January 2006 - U-Boot

[U-Boot-Users] Re: Problem with the cfi flash driver with a x8/x16 flash configuredinx8 mode.
by Txema Lopez 28 Feb '06

28 Feb '06

Alex BASTOS wrote: Thanks Alex, >Txema, > > > >>Does anybody test the cfi flash driver (cfi_flash.c) with a x8/x16 >>flash configured in x8 mode ? I'm trying to port U-Boot for a Lite5200B >>with a S29GL128N flash (CFI compliant) using the cfi flash driver. >>U-Boot detects the flash correctly but It's unable to erase or write the >>flash. In my case the following patch fix the bug but I'm not completly >>sure If It's only a bug with my configuration. Any suggestions? Could >>anybody test the patch in other architecture? >> >> > >I think your topic is very similar to this recent one: > >http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/19970 > > OK, I've checked it. >and previous similar patches have been submited and are still >pending (but it seems they are coming, read the whole thread) > > But not the patch that Wolfgang proposed, isn'it? >So, yes, til now cfi_flash didn't support x8 mode for AMD chips, >but it will. > >Best regards, > >Alex BASTOS > > > > > > I think the patch that Wolfgang proposed could be like that. I've tested it with a S29GL128N x8/x16 in x8 mode. It works fine. But I have no idea what would happen in x16 mode or others modes. I haven't the hardware to check it. --- ../tmp/u-boot/drivers/cfi_flash.c 2006-01-12 12:22:08.000000000 +0100 +++ drivers/cfi_flash.c 2006-01-12 17:21:19.000000000 +0100 @@ -107,9 +107,10 @@ #define AMD_STATUS_TOGGLE 0x40 #define AMD_STATUS_ERROR 0x20 -#define AMD_ADDR_ERASE_START 0x555 -#define AMD_ADDR_START 0x555 -#define AMD_ADDR_ACK 0x2AA + +#define AMD_ADDR_ERASE_START (0xAAA >> (info->portwidth-1)) +#define AMD_ADDR_START (0xAAA >> (info->portwidth-1)) +#define AMD_ADDR_ACK (0x555 >> (info->portwidth-1)) #define FLASH_OFFSET_CFI 0x55 #define FLASH_OFFSET_CFI_RESP 0x10 Best regards, Jose Maria Lopez.

2 1

[U-Boot-Users] Patch for drivers/cfi_flash.c; properly detect write errors.
by Marcus Hall 28 Feb '06

28 Feb '06

Hello.. It seems that there is a minor problem in cfi_flash.c that prevents it from detecting that a flash write error has occurred. The first change in the attached patch is trivial, it just makes the strings output for a protected and a non-protected sector have the same length so the output looks prettier.. The second change is in flash_full_status_check(). We have just called flash_status_check() and then did a switch() on the vendor. For intel, we *should* check if there are any error bits if the previous call returned ERR_OK (Otherwise we will have output an error message in flash_status_check() already.) The original code would only check for error bits if flash_status_check() returns ERR_TIMEOUT. -- Marcus Hall marcushall(a)lucent.com *** ../u-boot-1.1.3/drivers/cfi_flash.c Sat Aug 13 17:53:35 2005 --- u-boot-1.1.3/drivers/cfi_flash.c Tue Aug 23 14:47:06 2005 *************** *** 502,508 **** if ((i % 5) == 0) printf ("\n "); printf (" %08lX%s", ! info->start[i], info->protect[i] ? " (RO) " : " "); #endif } putc ('\n'); --- 502,508 ---- if ((i % 5) == 0) printf ("\n "); printf (" %08lX%s", ! info->start[i], info->protect[i] ? " (RO)" : " "); #endif } putc ('\n'); *************** *** 718,724 **** switch (info->vendor) { case CFI_CMDSET_INTEL_EXTENDED: case CFI_CMDSET_INTEL_STANDARD: ! if ((retcode != ERR_OK) && !flash_isequal (info, sector, 0, FLASH_STATUS_DONE)) { retcode = ERR_INVAL; printf ("Flash %s error at address %lx\n", prompt, --- 718,724 ---- switch (info->vendor) { case CFI_CMDSET_INTEL_EXTENDED: case CFI_CMDSET_INTEL_STANDARD: ! if ((retcode == ERR_OK) && !flash_isequal (info, sector, 0, FLASH_STATUS_DONE)) { retcode = ERR_INVAL; printf ("Flash %s error at address %lx\n", prompt,

2 1

[U-Boot-Users] [PATCH] cfi_flash.c patches
by Sangmoon Kim 28 Feb '06

28 Feb '06

Hi, The two patches attached are for drivers/cfi_flash.c. cfi_flash-protect.patch adds CFG_FLASH_PROTECT_CLEAR because for some flash memories(such as 28F320C3) all banks are protected after reset. cfi_flash-buffer.patch makes write_buff not to call flash_write_cfibuffer if buffer_size is1. Because for flash memories with buffer_size 1, buffer write is not supported. Regards, Sangmoon Kim

7 28

[U-Boot-Users] PATCH for drivers/cfi_flash.c
by Tolunay Orkun 28 Feb '06

28 Feb '06

This patch reworks the patch submitted on May 6, 2005: Ref: http://sourceforge.net/mailarchive/message.php?msg_id=11674386 This patch (cfi_flash.patch.tolunay) is prepared against current CVS version + Peter Pearse Patch dated March 15, 2005. Ref: http://sourceforge.net/mailarchive/message.php?msg_id=11164812 CHANGELOG: * Patch by Tolunay Orkun, 02 July 2005: Fixes for drivers/cfi_flash.c: - Fix wrong timeout value usage in flash_status_check() - Round write_tout up when converting to msec in flash_get_size() - Remove clearing flash status at the end of flash_write_cfibuffer() which sets Intel 28F640J3 flash back to command mode on CSB472 ~~~ In case Peter's Patch is rejected for some reason please apply the alternate patch (cfi_flash.patch.tolunay2) instead. It makes one additional fix that I had submitted in my original patch. CHANGELOG (Alternate) * Patch by Tolunay Orkun, 02 July 2005: Fixes for drivers/cfi_flash.c: - Fix wrong timeout value usage in flash_status_check() - Round write_tout up when converting to msec in flash_get_size() - Remove clearing flash status at the end of flash_write_cfibuffer() which sets Intel 28F640J3 flash back to command mode on CSB472 - Fix incorrect if condition in flash_full_status_check() Best regards, Tolunay Orkun Index: drivers/cfi_flash.c =================================================================== RCS file: /cvsroot/u-boot/u-boot/drivers/cfi_flash.c,v retrieving revision 1.18 diff -p -u -r1.18 cfi_flash.c --- drivers/cfi_flash.c 16 May 2005 15:23:24 -0000 1.18 +++ drivers/cfi_flash.c 2 Jul 2005 23:45:50 -0000 @@ -694,7 +694,7 @@ static int flash_status_check (flash_inf /* Wait for command completion */ start = get_timer (0); while (flash_is_busy (info, sector)) { - if (get_timer (start) > info->erase_blk_tout * CFG_HZ) { + if (get_timer (start) > tout) { printf ("Flash %s timeout at address %lx data %lx\n", prompt, info->start[sector], flash_read_long (info, sector, 0)); @@ -1113,8 +1113,9 @@ static ulong flash_get_size (ulong base, info->erase_blk_tout = (tmp * (1 << flash_read_uchar (info, FLASH_OFFSET_EMAX_TOUT))); tmp = 1 << flash_read_uchar (info, FLASH_OFFSET_WBTOUT); info->buffer_write_tout = (tmp * (1 << flash_read_uchar (info, FLASH_OFFSET_WBMAX_TOUT))); - tmp = 1 << flash_read_uchar (info, FLASH_OFFSET_WTOUT); - info->write_tout = (tmp * (1 << flash_read_uchar (info, FLASH_OFFSET_WMAX_TOUT))) / 1000; + tmp = (1 << flash_read_uchar (info, FLASH_OFFSET_WTOUT)) * + (1 << flash_read_uchar (info, FLASH_OFFSET_WMAX_TOUT)); + info->write_tout = tmp / 1000 + (tmp % 1000 ? 1 : 0); /* round up when converting to ms */ info->flash_id = FLASH_MAN_CFI; if ((info->interface == FLASH_CFI_X8X16) && (info->chipwidth == FLASH_CFI_BY8)) { info->portwidth >>= 1; /* XXX - Need to test on x8/x16 in parallel. */ @@ -1281,7 +1282,6 @@ static int flash_write_cfibuffer (flash_ info->buffer_write_tout, "buffer write"); } - flash_write_cmd (info, sector, 0, FLASH_CMD_CLEAR_STATUS); return retcode; } #endif /* CFG_FLASH_USE_BUFFER_WRITE */ Index: drivers/cfi_flash.c =================================================================== RCS file: /cvsroot/u-boot/u-boot/drivers/cfi_flash.c,v retrieving revision 1.18 diff -p -u -r1.18 cfi_flash.c --- drivers/cfi_flash.c 16 May 2005 15:23:24 -0000 1.18 +++ drivers/cfi_flash.c 2 Jul 2005 23:52:15 -0000 @@ -694,7 +694,7 @@ static int flash_status_check (flash_inf /* Wait for command completion */ start = get_timer (0); while (flash_is_busy (info, sector)) { - if (get_timer (start) > info->erase_blk_tout * CFG_HZ) { + if (get_timer (start) > tout) { printf ("Flash %s timeout at address %lx data %lx\n", prompt, info->start[sector], flash_read_long (info, sector, 0)); @@ -719,7 +719,7 @@ static int flash_full_status_check (flas case CFI_CMDSET_INTEL_EXTENDED: case CFI_CMDSET_INTEL_STANDARD: if ((retcode != ERR_OK) - && !flash_isequal (info, sector, 0, FLASH_STATUS_DONE)) { + || !flash_isequal (info, sector, 0, FLASH_STATUS_DONE)) { retcode = ERR_INVAL; printf ("Flash %s error at address %lx\n", prompt, info->start[sector]); @@ -1113,8 +1113,9 @@ static ulong flash_get_size (ulong base, info->erase_blk_tout = (tmp * (1 << flash_read_uchar (info, FLASH_OFFSET_EMAX_TOUT))); tmp = 1 << flash_read_uchar (info, FLASH_OFFSET_WBTOUT); info->buffer_write_tout = (tmp * (1 << flash_read_uchar (info, FLASH_OFFSET_WBMAX_TOUT))); - tmp = 1 << flash_read_uchar (info, FLASH_OFFSET_WTOUT); - info->write_tout = (tmp * (1 << flash_read_uchar (info, FLASH_OFFSET_WMAX_TOUT))) / 1000; + tmp = (1 << flash_read_uchar (info, FLASH_OFFSET_WTOUT)) * + (1 << flash_read_uchar (info, FLASH_OFFSET_WMAX_TOUT)); + info->write_tout = tmp / 1000 + (tmp % 1000 ? 1 : 0); /* round up when converting to ms */ info->flash_id = FLASH_MAN_CFI; if ((info->interface == FLASH_CFI_X8X16) && (info->chipwidth == FLASH_CFI_BY8)) { info->portwidth >>= 1; /* XXX - Need to test on x8/x16 in parallel. */ @@ -1281,7 +1282,6 @@ static int flash_write_cfibuffer (flash_ info->buffer_write_tout, "buffer write"); } - flash_write_cmd (info, sector, 0, FLASH_CMD_CLEAR_STATUS); return retcode; } #endif /* CFG_FLASH_USE_BUFFER_WRITE */

2 1

[U-Boot-Users] cfi_flash.c patch
by richard＠uclinux.net 28 Feb '06

28 Feb '06

Afternoon... Adding 2 spaces in cfi_flash.c makes the output of "flinfo" much more readable if there are protected and unprotected sectors: rick ---------------------------------------------------------------------------- CHANGELOG: * Patch by Richard Klingler, 09 Apr 2005: correct column layout of flinfo output. --- u-boot.orig/drivers/cfi_flash.c 2005-01-09 19:12:33.000000000 +0200 +++ u-boot/drivers/cfi_flash.c 2005-04-09 16:35:04.000000000 +0300 @@ -483,7 +483,7 @@ void flash_print_info (flash_info_t * in if ((i % 5) == 0) printf ("\n "); printf (" %08lX%s", - info->start[i], info->protect[i] ? " (RO) " : " "); + info->start[i], info->protect[i] ? " (RO) " : " "); #endif } putc ('\n');

3 3

[U-Boot-Users] PATCH: minor bugs fix in cfi_driver.c
by Pierre AUBERT 28 Feb '06

28 Feb '06

Please find attached a little patch against current CVS. CHANGELOG: * Patch by Pierre Aubert 09 Dec 2004 - Fix flash information strings alignement in cfi_driver.c - Fix monitor size for flash protection in cfi_driver.c Best regards diff -pur u-boot/drivers/cfi_flash.c u-boot-paub/drivers/cfi_flash.c --- u-boot/drivers/cfi_flash.c 2004-10-11 01:27:35.000000000 +0200 +++ u-boot-paub/drivers/cfi_flash.c 2004-12-09 10:11:31.000000000 +0100 @@ -334,7 +334,7 @@ unsigned long flash_init (void) #if (CFG_MONITOR_BASE >= CFG_FLASH_BASE) flash_protect (FLAG_PROTECT_SET, CFG_MONITOR_BASE, - CFG_MONITOR_BASE + CFG_MONITOR_LEN - 1, + CFG_MONITOR_BASE + monitor_flash_len - 1, &flash_info[0]); #endif @@ -477,7 +477,7 @@ void flash_print_info (flash_info_t * in if ((i % 5) == 0) printf ("\n "); printf (" %08lX%s", - info->start[i], info->protect[i] ? " (RO) " : " "); + info->start[i], info->protect[i] ? " (RO)" : " "); #endif } putc ('\n');

2 1

[U-Boot-Users] Subject: [PATCH] Git ignore generated files in sub-directory tools.
by Jon Loeliger 20 Feb '06

20 Feb '06

Signed-off-by: Jon Loeliger <jdl(a)freescale.com> --- tools/.gitignore | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) create mode 100644 tools/.gitignore ee1b1a7ee4f4b7540c8a016e41cb653ea20b8427 diff --git a/tools/.gitignore b/tools/.gitignore new file mode 100644 index 0000000..76ab6ae --- /dev/null +++ b/tools/.gitignore @@ -0,0 +1,10 @@ +# +# Generated files +# +bmp_logo +crc32.c +envcrc +environment.c +gen_eth_addr +img2srec +mkimage -- 0.99.9n

3 8

[U-Boot-Users] unexpected behaviour !!!
by Maximus 01 Feb '06

01 Feb '06

Hi, I have designed my own protocol and am trying to include it in u-boot. In the Recieve Handler i am trying to dereference the pkt. struct mypacket { unsigned int seqnum ; char data [24] ; } ; struct mypacket mypacket ; void myrecieve_handler (uchar * pkt, unsigned dest, unsigned src, unsigned len) { int seqnum ; uchar * q ; if (dest != 8888) return ; printf ("pkt => %u\n", pkt) ; printf ("len => %d\n", len) ; q = pkt ; for (i = 0 ; i < len ; i++) { printf ("%d:%u=> %c\n", i, q, *q) ; q++ ; } memcpy (&mypacket, pkt, sizeof (mypacket)) ; printf ("seq:%u\n", mypacket.seq) ; printf ("data:%s\n", mypacket.data) ; printf ("before\n") ; seqnum = * (int *) pkt ; /* <------------------------ Abort - Board Hangs */ printf ("after\n") ; } I am able to access the contents of pkt 1) byte by byte 2) memcpy But, as soon as i try to deference it either by doing 1) * (int *) pkt /***************** board hangs ********/ or 2) mp = (struct mypacket *) pkt printf ("%u\n", mp -> seqnum) ; /************* board Hangs *************/ I get the Error shown below. As, soon as seqnum = * (int *) pkt - is executed my board prints the message: Why is that ?. I dont understand this. i am using an OMAP 5912 OSK BOARD. Is it not possible to dereference pkt ?. data abort pc : [<2000d6b0>] lr : [<2000d6a8>] sp : 1ffbfc38 ip : ffffffff fp : 20015fc8 r10: 2001dc24 r9 : 20015fb8 r8 : 1ffbffdc r7 : 2001dc08 r6 : 0000001c r5 : 2001c40a r4 : 000022b8 r3 : 00000020 r2 : 00000001 r1 : 0000000a r0 : 00000000 Flags: nZCv IRQs off FIQs off Mode SVC_32 Resetting CPU ... Regards, Jo

3 3

[U-Boot-Users] SDRAM
by Måns Brorsson 01 Feb '06

01 Feb '06

Ok i hope i turned the irritating HTML of, now! The thing i missed is which cfg_ the U-Boot has to have. Should i put my sram address as CFG_SDRAM_BASE and underline CFG_SRAM_BASE? -- Måns Brorsson

3 2

[U-Boot-Users] eldk4.0 and LTT
by Frank 31 Jan '06

31 Jan '06

Is there a patch available for the 2.6.15 kernel and LTT. I tried to apply the 0.9.6 patch for the 2.6.10 kernel and it failed miserably... __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 3