[PATCH] ichspi reprogram_opcode_on_the_fly return code fix

[Carl-Daniel Hailfinger]

Fix ICH SPI reprogram_opcode_on_the_fly return code.
Helpful for SFDP debugging.

Signed-off-by: Carl-Daniel Hailfinger

Index: flashrom-sfdp_read_dummy_reduce_param_chunksize/ichspi.c
===================================================================
--- flashrom-ichspi_invalid_length_consistent/ichspi.c	(Revision 1528)
+++ flashrom-ichspi_invalid_length_consistent/ichspi.c	(Arbeitskopie)
@@ -416,7 +416,10 @@
    	msg_pdbg ("on-the-fly OPCODE (0x%02X) re-programmed, op-pos=%d\n", opcode, oppos);
    	return oppos;
    }
-	return -1;
+	/* This case only happens if the writecnt/readcnt combination is
+	 * impossible to achieve on ICH SPI.
+	 */
+	return SPI_INVALID_LENGTH;
 }

static int find_opcode(OPCODES *op, uint8_t opcode)
@@ -1001,7 +1004,11 @@
    if (opcode_index == -1) {
    	if (!ichspi_lock)
    		opcode_index = reprogram_opcode_on_the_fly(cmd, writecnt, readcnt);
-		if (opcode_index == -1) {
+		if (opcode_index == SPI_INVALID_LENGTH) {
+			msg_pdbg("OPCODE 0x%02x has unsupported length, will "
+				 "not execute.\n", cmd);
+			return SPI_INVALID_LENGTH;
+		} else if (opcode_index == -1) {
    		msg_pdbg("Invalid OPCODE 0x%02x, will not execute.\n",
    			 cmd);
    		return SPI_INVALID_OPCODE;

[PATCH] ichspi reprogram_opcode_on_the_fly return code fix

[Stefan Tauner]

On Sat, 05 May 2012 01:22:12 +0200
Carl-Daniel Hailfinger  wrote:
> Fix ICH SPI reprogram_opcode_on_the_fly return code.

i wouldnt call this a fix.
> Helpful for SFDP debugging.
>
> Signed-off-by: Carl-Daniel Hailfinger
> Index: flashrom-sfdp_read_dummy_reduce_param_chunksize/ichspi.c
>
> --- flashrom-ichspi_invalid_length_consistent/ichspi.c	(Revision 1528)
> +++ flashrom-ichspi_invalid_length_consistent/ichspi.c	(Arbeitskopie)
> @@ -416,7 +416,10 @@
>    	msg_pdbg ("on-the-fly OPCODE (0x%02X) re-programmed, op-pos=%d\n", opcode, oppos);
>    	return oppos;
>    }
> return -1;
> /* This case only happens if the writecnt/readcnt combination is
> * impossible to achieve on ICH SPI.
> */
> return SPI_INVALID_LENGTH;
>
> }

imho it would make more sense to have less comments + a cleaner code
flow instead ;) see attached patch.
> static int find_opcode(OPCODES *op, uint8_t opcode)
> @@ -1001,7 +1004,11 @@
>    if (opcode_index == -1) {
>    	if (!ichspi_lock)
>    		opcode_index = reprogram_opcode_on_the_fly(cmd, writecnt, readcnt);
> if (opcode_index == -1) {
> if (opcode_index == SPI_INVALID_LENGTH) {
> 	msg_pdbg("OPCODE 0x%02x has unsupported length, will "
> 		 "not execute.\n", cmd);
> 	return SPI_INVALID_LENGTH;
> } else if (opcode_index == -1) {
> msg_pdbg("Invalid OPCODE 0x%02x, will not execute.\n",
> 	 cmd);
> return SPI_INVALID_OPCODE;

[PATCH] ichspi reprogram_opcode_on_the_fly return code fix

[Carl-Daniel Hailfinger]

Am 06.05.2012 14:18 schrieb Stefan Tauner:
> imho it would make more sense to have less comments + a cleaner code
> flow instead ;) see attached patch.
>
> Signed-off-by: Stefan Tauner

Indeed.
Acked-by: Carl-Daniel Hailfinger

Regards,
Carl-Daniel

[PATCH] ichspi reprogram_opcode_on_the_fly return code fix

[Stefan Tauner]

On Sun, 06 May 2012 15:33:30 +0200
Carl-Daniel Hailfinger  wrote:
> Am 06.05.2012 14:18 schrieb Stefan Tauner:
> > imho it would make more sense to have less comments + a cleaner code
> > flow instead ;) see attached patch.
> >
> > Signed-off-by: Stefan Tauner
>
> Indeed.
> Acked-by: Carl-Daniel Hailfinger

:)
thank you. committed in r1531