diff mbox

AT25DF161 erase failed on ft2232_spi

Message ID 575E9C93.70002@hatimak.me
State New
Headers show

Commit Message

Hatim Kanchwala June 13, 2016, 11:44 a.m. UTC
On Sunday 12 June 2016 09:04 PM, Алексей Елсаков wrote:
> Hello, Flashrom developers!
>
> I'am trying to flash Atmel AT25DF161 EEPROM with ft2232h_spi programmer.
> My hardware is "FT2232H Breakout board" by DangerousPrototypes -
> http://dangerousprototypes.com/docs/FT2232_breakout_board
> IC is soldered to board with wires ~3cm length.
>
> using latest flashrom v0.9.9-r1954 form SVN, IC reads well, but erasing
> failed - see logs.
> after erasing fail, IC content looks like this:
>
>
> with SST 25VF512, Winbond 25X80VSIC, Macronix MX25L8005 this flashrom
> and programmer works well.
> what should I do?
>
Hi Alexey,

Based on the logs that you've sent, your chip had some write protection 
in place and flashrom successfully disabled it. The erase functions 
defined in flashrom for your chip is correct as per datasheet 
(https://www.adestotech.com/wp-content/uploads/doc3687.pdf).

My theory as to why erase failed for your chip is that some 64kB 
sector(s) must be locked down, meaning that they have become permanently 
read-only. Attempting to program or erase such a sector is a cause of 
command failure.

I have attached a patch that can help verify this theory. Please apply 
the patch locally, build, test and resend the verbose log to the mailing 
list.

Thanks.

Kind regards,
Hatim

>
> /--
> best regards,
> Alexey Elsakov /mailto:alexey@yelsakov.ru
>
>
> _______________________________________________
> flashrom mailing list
> flashrom@flashrom.org
> https://www.flashrom.org/mailman/listinfo/flashrom
>

Comments

Алексей Елсаков June 13, 2016, 2:03 p.m. UTC | #1
Hello, Hatim.

You suggestion was right. 11 sectors of my IC are locked down. :(
I tried to use layout excluding those sectors and succeeded :)
But to flash whole IC I have to buy a new one...

I  thought about locked sectors too, but was confused with changed data in
sector 0 yesterday. it is strange, but today data stay untouched.

In any case, we just checked, that flashrom reads and programs AT25DF161 well.

It  will be good, if flashrom had ability to check locked down sectors
at erase fails or as a command option for example...

You wrote 13 June 2016 14:44:19:

> On Sunday 12 June 2016 09:04 PM, Алексей Елсаков wrote:
>> Hello, Flashrom developers!
>>
>> I'am trying to flash Atmel AT25DF161 EEPROM with ft2232h_spi programmer.
>> My hardware is "FT2232H Breakout board" by DangerousPrototypes -
>> http://dangerousprototypes.com/docs/FT2232_breakout_board
>> IC is soldered to board with wires ~3cm length.
>>
>> using latest flashrom v0.9.9-r1954 form SVN, IC reads well, but erasing
>> failed - see logs.
>> after erasing fail, IC content looks like this:
>>
>>
>> with SST 25VF512, Winbond 25X80VSIC, Macronix MX25L8005 this flashrom
>> and programmer works well.
>> what should I do?
>>
> Hi Alexey,

> Based on the logs that you've sent, your chip had some write protection
> in place and flashrom successfully disabled it. The erase functions 
> defined in flashrom for your chip is correct as per datasheet 
> (https://www.adestotech.com/wp-content/uploads/doc3687.pdf).

> My theory as to why erase failed for your chip is that some 64kB 
> sector(s) must be locked down, meaning that they have become permanently
> read-only. Attempting to program or erase such a sector is a cause of 
> command failure.

> I have attached a patch that can help verify this theory. Please apply
> the patch locally, build, test and resend the verbose log to the mailing
> list.

> Thanks.

> Kind regards,
> Hatim

>>
>> /--
>> best regards,
>> Alexey Elsakov /mailto:alexey@yelsakov.ru
>>
>>
>> _______________________________________________
>> flashrom mailing list
>> flashrom@flashrom.org
>> https://www.flashrom.org/mailman/listinfo/flashrom
>>

best regards,
Alexey Elsakov mailto:alexey@yelsakov.ru
diff mbox

Patch

From 871811ebe82fb9f9be16b4102626552b9f845f42 Mon Sep 17 00:00:00 2001
In-Reply-To: <156892285.20160612183407@yelsakov.ru>
References: <156892285.20160612183407@yelsakov.ru>
From: Hatim Kanchwala <hatim@hatimak.me>
Date: Mon, 13 Jun 2016 16:43:34 +0530
Subject: [PATCH] Print sector lockdown registers for AT25DF161
To: alexey@yelsakov.ru
Cc: flashrom@flashrom.org

When pretty-printing status register byte 1 of AT25DF chips, if chip
is AT25DF161, also print the values of sector lockdown registers for
each of the 32 64kB sectors.

Signed-off-by: Hatim Kanchwala <hatim@hatimak.me>
---
 spi25_statusreg.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/spi25_statusreg.c b/spi25_statusreg.c
index 01a6862..fb0e8a4 100644
--- a/spi25_statusreg.c
+++ b/spi25_statusreg.c
@@ -14,26 +14,27 @@ 
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
  *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
  */
 
 #include "flash.h"
 #include "chipdrivers.h"
 #include "spi.h"
+#include "flashchips.h"
 
 /* === Generic functions === */
 int spi_write_status_enable(struct flashctx *flash)
 {
 	static const unsigned char cmd[JEDEC_EWSR_OUTSIZE] = { JEDEC_EWSR };
 	int result;
 
 	/* Send EWSR (Enable Write Status Register). */
 	result = spi_send_command(flash, sizeof(cmd), JEDEC_EWSR_INSIZE, cmd, NULL);
 
 	if (result)
 		msg_cerr("%s failed\n", __func__);
 
@@ -452,26 +453,61 @@  static void spi_prettyprint_status_register_atmel_at25_swp(uint8_t status)
 	}
 }
 
 int spi_prettyprint_status_register_at25df(struct flashctx *flash)
 {
 	uint8_t status = spi_read_status_register(flash);
 	spi_prettyprint_status_register_hex(status);
 
 	spi_prettyprint_status_register_atmel_at25_srpl(status);
 	spi_prettyprint_status_register_bit(status, 6);
 	spi_prettyprint_status_register_atmel_at25_epewpp(status);
 	spi_prettyprint_status_register_atmel_at25_swp(status);
 	spi_prettyprint_status_register_welwip(status);
+
+	if (flash->chip->model_id == ATMEL_AT25DF161)
+	{
+		int address, i, count, result;
+		unsigned char read_result, lockdown_status_sector[64], cmd[5];
+		cmd[0] = (unsigned char)0x35;
+		cmd[4] = (unsigned char)0x00;
+
+		for (address = 0x000000, i = 0, count = 0; address < 0x200000; address += 0x010000, i++)
+		{
+			cmd[1] = (unsigned char)(address >> 16) & 0xff;
+			cmd[2] = (unsigned char)(address >> 8) & 0xff;
+			cmd[3] = (unsigned char)address & 0xff;
+			result = spi_send_command(flash, sizeof(cmd), sizeof(unsigned char), cmd, &read_result);
+			if (result)
+			{
+				msg_cerr("%s failed during command execution (ATMEL_AT25DF161)\n", __func__);
+				return result;
+			}
+			if (i % 8 == 0)
+				msg_cdbg("0x%02x:", i);
+			msg_cdbg(" %02x%s", read_result, (i + 1) % 8 == 0 ? "\n": "");
+			lockdown_status_sector[address / 0x010000] = read_result;
+			if (read_result)
+				count++;
+		}
+
+		msg_cdbg("%d sector%s locked down permanently%s", count, (count == 1) ? "" : "s", (count == 0) ? "." : " :");
+		if (count)
+			for (i = 0; i < 64; i++)
+				if (lockdown_status_sector[i])
+					msg_cdbg(" %2d", i);
+		msg_cdbg("\n");
+	}
+
 	return 0;
 }
 
 int spi_prettyprint_status_register_at25df_sec(struct flashctx *flash)
 {
 	/* FIXME: We should check the security lockdown. */
 	msg_cdbg("Ignoring security lockdown (if present)\n");
 	msg_cdbg("Ignoring status register byte 2\n");
 	return spi_prettyprint_status_register_at25df(flash);
 }
 
 /* used for AT25F512, AT25F1024(A), AT25F2048 */
 int spi_prettyprint_status_register_at25f(struct flashctx *flash)
-- 
http://hatimak.me