Yahoo Groups archive
Thread
2005-04-29 by Ralf Knorr
Hi Chris, I'm writing an IAP bootloader for the LPC2114 and test it also on an LPC2106 and I saw, that the flash erase takes about 100ms. I think this depends on the charge pumps to get the internal stable programming voltage for the flash. I tested erasing one or all sectors and the time is nearly constant. If you use the keil IAP example there is also a for loop which takes about 300ms. This would explain your 400ms. Regards, Ralf
2005-04-29 by sig5534
> This would explain your 400ms. > Regards, Ralf Thanks Ralf. At least I know I am not crazy now. Boy that really sucks. I can understand it taking 100mS to build up a charge, but 300mS to burn 1024 bytes of flash? That's 38 seconds to burn all 128K. Very slow. Do you think there is anyway to do a page write kind of thing, or something else to reduce this time? Thanks, Chris.
2005-04-29 by Charles Manning
On Friday 29 April 2005 20:31, sig5534 wrote: > > This would explain your 400ms. > > Regards, Ralf > > Thanks Ralf. At least I know I am not crazy now. > > Boy that really sucks. I can understand it taking 100mS to build up a > charge, but 300mS to burn 1024 bytes of flash? That's 38 seconds to > burn all 128K. Very slow. > > Do you think there is anyway to do a page write kind of thing, or > something else to reduce this time? You can do the whole erase as a single IAP operation. This is takes only a short time (~1 sec) for 128kB on an LPC2129.
2005-04-29 by Ralf Knorr
Ok, I've just remeasured the following times on an LPC2106: Erase sector 1-14: 100ms Program 512 bytes starting from adress 0x2200: 250\ufffds My xtal is 14.7456MHz and I use PLL with multiplying factor 4, and Pll devider 2. VPBdivder is set to CPU clock /4. If you want to programm whole flash (0x0-0x1DFFFF) it should take about ~1s. Ralf sig5534 schrieb: > > This would explain your 400ms. > > Regards, Ralf > > Thanks Ralf. At least I know I am not crazy now. > > Boy that really sucks. I can understand it taking 100mS to build up a > charge, but 300mS to burn 1024 bytes of flash? That's 38 seconds to > burn all 128K. Very slow. > > Do you think there is anyway to do a page write kind of thing, or > something else to reduce this time? > > Thanks, Chris. > > > > > > ------------------------------------------------------------------------ > *Yahoo! Groups Links* > > * To visit your group on the web, go to: > http://groups.yahoo.com/group/lpc2000/ > > * To unsubscribe from this group, send an email to: > lpc2000-unsubscribe@yahoogroups.com > <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe> > > * Your use of Yahoo! Groups is subject to the Yahoo! Terms of > Service <http://docs.yahoo.com/info/terms/>. > > -- Mit freundlichen Gr\ufffd\ufffden / Best regards Ralf Knorr ___________________________________________________________________________________________ * Author: * Dipl.-Ing. Ralf Knorr (FH) Development: Area leader Software Bury GmbH & Co KG Robert-Koch-Str. 1-7 32584 L\ufffdhne Germany *Phone:* +49 (0) 5732 9706 0 *Fax:* +49 (0) 5732 9706 209 *Email:* knorr@... <mailto:knorr@...> *Internet:* http://www.thb.de <http://www.thb.de/> This email and any files transmitted with it are confidential and intended for the adressee only. If you are not the adressee you may not copy, forward, disclose or otherwise use it, or any part of it, in any form whatever. If you have received this e-mail in error please notify the sender and ensure that all copies of this e-mail and any files transmitted with it are deleted. Any views or opinions represented in this e-mail are solely those of the author and do not necessarily represented those of Bury GmbH &Co KG or its affiliate companies. ___________________________________________________________________________________________
2005-04-29 by sig5534
> Erase sector 1-14: 100ms > Program 512 bytes starting from adress 0x2200: 250µs Well I'm confused now. I am seeing 400mS to erase one sector and prog 1K of flash. Are you waiting to see if the prog is actually done? I am doing that, and the IAP_BUSY status does not go down until 400mS. I can also verify that if the power goes down in 300mS the programming does not get completed. The erase does, but not the prog. So my 1024 bytes to prog is certainly taking much longer than 500uS as you suggest. Any ideas? Thanks, Chris.
2005-04-29 by Ralf Knorr
Here is an extraction from my program:
/******************************************************************/
/* Program *data to flash_addr. number of bytes specified by size */
/* Return: IAP error code (0 when OK) */
/******************************************************************/
UINT32 gIAPProgram (void *flash_addr, void *data, UINT32 size)
{
struct iap_in iap; // IAP input parameters
UINT32 result[16]; // IAP results
WDMOD=0x00; // stop watchdog
VICIntEnClr = 0xFFFFFFFF; // disable all interrupts
iap.cmd = FLASH_CMD_PREP_SEC; // IAP
Command: Prepare Sectors for Write
iap.par[0] = get_secnum (flash_addr); // start sector
iap.par[1] = iap.par[0]; // end Sektor
iap_entry (&iap, result); // call IAP function
if (result[0]==CMD_SUCCESS) // an error occured?
{
iap.cmd = FLASH_CMD_CPY_RAM_TO_FLASH;
// IAP Command: Copy RAM to Flash
iap.par[0] = (UINT32) flash_addr; // destination-addr
iap.par[1] = (UINT32) data; // source-addr
iap.par[2] = size; // number of bytes
iap.par[3] = CCLK; // CPU clock
iap_entry (&iap, result); // call IAP function
}
gInit_Watchdog();
VICIntEnable = gEnabledInterrupts;
gCopySram32ToSram32BufferBinary(result,gIAP_Results,0x0000,16);
return (result[0]);
}
UINT8 gIAP_DataBuffer[512];
//somewhere in IAP MainLoop
{
...
#ifdef DEBUG_INFO
CplPortPin(PORT0,Port0_06);
#endif
i=gIAPErase(FlashSector01_START, FlashSector14_END);//
erasing time is ~100ms, independant if 1 or more sectors
#ifdef DEBUG_INFO
CplPortPin(PORT0,Port0_06);
#endif
...
#endif
#ifdef DEBUG_INFO
CplPortPin(PORT0,Port0_07);
#endif
i=gIAPProgram
((FlashSector01_START+OffsetProgrammedData), gIAP_DataBuffer, sizeof
(gIAP_DataBuffer));
#ifdef DEBUG_INFO
CplPortPin(PORT0,Port0_07);
#endif
...
}
Times are measured with scope on portpins
Ralf
sig5534 schrieb:
> > Erase sector 1-14: 100ms
> > Program 512 bytes starting from adress 0x2200: 250\ufffds
>
> Well I'm confused now. I am seeing 400mS to erase one sector and prog
> 1K of flash. Are you waiting to see if the prog is actually done? I
> am doing that, and the IAP_BUSY status does not go down until 400mS.
>
> I can also verify that if the power goes down in 300mS the programming
> does not get completed. The erase does, but not the prog.
>
> So my 1024 bytes to prog is certainly taking much longer than 500uS as
> you suggest.
>
> Any ideas?
>
> Thanks, Chris.
>
>
>
>
>
> ------------------------------------------------------------------------
> *Yahoo! Groups Links*
>
> * To visit your group on the web, go to:
> http://groups.yahoo.com/group/lpc2000/
>
> * To unsubscribe from this group, send an email to:
> lpc2000-unsubscribe@yahoogroups.com
> <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe>
>
> * Your use of Yahoo! Groups is subject to the Yahoo! Terms of
> Service <http://docs.yahoo.com/info/terms/>.
>
>
--
Mit freundlichen Gr\ufffd\ufffden / Best regards
Ralf Knorr
___________________________________________________________________________________________
* Author: *
Dipl.-Ing. Ralf Knorr (FH)
Development: Area leader Software
Bury GmbH & Co KG
Robert-Koch-Str. 1-7
32584 L\ufffdhne
Germany
*Phone:* +49 (0) 5732 9706 0
*Fax:* +49 (0) 5732 9706 209
*Email:* knorr@... <mailto:knorr@...>
*Internet:* http://www.thb.de <http://www.thb.de/>
This email and any files transmitted with it are confidential and
intended for the adressee only.
If you are not the adressee you may not copy, forward, disclose or
otherwise use it, or any part of it, in any form whatever.
If you have received this e-mail in error please notify the sender and
ensure that all copies of this e-mail and any files transmitted with it
are deleted.
Any views or opinions represented in this e-mail are solely those of the
author and do not necessarily represented those of Bury GmbH &Co KG or
its affiliate companies.
___________________________________________________________________________________________2005-04-29 by sig5534
Your code is not much different than mine, but I have a few more
things in there for checking busy.
void attrib_RAM FlashVaderData(void) {
Disable_INT(); // Disable INTs
// Prepare to erase
flashParams[0] = IAP_CMD_PREPARE_SECTORS; // prepare sector
(s)
flashParams[1] = CFG_FlashDataBlock_Sec; // start sector
flashParams[2] = CFG_FlashDataBlock_Sec; // end sector
callIAP();
IOSET=IO_USB_ACK; // toggle USB line
// Erase
WaitTillFlashReady();
flashParams[0]=IAP_CMD_ERASE_SECTORS;
flashParams[1]=CFG_FlashDataBlock_Sec;
flashParams[2]=CFG_FlashDataBlock_Sec;
flashParams[3]=CCLK/1000;// CCLK in KHz
callIAP();
IOCLR=IO_USB_ACK; // toggle USB line
// Prepare to program
WaitTillFlashReady();
flashParams[0]=IAP_CMD_PREPARE_SECTORS;
flashParams[1]=CFG_FlashDataBlock_Sec;
flashParams[2]=CFG_FlashDataBlock_Sec;
callIAP();
IOSET=IO_USB_ACK; // toggle USB line
// Program
WaitTillFlashReady();
flashParams[0]=IAP_CMD_COPY_RAM_TO_FLASH;
flashParams[1]=CFG_FlashDataBlock_Adr; // Dst FLASH adr
flashParams[2]=CFG_RamDataBlock_Adr; // Src RAM adr
flashParams[3]=1024;
flashParams[4]=CCLK/1000;
// CCLK in KHz
callIAP();
IOCLR=IO_USB_ACK; // toggle USB line
WaitTillFlashReady();
Enable_INT(); // Enable INTs
}
Essentially here are the 4 steps and their timing:
IAP_CMD_PREPARE_SECTORS (<4mS)
IAP_CMD_ERASE_SECTORS (400mS)
IAP_CMD_PREPARE_SECTORS (<4mS)
IAP_CMD_COPY_RAM_TO_FLASH (<8mS)
I am simply toggling a line IO_USB_ACK on the board and sampling with
the scope.
So all of my time is spent in doing the erase task on this single
sector - 400mSec. That's the problem.
Why do you think yours is 100mS and mine is 400mS?
Thanks, Chris.2005-04-29 by sig5534
Well I think I found an answer. My system is using a 24MHz xtal, and I am using the PLL to up it to 48MHz. As you see I was div the 48MHz by 1000 to give IAP the value in kHz. So I tried div the 48MHz by 2000 instead: The erase time dropped from 400mS to 200mS. Obviously this clock parameter is directly affecting the timing. Is it suppose to be the CCLK(48MHz) or the XtalOsc(24MHz) ? If your's is coming out at 100mS, seems like you would have to have yet another ratio, say div by 4. What is your PLL freq, Xtal Freq, and the value you are putting into the IAP param? Thanks, Chris.
2005-05-02 by Ralf Knorr
My XTAL freq is 14.7456 MHz Pll factor is 4 -> CPUclk=58,9824MHz PLL div is 2 -> 29,4912MHz VPB div is 4 -> 14.7456MHz But I saw maybe a problem in software, because I set the value to program flash to XTALfreq /1000. I've to recheck it when I'll finish my bootloader. Generally, the correct factor would be the CPUclk. Regards, Ralf sig5534 schrieb: > Well I think I found an answer. My system is using a 24MHz xtal, and I > am using the PLL to up it to 48MHz. > > As you see I was div the 48MHz by 1000 to give IAP the value in kHz. > So I tried div the 48MHz by 2000 instead: > > The erase time dropped from 400mS to 200mS. > > Obviously this clock parameter is directly affecting the timing. > > Is it suppose to be the CCLK(48MHz) or the XtalOsc(24MHz) ? > > If your's is coming out at 100mS, seems like you would have to have yet > another ratio, say div by 4. > > What is your PLL freq, Xtal Freq, and the value you are putting into > the IAP param? > > Thanks, Chris. > > > > > ------------------------------------------------------------------------ > *Yahoo! Groups Links* > > * To visit your group on the web, go to: > http://groups.yahoo.com/group/lpc2000/ > > * To unsubscribe from this group, send an email to: > lpc2000-unsubscribe@yahoogroups.com > <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe> > > * Your use of Yahoo! Groups is subject to the Yahoo! Terms of > Service <http://docs.yahoo.com/info/terms/>. > > -- Mit freundlichen Gr\ufffd\ufffden / Best regards Ralf Knorr ___________________________________________________________________________________________ * Author: * Dipl.-Ing. Ralf Knorr (FH) Development: Area leader Software Bury GmbH & Co KG Robert-Koch-Str. 1-7 32584 L\ufffdhne Germany *Phone:* +49 (0) 5732 9706 0 *Fax:* +49 (0) 5732 9706 209 *Email:* knorr@... <mailto:knorr@...> *Internet:* http://www.thb.de <http://www.thb.de/> This email and any files transmitted with it are confidential and intended for the adressee only. If you are not the adressee you may not copy, forward, disclose or otherwise use it, or any part of it, in any form whatever. If you have received this e-mail in error please notify the sender and ensure that all copies of this e-mail and any files transmitted with it are deleted. Any views or opinions represented in this e-mail are solely those of the author and do not necessarily represented those of Bury GmbH &Co KG or its affiliate companies. ___________________________________________________________________________________________