Skip to contentSkip to conversation

Git tree updated, clocks and PM code reset to mainline

DISCUSSION Tony Lindgren 4 participants 7 messages Apr 3, 2009 – Apr 6, 2009

Subsystems (4)
lore
Share & tools
lore

Tony Lindgren

author maintainer
#0 lore raw

Hi all,

So after playing with merging the current mainline code into linux-omap, I ended up resetting all the clock and PM code to be the same as in the mainline kernel. This allows Paul and Kevin do their branches directly against the mainline, and we'll have those patches ready to go to mainline for the next merge window.

This means that the PM and some clock features will be broken until we can merge in Paul's clock branch and Kevin's pm branch.

Please everybody take a look at the big commit that resets the code, it's commit eba05254cb561dc27d5664503f91f7c21954e648. I've made sure things build for all omap configs, and booted several boards. But I may have accidentally left out something critical too..

Anyways, anything that we need to merge back should be posted to this list as patches against the mainline tree. From now on, let's try to make all the patches against the mainline tree in a way where there are no cross-patch dependencies.

For the other stuff that I've reset to mainline, please grep for "REMOVE OMAP LEGACY CODE" in git log of the commits.

I've also branched omap-2.6.29 to have something stable until we have things working properly again.

Cheers,

Tony

#1 lore raw
Quoted 8 lines · On Fri, 3 Apr 2009 02:13:33 +0200 ext Tony Lindgren <[email protected]> wrote:

This means that the PM and some clock features will be broken until we can merge in Paul's clock branch and Kevin's pm branch.

Please everybody take a look at the big commit that resets the code, it's commit eba05254cb561dc27d5664503f91f7c21954e648. I've made sure things build for all omap configs, and booted several boards. But I may have accidentally left out something critical too..

Just a test report. N810 is not booting anymore due an oops and Beagle hangs when calibrating the delay loop.

Both using their defconfigs + CONFIG_DEBUG_LL=y. N810 boot log below.

Jarkko

<1>Unable to handle kernel NULL pointer dereference at virtual address 00000000 <1>pgd = c0004000
<1>[00000000] *pgd=00000000

Internal error: Oops: 5 [#1]
Modules linked in:
CPU: 0    Not tainted  (2.6.29-omap1 #37)
PC is at propagate_rate+0x10/0x60
LR is at omap2_clk_init+0x30/0x218
pc : [<c0037378>]    lr : [<c000e4c8>]    psr: 400001d3

sp : c0335f90 ip : 00000003 fp : 00000000
r10: 80023f4c r9 : 4107b362 r8 : 80023fe8
r7 : c0338f94 r6 : c0025eec r5 : c0025ef0 r4 : ffffffe0 r3 : 0124f800 r2 : 00000000 r1 : 00000040 r0 : c033900c
Flags: nZcv IRQs off FIQs off Mode SVC_32 ISA ARM Segment kernel
Control: 00c5387f Table: 80004000 DAC: 00000017
Process swapper (pid: 0, stack limit = 0xc0334260)

Stack: (0xc0335f90 to 0xc0336000)

5f80: c047806a c0338ff4 c0025ef0 c000e4c8
5fa0: c0337c60 80023fe8 00000000 c0025ef0 c0025eec c0337c60 80023fe8 c000dd48 5fc0: c035c390 c000e6cc fffffed4 c000b5b4 00000000 c0008938 c0008450 00000000 5fe0: 00000000 c0025ef0 00000000 00c5387d c035c420 80008034 00000000 00000000 [<c0037378>] (propagate_rate+0x10/0x60) from [<c000e4c8>] (omap2_clk_init+0x30/0x218) [<c000e4c8>] (omap2_clk_init+0x30/0x218) from [<c000dd48>] (omap2_init_common_hw+0x20/0x3c) [<c000dd48>] (omap2_init_common_hw+0x20/0x3c) from [<c000e6cc>] (nokia_n800_init_irq+0x10/0x60) [<c000e6cc>] (nokia_n800_init_irq+0x10/0x60) from [<c000b5b4>] (init_IRQ+0x34/0x44) [<c000b5b4>] (init_IRQ+0x34/0x44) from [<c0008938>] (start_kernel+0x138/0x2bc) [<c0008938>] (start_kernel+0x138/0x2bc) from [<80008034>] (0x80008034)

Code: e92d4030 e5b02018 e2424020 e24dd004 (e5943020) 

<4>---[ end trace 1b75b31a2719ed1c ]---
<0>Kernel panic - not syncing: Attempted to kill the idle task! [<c002e6bc>] (unwind_backtrace+0x0/0xd4) from [<c004ef00>] (panic+0x38/0x11c) [<c004ef00>] (panic+0x38/0x11c) from [<c00513f4>] (do_exit+0x204/0x65c) [<c00513f4>] (do_exit+0x204/0x65c) from [<c002cc64>] (die+0x120/0x140) [<c002cc64>] (die+0x120/0x140) from [<c002f5b4>] (__do_kernel_fault+0x68/0x80) [<c002f5b4>] (__do_kernel_fault+0x68/0x80) from [<c002f8d4>] (do_page_fault+0x1f4/0x214) [<c002f8d4>] (do_page_fault+0x1f4/0x214) from [<c00281f0>] (do_DataAbort+0x34/0x9c) [<c00281f0>] (do_DataAbort+0x34/0x9c) from [<c002898c>] (__dabt_svc+0x4c/0x60)

Exception stack(0xc0335f48 to 0xc0335f90)

5f40: c033900c 00000040 00000000 0124f800 ffffffe0 c0025ef0
5f60: c0025eec c0338f94 80023fe8 4107b362 80023f4c 00000000 00000003 c0335f90 5f80: c000e4c8 c0037378 400001d3 ffffffff
[<c002898c>] (__dabt_svc+0x4c/0x60) from [<c000e4c8>] (omap2_clk_init+0x30/0x218) [<c000e4c8>] (omap2_clk_init+0x30/0x218) from [<c0025ef0>] (command_line+0x0/0x400) [<c0025ef0>] (command_line+0x0/0x400) from [<c0337c60>] (0xc0337c60)

author maintainer
#2 lore raw
Quoted 11 lines · Jarkko Nikula
Quoted 8 lines · On Fri, 3 Apr 2009 02:13:33 +0200 ext Tony Lindgren <[email protected]> wrote:

This means that the PM and some clock features will be broken until we can merge in Paul's clock branch and Kevin's pm branch.

Please everybody take a look at the big commit that resets the code, it's commit eba05254cb561dc27d5664503f91f7c21954e648. I've made sure things build for all omap configs, and booted several boards. But I may have accidentally left out something critical too..

Quoted 4 lines

Just a test report. N810 is not booting anymore due an oops and Beagle hangs when calibrating the delay loop.

Both using their defconfigs + CONFIG_DEBUG_LL=y. N810 boot log below.

It sounds like the Beagle issue depends on the compiler where 2007q3 codesourcery one won't boot for some reason.

Tony

Quoted 47 lines

Jarkko

<1>Unable to handle kernel NULL pointer dereference at virtual address 00000000 <1>pgd = c0004000
<1>[00000000] *pgd=00000000

Internal error: Oops: 5 [#1]
Modules linked in:
CPU: 0    Not tainted  (2.6.29-omap1 #37)
PC is at propagate_rate+0x10/0x60
LR is at omap2_clk_init+0x30/0x218
pc : [<c0037378>]    lr : [<c000e4c8>]    psr: 400001d3

sp : c0335f90 ip : 00000003 fp : 00000000
r10: 80023f4c r9 : 4107b362 r8 : 80023fe8
r7 : c0338f94 r6 : c0025eec r5 : c0025ef0 r4 : ffffffe0 r3 : 0124f800 r2 : 00000000 r1 : 00000040 r0 : c033900c
Flags: nZcv IRQs off FIQs off Mode SVC_32 ISA ARM Segment kernel
Control: 00c5387f Table: 80004000 DAC: 00000017
Process swapper (pid: 0, stack limit = 0xc0334260)

Stack: (0xc0335f90 to 0xc0336000)

5f80: c047806a c0338ff4 c0025ef0 c000e4c8
5fa0: c0337c60 80023fe8 00000000 c0025ef0 c0025eec c0337c60 80023fe8 c000dd48 5fc0: c035c390 c000e6cc fffffed4 c000b5b4 00000000 c0008938 c0008450 00000000 5fe0: 00000000 c0025ef0 00000000 00c5387d c035c420 80008034 00000000 00000000 [<c0037378>] (propagate_rate+0x10/0x60) from [<c000e4c8>] (omap2_clk_init+0x30/0x218) [<c000e4c8>] (omap2_clk_init+0x30/0x218) from [<c000dd48>] (omap2_init_common_hw+0x20/0x3c) [<c000dd48>] (omap2_init_common_hw+0x20/0x3c) from [<c000e6cc>] (nokia_n800_init_irq+0x10/0x60) [<c000e6cc>] (nokia_n800_init_irq+0x10/0x60) from [<c000b5b4>] (init_IRQ+0x34/0x44) [<c000b5b4>] (init_IRQ+0x34/0x44) from [<c0008938>] (start_kernel+0x138/0x2bc) [<c0008938>] (start_kernel+0x138/0x2bc) from [<80008034>] (0x80008034)

Code: e92d4030 e5b02018 e2424020 e24dd004 (e5943020) 

<4>---[ end trace 1b75b31a2719ed1c ]---
<0>Kernel panic - not syncing: Attempted to kill the idle task! [<c002e6bc>] (unwind_backtrace+0x0/0xd4) from [<c004ef00>] (panic+0x38/0x11c) [<c004ef00>] (panic+0x38/0x11c) from [<c00513f4>] (do_exit+0x204/0x65c) [<c00513f4>] (do_exit+0x204/0x65c) from [<c002cc64>] (die+0x120/0x140) [<c002cc64>] (die+0x120/0x140) from [<c002f5b4>] (__do_kernel_fault+0x68/0x80) [<c002f5b4>] (__do_kernel_fault+0x68/0x80) from [<c002f8d4>] (do_page_fault+0x1f4/0x214) [<c002f8d4>] (do_page_fault+0x1f4/0x214) from [<c00281f0>] (do_DataAbort+0x34/0x9c) [<c00281f0>] (do_DataAbort+0x34/0x9c) from [<c002898c>] (__dabt_svc+0x4c/0x60)

Exception stack(0xc0335f48 to 0xc0335f90)

5f40: c033900c 00000040 00000000 0124f800 ffffffe0 c0025ef0
5f60: c0025eec c0338f94 80023fe8 4107b362 80023f4c 00000000 00000003 c0335f90 5f80: c000e4c8 c0037378 400001d3 ffffffff
[<c002898c>] (__dabt_svc+0x4c/0x60) from [<c000e4c8>] (omap2_clk_init+0x30/0x218) [<c000e4c8>] (omap2_clk_init+0x30/0x218) from [<c0025ef0>] (command_line+0x0/0x400) [<c0025ef0>] (command_line+0x0/0x400) from [<c0337c60>] (0xc0337c60)

#3 lore raw
In reply to Jarkko Nikula
Quoted 11 lines · Jarkko Nikula
Quoted 8 lines · On Fri, 3 Apr 2009 02:13:33 +0200 ext Tony Lindgren <[email protected]> wrote:

This means that the PM and some clock features will be broken until we can merge in Paul's clock branch and Kevin's pm branch.

Please everybody take a look at the big commit that resets the code, it's commit eba05254cb561dc27d5664503f91f7c21954e648. I've made sure things build for all omap configs, and booted several boards. But I may have accidentally left out something critical too..

Quoted 4 lines

Just a test report. N810 is not booting anymore due an oops and Beagle hangs when calibrating the delay loop.

Both using their defconfigs + CONFIG_DEBUG_LL=y. N810 boot log below.

I see the same with N800, same configs as Jarkko had:

<6>SRAM: Mapped pa 0x40200000 to va 0xe3000000 size: 0x100000 Built 1 zonelists in Zone order, mobility grouping on. Total pages:
32512
<5>Kernel command line: root=1f03 rootfstype=jffs2 <6>NR_IRQS:368
<1>Unable to handle kernel NULL pointer dereference at virtual address 00000000
<1>pgd = c0004000
<1>[00000000] *pgd=00000000

Internal error: Oops: 5 [#1]                                                    
Modules linked in:                                                              
CPU: 0    Not tainted  (2.6.29-omap1 #118)                                      
PC is at propagate_rate+0x34/0x54                                               
LR is at omap2_clk_init+0x28/0x1c8                                              
pc : [<c003494c>]    lr : [<c000dd0c>]    psr: 400001d3                         

sp : c0313f90 ip : 00000003 fp : 00000000
r10: 80021a04 r9 : 4107b362 r8 : 80021a38
r7 : c0315c60 r6 : c0023ee8 r5 : c0317288 r4 : ffffffe0 r3 : 00000000 r2 : c0317268 r1 : 00000040 r0 : c0317288
Flags: nZcv IRQs off FIQs off Mode SVC_32 ISA ARM Segment kernel
Control: 00c5387f Table: 80004000 DAC: 00000017
Process swapper (pid: 0, stack limit = 0xc0312260)

Stack: (0xc0313f90 to 0xc0314000)                                               

3f80: c0316cc8 c03172e8 c0023ee8
c000dd0c
3fa0: 00000000 00000001 00000000 c0339000 c0023ee8 c0315c60 80021a38 c000d5c0
3fc0: c0023eec c000e3d4 c0023eec c000b110 c0023eec c00089c8 c00084bc 00000000
3fe0: 00000000 c0023eec 00000000 00c5387d c0339120 80008034 00000000 00000000
[<c003494c>] (propagate_rate+0x34/0x54) from [<c000dd0c>] (omap2_clk_init+0x28/0
x1c8)
[<c000dd0c>] (omap2_clk_init+0x28/0x1c8) from [<c000d5c0>] (omap2_init_common_hw
+0x20/0x38)
[<c000d5c0>] (omap2_init_common_hw+0x20/0x38) from [<c000e3d4>] (nokia_n800_init

_irq+0xc/0x50)                                                                  
[<c000e3d4>] (nokia_n800_init_irq+0xc/0x50) from [<c000b110>]
(init_IRQ+0x30/0x4

0)
[<c000b110>] (init_IRQ+0x30/0x40) from [<c00089c8>] (start_kernel+0x120/0x288)
[<c00089c8>] (start_kernel+0x120/0x288) from [<80008034>] (0x80008034)

Code: e1a00004 ebfffff4 e5943020 e2434020 (e5943020)                            

<4>---[ end trace 1b75b31a2719ed1c ]---
<0>Kernel panic - not syncing: Attempted to kill the idle task! [<c002c5e4>] (unwind_backtrace+0x0/0xe0) from [<c004abb8>] (panic+0x40/0x110)
[<c004abb8>] (panic+0x40/0x110) from [<c004d4d4>] (do_exit+0x60/0x644) [<c004d4d4>] (do_exit+0x60/0x644) from [<c002ab70>] (die+0x12c/0x14c) [<c002ab70>] (die+0x12c/0x14c) from [<c002d450>] (__do_kernel_fault+0x64/0x74)
[<c002d450>] (__do_kernel_fault+0x64/0x74) from [<c002d65c>] (do_page_fault+0x1f
c/0x218)
[<c002d65c>] (do_page_fault+0x1fc/0x218) from [<c002620c>] (do_DataAbort+0x34/0x
98)
[<c002620c>] (do_DataAbort+0x34/0x98) from [<c002692c>] (__dabt_svc+0x4c/0x60)

Exception stack(0xc0313f48 to 0xc0313f90)                                       

3f40: c0317288 00000040 c0317268 00000000 ffffffe0
c0317288
3f60: c0023ee8 c0315c60 80021a38 4107b362 80021a04 00000000 00000003 c0313f90
3f80: c000dd0c c003494c 400001d3 ffffffff
[<c002692c>] (__dabt_svc+0x4c/0x60) from [<c000dd0c>] (omap2_clk_init+0x28/0x1c8
)
[<c000dd0c>] (omap2_clk_init+0x28/0x1c8) from [<c0339000>] (0xc0339000)

Signature · 1 line
Kalle Valo
maintainer
#4 lore raw

Hello Kalle, Jarkko,

Quoted 8 lines · Kalle Valo
Quoted 4 lines · Jarkko Nikula

Just a test report. N810 is not booting anymore due an oops and Beagle hangs when calibrating the delay loop.

Both using their defconfigs + CONFIG_DEBUG_LL=y. N810 boot log below.

I see the same with N800, same configs as Jarkko had:

Evidently the 24xx clock code/data was never actually tested. Does this patch work for you?

regards,

- Paul

From: Paul Walmsley <[email protected]>
Date: Sat, 4 Apr 2009 16:26:39 -0600

OMAP2 clock fixes rollup: test

This test patch includes three fixes to the OMAP2 clock code and data to get N8x0 booting again.

---
 arch/arm/mach-omap2/clock24xx.c |   14 +++++---------
 1 files changed, 5 insertions(+), 9 deletions(-)
arch/arm/mach-omap2/clock24xx.c +5 -9
index 1e839c5..7625d06 100644
@@ -60,8 +60,8 @@ struct omap_clk {
 		},			\
 	}
 
-#define CK_243X	(1 << 0)
-#define CK_242X	(1 << 1)
+#define CK_243X		RATE_IN_243X
+#define CK_242X		RATE_IN_242X
 
 static struct omap_clk omap24xx_clks[] = {
 	/* external root sources */
@@ -711,7 +711,7 @@ int __init omap2_clk_init(void)
 {
 	struct prcm_config *prcm;
 	struct omap_clk *c;
-	u32 clkrate, cpu_mask;
+	u32 clkrate;
 
 	if (cpu_is_omap242x())
 		cpu_mask = RATE_IN_242X;
@@ -720,6 +720,8 @@ int __init omap2_clk_init(void)
 
 	clk_init(&omap2_clk_functions);
 
+	clk_init_one(&osc_ck);
+	clk_init_one(&sys_ck);
 	osc_ck.rate = omap2_osc_clk_recalc(&osc_ck);
 	propagate_rate(&osc_ck);
 	sys_ck.rate = omap2_sys_clk_recalc(&sys_ck);
@@ -728,12 +730,6 @@ int __init omap2_clk_init(void)
 	for (c = omap24xx_clks; c < omap24xx_clks + ARRAY_SIZE(omap24xx_clks); c++)
 		clk_init_one(c->lk.clk);
 
-	cpu_mask = 0;
-	if (cpu_is_omap2420())
-		cpu_mask |= CK_242X;
-	if (cpu_is_omap2430())
-		cpu_mask |= CK_243X;
-
 	for (c = omap24xx_clks; c < omap24xx_clks + ARRAY_SIZE(omap24xx_clks); c++)
 		if (c->cpu & cpu_mask) {
 			clkdev_add(&c->lk);
Signature · 1 line
1.6.2.1.307.g91408
#5 lore raw
Quoted 3 lines · On Sun, 5 Apr 2009 00:33:25 +0200 ext Paul Walmsley <[email protected]> wrote:

Evidently the 24xx clock code/data was never actually tested. Does this patch work for you?

Thanks Paul! Your patch makes N810 booting up to retu initialization (kernel BUG at drivers/cbus/retu.c:90!) but I don't think so this has anything to do with the clock changes.

I noticed also that N810 hangs like Beagle in delay loop calibration when compiled with 2007q3-53. For <= ARM11 I still use good old 2005q3 :-)

Jarkko

maintainer
#6 lore raw
Quoted 9 lines · Jarkko Nikula
Quoted 3 lines · On Sun, 5 Apr 2009 00:33:25 +0200 ext Paul Walmsley <[email protected]> wrote:

Evidently the 24xx clock code/data was never actually tested. Does this patch work for you?

Thanks Paul! Your patch makes N810 booting up to retu initialization (kernel BUG at drivers/cbus/retu.c:90!) but I don't think so this has anything to do with the clock changes.

Cool, thanks for testing, Jarkko. Will queue cleaned-up versions of these fixes into the omap-clock-test branch. I test-booted them on the N800 here and didn't see the retu bug, but maybe N800 interacts with it differently?

Quoted 3 lines

I noticed also that N810 hangs like Beagle in delay loop calibration when compiled with 2007q3-53. For <= ARM11 I still use good old 2005q3 :-)

Hehe.

- Paul