2011-04-08 09:03:45

by Paul Bolle

Subject: [PATCH] selinux: downgrade two printks to KERN_DEBUG

A recent (Fedora Rawhide) selinux policy update triggers this error:
filename_trans_read: nel=0

But there seems to be no reason why filename_trans_read() prints this
message (and a similar message) at KERN_ERR level. So downgrade it to
KERN_DEBUG level.

Signed-off-by: Paul Bolle <[email protected]>
---
Compile tested only.

security/selinux/ss/policydb.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c
index e7b850a..8e9cc8d10 100644
--- a/security/selinux/ss/policydb.c
+++ b/security/selinux/ss/policydb.c
@@ -1819,7 +1819,7 @@ static int filename_trans_read(struct policydb *p, void *fp)
goto out;
nel = le32_to_cpu(buf[0]);

- printk(KERN_ERR "%s: nel=%d\n", __func__, nel);
+ printk(KERN_DEBUG "%s: nel=%d\n", __func__, nel);

last = p->filename_trans;
while (last && last->next)
@@ -1857,7 +1857,7 @@ static int filename_trans_read(struct policydb *p, void *fp)
goto out;
name[len] = 0;

- printk(KERN_ERR "%s: ft=%p ft->name=%p ft->name=%s\n", __func__, ft, ft->name, ft->name);
+ printk(KERN_DEBUG "%s: ft=%p ft->name=%p ft->name=%s\n", __func__, ft, ft->name, ft->name);

rc = next_entry(buf, fp, sizeof(u32) * 4);
if (rc)
--
1.7.4.2

2011-04-08 13:00:29

by Eric Paris

Subject: Re: [PATCH] selinux: downgrade two printks to KERN_DEBUG

They shouldn't be there at all and it was a big screw up on my part that
they were left in the committed code. I already applied but haven't
sent or pushed a patch to remove these. Thanks Paul!

-Eric

On Fri, 2011-04-08 at 11:03 +0200, Paul Bolle wrote:
> A recent (Fedora Rawhide) selinux policy update triggers this error:
> filename_trans_read: nel=0
>
> But there seems to be no reason why filename_trans_read() prints this
> message (and a similar message) at KERN_ERR level. So downgrade it to
> KERN_DEBUG level.
>
> Signed-off-by: Paul Bolle <[email protected]>
> ---
> Compile tested only.
>
> security/selinux/ss/policydb.c | 4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c
> index e7b850a..8e9cc8d10 100644
> --- a/security/selinux/ss/policydb.c
> +++ b/security/selinux/ss/policydb.c
> @@ -1819,7 +1819,7 @@ static int filename_trans_read(struct policydb *p, void *fp)
> goto out;
> nel = le32_to_cpu(buf[0]);
>
> - printk(KERN_ERR "%s: nel=%d\n", __func__, nel);
> + printk(KERN_DEBUG "%s: nel=%d\n", __func__, nel);
>
> last = p->filename_trans;
> while (last && last->next)
> @@ -1857,7 +1857,7 @@ static int filename_trans_read(struct policydb *p, void *fp)
> goto out;
> name[len] = 0;
>
> - printk(KERN_ERR "%s: ft=%p ft->name=%p ft->name=%s\n", __func__, ft, ft->name, ft->name);
> + printk(KERN_DEBUG "%s: ft=%p ft->name=%p ft->name=%s\n", __func__, ft, ft->name, ft->name);
>
> rc = next_entry(buf, fp, sizeof(u32) * 4);
> if (rc)

2011-05-08 10:46:54

by Paul Bolle

Subject: Re: [PATCH] selinux: downgrade two printks to KERN_DEBUG

Eric,

On Fri, 2011-04-08 at 09:00 -0400, Eric Paris wrote:
> They shouldn't be there at all and it was a big screw up on my part that
> they were left in the committed code. I already applied but haven't
> sent or pushed a patch to remove these. Thanks Paul!

0) What's the status of your patch? It is not part of v2.6.39-rc6, is
it?

1) See, I just booted v2.6.39-rc6 and over a thousand of these messages
blew away the messages from the first (about) ten seconds of my first
boot of that kernel:
> $ dmesg | grep filename_trans_read | wc -l
> 1277

As is to be expected, this happened just when I wanted to see how I
apparently misconfigured something by looking at the messages the kernel
prints in those first few seconds ...


Paul