Skip to contentSkip to conversation

Format of MMUCSR0 on 8540

U-Boot-Users Charles J Gillan 3 participants 6 messages Nov 18, 2005

lore
Share & tools
lore
#0 lore raw

I downloaded U-boot from CVS on Nov 14th and have compiled and run this on a TQM8540 board (with a Rev 2 MPC8540 chip).

However, I started stepping through this with the CodeWarrior debugger (and a FreeScale USB TAP - COP version) find that the "isync" in the
following code from start.S is actually causing Interrupt 14 to be executed (TLBInstructionException).

	/*
	 * Invalidate MMU L1/L2
	 *
	 * Note: There is a fixup earlier for Errata CPU4 on
	 * Rev 1 parts that must precede this MMU invalidation.
	 */
	li	r2, 0x001e
	mtspr	MMUCSR0, r2
	isync  <------------------- Interrupt

Without CodeWarrior, I do eventually see the U-Boot command line appear on the serial port (however it did take several seconds after power on, more than I expected).

Checking the 8540RM manual, I see that there appear to be only two bits that need to be set to invalidate cache (page 633 on MMUCSR0).

Can anyone on the explain the presence of

       0x001e

as opposed to

       0x0006 

in the above code?

Any suggestions as to why this exception arises. Has anyone seen it with An Abatron BDI2000 or not - as the case may be.

Thanks,

Charles.

--------------------------------------------------------------------------
?
Dr Charles J Gillan
The Institute of Electronics, Communications and

       Information Technology (ECIT),???????????????

Queen's University Belfast, Titanic Quarter
Queen?s Road, Queen?s Island, Belfast, BT3 9DT
Northern Ireland, UK
?
Tel: +44 (0) 2890 971847
Fax: +44 (0) 2890 971702
??
--------------------------------------------------------------------------

#1 lore raw
Quoted 29 lines · Charles J Gillan

I downloaded U-boot from CVS on Nov 14th and have compiled and run this on a TQM8540 board (with a Rev 2 MPC8540 chip).

However, I started stepping through this with the CodeWarrior debugger (and a FreeScale USB TAP - COP version) find that the "isync" in the
following code from start.S is actually causing Interrupt 14 to be executed (TLBInstructionException).

	/*
	 * Invalidate MMU L1/L2
	 *
	 * Note: There is a fixup earlier for Errata CPU4 on
	 * Rev 1 parts that must precede this MMU invalidation.
	 */
	li	r2, 0x001e
	mtspr	MMUCSR0, r2
	isync  <------------------- Interrupt

Without CodeWarrior, I do eventually see the U-Boot command line appear on the serial port (however it did take several seconds after power on, more than I expected).

Checking the 8540RM manual, I see that there appear to be only two bits that need to be set to invalidate cache (page 633 on MMUCSR0).

Can anyone on the explain the presence of

Quoted 7 lines
       0x001e

as opposed to

       0x0006

in the above code?

It's flash invalidating ALL of the TLB arrays. A value of 0x6 will only do the L2 arrays.

Quoted 3 lines

Any suggestions as to why this exception arises. Has anyone seen it with
An Abatron BDI2000 or not - as the case may be.

Can you see what's in the following registers, ESR, SRR0, SRR1?

Also, if I remember correctly there are some patches to u-boot for it to work properly with the CodeWarrior JTAG debuggers.

- kumar

#2 lore raw

Thanks Kumar, for the information.

> Can you see what's in the following registers, ESR, SRR0, SRR1?

I stopped the code at the interrupt handler and see the following:

ESR = 0x00000000
SRR0 = 0xFFFC0E00 (the address set in IVOR14 was 0x0e00,
TEXT_BASE = FFFC0000)
SRR1 = 0x00000200

There is no information on page 6-18 of the 8540RM where these registers are defined. Does the SRR1 value have significance?

> Also, if I remember correctly there are some patches to u-boot for

> it to work properly with the CodeWarrior JTAG debuggers.

From where would I get these? Only FreeScale/Metrowerks ?

Charles.

-----Original Message-----
From: Kumar Gala [mailto:galak at kernel.crashing.org]
Sent: 18 November 2005 15:43
To: C.Gillan at ecit.qub.ac.uk
Cc: u-boot-users at lists.sourceforge.net
Subject: Re: [U-Boot-Users] Format of MMUCSR0 on 8540

Quoted 29 lines · Charles J Gillan

I downloaded U-boot from CVS on Nov 14th and have compiled and run this on a TQM8540 board (with a Rev 2 MPC8540 chip).

However, I started stepping through this with the CodeWarrior debugger (and a FreeScale USB TAP - COP version) find that the "isync" in the
following code from start.S is actually causing Interrupt 14 to be executed (TLBInstructionException).

	/*
	 * Invalidate MMU L1/L2
	 *
	 * Note: There is a fixup earlier for Errata CPU4 on
	 * Rev 1 parts that must precede this MMU invalidation.
	 */
	li	r2, 0x001e
	mtspr	MMUCSR0, r2
	isync  <------------------- Interrupt

Without CodeWarrior, I do eventually see the U-Boot command line appear on the serial port (however it did take several seconds after power on, more than I expected).

Checking the 8540RM manual, I see that there appear to be only two bits that need to be set to invalidate cache (page 633 on MMUCSR0).

Can anyone on the explain the presence of

Quoted 7 lines
       0x001e

as opposed to

       0x0006

in the above code?

It's flash invalidating ALL of the TLB arrays. A value of 0x6 will only do the L2 arrays.

Quoted 3 lines

Any suggestions as to why this exception arises. Has anyone seen it with
An Abatron BDI2000 or not - as the case may be.

Can you see what's in the following registers, ESR, SRR0, SRR1?

Also, if I remember correctly there are some patches to u-boot for it to work properly with the CodeWarrior JTAG debuggers.

- kumar

#3 lore raw
Quoted 9 lines · Charles J Gillan

Thanks Kumar, for the information.

> Can you see what's in the following registers, ESR, SRR0, SRR1?

I stopped the code at the interrupt handler and see the
following:

Quoted 7 lines
	ESR  = 0x00000000
      SRR0 = 0xFFFC0E00  (the address set in IVOR14 was 0x0e00,
                          TEXT_BASE = FFFC0000)
      SRR1 = 0x00000200

      There is no information on page 6-18 of the 8540RM where these
      registers are defined. Does the SRR1 value have significance?

You need the e500 Core UM for these and all CPU registers. These regs are the exception save/restore registers. SRR0 is the address of the instruction that faulted and SRR1 is the MSR when the fault occurred.

Are you able to dump the TLBs out with the debugger you have? If so that would be useful.

Quoted 5 lines

> Also, if I remember correctly there are some patches to u-boot for

Quoted 1 line

it to work properly with the CodeWarrior JTAG debuggers.

From where would I get these? Only FreeScale/Metrowerks ?

Yeah. You'll have to get it from Freescale/Metrowerks. If you download the Freescale Linux "BSP" for 8540 it should have the patches.

- kumar

Quoted 73 lines

-----Original Message-----
From: Kumar Gala [mailto:galak at kernel.crashing.org]
Sent: 18 November 2005 15:43
To: C.Gillan at ecit.qub.ac.uk
Cc: u-boot-users at lists.sourceforge.net
Subject: Re: [U-Boot-Users] Format of MMUCSR0 on 8540

Quoted 29 lines · Charles J Gillan

I downloaded U-boot from CVS on Nov 14th and have compiled and run this on a TQM8540 board (with a Rev 2 MPC8540 chip).

However, I started stepping through this with the CodeWarrior debugger (and a FreeScale USB TAP - COP version) find that the "isync" in the
following code from start.S is actually causing Interrupt 14 to be executed (TLBInstructionException).

	/*
	 * Invalidate MMU L1/L2
	 *
	 * Note: There is a fixup earlier for Errata CPU4 on
	 * Rev 1 parts that must precede this MMU invalidation.
	 */
	li	r2, 0x001e
	mtspr	MMUCSR0, r2
	isync  <------------------- Interrupt

Without CodeWarrior, I do eventually see the U-Boot command line appear on the serial port (however it did take several seconds after power on, more than I expected).

Checking the 8540RM manual, I see that there appear to be only two bits that need to be set to invalidate cache (page 633 on MMUCSR0).

Can anyone on the explain the presence of

Quoted 7 lines
       0x001e

as opposed to

       0x0006

in the above code?

It's flash invalidating ALL of the TLB arrays. A value of 0x6 will only do the L2 arrays.

Quoted 3 lines

Any suggestions as to why this exception arises. Has anyone seen it with
An Abatron BDI2000 or not - as the case may be.

Can you see what's in the following registers, ESR, SRR0, SRR1?

Also, if I remember correctly there are some patches to u-boot for it to work properly with the CodeWarrior JTAG debuggers.

- kumar

-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc. Get Certified Today Register for a JBoss Training Course. Free Certification Exam for All Training Attendees Through End of 2005. For more info visit:
http://ads.osdn.com/?ad_id=7628&alloc_id=16845&op=click
_______________________________________________
U-Boot-Users mailing list
U-Boot-Users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/u-boot-users

#4 lore raw
Quoted 4 lines · In message <[email protected]> you wrote:
Quoted 2 lines

From where would I get these? Only FreeScale/Metrowerks ?

Yeah. You'll have to get it from Freescale/Metrowerks. If you

Why doesn't Freescale post this stuff here on the list?

Best regards,

Wolfgang Denk

Signature · 4 lines
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"Why waste negative entropy on comments, when you could use the  same
entropy to create bugs instead?"                        - Steve Elias
#5 lore raw
Quoted 8 lines · Wolfgang Denk

In message <977AA505-6DBA-4F85-

Quoted 4 lines · ACB8-37B452CF9417 at kernel.crashing.org> you wrote:
Quoted 2 lines

From where would I get these? Only FreeScale/Metrowerks ?

Yeah. You'll have to get it from Freescale/Metrowerks. If you

Why doesn't Freescale post this stuff here on the list?

You would have to ask Freescale that :)

Ron, Neil you guys around??

- kumar