BUG and WARN kernel log levels

BUG and WARN kernel log levels

[Kees Cook]

Hi,

So, I noticed that asm-gemeric/bug.h defines BUG() without a log level:

#ifndef HAVE_ARCH_BUG
#define BUG() do { \
       printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \

Seems like it should have one?

Also, I think we might want to examine WARN() a bit... it doesn't have
a log level either, but only a fraction of callers set one:

$ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep -v KERN_ | wc -l
2735

$ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep KERN_ | wc -l
77

If I'm reading checkpatch.pl correctly, it doesn't warn about missing
log levels on WARN calls, but I think it should.

How do you think is best to clean this up?

Mainly, I'd like to add a format string to BUG, or introduce a new
BUGish call that takes a format...

-Kees

-- 
Kees Cook
Nexus Security

Re: BUG and WARN kernel log levels

[Joe Perches]

On Mon, 2016-08-15 at 11:53 -0700, Kees Cook wrote:
> Hi,
> 
> So, I noticed that asm-gemeric/bug.h defines BUG() without a log level:
> 
> #ifndef HAVE_ARCH_BUG
> #define BUG() do { \
>        printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \
> 
> Seems like it should have one?
> 
> Also, I think we might want to examine WARN() a bit... it doesn't have
> a log level either, but only a fraction of callers set one:
> 
> $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep -v KERN_ | wc -l
> 2735
> 
> $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep KERN_ | wc -l
> 77
> 
> If I'm reading checkpatch.pl correctly, it doesn't warn about missing
> log levels on WARN calls, but I think it should.
> 
> How do you think is best to clean this up?
> 
> Mainly, I'd like to add a format string to BUG, or introduce a new
> BUGish call that takes a format...

I once suggested something similar awhile ago.
https://lkml.org/lkml/2008/7/8/261

I think it's best to remove any KERN_<LEVEL> from the use of
all the WARN variants and add it to the WARN definitions.

Same with BUG.

Re: BUG and WARN kernel log levels

[Kees Cook]

On Mon, Aug 15, 2016 at 12:00 PM, Joe Perches <joe@perches.com> wrote:
> On Mon, 2016-08-15 at 11:53 -0700, Kees Cook wrote:
>> Hi,
>>
>> So, I noticed that asm-gemeric/bug.h defines BUG() without a log level:
>>
>> #ifndef HAVE_ARCH_BUG
>> #define BUG() do { \
>>        printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \
>>
>> Seems like it should have one?
>>
>> Also, I think we might want to examine WARN() a bit... it doesn't have
>> a log level either, but only a fraction of callers set one:
>>
>> $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep -v KERN_ | wc -l
>> 2735
>>
>> $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep KERN_ | wc -l
>> 77
>>
>> If I'm reading checkpatch.pl correctly, it doesn't warn about missing
>> log levels on WARN calls, but I think it should.
>>
>> How do you think is best to clean this up?
>>
>> Mainly, I'd like to add a format string to BUG, or introduce a new
>> BUGish call that takes a format...
>
> I once suggested something similar awhile ago.
> https://lkml.org/lkml/2008/7/8/261
>
> I think it's best to remove any KERN_<LEVEL> from the use of
> all the WARN variants and add it to the WARN definitions.

Yeah, that's what I was thinking too. It does mean that the format
needs to be a const char string, though. I haven't checked all of them
but most seem to be. It's an easy fix to move them to "%s" as needed,
though.

> Same with BUG.

Yeah, though for full effect, it needs to be plumbed into each
architecture's BUG handler. Mainly what I don't like is that if I do
this on x86:

pr_err("eek, terrible thing\n");
BUG();

I get a dmesg that read:

eek, terrible thing
=== cut here ===
...traceback, etc

I'd like the "eek" part to be inside the "cut here". And I'd like BUGs
to be able to be more verbose.

-Kees

-- 
Kees Cook
Nexus Security

Re: BUG and WARN kernel log levels

[Joe Perches]

On Mon, 2016-08-15 at 13:28 -0700, Kees Cook wrote:
> On Mon, Aug 15, 2016 at 12:00 PM, Joe Perches <joe@perches.com> wrote:
> > On Mon, 2016-08-15 at 11:53 -0700, Kees Cook wrote:
> > > So, I noticed that asm-gemeric/bug.h defines BUG() without a log level:
> > > 
> > > #ifndef HAVE_ARCH_BUG
> > > #define BUG() do { \
> > >        printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \
> > > 
> > > Seems like it should have one?
> > > 
> > > Also, I think we might want to examine WARN() a bit... it doesn't have
> > > a log level either, but only a fraction of callers set one:
> > > 
> > > $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep -v KERN_ | wc -l
> > > 2735
> > > 
> > > $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep KERN_ | wc -l
> > > 77
> > > 
> > > If I'm reading checkpatch.pl correctly, it doesn't warn about missing
> > > log levels on WARN calls, but I think it should.
> > > 
> > > How do you think is best to clean this up?
> > > 
> > > Mainly, I'd like to add a format string to BUG, or introduce a new
> > > BUGish call that takes a format...
> > I once suggested something similar awhile ago.
> > https://lkml.org/lkml/2008/7/8/261
> > 
> > I think it's best to remove any KERN_ from the use of
> > all the WARN variants and add it to the WARN definitions.
> Yeah, that's what I was thinking too. It does mean that the format
> needs to be a const char string, though.

No it doesn't mean that.

Prefixing KERN_WARNING to the const char[] types and using
KERN_WARNING "%pV" could work for the non const char[] types

Or maybe use the same KERN_WARNING "%pV" for simpler code.

> > Same with BUG.
> Yeah, though for full effect, it needs to be plumbed into each
> architecture's BUG handler. Mainly what I don't like is that if I do
> this on x86:
> 
> pr_err("eek, terrible thing\n");
> BUG();
> 
> I get a dmesg that read:
> 
> eek, terrible thing
> === cut here ===
> ...traceback, etc
> 
> I'd like the "eek" part to be inside the "cut here". And I'd like BUGs
> to be able to be more verbose.

Maybe add BUG_MSG/BUG_ON_MSG or some such.

Re: BUG and WARN kernel log levels

[Joe Perches]

On Mon, 2016-08-15 at 12:00 -0700, Joe Perches wrote:
> On Mon, 2016-08-15 at 11:53 -0700, Kees Cook wrote:
> > 
> > Hi,
> > 
> > So, I noticed that asm-gemeric/bug.h defines BUG() without a log level:
> > 
> > #ifndef HAVE_ARCH_BUG
> > #define BUG() do { \
> >        printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \
> > 
> > Seems like it should have one?
> > 
> > Also, I think we might want to examine WARN() a bit... it doesn't have
> > a log level either, but only a fraction of callers set one:
> > 
> > $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep -v KERN_ | wc -l
> > 2735
> > 
> > $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep KERN_ | wc -l
> > 77
> > 
> > If I'm reading checkpatch.pl correctly, it doesn't warn about missing
> > log levels on WARN calls, but I think it should.
> > 
> > How do you think is best to clean this up?
> > 
> > Mainly, I'd like to add a format string to BUG, or introduce a new
> > BUGish call that takes a format...
> I once suggested something similar awhile ago.
> https://lkml.org/lkml/2008/7/8/261

And here I submitted patches:
https://lkml.org/lkml/2010/10/30/176

Re: BUG and WARN kernel log levels

[Kees Cook]

On Wed, Aug 17, 2016 at 9:36 AM, Joe Perches <joe@perches.com> wrote:
> On Mon, 2016-08-15 at 12:00 -0700, Joe Perches wrote:
>> On Mon, 2016-08-15 at 11:53 -0700, Kees Cook wrote:
>> >
>> > Hi,
>> >
>> > So, I noticed that asm-gemeric/bug.h defines BUG() without a log level:
>> >
>> > #ifndef HAVE_ARCH_BUG
>> > #define BUG() do { \
>> >        printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __func__); \
>> >
>> > Seems like it should have one?
>> >
>> > Also, I think we might want to examine WARN() a bit... it doesn't have
>> > a log level either, but only a fraction of callers set one:
>> >
>> > $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep -v KERN_ | wc -l
>> > 2735
>> >
>> > $ git grep -E 'WARN(_TAINT|)(_RATELIMIT|_ONCE|)\([^\)]' | grep KERN_ | wc -l
>> > 77
>> >
>> > If I'm reading checkpatch.pl correctly, it doesn't warn about missing
>> > log levels on WARN calls, but I think it should.
>> >
>> > How do you think is best to clean this up?
>> >
>> > Mainly, I'd like to add a format string to BUG, or introduce a new
>> > BUGish call that takes a format...
>> I once suggested something similar awhile ago.
>> https://lkml.org/lkml/2008/7/8/261
>
> And here I submitted patches:
> https://lkml.org/lkml/2010/10/30/176

Ah, I see some of this series landed. I see commits scattered in the
tree, but not as many as you sent, I think. What still remains?

-Kees

-- 
Kees Cook
Nexus Security

Re: BUG and WARN kernel log levels

[Joe Perches]

On Wed, 2016-08-17 at 14:19 -0700, Kees Cook wrote:
> On Wed, Aug 17, 2016 at 9:36 AM, Joe Perches <joe@perches.com> wrote:
[]
> > And here I submitted patches:
> > https://lkml.org/lkml/2010/10/30/176
> Ah, I see some of this series landed. I see commits scattered in the
> tree, but not as many as you sent, I think. What still remains?

No idea.

There wasn't any enthusiasm to apply the asm-generic
patch so I didn't push it.