Re: [PATCH] selinux: Use pr_fmt to prefix "SELinux: "

2018-07-02 Thread Paul Moore
On Mon, Jul 2, 2018 at 10:10 PM Joe Perches wrote: > On Mon, 2018-07-02 at 21:15 -0400, Paul Moore wrote: > > On Mon, Jul 2, 2018 at 5:31 PM Joe Perches wrote: > > > On Mon, 2018-07-02 at 16:51 -0400, Paul Moore wrote: > > > > On Wed, Jun 20, 2018 at 2:39 AM Joe Perches wrote: > > > > > pr_fmt

Re: [PATCH] selinux: Use pr_fmt to prefix "SELinux: "

2018-07-02 Thread Paul Moore
On Mon, Jul 2, 2018 at 5:31 PM Joe Perches wrote: > On Mon, 2018-07-02 at 16:51 -0400, Paul Moore wrote: > > On Wed, Jun 20, 2018 at 2:39 AM Joe Perches wrote: > > > pr_fmt can be used with the pr_ macros to prefix > > > arbitrary content to logging messages. > > > > > > So add '#define

Re: [PATCH 1/1] mcstrans: fix memory leaks reported by clang's static analyzer

2018-07-02 Thread William Roberts
On Mon, Jul 2, 2018 at 11:38 AM, Nicolas Iooss wrote: > On Sun, Jul 1, 2018 at 10:51 PM, William Roberts > wrote: >> I see lots of repeating blocks, would it make more sense to goto an >> error label and free them then return -1? > > Both trans_context() and untrans_context() currently define

Re: [PATCH] selinux: Use pr_fmt to prefix "SELinux: "

2018-07-02 Thread Paul Moore
On Wed, Jun 20, 2018 at 2:39 AM Joe Perches wrote: > pr_fmt can be used with the pr_ macros to prefix > arbitrary content to logging messages. > > So add '#define pr_fmt(fmt) "SELinux: " fmt' to selinux files > that use pr_ and remove embedded "SELinux: " prefixes > from the format strings. > >

Re: [PATCH 1/1] mcstrans: fix memory leaks reported by clang's static analyzer

2018-07-02 Thread Nicolas Iooss
On Sun, Jul 1, 2018 at 10:51 PM, William Roberts wrote: > I see lots of repeating blocks, would it make more sense to goto an > error label and free them then return -1? Both trans_context() and untrans_context() currently define "char *ltrans = NULL, *utrans = NULL;" and "char *lrange = NULL,

[RFC PATCH v2] ipv6: make ipv6_renew_options() interrupt/kernel safe

2018-07-02 Thread Paul Moore
From: Paul Moore At present the ipv6_renew_options_kern() function ends up calling into access_ok() which is problematic if done from inside an interrupt as access_ok() calls WARN_ON_IN_IRQ() on some (all?) architectures (x86-64 is affected). Example warning/backtrace is shown below: WARNING:

Re: [PATCH] selinux: move user accesses in selinuxfs out of locked regions

2018-07-02 Thread Jann Horn via Selinux
On Fri, Jun 29, 2018 at 2:38 AM Paul Moore wrote: > > On Thu, Jun 28, 2018 at 8:23 PM Paul Moore wrote: > > On Tue, Jun 26, 2018 at 8:15 AM Stephen Smalley wrote: > > > On 06/25/2018 12:34 PM, Jann Horn wrote: > > > > If a user is accessing a file in selinuxfs with a pointer to a userspace > >

[RFC net-next 01/15] net: Reserve protocol numbers for LoRa

2018-07-02 Thread Andreas Färber
LoRa is a long-range, low-power wireless network technology by Semtech. It serves as base for LoRaWAN as well as multiple proprietary protocols. AF_LORA PF_LORA ARPHRD_LORA ETH_P_LORA Signed-off-by: Andreas Färber --- include/linux/socket.h | 4 +++- include/uapi/linux/if_arp.h

Re: [RFC PATCH] ipv6: make ipv6_renew_options() interrupt/kernel safe

2018-07-02 Thread Paul Moore
On July 1, 2018 11:01:04 PM Paul Moore wrote: > From: Paul Moore > > At present the ipv6_renew_options_kern() function ends up calling into > access_ok() which is problematic if done from inside an interrupt as > access_ok() calls WARN_ON_IN_IRQ() on some (all?) architectures > (x86-64 is