Skip to contentSkip to conversation

regulator: fix fatal kernel-doc error

PATCH In review Randy Dunlap 3 participants 8 messages Sep 17, 2013 – Sep 19, 2013 Voltage and current regulator framework

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 regulator: fix fatal kernel-doc error

From: Randy Dunlap <[email protected]>

Fix fatal kernel-doc error in <linux/regulator/driver.h>:

Error(include/linux/regulator/driver.h:52): cannot understand prototype: 'struct regulator_linear_range '

Cc Liam Girdwood <[email protected]>
Cc Mark Brown <[email protected]>
---
 include/linux/regulator/driver.h |    2 ++
 1 file changed, 2 insertions(+)
include/linux/regulator/driver.h +2 -0
@@ -40,6 +40,8 @@ enum regulator_status {
 };
 
 /**
+ * struct regulator_linear_range - specify voltage ranges
+ *
  * Specify a range of voltages for regulator_map_linar_range() and
  * regulator_list_linear_range().
  *
maintainer
#1 lore raw
Quoted 5 lines · Randy Dunlap

From: Randy Dunlap <[email protected]>

Fix fatal kernel-doc error in <linux/regulator/driver.h>:

Error(include/linux/regulator/driver.h:52): cannot understand prototype: 'struct regulator_linear_range '

Applied with the first line redone, but this seems like really terrible quality of implementation for the kernel-doc stuff - it shouldn't explode over something readily copable with like this. Should we also not pick this sort of thing up in -next?

#2 lore raw
Quoted 11 lines · Mark Brown
Quoted 5 lines · Randy Dunlap

From: Randy Dunlap <[email protected]>

Fix fatal kernel-doc error in <linux/regulator/driver.h>:

Error(include/linux/regulator/driver.h:52): cannot understand prototype: 'struct regulator_linear_range '

Applied with the first line redone, but this seems like really terrible quality of implementation for the kernel-doc stuff - it shouldn't explode over something readily copable with like this. Should we also not pick this sort of thing up in -next?

/** means kernel-doc syntax and it was not in proper kernel-doc format, but maybe it could just be a Warning instead of a fatal Error.

Ideally we should pick it up in -next, of course, but I doubt that anyone is running kernel-doc on linux-next. I used to do that, but there are too many errors/warnings. I suppose that I could just concentrate on (fatal) Errors in linux-next and ignore the Warnings.

Signature · 1 line
~Randy
maintainer
#3 lore raw
Quoted 9 lines · Randy Dunlap
Quoted 4 lines · Mark Brown

Applied with the first line redone, but this seems like really terrible quality of implementation for the kernel-doc stuff - it shouldn't explode over something readily copable with like this. Should we also not pick this sort of thing up in -next?

/** means kernel-doc syntax and it was not in proper kernel-doc format, but maybe it could just be a Warning instead of a fatal Error.

Yes, that's what I'm saying - it really doesn't seem like something that should be a fatal error, that's far too fragile.

Quoted 4 lines

Ideally we should pick it up in -next, of course, but I doubt that anyone is running kernel-doc on linux-next. I used to do that, but there are too many errors/warnings. I suppose that I could just concentrate on (fatal) Errors in linux-next and ignore the Warnings.

Could these checks be added to Fengguang's tester? If nobody's looking at the output (I've never found the processed output useful myself) or running the tool then it's a bit worrying...

#4 lore raw
Quoted 9 lines · Mark Brown
Quoted 4 lines · Randy Dunlap

On 09/17/13 03:32, Mark Brown wrote:

/** means kernel-doc syntax and it was not in proper kernel-doc format, but maybe it could just be a Warning instead of a fatal Error.

Yes, that's what I'm saying - it really doesn't seem like something that should be a fatal error, that's far too fragile.

Quoted 8 lines
Quoted 4 lines

Ideally we should pick it up in -next, of course, but I doubt that anyone is running kernel-doc on linux-next. I used to do that, but there are too many errors/warnings. I suppose that I could just concentrate on (fatal) Errors in linux-next and ignore the Warnings.

Could these checks be added to Fengguang's tester? If nobody's looking at the output (I've never found the processed output useful myself) or running the tool then it's a bit worrying...

Yes, I'm running "make htmldocs" checks and send out all new warnings:

 Sep 03 To Mark Brown   (  19:0) [regulator:topic/devm 5/15] Warning(drivers/regulator/core.c:3750): No description found for parameter 'dev'
 Sep 06 To Miklos Szere (  20:0) [vfs:for-miklos 34/43] Warning(fs/dcache.c:1167): No description found for parameter 'data'
 Sep 10 To Waiman Long  (  22:0) [vfs:for-next 9/10] Warning(fs/dcache.c:102): No description found for parameter 'lock'
 Sep 11 To Lars-Peter C (  25:0) [xlnx:master-next 96/99] Warning(drivers/gpu/drm/drm_edid.c:1061): No description found for parameter 'data'    
 Sep 12 To Brian Norris (  13:0) [l2-mtd:master 9/11] Warning(include/linux/mtd/nand.h:587): Excess struct/union/enum/typedef member 'ecclayout'

Does that sound enough?

Thanks,
Fengguang

maintainer
#5 lore raw
Quoted 9 lines · Fengguang Wu
Quoted 3 lines · Mark Brown

Could these checks be added to Fengguang's tester? If nobody's looking at the output (I've never found the processed output useful myself) or running the tool then it's a bit worrying...

Yes, I'm running "make htmldocs" checks and send out all new warnings:

Does that sound enough?

I'd expect it to be but then it seems like the issue that Randy found wasn't picked up... not sure if there's some other build that would cover it or something.

#6 lore raw
Quoted 7 lines · Mark Brown
Quoted 5 lines · Fengguang Wu
Quoted 3 lines · Mark Brown

Could these checks be added to Fengguang's tester? If nobody's looking at the output (I've never found the processed output useful myself) or running the tool then it's a bit worrying...

Quoted 7 lines
Quoted 3 lines

Yes, I'm running "make htmldocs" checks and send out all new warnings:

Does that sound enough?

I'd expect it to be but then it seems like the issue that Randy found wasn't picked up... not sure if there's some other build that would cover it or something.

Yes.. it was missed because the script used to only grep for "Warning(.*)" patterns and ignored the "Error(.*)" ones. Fixed since last email.

Thanks,
Fengguang

maintainer
#7 lore raw
Quoted 9 lines · Fengguang Wu
Quoted 3 lines · Mark Brown

I'd expect it to be but then it seems like the issue that Randy found wasn't picked up... not sure if there's some other build that would cover it or something.

Yes.. it was missed because the script used to only grep for "Warning(.*)" patterns and ignored the "Error(.*)" ones. Fixed since last email.

Ah, great - thanks!