Skip to contentSkip to conversation

splash: Use splashfile instead of location->name

PATCH Superseded Jonas Mark (BT-FIR/ENG1) 2 participants 2 messages Feb 6, 2019

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/2 splash: Use splashfile instead of location->name

Hi Tomas,

Quoted 12 lines
Quoted 1 line

I conclude that no upstream board is actually affected. They would only

be affecte if they would use SPLASH_STORAGE_FIT.

You are right, indeed upstream boards should be ok. However, some other non upstream boards I know of would be affected.

So my initial question still stands, could existing implementation be kept as fallback?

(quoted diff) +2 -0
+	if (!splash_file)
+		splash_file = location->name;
Quoted 2 lines

location->name has worked fine for those cases, so it would be nice to keep that option intact.

I think that the .name field of struct splash_location is actually intended to hold a human readable name for the storage itself. That the code so far uses the storage location name synonymously with the splash screen name is unfortunate. That's why we decided to fix this. I am convinced that the patch actually makes it easier for others to start reading splash screens from a FIT image.

Why do I think so? Because the behavior is now consistent to what splash_load_fs() does. That is, if there is an environment variable "splashfile" the splash screen name is taken from there. Else, the fixed default name "splash.bmp" is used.

Regards,
Mark

Building Technologies, Panel Software Fire (BT-FIR/ENG1)
Bosch Sicherheitssysteme GmbH | Postfach 11 11 | 85626 Grasbrunn | GERMANY | www.boschsecurity.com

Sitz: Stuttgart, Registergericht: Amtsgericht Stuttgart HRB 23118
Aufsichtsratsvorsitzender: Christian Fischer; Geschäftsführung: Tanja Rückert, Andreas Bartz, Thomas Quante, Bernhard Schuster

#1 lore raw

PATCH v2 1/2 splash: Use splashfile instead of location->name

Hi Mark,

Quoted 7 lines · On 2/6/19 4:30 PM, Jonas Mark (BT-FIR/ENG1) wrote:
Quoted 7 lines

So my initial question still stands, could existing implementation be kept as fallback?

(quoted diff) +2 -0
+	if (!splash_file)
+		splash_file = location->name;
Quoted 8 lines
Quoted 2 lines

location->name has worked fine for those cases, so it would be nice to keep that option intact.

I think that the .name field of struct splash_location is actually intended to hold a human readable name for the storage itself. That the code so far uses the storage location name synonymously with the splash screen name is unfortunate. That's why we decided to fix this. I am convinced that the patch actually makes it easier for others to start reading splash screens from a FIT image.

If I understand the original issue you had correctly, could it also have been fixed by renaming splash node in FIT image?

Quoted 5 lines

Why do I think so? Because the behavior is now consistent to what splash_load_fs() does. That is, if there is an environment variable "splashfile" the splash screen name is taken from there. Else, the fixed default name "splash.bmp" is used.

Symmetry is nice, so ok. But still, it now rules out the possiblity to use the splash name in the FIT without defining additional "splashfile" to the env.

BR,

Tomas

Quoted 10 lines

Regards,
Mark

Building Technologies, Panel Software Fire (BT-FIR/ENG1)
Bosch Sicherheitssysteme GmbH | Postfach 11 11 | 85626 Grasbrunn | GERMANY | www.boschsecurity.com

Sitz: Stuttgart, Registergericht: Amtsgericht Stuttgart HRB 23118
Aufsichtsratsvorsitzender: Christian Fischer; Geschäftsführung: Tanja Rückert, Andreas Bartz, Thomas Quante, Bernhard Schuster