break if multiple ARM architectures are selected?

I am not sure whether I am following you completely.

Quoted 28 lines · Hiremath, Vaibhav

Quoted 17 lines · Tony Lindgren

Quoted 16 lines · Hiremath, Vaibhav

arch/arm/plat-omap/include/plat/cpu.h +8 -0

@@ -42,7 +42,14 @@
 #define OMAP2_DEVICE_TYPE_GP           3
 #define OMAP2_DEVICE_TYPE_BAD          4
 
+#ifdef CONFIG_ARCH_OMAP2PLUS
 int omap_type(void);
+#else
+inline static int omap_type(void)
+{
+	/* Always return GP, since it is not being used anyway for omap1 */
+	return OMAP2_DEVICE_TYPE_GP;
+}
+#endif
 

Quoted 7 lines

arch/arm/plat-omap/include/plat/cpu.h

Quoted 2 lines

arch/arm/plat-omap/include/plat/cpu.h +0 -0

 /*
  * omap_rev bits:

Just to follow-up on this, looks like we need to postpone this for v3.7. While at it, we need to make sure all soc_is_omapxxxx() are false by default.

Just FYI, I had submitted V2 of this patch.

Quoted 8 lines

Quoted 4 lines

What you're suggesting above would also depend on some MULTI_ARCH option set if multiple ARM architectures are selected. Above patch would only optimize things when ARCH_OMAP2PLUS alone is selected.

Isn't this only applicable for ARCH_OMAP2PLUS and OMAP1?? And above patch take cares of both, as OMAP1 and 2 will not be build together, since compiler flags itself will be different.

Right, but we also need to deal with

Quoted 3 lines

Can you elaborate on how things will break if multiple ARM architectures are selected?

Quoted 1 line

I am not sure whether I am following you completely.

We have quite a few initcalls that exit early with various soc_is, cpu_is, cpu_class_is tests that would fail for multi zImage kernels. Some of these have been patched away with additional calls in MACHINE_START, but some still remain.

We probably always need to return false for all soc_is_omapxxxx() unless the SoC detection is initialized in id.c.

Quoted 6 lines · Tony Lindgren

We have quite a few initcalls that exit early with various soc_is, cpu_is, cpu_class_is tests that would fail for multi zImage kernels. Some of these have been patched away with additional calls in MACHINE_START, but some still remain.

Let me check on this inside code but if you have any quick reference, please let me know, I can fix it and submit the patch.