you or the "DT guys" can say, can convince me that we should reverse
that course.
I don't care how it works, but enabling and using PWM on an
out-of-the-box RPi should not require a reboot.
Over&Out
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org
without calling ->attach())
That way you would know you loaded the right driver, rather
than have to guess.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malic
In message
, Warner Losh writes:
>> That also removes a couple of boilerplate lines from all FDT device
>> drivers.
>
>That would be useful, but the status bit is so much more than on/off.
The boilerplate code in all the ->probe() functions don't seem to th
uded me, apart from some people pointing out that "somebody
should write XYZ" for severely underdocumented values of XYZ.
And now I *really* think this thread should end...
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
Free
ld avoid that
boilerplate in all the drivers, if it was done one level up with
a consistent device_printf() and not calling attach() at all.
What I don't know is how noisy that would be in practice ?
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RF
In message
, Warner
Losh writes:
>--089e08222b4ce7f8a4056372a0d5
>I'm actually thinking the right thing is to remove the checks everywhere.
>
>We then create a couple of new simplebus methods:
+1
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.
then, you are wasting everybodys time pointing accusingly
into your book of unwritten rules.
Poul-Henning
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice wh
ot on every other armv7/arm64 platform, just not there incrementally at
>runtime. Some work would be needed to make that happen, but I think I have a
>reasonable design that I've started to socialize... You're welcome to join in
>in making it happen, if you like
With the l
together to extend arm support toward what
>you want to do.
Empathetically NO!
(But feel free to ask me again, if my driver ever in any way prevents
you from committing your changes to the RPi platform.)
Poul-Henning
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.
gt;
>Yea, otherwise we could kill armv6 completely.
But it is only the original RPi is armv6, right ?
I think we would be OK as long as we support RPi2 and RPi3.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer | B
t", and there are
indeed hardware out there where that is required for various reason.
But trival (and popular) cases "should just work".
>It's a tricky ball of wax that we've neglected for too long.
Indeed, and thanks for doing so.
--
Poul-Henning Kamp | UNI
best practices and guidelines are unwriteen, and
>they're not always implemented on older platforms. And it's the
>problem from which this situation has risen.
With the added cherry on top that RPi is a horrible platform which
nobody loves - except thousands of teachers, students, hacke
ck some shrewed guesses
and finally in-lab measurement of what actually transpired when I
frobbed registers.
But more importantly, I have no idea what servies *a* clock
manager offers, through which apis and to what clients and
at what level of abstraction and flexibility ?
--
Poul-Henning Kamp | UN
oo weird to
>be reference. TI's AM335x or iMX would be a good candidate I think
>They are complex enough to illustrate all the concepts yet not super
>complex.
I fully agree that RPi should *not* be our reference platform.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
In message <20180130094251.ga85...@bluezbox.com>, Oleksandr Tymoshenko writes:
>Poul-Henning Kamp (p...@phk.freebsd.dk) wrote:
>> But only if you add pwm.dtbo to config.txt and reboot, because
>> I see nothing indicating that we will put it there in the
>> rele
w.
So the essential architectural question seems to be:
Can we load the pwm overlay by default, but still use the
PWM-pins for something else until/if the pwm driver is kldloaded ?
If not, we're right back to FreeBSD 1.x semantics again.
--
Poul-Henning Kamp | UNIX since Zilog Z
Author: phk
Date: Sun Apr 8 19:23:50 2018
New Revision: 332297
URL: https://svnweb.freebsd.org/changeset/base/332297
Log:
Pedantic polishing of code to please FlexeLint.
Approved by: ed
Modified:
head/sys/teken/libteken/teken.3
head/sys/teken/teken.c
head/sys/teken/teken.h
head/s
Author: phk
Date: Mon Nov 27 09:57:37 2017
New Revision: 326246
URL: https://svnweb.freebsd.org/changeset/base/326246
Log:
Truncate negative lengths to zero
Modified:
head/lib/libc/stdio/xprintf_hexdump.c
Modified: head/lib/libc/stdio/xprintf_hexdump.c
===
DERED is @gibbs baby ?
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
___
nt of BIO_ORDERED...
I think ordering is strictly a consumer responsibility for exactly
(and then some) of the reasons you mention.
"End to end principle in systems design" and all that...
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC
-License-Identifier: BSD-3-Clause
+ * SPDX-License-Identifier: BSD-2-Clause
*
- * Copyright (c) 2002 David E. O'Brien. All rights reserved.
- * Copyright (c) 1991, 1993
- * The Regents of the University of California. All rights reserved.
+ * Copyright (c) 2017 Poul-Henning Kamp. All r
Author: phk
Date: Fri Dec 29 15:38:43 2017
New Revision: 327345
URL: https://svnweb.freebsd.org/changeset/base/327345
Log:
Add a crude workaround for ports with flavors
Modified:
head/tools/tools/sysbuild/sysbuild.sh
Modified: head/tools/tools/sysbuild/sysbuild.sh
===
Author: phk
Date: Wed Jan 3 09:33:59 2018
New Revision: 327515
URL: https://svnweb.freebsd.org/changeset/base/327515
Log:
Eliminate a paranthesis which is both unneeded and causing trouble.
Modified:
head/sys/sys/_stdarg.h
Modified: head/sys/sys/_stdarg.h
===
In message
, Warner Losh writes:
>Deleting lmc(4) is probably the appropriate fix.
>Inc is for very old interconnect to telco... So maybe that's safe to delete.
As the guy who put lmc(4) in the tree I concur.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@
aces in US.
I don't think they could do SS7 or ISDN like the MUSYCC chip could.
But nobody uses T1/E1/T3/E3 lines any more for anything...
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
-Clause-FreeBSD
+ *
+ * Copyright (c) 2017 Poul-Henning Kamp
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the
Author: phk
Date: Sun Jan 21 21:27:41 2018
New Revision: 328223
URL: https://svnweb.freebsd.org/changeset/base/328223
Log:
Rename rpi_pwm to bcm283x_pwm, and build it on armv[67] and arm64.
Truncate ratio if period is lowered.
Tested on Rpi2 and Rpi3.
Rpi3 requires DTB->DTS->edit-
Author: phk
Date: Sat Dec 10 09:10:48 2016
New Revision: 309790
URL: https://svnweb.freebsd.org/changeset/base/309790
Log:
Add Genesys Logic USB hub.
Make two other USB hub descriptions more precise.
Modified:
head/sys/dev/usb/usbdevs
Modified: head/sys/dev/usb/usbdevs
=
We tried to be future compatible, and UFS2
will look four different places for the superblock: +64k, +8k, +0k
and +256k, (the latter named SBLOCK_PIGGY because somebody senior
thought even 64k was an outrageous waste of space :-)
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org
: 2 months
>> Sponsored by: Dell EMC Isilon
>
>This can't be changed. It's baked into the BSD disklabel format.
No it is not, it is baked into FFS, and for UFS2 0, 8, 64 and 256K works.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org |
Author: phk
Date: Wed May 3 20:41:26 2017
New Revision: 317771
URL: https://svnweb.freebsd.org/changeset/base/317771
Log:
Flush stdout before yes/no confirmations, to force question
through pipes/tee(1)/whatever
Modified:
head/usr.sbin/pkg/pkg.c
Modified: head/usr.sbin/pkg/pkg.c
=
Author: phk
Date: Thu May 25 21:59:19 2017
New Revision: 318903
URL: https://svnweb.freebsd.org/changeset/base/318903
Log:
Try to keep up with the ports system Makefiles.
Modified:
head/tools/tools/sysbuild/sysbuild.sh
Modified: head/tools/tools/sysbuild/sysbuild.sh
=
Author: phk
Date: Tue Aug 18 20:19:48 2015
New Revision: 286904
URL: https://svnweb.freebsd.org/changeset/base/286904
Log:
Update sysbuild to new ports infrastructure.
Modified:
head/tools/tools/sysbuild/sysbuild.sh
Modified: head/tools/tools/sysbuild/sysbuild.sh
In message <20151212121209.ga60...@freebsd.org>, Alexey Dokuchaev writes:
>+1, I'm also used to "dd'ing zeros" trick.
The modern way to do that is to reinitialize the on-drive encryption
to a new key.
--
Poul-Henning Kamp | UNIX since Zil
Author: phk
Date: Fri Sep 25 07:37:00 2015
New Revision: 288212
URL: https://svnweb.freebsd.org/changeset/base/288212
Log:
Fix two cases where "const" were washed off pointers with strchr(3)
Modified:
head/sbin/natd/natd.c
Modified: head/sbin/natd/natd.c
=
Author: phk
Date: Fri Oct 2 09:23:14 2015
New Revision: 288484
URL: https://svnweb.freebsd.org/changeset/base/288484
Log:
Fail the sbuf if vsnprintf(3) fails.
Modified:
head/sys/kern/subr_sbuf.c
Modified: head/sys/kern/subr_sbuf.c
Author: phk
Date: Sat Oct 10 17:32:06 2015
New Revision: 289116
URL: https://svnweb.freebsd.org/changeset/base/289116
Log:
Remove a double "as well as"
Modified:
head/share/man/man4/vtnet.4
Modified: head/share/man/man4/vtnet.4
Author: phk
Date: Tue Oct 20 12:49:38 2015
New Revision: 289627
URL: https://svnweb.freebsd.org/changeset/base/289627
Log:
Allow osreldate and osrelease to be set per jail
Modified:
head/etc/rc.d/jail
Modified: head/etc/rc.d/jail
==
Author: phk
Date: Mon Mar 21 08:30:58 2016
New Revision: 297146
URL: https://svnweb.freebsd.org/changeset/base/297146
Log:
Unquote NANO_PACKAGE_LIST so that it actually can be a list and so
that the default value of "*" works.
Modified:
head/tools/tools/nanobsd/defaults.sh
Modified: head/t
uld even trust CPU report about its frequency as the last resort.
Usually the errors will be magnificient, so a trivial sanity-check will
catch them. Don't leave home without it.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
p...@freebsd.org | TCP/IP since RFC 956
FreeBSD
;timecounter source that is emulated correctly, but you don't know which
>it is. There might be none.
Indeed: One clock or three, never two.
John R. Vig's famous tutorial is the ultimate collection of wisdom
about care and feeding of quartz:
http://www.ieee-uffc.org/frequen
Author: phk
Date: Sat Apr 9 10:53:21 2016
New Revision: 297737
URL: https://svnweb.freebsd.org/changeset/base/297737
Log:
Create /usr/local/etc before filling things into it.
Only run mtree if the NANO_CUST_FILES_MTREE is nonempty
Also umount the packages on non-signal failure
Modifi
201 - 242 of 242 matches
Mail list logo
