RE: [ECOS] flash erase - network stack

["Nicolas Brouard" <nicolas dot brouard at silicomp dot ca>]

With Per Hebdlom's patch, my application doesn't work better. This patch
does a cyg_drv_isr_lock/unlock around the six bus cycle operation to erase
sector.
I tried with cyg_drv_isr_lock at the beginning of flash_erase_block function
and cyg_drv_isr_unlock at the end, and it works very well. Of course, I'll
have a big delay when I ping my application.
I'm reading the flash data sheet to be more precise in the position of
cyg_drv_isr_lock/unlock functions.

Is it a good way to use cyg_drv_isr_lock/unlock functions to disable
interrupt during flash erase? Or must I use cyg_interrupt_disable?

-----Original Message-----
From: Gary Thomas [mailto:gary@mlbassoc.com]
Sent: Tuesday, November 09, 2004 7:56 AM
To: Nicolas Brouard
Cc: eCos Discussion
Subject: RE: [ECOS] flash erase - network stack


On Tue, 2004-11-09 at 05:25, Nicolas Brouard wrote:
> Thank you for this explanation. I understand I can't debug my application
> while the flash is erased because my Redboot is in the FLASH. But, if I
> execute my application without debugging it, I think no Redboot code is
> executing:

Not necessarily.  Depending on the platform, you may still be executing
some instructions from FLASH, even without debugging.

Try disabling interrupts during the FLASH operations to see if it helps.

>
> RedBoot> load myapp.srec
> Using default protocol (TFTP)
> Entry point: 0x00020000, address range: 0x00020000-0x000895a4
> RedBoot> go
> ...etc...
>
> I have the same problem when I ping the application.
>
> ... Erase from 0x7c3e0000-0x7c6247e4:
> ........$T04thread:00000003;40:7c017eb0;01:000dc5d4;#24
>
> I think this happens because of interrupts and not because RedBoot code
from
> Flash is executed. Do you?
>
> I will try the patch of Per Hedblom.
>
> -----Original Message-----
> From: Gary Thomas [mailto:gary@mlbassoc.com]
> Sent: Monday, November 08, 2004 6:43 PM
> To: Nicolas Brouard
> Cc: eCos Discussion
> Subject: RE: [ECOS] flash erase - network stack
>
>
> On Mon, 2004-11-08 at 14:50, Nicolas Brouard wrote:
> > The code I execute is in RAM.
> > I can debug my code using RedBoot in FLASH with network debugging.
> > I put breakpoints in the function ip_input (ip_input.c) and in the
> function
> > icmp_input (ip_icmp.c)
> > I ping my application (not RedBoot). GDB breaks on input and on
> icmp_input.
> > I continue step by step then I go to icmp_reflect function, etc...
> > If I do that during erasing flash, I have a SIGILL signal (Illegal
> > instruction).
>
> Precisely :-)  When you hit a breakpoint, single step or even just print
> a message, you'll be executing RedBoot from the FLASH.  Once you start
> the erase, execution is no longer possible and you'll get a trap.
>
> The best solution is to run RedBoot in ROMRAM mode - where it never
> actually executes from FLASH [directly].  An alternative would be to
> disable interrupts during FLASH operations.
>
> >
> > -----Original Message-----
> > From: Gary Thomas [mailto:gary@mlbassoc.com]
> > Sent: Monday, November 08, 2004 4:19 PM
> > To: Nicolas Brouard
> > Cc: eCos Discussion
> > Subject: Re: [ECOS] flash erase - network stack
> >
> >
> > On Mon, 2004-11-08 at 14:12, Nicolas Brouard wrote:
> > > Hi,
> > >
> > > I have a problem with the Ethernet Stack (FreeBSD) during flash
erasing.
> > > My board is based on a MPC860. My application initializes the FreeBSD
> > stack
> > > and then erases several sectors of flash (2.5 Mo).
> > > Everything is ok if there is no Ethernet traffic.
> > > But, if I ping my board during flash erasing, the program scratches
> > > immediately:
> > > ... Erase from 0x7c3e0000-0x7c6247e4:
> > > ........$T04thread:00000003;40:7c017eb0;01:000dc5d4;#24
> > >
> > > The flash driver I use is: amd/am29xxxxx variant am29lv640
> >
> > It looks like you are trying to use (execute from) the FLASH while
> > executing the erase.  You don't happen to be using RedBoot in FLASH
> > with network debugging while running your code, do you?
> >
> > --
> > Gary Thomas <gary@mlbassoc.com>
> > MLB Associates
> >
> >
> > --
> > Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
> > and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
> >
> --
> Gary Thomas <gary@mlbassoc.com>
> MLB Associates
>
--
Gary Thomas <gary@mlbassoc.com>
MLB Associates




-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss