From: Johannes Berg <johannes@sipsolutions.net> Date: 2014-09-10 13:57:17
On Wed, 2014-09-10 at 19:21 +0530, Sudip Mukherjee wrote:
as pr_* macros are more preffered over printk, so printk replaced
with corresponding pr_* macros.
Are you simply running checkpatch on every file and decided to do
something about it? :)
I'll let Takashi decide whether to take this or not as I no longer care
about this code, but IMHO this changes is completely pointless since you
don't also clean up the code to have a common prefix with #define pr_fmt
and then clean up the callers etc.
There's a reason pr_* is preferred, but random code changes like this
aren't it, I think.
johannes
On Wed, Sep 10, 2014 at 03:57:04PM +0200, Johannes Berg wrote:
On Wed, 2014-09-10 at 19:21 +0530, Sudip Mukherjee wrote:
quoted
as pr_* macros are more preffered over printk, so printk replaced
with corresponding pr_* macros.
Are you simply running checkpatch on every file and decided to do
something about it? :)
i am running checkpatch on the patch generated. if i am doing checkpatch
cleanups then that i do it only in the staging.
only exception : printk .. :)
I'll let Takashi decide whether to take this or not as I no longer care
about this code, but IMHO this changes is completely pointless since you
don't also clean up the code to have a common prefix with #define pr_fmt
and then clean up the callers etc.
i mentioned in the comment that in a future patch we can have pr_fmt,
it was not done in this patch since the changes for this patch is
generated by a script and not manually.
if Takashi accepts this then the next patch will have pr_fmt.
thanks
sudip
There's a reason pr_* is preferred, but random code changes like this
aren't it, I think.
johannes
At Wed, 10 Sep 2014 20:02:04 +0530,
Sudip Mukherjee wrote:
On Wed, Sep 10, 2014 at 03:57:04PM +0200, Johannes Berg wrote:
quoted
On Wed, 2014-09-10 at 19:21 +0530, Sudip Mukherjee wrote:
quoted
as pr_* macros are more preffered over printk, so printk replaced
with corresponding pr_* macros.
Are you simply running checkpatch on every file and decided to do
something about it? :)
i am running checkpatch on the patch generated. if i am doing checkpatch
cleanups then that i do it only in the staging.
only exception : printk .. :)
quoted
I'll let Takashi decide whether to take this or not as I no longer care
about this code, but IMHO this changes is completely pointless since you
don't also clean up the code to have a common prefix with #define pr_fmt
and then clean up the callers etc.
i mentioned in the comment that in a future patch we can have pr_fmt,
it was not done in this patch since the changes for this patch is
generated by a script and not manually.
if Takashi accepts this then the next patch will have pr_fmt.
If you're going to work on it, please give a patch series and let me
merge once. There is no good merit to merge a half-baked piece by
piece.
Regarding the changes you've made: so far, I've merged two such
patches just because it's a good exercise for newbies. You've played
it and experienced it enough. So it's time to go up to a higher
stage, more "real" fixes.
For example, if you are still interested in printk stuff, try to
change the calls to dev_err() and co. Of course, this needs more
understanding of the code you'll handle, which object is passed for
which messages.
thanks,
Takashi
On Wed, Sep 10, 2014 at 04:43:03PM +0200, Takashi Iwai wrote:
At Wed, 10 Sep 2014 20:02:04 +0530,
Sudip Mukherjee wrote:
quoted
On Wed, Sep 10, 2014 at 03:57:04PM +0200, Johannes Berg wrote:
quoted
On Wed, 2014-09-10 at 19:21 +0530, Sudip Mukherjee wrote:
quoted
as pr_* macros are more preffered over printk, so printk replaced
with corresponding pr_* macros.
Are you simply running checkpatch on every file and decided to do
something about it? :)
i am running checkpatch on the patch generated. if i am doing checkpatch
cleanups then that i do it only in the staging.
only exception : printk .. :)
quoted
I'll let Takashi decide whether to take this or not as I no longer care
about this code, but IMHO this changes is completely pointless since you
don't also clean up the code to have a common prefix with #define pr_fmt
and then clean up the callers etc.
i mentioned in the comment that in a future patch we can have pr_fmt,
it was not done in this patch since the changes for this patch is
generated by a script and not manually.
if Takashi accepts this then the next patch will have pr_fmt.
If you're going to work on it, please give a patch series and let me
merge once. There is no good merit to merge a half-baked piece by
piece.
Regarding the changes you've made: so far, I've merged two such
patches just because it's a good exercise for newbies. You've played
it and experienced it enough. So it's time to go up to a higher
stage, more "real" fixes.
can you please give me some hint of fixes that can be attempted by
newbies. except printk :)
so far i have started with fixing the sparse warnings in staging.
For example, if you are still interested in printk stuff, try to
change the calls to dev_err() and co. Of course, this needs more
understanding of the code you'll handle, which object is passed for
which messages.
At Wed, 10 Sep 2014 20:37:50 +0530,
Sudip Mukherjee wrote:
On Wed, Sep 10, 2014 at 04:43:03PM +0200, Takashi Iwai wrote:
quoted
At Wed, 10 Sep 2014 20:02:04 +0530,
Sudip Mukherjee wrote:
quoted
On Wed, Sep 10, 2014 at 03:57:04PM +0200, Johannes Berg wrote:
quoted
On Wed, 2014-09-10 at 19:21 +0530, Sudip Mukherjee wrote:
quoted
as pr_* macros are more preffered over printk, so printk replaced
with corresponding pr_* macros.
Are you simply running checkpatch on every file and decided to do
something about it? :)
i am running checkpatch on the patch generated. if i am doing checkpatch
cleanups then that i do it only in the staging.
only exception : printk .. :)
quoted
I'll let Takashi decide whether to take this or not as I no longer care
about this code, but IMHO this changes is completely pointless since you
don't also clean up the code to have a common prefix with #define pr_fmt
and then clean up the callers etc.
i mentioned in the comment that in a future patch we can have pr_fmt,
it was not done in this patch since the changes for this patch is
generated by a script and not manually.
if Takashi accepts this then the next patch will have pr_fmt.
If you're going to work on it, please give a patch series and let me
merge once. There is no good merit to merge a half-baked piece by
piece.
Regarding the changes you've made: so far, I've merged two such
patches just because it's a good exercise for newbies. You've played
it and experienced it enough. So it's time to go up to a higher
stage, more "real" fixes.
can you please give me some hint of fixes that can be attempted by
newbies. except printk :)
You should study the code at first. There is nothing you can "fix"
without understanding the code. So, pick up a driver you're
interested in. You may or may not find bugs there. Or, if you see /
know any bug, try to join debugging.
Takashi
On Wed, Sep 10, 2014 at 05:38:10PM +0200, Takashi Iwai wrote:
At Wed, 10 Sep 2014 20:37:50 +0530,
Sudip Mukherjee wrote:
quoted
On Wed, Sep 10, 2014 at 04:43:03PM +0200, Takashi Iwai wrote:
quoted
At Wed, 10 Sep 2014 20:02:04 +0530,
Sudip Mukherjee wrote:
quoted
On Wed, Sep 10, 2014 at 03:57:04PM +0200, Johannes Berg wrote:
quoted
On Wed, 2014-09-10 at 19:21 +0530, Sudip Mukherjee wrote:
quoted
as pr_* macros are more preffered over printk, so printk replaced
with corresponding pr_* macros.
Are you simply running checkpatch on every file and decided to do
something about it? :)
i am running checkpatch on the patch generated. if i am doing checkpatch
cleanups then that i do it only in the staging.
only exception : printk .. :)
quoted
I'll let Takashi decide whether to take this or not as I no longer care
about this code, but IMHO this changes is completely pointless since you
don't also clean up the code to have a common prefix with #define pr_fmt
and then clean up the callers etc.
i mentioned in the comment that in a future patch we can have pr_fmt,
it was not done in this patch since the changes for this patch is
generated by a script and not manually.
if Takashi accepts this then the next patch will have pr_fmt.
If you're going to work on it, please give a patch series and let me
merge once. There is no good merit to merge a half-baked piece by
piece.
Regarding the changes you've made: so far, I've merged two such
patches just because it's a good exercise for newbies. You've played
it and experienced it enough. So it's time to go up to a higher
stage, more "real" fixes.
can you please give me some hint of fixes that can be attempted by
newbies. except printk :)
You should study the code at first. There is nothing you can "fix"
without understanding the code. So, pick up a driver you're
interested in. You may or may not find bugs there. Or, if you see /
know any bug, try to join debugging.
Takashi
one of my patch has alredy changed printk to pr_* in ctxfi. but that can be improved
to dev_*. since this has to be done manully and not through script , so should i send the
patch for one file at a time or for all of them together in a single patch?
thanks
sudip