Skip to contentSkip to conversation

CFI driver AMD Command Set Top boot geometry reversal, etc.

PATCH New Tolunay Orkun 2 participants 5 messages Nov 9, 2006 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 CFI driver AMD Command Set Top boot geometry reversal, etc.

This patch:

* Adds support for AMD command set Top Boot flash geometry reversal
* Adds support for displaying command set, manufacturer id and device ids (flinfo)
* Makes flinfo output to be consistent when CFG_FLASH_EMPTY_INFO defined
* Removes outdated change history (refer to git log instead)

---

drivers/cfi_flash.c | 153 +++++++++++++++++++++++++++++++++------------------
include/flash.h | 7 ++

2 files changed, 105 insertions(+), 55 deletions(-)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: cfi_driver.patch
Type: text/x-patch
Size: 9794 bytes
Desc: not available
Url : http://lists.denx.de/pipermail/u-boot/attachments/20061108/b740e9c2/attachment.bin

#1 lore raw

PATCH 1/1 CFI driver AMD Command Set Top boot geometry reversal, etc.

Quoted 7 lines · Tolunay Orkun

This patch:

* Adds support for AMD command set Top Boot flash geometry reversal
* Adds support for displaying command set, manufacturer id and device ids (flinfo)
* Makes flinfo output to be consistent when CFG_FLASH_EMPTY_INFO defined
* Removes outdated change history (refer to git log instead)

This patch looks good, but I get this from flinfo:

Bank # 1: CFI conformant FLASH (16 x 16) Size: 8 MB in 135 Sectors
AMD Standard command set, Manufacturer ID: 0xFF, Device ID: 0xFF Erase timeout: 16384 ms, write timeout: 1 ms

Are these the correct values for manufacturer ID and device ID?

Signature · 2 lines
Timur Tabi
Linux Kernel Developer @ Freescale
#2 lore raw

PATCH 1/1 CFI driver AMD Command Set Top boot geometry reversal, etc.

Quoted 10 lines · Timur Tabi
Quoted 8 lines · Tolunay Orkun

This patch:

* Adds support for AMD command set Top Boot flash geometry reversal
* Adds support for displaying command set, manufacturer id and device ids (flinfo)
* Makes flinfo output to be consistent when CFG_FLASH_EMPTY_INFO
defined
* Removes outdated change history (refer to git log instead)

Quoted 8 lines

This patch looks good, but I get this from flinfo:

Bank # 1: CFI conformant FLASH (16 x 16) Size: 8 MB in 135 Sectors
AMD Standard command set, Manufacturer ID: 0xFF, Device ID: 0xFF Erase timeout: 16384 ms, write timeout: 1 ms

Are these the correct values for manufacturer ID and device ID?

Thanks for testing the patch. Not correct. It is cosmetic in your case since your flash CFI is version 1.1 or later.

It looks like I cannot access JEDEC IDs while still in CFI Query mode in AMD style flash (unlike Intel). Bummer :( I have to explicitly send the autoselect command to read out and switch back to CFI Query mode again.

I will improve the patch. Also, while browsing Spansion web site, I came across with 3 byte Device IDs. So, we have to improve on that as well.

Best regards,
Tolunay

#3 lore raw

PATCH 1/1 CFI driver AMD Command Set Top boot geometry reversal, etc.

Quoted 2 lines · Tolunay Orkun

I will improve the patch. Also, while browsing Spansion web site, I came across with 3 byte Device IDs. So, we have to improve on that as well.

Well, I wouldn't object to leaving that work for another time. I'm sure there is tons of flash info that could be collected and displayed. I'd rather have the current patch added to U-Boot this week, then to wait until next week for an expanded one.

Of course, that decision is really Wolfgang's.

FYI, your patch also works on my 8360 board, which has uniform sectors.

Signature · 2 lines
Timur Tabi
Linux Kernel Developer @ Freescale
#5 lore raw

PATCH 1/1 CFI driver AMD Command Set Top boot geometry reversal, etc.

In reply to a message not in the archive
Quoted 2 lines · Stefan Roese

Yes, I second Timurs opinion. So Tolunay, please send your updated patch to also correctly display the ID's and we'll try to get this in pretty soon.

Actually, what I meant to say is the Tolunay's patch, as it is *now*, is good enough for now. But again, that's just my opinion.

Signature · 2 lines
Timur Tabi
Linux Kernel Developer @ Freescale