[ANNOUNCE] 4.14.87-rt49

[ANNOUNCE] 4.14.87-rt49

[Steven Rostedt]

Dear RT Folks,

I'm pleased to announce the 4.14.87-rt49 stable release.

Sorry for taking so long, I've just been swamped (like everyone else)

*** PLEASE NOTE ***

This release is **NOT** just an update to the new stable 4.14.87 version
as there was a small RT specific change that has been made.

I had to make an update to handle a backport of adding
preempt_disable() around the console_unlock() code. That patch is below.

You can get this release via the git tree at:

  git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-stable-rt.git

  branch: v4.14-rt
  Head SHA1: c3fb62599cb0bf9ea0223433ccf7c0c37056c392


Or to build 4.14.87-rt49 directly, the following patches should be applied:

  http://www.kernel.org/pub/linux/kernel/v4.x/linux-4.14.tar.xz

  http://www.kernel.org/pub/linux/kernel/v4.x/patch-4.14.87.xz

  http://www.kernel.org/pub/linux/kernel/projects/rt/4.14/patch-4.14.87-rt49.patch.xz




Enjoy,

-- Steve


From 0ab3459b0a8b8d19054aa813611c8dc22ce04af8 Mon Sep 17 00:00:00 2001
From: "Steven Rostedt (VMware)" <rostedt@goodmis.org>
Date: Fri, 14 Dec 2018 21:36:55 -0500
Subject: [PATCH] printk: Do not disable preemption when calling consoles in RT

console_unlock() may call consoles that call spin_lock(), where spin_lock is
a mutex in PREEMPT_RT. Although mainline may be worried about preempted
holders of the console lock from preventing printk() output, the RT kernel
does not.

Use migrate_disable() to disable preemption when !RT, but we really don't
care if the task stays on the CPU or not for RT. It just has to be able to
schedule.

Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
---
 kernel/printk/printk.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 8a7c4f79fca0..029ae3caa80e 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -1811,7 +1811,7 @@ asmlinkage int vprintk_emit(int facility, int level,
 		 * console_sem which would prevent anyone from printing to
 		 * console
 		 */
-		preempt_disable();
+		migrate_disable();
 		/*
 		 * Try to acquire and then immediately release the console
 		 * semaphore.  The release will print out buffers and wake up
@@ -1819,7 +1819,7 @@ asmlinkage int vprintk_emit(int facility, int level,
 		 */
 		if (may_trylock && console_trylock())
 			console_unlock();
-		preempt_enable();
+		migrate_enable();
 	}
 
 	return printed_len;
-- 
2.19.2

Re: [ANNOUNCE] 4.14.87-rt49

[Sebastian Andrzej Siewior]

On 2018-12-18 15:40:34 [-0500], Steven Rostedt wrote:
> 
> Dear RT Folks,
> 
> I'm pleased to announce the 4.14.87-rt49 stable release.

This contains also commit
  e8499ab5b93ac ("x86/fpu: Disable bottom halves while loading FPU registers")

which means you need commit
  [PATCH] x86/fpu: Disable preemption around local_bh_disable()
  https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git/plain/patches/x86-fpu-Disable-preemption-around-local_bh_disable.patch?h=linux-4.19.y-rt-patches

Sebastian

Re: [ANNOUNCE] 4.14.87-rt49

[Steven Rostedt]

On Tue, 18 Dec 2018 23:08:36 +0100
Sebastian Andrzej Siewior <bigeasy@linutronix.de> wrote:

> On 2018-12-18 15:40:34 [-0500], Steven Rostedt wrote:
> > 
> > Dear RT Folks,
> > 
> > I'm pleased to announce the 4.14.87-rt49 stable release.  
> 
> This contains also commit
>   e8499ab5b93ac ("x86/fpu: Disable bottom halves while loading FPU registers")
> 
> which means you need commit
>   [PATCH] x86/fpu: Disable preemption around local_bh_disable()
>   https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git/plain/patches/x86-fpu-Disable-preemption-around-local_bh_disable.patch?h=linux-4.19.y-rt-patches
> 

I plan on doing a backport soon. My tests didn't bomb without this
patch :-) so i released this stable. But this just shows I need to do a
backport of your updates ASAP.

-- Steve