Linux-rt-users archive on lore.kernel.org
 help / color / Atom feed
From: Gene Heskett <gheskett@shentel.net>
To: "linux-rt-users" <linux-rt-users@vger.kernel.org>
Subject: Re: [ANNOUNCE] v5.4.10-rt5
Date: Tue, 14 Jan 2020 18:10:48 -0500
Message-ID: <202001141810.49000.gheskett@shentel.net> (raw)
In-Reply-To: <CAD56B7dVBuvrpVCz8ZCvc6GuY=W8rhg=Dmny1+j2ffrdz2SX4w@mail.gmail.com>

On Tuesday 14 January 2020 16:26:02 Paul Thomas wrote:

> On Tue, Jan 14, 2020 at 4:04 PM Gene Heskett <gheskett@shentel.net> 
wrote:
> > On Tuesday 14 January 2020 12:15:37 Paul Thomas wrote:
> > > On Tue, Jan 14, 2020 at 11:24 AM Gene Heskett
> > > <gheskett@shentel.net>
> >
> > wrote:
> > > > On Tuesday 14 January 2020 08:54:30 Paul Thomas wrote:
> > > > > On Sun, Jan 12, 2020 at 9:44 PM Gene Heskett
> > > > > <gheskett@shentel.net>
> > > >
> > > > wrote:
> > > > > > On Friday 10 January 2020 15:14:30 Sebastian Andrzej Siewior
> >
> > wrote:
> > > > > > > Dear RT folks!
> > > > > > >
> > > > > > > I'm pleased to announce the v5.4.10-rt5 patch set.
> > > > > > >
> > > > > > > Changes since v5.4.10-rt4:
> > > > > > >
> > > > > > >   - Dick Hollenbeck reported that the printk rework had a
> > > > > > > negative impact on the 8250 driver if not used as a
> > > > > > > console. Patch by John Ogness.
> > > > > > >
> > > > > > > Known issues
> > > > > > >      - None
> > > > > > >
> > > > > > > The delta patch against v5.4.10-rt4 is appended below and
> > > > > > > can be found here:
> > > > > > >
> > > > > > >
> > > > > > > https://cdn.kernel.org/pub/linux/kernel/projects/rt/5.4/in
> > > > > > >cr/p atch -5.4 .10-rt4-rt5.patch.xz
> > > > > > >
> > > > > > > You can get this release via the git tree at:
> > > > > > >
> > > > > > >
> > > > > > > git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-
> > > > > > >deve l.gi t v5.4.10-rt5
> > > > > >
> > > > > > I grabbed a copy of this, then turned off anything not
> > > > > > needed to run LinuxCNC on an rpi4 in armhf config.  Then I
> > > > > > noted its make had a make deb-pkg option.  But it isn't
> > > > > > making what I told it to make as I'm sitting here watching
> > > > > > it ALL go by. Did I miss something in  a make menuconfig
> > > > > > that triggered a make all? It was supposedly set for armhf.
> > > > > > But I'm even seeing x86 stuff go by. Obviously I know just
> > > > > > enough to screw it up. Point me at the right docs for arm
> > > > > > builds please.
> > > > > >
> > > > > > Hi Gene,
> > > > >
> > > > > I recently stumbled into deb-pkg, and it's awesome building up
> > > > > the source and standard debian packages all in one command,
> > > > > but it takes the same arguments as a standard make. For rpi4
> > > > > this is arm64 not armhf so you probably want something like
> > > > > below to configure for arm64
> > > >
> > > > armhf (32 bit) runs fine in on an RPI4, been doing it for a
> > > > couple months now. With a 32 bit stack frame, context switch
> > > > latencies are several microseconds better which linuxcnc loves.
> > > > However with mesa interface cards handling all the
> > > > sub-millisecond stuffs and using preempt-rt running in
> > > > userspace, it all works quite well as the  max latency stays
> > > > below 200 microseconds 99.99% of the time. Thats well within the
> > > > servo-thread time of 1 millisecond the mesa cards need.
> > >
> > > Ah, makes sense.
> > >
> > > > > If you are doing this on the rpi4 you don't need the
> > > > > CROSS_COMPILE= option.[1][2] make -j4 ARCH=arm64
> > > > > CROSS_COMPILE=aarch64-elf- defconfig
> > > > > Or if the kernel you're already running has a /proc/config.gz
> > > > > you can use that directly (zcat /proc/config.gz > .config).
> > > > > Then you can use the main make.
> > > >
> > > > The kernel I an running was built, by me (queue the raspberry
> > > > noises), with only the ondemand governer so while its working,
> > > > its stuck in low gear. And the pi has plenty of cooling.
> > > >
> > > > that kernel running now is a 4.19.71-rt24-v7l+ #1 SMP PREEMPT
> > > > RT, but there has since been added another video speedup option,
> > > > so ATM I am trying to build a 4.19.94-rt38 announced last week.
> > > > But the git clone I pulled doesn't have the rt patch. And
> > > > pulling that patch with curl today fails:
> > > >
> > > > pi@rpi4:/media/pi/workspace $ curl
> > > > http://www.kernel.org/pub/linux/kernel/projects/rt/4.19/patch-4.
> > > >19.9 4-rt38.patch.xz
> > >
> > > Hmm, this seems fine for me:
> > >  wget
> > > https://cdn.kernel.org/pub/linux/kernel/v4.x/linux-4.19.94.tar.xz
> > > tar xf linux-4.19.94.tar.xz
> > > cd linux-4.19.94/
> > >  wget
> > > http://www.kernel.org/pub/linux/kernel/projects/rt/4.19/patch-4.19
> > >.94- rt38.patch.xz
> > > unxz patch-4.19.94-rt38.patch.xz
> > > patch -p1 < patch-4.19.94-rt38.patch
> >
> > I blew away my git clone and followed this, which all worked to
> > here,
> >
> > but:
> > >  make ARCH=armhf defconfig
>
That was the fix. make -j4 zImage underway now.

> Sorry that was a cut/paste typo, there is no arch/armhf just arm, so
> you want make ARCH=arm defconfig
> I had tried the command that's why it was in my history and thus the
> copy/paste mistake. I believe how this works is the kernel just has
> CONFIG_AEABI=y as the ABI config, then userspace compilers have to
> decide if they want to compile for a hardware floating point or not,
> and what debian has decided is that they would have a special
> architecture just for hardfloat:
> https://wiki.debian.org/ArmHardFloatPort
>
> thanks,
> Paul

I'll put this on a fresh 64Gig card, that way I'll have a fallback.

Thank you very much Paul.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>

      reply index

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-10 20:14 Sebastian Andrzej Siewior
2020-01-13  2:22 ` Gene Heskett
2020-01-14  8:01   ` Alexander Dahl
2020-01-14 13:56   ` Paul Thomas
     [not found]   ` <CAD56B7cgPfbcx8MNtYosOwhPi6fy+fqdauLsayR65dAg1yaYSg@mail.gmail.com>
2020-01-14 16:23     ` Gene Heskett
2020-01-14 17:15       ` Paul Thomas
2020-01-14 21:03         ` Gene Heskett
2020-01-14 21:26           ` Paul Thomas
2020-01-14 23:10             ` Gene Heskett [this message]

Reply instructions:

You may reply publically to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=202001141810.49000.gheskett@shentel.net \
    --to=gheskett@shentel.net \
    --cc=linux-rt-users@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-rt-users archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-rt-users/0 linux-rt-users/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-rt-users linux-rt-users/ https://lore.kernel.org/linux-rt-users \
		linux-rt-users@vger.kernel.org
	public-inbox-index linux-rt-users

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-rt-users


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git