Skip to contentSkip to conversation

linux-2.6.15.4/drivers/i2c/i2c-core.c.bak: EPERM -> EINVAL

KJ walter harms 2 participants 4 messages Feb 25, 2006 Kernel janitors

lore
Share & tools
lore
#0 lore raw

Hi list,
it was noticed that some driver return EPERM instead of EINVAL. EPERM indicates no enought privilege to access a function. This will fixit an return (i hope) a 'better' error msg.

re,
walter

--- linux-2.6.15.4/drivers/i2c/i2c-core.c.bak 2006-02-25
13:25:48.000000000 +0100
+++ linux-2.6.15.4/drivers/i2c/i2c-core.c 2006-02-25 13:24:44.000000000
+0100
@@ -522,7 +522,7 @@
else {
pr_debug("i2c-core: %s used one too many times\n",
__FUNCTION__);
- return -EPERM;
+ return -EINVAL;
}
}

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#1 lore raw

Review

Quoted 4 lines · walter harms

pr_debug("i2c-core: %s used one too many times\n",
__FUNCTION__);

(quoted diff) +1 -1
-			return -EPERM;
+			return -EINVAL;

I think EBUSY would be more apprioriate here. I'm not sure, though. include/asm-generic/errno-base.h has a short description for each errno.

Tim
_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#2 lore raw
In reply to walter harms

hi all,
imho EBUSY would indicate someone else is using it. But < 0 should never happen (integer overflow ?). even BUG() would be ok but the author decided otherwise and i do not what to change that.

re,
walter

tim: do you know this ?
http://dev1.boomersdomain.com/cgi-bin/store_pedalinghistory_2/01068.html

Quoted 12 lines · Tim Cooijmans

Review

Quoted 4 lines · walter harms

pr_debug("i2c-core: %s used one too many times\n",
__FUNCTION__);

(quoted diff) +1 -1
-			return -EPERM;
+			return -EINVAL;

I think EBUSY would be more apprioriate here. I'm not sure, though. include/asm-generic/errno-base.h has a short description for each errno.

Tim

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#3 lore raw
In reply to walter harms
Quoted 4 lines · walter harms

hi all,
imho EBUSY would indicate someone else is using it. But < 0 should never happen (integer overflow ?). even BUG() would be ok but the author decided otherwise and i do not what to change that.

Oh wait, I thought the function would try to set something up and fail at that. Didn't see it was only trying to release stuff.

Quoted 2 lines

tim: do you know this ?
http://dev1.boomersdomain.com/cgi-bin/store_pedalinghistory_2/01068.html

Hahahaha, yes, that does look familiar.

Tim
_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors