Skip to contentSkip to conversation

PM branch updated to v2.6.35, SRF dropped

PATCH New Kevin Hilman 5 participants 13 messages Aug 4, 2010 – Aug 9, 2010

Subsystems (3)
Arm (32-bit) Arm sub-architectures OMAP2+ support
lore
Share & tools
lore

Kevin Hilman

author
#0 lore raw

Hello,

It's been a while since a PM branch update, thanks to a nice summer vacation for me. :)

The PM branch has now been updated to v2.6.35 plus Tony's for-next branch.

This is mostly a rebase, but the most important change is that I'm finally dropping SRF as promised[1]. The new DVFS infrastructure is now in the RFC/PATCH phase so I'm no longer going to keep dragging SRF along. Note that this means that ***DVFS no longer works in the PM branch***. If this bothers you, then please consider helping review and test the new DVFS infrastructure recently posted by Thara so we can get the new, upstream-ready DVFS code upstream ASAP.

I've only tested the new PM branch on a couple boards to avoid any further delay getting this version out, update, so hopefully others will help test.

I've tested RET on idle & suspend, OFF in idle & suspend on:

3430/omap3evm: works

3630/Zoom3: works
- Known problem: console hangs on UART wakeup when wakeups are enabled using echo enabled > /sys/devices/platform/serial8250.0/tty/ttyS0/power/wakeup but I haven't debugged this further.

Kevin

[1] http://marc.info/?l=linux-omap&m=127206504624556&w=2

#1 lore raw

Kevin Hilman had written, on 08/04/2010 05:50 PM, the following:
[...]

Quoted 4 lines

I've only tested the new PM branch on a couple boards to avoid any further delay getting this version out, update, so hopefully others will help test.

Quoted 8 lines

I've tested RET on idle & suspend, OFF in idle & suspend on:

3430/omap3evm: works

3630/Zoom3: works
- Known problem: console hangs on UART wakeup when wakeups are enabled using echo enabled > /sys/devices/platform/serial8250.0/tty/ttyS0/power/wakeup but I haven't debugged this further.

Just sharing results from SDP testing omap3_pm_defconfig + earlyprintk:
SDP3430 (with denx.com u-boot, nor boot): does not boot to shell full Bootlog: http://pastebin.mozilla.org/761964

SDP3630 (with TI u-boot, xloader, mmc boot): ret works, uart wakeup fails for off mode (I am guessing this is similar to Kevin's results) full Bootlog: http://pastebin.mozilla.org/761966

OMAP4 Blaze (with TI u-boot, xloader, mmc boot): does not boot, hell lot of crashes - note, need to explicitly enable MACH_OMAP4430_SDP in .config full Bootlog: http://pastebin.mozilla.org/761968

[...]

Signature · 2 lines
Regards,
Nishanth Menon
#4 lore raw
In reply to Kevin Hilman
Quoted 10 lines

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Hilman
Sent: Thursday, August 05, 2010 4:21 AM
To: [email protected]
Subject: PM branch updated to v2.6.35, SRF dropped

Hello,

Hi Kevin,

Can you pull in this patch into your branch as well?
http://marc.info/?l=linux-usb&m=127858711304301&w=2
Without this multi-omap build seems to be broken.

Also with omap_4430sdp_defconfig, I see these compile errors arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' make[1]: *** [arch/arm/kernel/entry-armv.o] Error 1
make: *** [arch/arm/kernel] Error 2

Doing a git log on entry-armv.S shows me a top commit which might be an issue if conflicts are'nt resolved well.

commit 7b70c4275f28702b76b273c8534c38f8313812e9
Merge: ceb0885... a20df56...
Author: Russell King <[email protected]>
Date: Sat Jul 31 14:20:16 2010 +0100

Merge branch 'devel-stable' into devel

Conflicts:
arch/arm/kernel/entry-armv.S
arch/arm/kernel/setup.c
arch/arm/mm/init.c

Maybe this is an issue in Tony's for-next as well. Haven't tested it though.

regards,
Rajendra

Quoted 43 lines

It's been a while since a PM branch update, thanks to a nice summer vacation for me. :)

The PM branch has now been updated to v2.6.35 plus Tony's for-next branch.

This is mostly a rebase, but the most important change is that I'm finally dropping SRF as promised[1]. The new DVFS infrastructure is now
in the RFC/PATCH phase so I'm no longer going to keep dragging SRF along. Note that this means that ***DVFS no longer works in the PM branch***.
If this bothers you, then please consider helping review and test the new DVFS infrastructure recently posted by Thara so we can get the new,
upstream-ready DVFS code upstream ASAP.

I've only tested the new PM branch on a couple boards to avoid any further delay getting this version out, update, so hopefully others will
help test.

I've tested RET on idle & suspend, OFF in idle & suspend on:

3430/omap3evm: works

3630/Zoom3: works
- Known problem: console hangs on UART wakeup when wakeups are enabled using
echo enabled >
/sys/devices/platform/serial8250.0/tty/ttyS0/power/wakeup
but I haven't debugged this further.

Kevin

[1] http://marc.info/?l=linux-omap&m=127206504624556&w=2
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

author
#2 lore raw
In reply to Nishanth Menon
Quoted 11 lines · Nishanth Menon

Kevin Hilman had written, on 08/04/2010 05:50 PM, the following:
[...]

Quoted 8 lines

I've only tested the new PM branch on a couple boards to avoid any further delay getting this version out, update, so hopefully others will help test.

I've tested RET on idle & suspend, OFF in idle & suspend on:

3430/omap3evm: works

Quoted 8 lines
Quoted 4 lines

3630/Zoom3: works
- Known problem: console hangs on UART wakeup when wakeups are enabled using echo enabled > /sys/devices/platform/serial8250.0/tty/ttyS0/power/wakeup but I haven't debugged this further.

Just sharing results from SDP testing omap3_pm_defconfig + earlyprintk:
SDP3430 (with denx.com u-boot, nor boot): does not boot to shell full Bootlog: http://pastebin.mozilla.org/761964

Does the same setup boot fine on l-o for-next, or l-o master?

Quoted 8 lines

SDP3630 (with TI u-boot, xloader, mmc boot): ret works, uart wakeup fails for off mode (I am guessing this is similar to Kevin's results) full Bootlog: http://pastebin.mozilla.org/761966

OMAP4 Blaze (with TI u-boot, xloader, mmc boot): does not boot, hell lot of crashes - note, need to explicitly enable MACH_OMAP4430_SDP in .config
full Bootlog: http://pastebin.mozilla.org/761968

yeah, I don't expect PM branch to work on OMAP4 until we get the OMAP4 hwmod core upstream.

Kevin

author
#5 lore raw
In reply to Nayak, Rajendra
Quoted 7 lines · Rajendra
Quoted 7 lines

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Hilman
Sent: Thursday, August 05, 2010 4:21 AM
To: [email protected]
Subject: PM branch updated to v2.6.35, SRF dropped

Quoted 8 lines
Quoted 2 lines

Hello,

Hi Kevin,

Can you pull in this patch into your branch as well?
http://marc.info/?l=linux-usb&m=127858711304301&w=2
Without this multi-omap build seems to be broken.

Done.

Quoted 17 lines

Also with omap_4430sdp_defconfig, I see these compile errors arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' make[1]: *** [arch/arm/kernel/entry-armv.o] Error 1
make: *** [arch/arm/kernel] Error 2

Doing a git log on entry-armv.S shows me a top commit which might be an issue if conflicts are'nt resolved well.

commit 7b70c4275f28702b76b273c8534c38f8313812e9
Merge: ceb0885... a20df56...
Author: Russell King <[email protected]>
Date: Sat Jul 31 14:20:16 2010 +0100

Merge branch 'devel-stable' into devel

Quoted 7 lines

Conflicts:
arch/arm/kernel/entry-armv.S
arch/arm/kernel/setup.c
arch/arm/mm/init.c

Maybe this is an issue in Tony's for-next as well. Haven't tested it though.

Yeah, I'm guessing this an issue in for-next, and probably l-o master too.

Kevin

Quoted 44 lines
Quoted 44 lines

It's been a while since a PM branch update, thanks to a nice summer vacation for me. :)

The PM branch has now been updated to v2.6.35 plus Tony's for-next branch.

This is mostly a rebase, but the most important change is that I'm finally dropping SRF as promised[1]. The new DVFS infrastructure is now
in the RFC/PATCH phase so I'm no longer going to keep dragging SRF along. Note that this means that ***DVFS no longer works in the PM branch***.
If this bothers you, then please consider helping review and test the new DVFS infrastructure recently posted by Thara so we can get the new,
upstream-ready DVFS code upstream ASAP.

I've only tested the new PM branch on a couple boards to avoid any further delay getting this version out, update, so hopefully others will
help test.

I've tested RET on idle & suspend, OFF in idle & suspend on:

3430/omap3evm: works

3630/Zoom3: works
- Known problem: console hangs on UART wakeup when wakeups are enabled using
echo enabled >
/sys/devices/platform/serial8250.0/tty/ttyS0/power/wakeup
but I haven't debugged this further.

Kevin

[1] http://marc.info/?l=linux-omap&m=127206504624556&w=2
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#3 lore raw
In reply to Kevin Hilman

Kevin Hilman had written, on 08/05/2010 05:29 PM, the following:

Quoted 17 lines
Quoted 15 lines · Nishanth Menon

Kevin Hilman had written, on 08/04/2010 05:50 PM, the following:
[...]

Quoted 12 lines

I've only tested the new PM branch on a couple boards to avoid any further delay getting this version out, update, so hopefully others will help test.

I've tested RET on idle & suspend, OFF in idle & suspend on:

3430/omap3evm: works

3630/Zoom3: works
- Known problem: console hangs on UART wakeup when wakeups are enabled using echo enabled > /sys/devices/platform/serial8250.0/tty/ttyS0/power/wakeup but I haven't debugged this further.

Quoted 5 lines
Quoted 3 lines

Just sharing results from SDP testing omap3_pm_defconfig + earlyprintk:
SDP3430 (with denx.com u-boot, nor boot): does not boot to shell full Bootlog: http://pastebin.mozilla.org/761964

Does the same setup boot fine on l-o for-next, or l-o master?

yep

l-o master (omap3_defconfig):
Log: http://pastebin.mozilla.org/762620

l-o for-next (omap3_defconfig):
build fails with:
drivers/usb/musb/omap2430.c: In function 'musb_platform_init':

drivers/usb/musb/omap2430.c:198: error: 'AE5_2430_USB0HS_STP' undeclared (first use in this function) drivers/usb/musb/omap2430.c:198: error: (Each undeclared identifier is reported only once drivers/usb/musb/omap2430.c:198: error: for each function it appears in.)

so disabled CONFIG_ARCH_OMAP2430 and CONFIG_MACH_OMAP_2430SDP in .config for quick bootup.

Log: http://pastebin.mozilla.org/762637

Ref:
remotes/origin/master 8428498 Linux-omap rebuilt: arm
changes merged to mainline, final omap changes for 2.6.36
remotes/origin/for-next d21872b Merge branch 'v2.6.35-omap-mailbox-for-next' of git://gitorious.org/~doyu/lk/mainline into omap-for-linus

Quoted 5 lines
Quoted 4 lines

SDP3630 (with TI u-boot, xloader, mmc boot): ret works, uart wakeup fails for off mode (I am guessing this is similar to Kevin's results) full Bootlog: http://pastebin.mozilla.org/761966

Quoted 7 lines
Quoted 4 lines

OMAP4 Blaze (with TI u-boot, xloader, mmc boot): does not boot, hell lot of crashes - note, need to explicitly enable MACH_OMAP4430_SDP in .config
full Bootlog: http://pastebin.mozilla.org/761968

yeah, I don't expect PM branch to work on OMAP4 until we get the OMAP4 hwmod core upstream.

:(

Signature · 2 lines
Regards,
Nishanth Menon
maintainer
#6 lore raw

PATCH 1/1 omap: Use CONFIG_SMP for test_for_ipi and test_for_ltirq belong (Re: PM branch updated to v2.6.35, SRF dropped)

In reply to Kevin Hilman
Quoted 23 lines · Kevin Hilman
Quoted 23 lines

Also with omap_4430sdp_defconfig, I see these compile errors arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' make[1]: *** [arch/arm/kernel/entry-armv.o] Error 1
make: *** [arch/arm/kernel] Error 2

Doing a git log on entry-armv.S shows me a top commit which might be an issue if conflicts are'nt resolved well.

commit 7b70c4275f28702b76b273c8534c38f8313812e9
Merge: ceb0885... a20df56...
Author: Russell King <[email protected]>
Date: Sat Jul 31 14:20:16 2010 +0100

Merge branch 'devel-stable' into devel

Conflicts:
arch/arm/kernel/entry-armv.S
arch/arm/kernel/setup.c
arch/arm/mm/init.c

Quoted 5 lines
Quoted 2 lines

Maybe this is an issue in Tony's for-next as well. Haven't tested it though.

Yeah, I'm guessing this an issue in for-next, and probably l-o master too.

Noticed that with omap3_defconfig with CONFIG_SMP enabled. Does the following work for you?

Tony

From: Tony Lindgren <[email protected]>
Date: Thu, 5 Aug 2010 13:18:20 +0300
Subject: [PATCH] omap: Use CONFIG_SMP for test_for_ipi and test_for_ltirq belong

Otherwise we get the following error when enabling CONFIG_SMP for omap3_defconfig:

arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ltirq r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ltirq r0,r6,r5,lr'

Signed-off-by Tony Lindgren <[email protected]>
arch/arm/mach-omap2/include/mach/entry-macro.S +4 -2
index 50fd749..06e64e1 100644
@@ -177,7 +177,10 @@ omap_irq_base:	.word	0
 		cmpne   \irqnr, \tmp
 		cmpcs   \irqnr, \irqnr
 		.endm
+#endif
+#endif	/* MULTI_OMAP2 */
 
+#ifdef CONFIG_SMP
 		/* We assume that irqstat (the raw value of the IRQ acknowledge
 		 * register) is preserved from the macro above.
 		 * If there is an IPI, we immediately signal end of interrupt

@@ -205,8 +208,7 @@ omap_irq_base:	.word	0
 		streq	\irqstat, [\base, #GIC_CPU_EOI]
 		cmp	\tmp, #0
 		.endm
-#endif
-#endif	/* MULTI_OMAP2 */
+#endif	/* CONFIG_SMP */
 
 		.macro	irq_prio_table
 		.endm
maintainer
#7 lore raw

PATCH 1/1 omap: Fix sev instruction usage for multi-omap

Quoted 24 lines · Tony Lindgren
Quoted 23 lines · Kevin Hilman
Quoted 23 lines

Also with omap_4430sdp_defconfig, I see these compile errors arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' make[1]: *** [arch/arm/kernel/entry-armv.o] Error 1
make: *** [arch/arm/kernel] Error 2

Doing a git log on entry-armv.S shows me a top commit which might be an issue if conflicts are'nt resolved well.

commit 7b70c4275f28702b76b273c8534c38f8313812e9
Merge: ceb0885... a20df56...
Author: Russell King <[email protected]>
Date: Sat Jul 31 14:20:16 2010 +0100

Merge branch 'devel-stable' into devel

Conflicts:
arch/arm/kernel/entry-armv.S
arch/arm/kernel/setup.c
arch/arm/mm/init.c

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Maybe this is an issue in Tony's for-next as well. Haven't tested it though.

Yeah, I'm guessing this an issue in for-next, and probably l-o master too.

Noticed that with omap3_defconfig with CONFIG_SMP enabled. Does the following work for you?

Here's a related patch that allows CONFIG_SMP to compile with omap3_defconfig. Booting still won't work before some arm generic code is changed.

Regards,

Tony

Quoted 1 line

From f931fb147f2a3cf4c4b7646e5f270c241ab4aad1 Mon Sep 17 00:00:00 2001

From: Tony Lindgren <[email protected]>
Date: Thu, 5 Aug 2010 13:28:42 +0300
Subject: [PATCH] omap: Fix sev instruction usage for multi-omap

Otherwise we get the following error with omap3_defconfig and CONFIG_SMP:

Error: selected processor does not support `sev'

Signed-off-by Tony Lindgren <[email protected]>
arch/arm/mach-omap2/Makefile +1 -0
index 63b2d88..88d3a1e 100644
@@ -25,6 +25,7 @@ obj-$(CONFIG_LOCAL_TIMERS)		+= timer-mpu.o
 obj-$(CONFIG_HOTPLUG_CPU)		+= omap-hotplug.o
 obj-$(CONFIG_ARCH_OMAP4)		+= omap44xx-smc.o omap4-common.o
 
+AFLAGS_omap-headsmp.o			:=-Wa,-march=armv7-a
 AFLAGS_omap44xx-smc.o			:=-Wa,-march=armv7-a
 
 # Functions loaded to SRAM
arch/arm/mach-omap2/omap-smp.c +1 -2
index af3c20c..9e9f70e 100644
@@ -102,8 +102,7 @@ static void __init wakeup_secondary(void)
 	 * Send a 'sev' to wake the secondary core from WFE.
 	 * Drain the outstanding writes to memory
 	 */
-	dsb();
-	set_event();
+	dsb_sev();
 	mb();
 }
arch/arm/plat-omap/include/plat/smp.h +0 -7
index 6a3ff65..5177a9c 100644
@@ -19,13 +19,6 @@
 
 #include <asm/hardware/gic.h>
 
-/*
- * set_event() is used to wake up secondary core from wfe using sev. ROM
- * code puts the second core into wfe(standby).
- *
- */
-#define set_event()	__asm__ __volatile__ ("sev" : : : "memory")
-
 /* Needed for secondary core boot */
 extern void omap_secondary_startup(void);
 extern u32 omap_modify_auxcoreboot0(u32 set_mask, u32 clear_mask);
#8 lore raw

RE: [PATCH] omap: Fix sev instruction usage for multi-omap

Quoted 37 lines

-----Original Message-----
From: Tony Lindgren [mailto:[email protected]]
Sent: Friday, August 06, 2010 12:35 PM
To: Kevin Hilman
Cc: Nayak, Rajendra; [email protected]
Subject: [PATCH] omap: Fix sev instruction usage for multi-omap

Quoted 5 lines · Tony Lindgren
Quoted 4 lines · Kevin Hilman
Quoted 4 lines

Also with omap_4430sdp_defconfig, I see these compile errors arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction

`test_for_ipi r0,r6,r5,lr'

Quoted 1 line
Quoted 1 line
Quoted 1 line

arch/arm/kernel/entry-armv.S:48: Error: bad instruction

`test_for_ipi r0,r6,r5,lr'

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

make[1]: *** [arch/arm/kernel/entry-armv.o] Error 1
make: *** [arch/arm/kernel] Error 2

Doing a git log on entry-armv.S shows me a top commit

which might

Quoted 14 lines
Quoted 14 lines
Quoted 14 lines

be an issue if conflicts are'nt resolved well.

commit 7b70c4275f28702b76b273c8534c38f8313812e9
Merge: ceb0885... a20df56...
Author: Russell King <[email protected]>
Date: Sat Jul 31 14:20:16 2010 +0100

Merge branch 'devel-stable' into devel

Conflicts:
arch/arm/kernel/entry-armv.S
arch/arm/kernel/setup.c
arch/arm/mm/init.c

Quoted 10 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line

Maybe this is an issue in Tony's for-next as well.

Haven't tested

Quoted 3 lines
Quoted 3 lines
Quoted 1 line

it though.

Yeah, I'm guessing this an issue in for-next, and

probably l-o master

Quoted 4 lines
Quoted 1 line

too.

Noticed that with omap3_defconfig with CONFIG_SMP enabled. Does the following work for you?

With this patch, I went past the previous break but hit this

/tmp/ccUTrImV.s: Assembler messages:
/tmp/ccUTrImV.s:140: Error: selected processor does not support `sev' make[1]: *** [arch/arm/mach-omap2/omap-smp.o] Error 1
make: *** [arch/arm/mach-omap2] Error 2

Quoted 4 lines

Here's a related patch that allows CONFIG_SMP to compile with omap3_defconfig. Booting still won't work before some arm generic code is changed.

Now with this, omap4 build using omap_4430sdp_defconfig seems to go through, but like you said, boot is still an issue.

regards,
Rajendra

Quoted 5 lines

Regards,

Tony

maintainer
#9 lore raw

PATCH 1/1 omap: Fix omap_4430sdp_defconfig for make oldconfig

Quoted 3 lines · Nayak, Rajendra

Now with this, omap4 build using omap_4430sdp_defconfig seems to go through, but like you said, boot is still an issue.

Hmm, I guess these issues pop up if you do yes "" | make oldconfig.

Can you try the patch below? After applying the patch, you need to:

$ rm .config
$ cp arch/arm/configs/omap_4430sdp_defconfig .config
$ yes "" | ARCH=arm make oldconfig

Then omap2 and 3 won't get selected by make oldconfig.

Regards,

Tony

Quoted 1 line

From 96214d5365a10f370e7690a3aa2dc24dbca39e79 Mon Sep 17 00:00:00 2001

From: Tony Lindgren <[email protected]>
Date: Fri, 6 Aug 2010 14:40:46 +0300
Subject: [PATCH] omap: Fix omap_4430sdp_defconfig for make oldconfig

We don't want to select select the other omaps at this point because otherwise we would have to disable CONFIG_SMP.

Signed-off-by Tony Lindgren <[email protected]>
arch/arm/configs/omap_4430sdp_defconfig +3 -0
index 63e0c2d..14c1e18 100644
@@ -13,6 +13,9 @@ CONFIG_MODULE_SRCVERSION_ALL=y
 # CONFIG_BLK_DEV_BSG is not set
 CONFIG_ARCH_OMAP=y
 CONFIG_ARCH_OMAP4=y
+# CONFIG_ARCH_OMAP2PLUS_TYPICAL is not set
+# CONFIG_ARCH_OMAP2 is not set
+# CONFIG_ARCH_OMAP3 is not set
 # CONFIG_OMAP_MUX is not set
 CONFIG_OMAP_32K_TIMER=y
 CONFIG_OMAP_DM_TIMER=y
#10 lore raw

RE: [PATCH] omap: Fix omap_4430sdp_defconfig for make oldconfig

Quoted 16 lines

-----Original Message-----
From: Tony Lindgren [mailto:[email protected]]
Sent: Friday, August 06, 2010 5:20 PM
To: Nayak, Rajendra
Cc: Kevin Hilman; [email protected]
Subject: [PATCH] omap: Fix omap_4430sdp_defconfig for make oldconfig

Quoted 3 lines · Nayak, Rajendra

Now with this, omap4 build using omap_4430sdp_defconfig seems to go through, but like you said, boot is still an issue.

Hmm, I guess these issues pop up if you do yes "" | make oldconfig.

Quoted 8 lines

Can you try the patch below? After applying the patch, you need to:

$ rm .config
$ cp arch/arm/configs/omap_4430sdp_defconfig .config
$ yes "" | ARCH=arm make oldconfig

Then omap2 and 3 won't get selected by make oldconfig.

Yes, this seems to work.

Quoted 5 lines

Regards,

Tony

#11 lore raw

Re: [PATCH] omap: Fix sev instruction usage for multi-omap

In reply to Tony Lindgren
Quoted 34 lines · Tony Lindgren
Quoted 24 lines · Tony Lindgren
Quoted 23 lines · Kevin Hilman
Quoted 23 lines

Also with omap_4430sdp_defconfig, I see these compile errors arch/arm/kernel/entry-armv.S: Assembler messages:
arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' arch/arm/kernel/entry-armv.S:48: Error: bad instruction test_for_ipi r0,r6,r5,lr' make[1]: *** [arch/arm/kernel/entry-armv.o] Error 1
make: *** [arch/arm/kernel] Error 2

Doing a git log on entry-armv.S shows me a top commit which might be an issue if conflicts are'nt resolved well.

commit 7b70c4275f28702b76b273c8534c38f8313812e9
Merge: ceb0885... a20df56...
Author: Russell King <[email protected]>
Date: Sat Jul 31 14:20:16 2010 +0100

Merge branch 'devel-stable' into devel

Conflicts:
arch/arm/kernel/entry-armv.S
arch/arm/kernel/setup.c
arch/arm/mm/init.c

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Maybe this is an issue in Tony's for-next as well. Haven't tested it though.

Yeah, I'm guessing this an issue in for-next, and probably l-o master too.

Noticed that with omap3_defconfig with CONFIG_SMP enabled. Does the following work for you?

Quoted 7 lines

Here's a related patch that allows CONFIG_SMP to compile with omap3_defconfig. Booting still won't work before some arm generic code is changed.

Regards,

Tony

Tony,

I also did similar thing these days. And yes, with these 2 CONFIG_SMP related patches in omap3_defconfig, I can built a single for omap2/3/4 with CONFIG_SMP=y.

But the kernel still doesn't boot on both omap3 beagle board and omap4 panda board on my side.

I'm very glad to help this out.

Thanks,

Signature · 5 lines
Bryan Wu <[email protected]>
Kernel Developer    +86.138-1617-6545 Mobile
Ubuntu Kernel Team | Hardware Enablement Team
Canonical Ltd.      www.canonical.com
Ubuntu - Linux for human beings | www.ubuntu.com
maintainer
#12 lore raw

Re: [PATCH] omap: Fix sev instruction usage for multi-omap

Quoted 8 lines · Bryan Wu

I also did similar thing these days. And yes, with these 2 CONFIG_SMP related patches in omap3_defconfig, I can built a single for omap2/3/4 with CONFIG_SMP=y.

But the kernel still doesn't boot on both omap3 beagle board and omap4 panda board on my side.

I'm very glad to help this out.

Yeah let's try to figure out what needs to be done to boot it. To me it looks like we should be able to get omap3 & 4 kernel working with CONFIG_SMP with (hopefully) minor changes.

Then getting omap2 working will require more work as some instructions require CONFIG_CPU_32v6K extensions, and 24xx does not support that.

Regards,

Tony