Suggestion on flash init
Iark,
Just for reference, some time ago I posted two patches which may be useful for you in this case. Those weren't accepted yet, but this worked for me.
First, x8 bit support for AMD.
http://sourceforge.net/mailarchive/message.php?msg_id=12799051
http://article.gmane.org/gmane.comp.boot-loaders.u-boot/18621
I also posted another one which allows buffer writing on Spansion flashes
http://sourceforge.net/mailarchive/message.php?msg_id=12797909
http://article.gmane.org/gmane.comp.boot-loaders.u-boot/18620
Patches are included on those messages.
Best regards,
Alex
Quoted 227 lines
5. Re: Suggestion on flash init (Wang Jian)
7. Re: Suggestion on flash init (Wolfgang Denk)
8. Re: Suggestion on flash init (Wang Jian)
9. Re: Suggestion on flash init (Wolfgang Denk)Message: 5
Date: Tue, 03 Jan 2006 21:16:03 +0800
From: Wang Jian <[email protected]>
To: Wang Jian <[email protected]>Subject: Re: [U-Boot-Users] Suggestion on flash init
Cc: u-boot-users at lists.sourceforge.netHi Denk,
Me again.
In my situation, S29AL016M works in x8 mode, which has different command sequence from in x16 mode.
The following are from datasheet.
Providing
vu_char*addr = (vu_char*) info->start[sect];x16 mode sector erase
addr[0x555] =0xAA;
addr[0x2AA] =0x55;
addr[0x555] =0x80;
addr[0x555] =0xAA;
addr[0x2AA] =0x55;
addr[0x000] =0x30;
udelay(50); // at least 50 us before polling for statusx16 mode program
addr[0x555] =0xAA;
addr[0x2AA] =0x55;
addr[0x555] =0xA0;
addr[offset] = data;x8 mode sector erase
addr[0xAAA] =0xAA;
addr[0x555] =0x55;
addr[0xAAA] =0x80;
addr[0xAAA] =0xAA;
addr[0x555] =0x55;
addr[0x000] =0x30;
udelay(50); // at least 50 us before polling for statusx8 mode program
addr[0xAAA] =0xAA;
addr[0x555] =0x55;
addr[0xAAA] =0xA0;
addr[offset] = data;I have little knowledge about other flash chips, so it is not easy for me to write clean patch for
cfi_flash.c, although I want to.Regards
Quoted 9 lines · Wang Jian
Hi Wolfgang Denk,
Quoted 4 lines · Wolfgang Denk
Quoted 3 lines · In message <[email protected]> you wrote:
I have tested
cfi_flash.cfor a while. Unfortuanately, it fails for Spansion S29AL016M. The chip can be recognized, chip information isall
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
correct, but the erase and write routines fail.
Then please (lets) fix the problem instead of inventing the wheel
again.
Quoted 2 lines
After a bad sleep, I read the manual again. This time, I find the
cfi_flash.cQuoted 3 lines
DOES do as what manual says. I will set some debug information and find what happend. Stay tune.
Signature · 12 lines
lark --__--__-- Message: 7 To: Wang Jian <[email protected]> cc: u-boot-users at lists.sourceforge.net From: Wolfgang Denk <[email protected]> Subject: Re: [U-Boot-Users] Suggestion on flash init Date: Tue, 03 Jan 2006 15:08:42 +0100Quoted 2 lines · In message <[email protected]> you wrote:
=20
In my situation, S29AL016M works in x8 mode, which has different comman=d
Quoted 1 line
sequence from in x16 mode.
If you look closer it is not so different at all.
Quoted 7 lines
x16 mode sector erase
addr[0x555] =3D0xAA;
addr[0x2AA] =3D0x55;
addr[0x555] =3D0x80;
addr[0x555] =3D0xAA;
addr[0x2AA] =3D0x55;
addr[0x000] =3D0x30;...
Quoted 7 lines
x8 mode sector erase
addr[0xAAA] =3D0xAA;
addr[0x555] =3D0x55;
addr[0xAAA] =3D0x80;
addr[0xAAA] =3D0xAA;
addr[0x555] =3D0x55;
addr[0x000] =3D0x30;
0x555=3D0xAAA>> 10x2AA=3D0x555>> 1Quoted 2 lines
I have little knowledge about other flash chips, so it is not easy for me to write clean patch for
cfi_flash.c, although I want to.The code is actually the very same, just taking into account the different addressing mode.
Best regards,
Wolfgang Denk
--=20
Software Engineering: Embedded and Realtime Systems, Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de Another Armenia, Belgium ... the weak innocents who always seem to be located on a natural invasion route.-- Kirk, "Errand of Mercy", stardate 3198.4--__--__--
Message: 8
Date: Tue, 03 Jan 2006 23:34:51 +0800
From: Wang Jian <[email protected]>
To: Wolfgang Denk <[email protected]>
Subject: Re: [U-Boot-Users] Suggestion on flash init
Cc: u-boot-users at lists.sourceforge.netHi Wolfgang Denk,
Quoted 3 lines · Wolfgang Denk
Quoted 2 lines · In message <[email protected]> you wrote:
In my situation, S29AL016M works in x8 mode, which has different
command
Quoted 20 lines
Quoted 1 line
sequence from in x16 mode.
If you look closer it is not so different at all.
Quoted 7 lines
x16 mode sector erase
addr[0x555] =0xAA;
addr[0x2AA] =0x55;
addr[0x555] =0x80;
addr[0x555] =0xAA;
addr[0x2AA] =0x55;
addr[0x000] =0x30;...
Quoted 7 lines
x8 mode sector erase
addr[0xAAA] =0xAA;
addr[0x555] =0x55;
addr[0xAAA] =0x80;
addr[0xAAA] =0xAA;
addr[0x555] =0x55;
addr[0x000] =0x30;Quoted 9 lines
0x555=0xAAA>> 10x2AA=0x555>> 1Quoted 2 lines
I have little knowledge about other flash chips, so it is not easy for me to write clean patch for
cfi_flash.c, although I want to.The code is actually the very same, just taking into account the different addressing mode.
I am not sure if other kind of AMD flash chips use such x8/x16 command sequence. I am afraid that I will fix one thing and break other things.
Signature · 12 lines
lark --__--__-- Message: 9 To: Wang Jian <[email protected]> cc: u-boot-users at lists.sourceforge.net From: Wolfgang Denk <[email protected]> Subject: Re: [U-Boot-Users] Suggestion on flash init Date: Tue, 03 Jan 2006 16:44:04 +0100Quoted 3 lines · In message <[email protected]> you wrote:
=20
I am not sure if other kind of AMD flash chips use such x8/x16 command sequence. I am afraid that I will fix one thing and break other things.They all do. At least all I can remember.
Best regards,
Wolfgang Denk
--=20
Software Engineering: Embedded and Realtime Systems, Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de Peace was the way.
-- Kirk, "The City on the Edge of Forever", stardate unknown