[PATCH] Do not crash with forced chips

[Carl-Daniel Hailfinger]

Avoid printing the chip locks if chip detection was forced because lock
access may involve flash chip registers which will not be mapped.

Signed-off-by: Carl-Daniel Hailfinger

Index: flashrom-no_force_register_access/flashrom.c
===================================================================
--- flashrom-no_force_register_access/flashrom.c	(Revision 1238)
+++ flashrom-no_force_register_access/flashrom.c	(Arbeitskopie)
@@ -1163,8 +1163,12 @@
           flash->vendor, flash->name, flash->total_size,
           flashbuses_to_text(flash->bustype), base);

-	if (flash->printlock)
-		flash->printlock(flash);
+	/* Flash registers will not be mapped if the chip was forced. Lock info
+	 * may be stored in registers, so avoid lock info printing.
+	 */
+	if (!force)
+		if (flash->printlock)
+			flash->printlock(flash);

return flash;
 }

[Uwe Hermann]

On Thu, Dec 02, 2010 at 01:55:53PM +0100, Carl-Daniel Hailfinger wrote:
> Avoid printing the chip locks if chip detection was forced because lock
> access may involve flash chip registers which will not be mapped.
>
> Signed-off-by: Carl-Daniel Hailfinger

Acked-by: Uwe Hermann

Looks good. Maybe add a small message to inform the user about this?

Uwe.

[Carl-Daniel Hailfinger]

On 02.12.2010 21:56, Uwe Hermann wrote:
> On Thu, Dec 02, 2010 at 01:55:53PM +0100, Carl-Daniel Hailfinger wrote:
> > Avoid printing the chip locks if chip detection was forced because lock
> > access may involve flash chip registers which will not be mapped.
> >
> > Signed-off-by: Carl-Daniel Hailfinger
>
> Acked-by: Uwe Hermann

Thanks, committed in r1240.
> Looks good. Maybe add a small message to inform the user about this?

Not sure. This is for forced detection only, and there you can't read
the locking status for most chips anyway because the chip didn't respond
in the first place. And for pure read of a chip it is not clear whether
printing locks would make sense.

Regards,
Carl-Daniel

[Uwe Hermann]

On Thu, Dec 02, 2010 at 11:02:02PM +0100, Carl-Daniel Hailfinger wrote:
> On 02.12.2010 21:56, Uwe Hermann wrote:
> > On Thu, Dec 02, 2010 at 01:55:53PM +0100, Carl-Daniel Hailfinger wrote:
> > > Avoid printing the chip locks if chip detection was forced because lock
> > > access may involve flash chip registers which will not be mapped.
> > >
> > > Signed-off-by: Carl-Daniel Hailfinger
> >
> > Acked-by: Uwe Hermann
>
> Thanks, committed in r1240.
> > Looks good. Maybe add a small message to inform the user about this?
>
> Not sure. This is for forced detection only, and there you can't read
> the locking status for most chips anyway because the chip didn't respond
> in the first place. And for pure read of a chip it is not clear whether
> printing locks would make sense.

No, I meant to print something like this:

This is a forced read, not trying to print chip lock status.

Trying to print any lock info would indeed make no sense.

Uwe.