[flashrom] [PATCH] Return code for spi_read_status_register

[Stefan Tauner]

On Sun, 03 Jun 2012 16:51:18 +0200
Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006 at gmx.net> wrote:

> spi_read_status_register now returns success/failure and stores the
> status register value via call-by-reference. That way, status register
> reading failure (only possible if the programmer couldn't run RDSR) can
> be detected and acted upon in the future.

with this patch this failure is largely ignored and i am not sure that
this does not change behavior due to the 0xff status value that is
returned on errors.

> Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006 at gmx.net>
> 
> Index: flashrom-spi_rdsr_errorcheck/it87spi.c
> ===================================================================
> --- flashrom-spi_rdsr_errorcheck/it87spi.c	(Revision 1539)
> +++ flashrom-spi_rdsr_errorcheck/it87spi.c	(Arbeitskopie)
> @@ -321,6 +321,7 @@
>  {
>  	unsigned int i;
>  	int result;
> +	uint8_t status;
>  	chipaddr bios = flash->virtual_memory;
>  
>  	result = spi_write_enable(flash);
> @@ -335,7 +336,7 @@
>  	/* Wait until the Write-In-Progress bit is cleared.
>  	 * This usually takes 1-10 ms, so wait in 1 ms steps.
>  	 */
> -	while (spi_read_status_register(flash) & SPI_SR_WIP)
> +	while (spi_read_status_register(flash, &status), status & SPI_SR_WIP)
>  		programmer_delay(1000);
>  	return 0;

for example here it would be easy to store the return value of
spi_read_status_register and return it instead of the fixed 0

>  }
> Index: flashrom-spi_rdsr_errorcheck/a25.c
> ===================================================================
> --- flashrom-spi_rdsr_errorcheck/a25.c	(Revision 1539)
> +++ flashrom-spi_rdsr_errorcheck/a25.c	(Arbeitskopie)
> @@ -33,7 +33,7 @@
>  {
>  	uint8_t status;
>  
> -	status = spi_read_status_register(flash);
> +	spi_read_status_register(flash, &status);
>  	msg_cdbg("Chip status register is %02x\n", status);
here the output might change depending on what the programmers'
send_spi_command routine "return" in case of errors etc.

in general i like the idea very much, but i would rather see a
"complete" patch, that does acts appropriately on errors... try to
convince me if you disagree :)
-- 
Kind regards/Mit freundlichen Grüßen, Stefan Tauner