Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Tue, Dec 12, 2017 at 10:05 AM, Rodney W. Grimes < free...@pdx.rh.cn85.dnsmgr.net> wrote: > Please stop dismissing concerns, especially if raised by more > than one person, as whining, that is not helpful. > Except in this case it was totally useless. Totally. It was an emotional reaction devoid of actual technical merit. Therefore, it's whining. Warner ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
> On Tue, 12 Dec 2017 07:40:21 -0800 (PST) > "Rodney W. Grimes"wrote: > > > > On Tue, 12 Dec 2017 06:18:36 + > > > Alexey Dokuchaev wrote: > > > > > > > On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: > > > > > I think this name might confuse people looking for "syscons". Can it > > > > > be > > > > > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > > > > > > > > +1 for rename, it's very confusing now. > > > > ++1 rename it or expect irrate users > > So where exactly in the confusion ? > > Is it because your tab completion when doing cd sys/dev/sysc will > now ask you to type more char ? > This is the only place where "syscons" (not syscon) exists. My appologies, I did not notice that there infact is a delta of syscon vs syscons. I thought that the name was an exact match. I still not real happy with it, but my objection is retracted. > > > > > > > > ./danfe > > > > > > It's not confusing. The spec is clear it's named "syscon" so our > > > driver must be named syscon, naming it otherwise would be confusing. > > > What's confusing is the directory name for "sc" named syscons, it's > > > config options are named SC_* and the device is named sc. > > > > The spec is wrong for dictating what any implementation must > > call this, the spec does not own the implementation name space. > > > > It is called called syscons and uses the variables and driver > > names sc beacuse that is what we did back in the 90's, drivers > > typically had 2 character names. > > > > > The only valuable remark in this thread is bde's one saying it's not > > > documented etc ... and yes we do sucks in this area in the arm world. > > > > Thank you for your dismissal of input from the community, does that > > mean your gong to ignore them? > > > > > Anyhow, I caught a discussion between mmel@ and kevans@ on IRC saying > > > that we will move the driver into extres after adding some standard > > > stuff common to extres framework. > > > > FYI why is this so bad an idea to call it syscons: > > man syscons > > How do you plan to resolve that name space conflict? > > AND this man page has existed since 1.0, So 25 years of > > finger memory, your just not gona do well trying to change that. > > The new driver name is "syscon" not "syscons", there is no name space > conflict. Again, I am sorry, I did not notice this and thought we had a direct conflict. > > Again, we will move it to sys/dev/extres, now can we stop whining > about small details like that and do some actual work please ? I am glad it is moving, but that was never the reason that someone raised issue. Please stop dismissing concerns, especially if raised by more than one person, as whining, that is not helpful. -- Rod Grimes rgri...@freebsd.org ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Tue, 2017-12-12 at 07:40 -0800, Rodney W. Grimes wrote: > > > > On Tue, 12 Dec 2017 06:18:36 + > > Alexey Dokuchaevwrote: > > > > > > > > On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: > > > > > > > > I think this name might confuse people looking for "syscons". Can it be > > > > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > > > +1 for rename, it's very confusing now. > ++1 rename it or expect irrate users > > > > > > > > > > > > ./danfe > > It's not confusing. The spec is clear it's named "syscon" so our > > driver must be named syscon, naming it otherwise would be confusing. > > What's confusing is the directory name for "sc" named syscons, it's > > config options are named SC_* and the device is named sc. > The spec is wrong for dictating what any implementation must > call this, the spec does not own the implementation name space. > > It is called called syscons and uses the variables and driver > names sc beacuse that is what we did back in the 90's, drivers > typically had 2 character names. > > > > > The only valuable remark in this thread is bde's one saying it's not > > documented etc ... and yes we do sucks in this area in the arm world. > Thank you for your dismissal of input from the community, does that > mean your gong to ignore them? > > > > > Anyhow, I caught a discussion between mmel@ and kevans@ on IRC saying > > that we will move the driver into extres after adding some standard > > stuff common to extres framework. > FYI why is this so bad an idea to call it syscons: > man syscons > How do you plan to resolve that name space conflict? > AND this man page has existed since 1.0, So 25 years of > finger memory, your just not gona do well trying to change that. > > So a driver named "syscon" that implements the FDT device named "syscon" is somehow a bad idea because it differs from an existing driver directory name by only one character, but you're not bothered by alc/ale, or bce/bfe/bge, or iwi/iwn, or mlx/mly, or any other of the literally dozens of existing drivers whose directory names differ by only 1 character. In addition you defend naming options related to "syscons" as "sc" because 25 years ago, "drivers typically have 2-character names", and I guess that fact magically connects "device sc" to "dev/syscons" in peoples' minds, because... phase of the moon, or what??? This whole discussion feels like it was scripted by Salvador Dali and produced by David Lynch. -- Ian ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Tue, Dec 12, 2017 at 8:58 AM, Emmanuel Vadotwrote: > Again, we will move it to sys/dev/extres, now can we stop whining > about small details like that and do some actual work please ? while true; do echo +1 ; done You have no idea how demotivating chicken !$!% complaining about a name that's perfectly fine can be. Unless it causes and actual problem, don't complain. Things are named similar to other things all the time and we somehow cope when we discover they aren't exactly the same thing. This whole conversation is the very definition of a bike shed: criticism about one minor detail on trumped up reasons by people that did none of the work. Warner ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Tue, 12 Dec 2017 07:40:21 -0800 (PST) "Rodney W. Grimes"wrote: > > On Tue, 12 Dec 2017 06:18:36 + > > Alexey Dokuchaev wrote: > > > > > On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: > > > > I think this name might confuse people looking for "syscons". Can it be > > > > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > > > > > > +1 for rename, it's very confusing now. > > ++1 rename it or expect irrate users So where exactly in the confusion ? Is it because your tab completion when doing cd sys/dev/sysc will now ask you to type more char ? This is the only place where "syscons" (not syscon) exists. > > > > > > ./danfe > > > > It's not confusing. The spec is clear it's named "syscon" so our > > driver must be named syscon, naming it otherwise would be confusing. > > What's confusing is the directory name for "sc" named syscons, it's > > config options are named SC_* and the device is named sc. > > The spec is wrong for dictating what any implementation must > call this, the spec does not own the implementation name space. > > It is called called syscons and uses the variables and driver > names sc beacuse that is what we did back in the 90's, drivers > typically had 2 character names. > > > The only valuable remark in this thread is bde's one saying it's not > > documented etc ... and yes we do sucks in this area in the arm world. > > Thank you for your dismissal of input from the community, does that > mean your gong to ignore them? > > > Anyhow, I caught a discussion between mmel@ and kevans@ on IRC saying > > that we will move the driver into extres after adding some standard > > stuff common to extres framework. > > FYI why is this so bad an idea to call it syscons: > man syscons > How do you plan to resolve that name space conflict? > AND this man page has existed since 1.0, So 25 years of > finger memory, your just not gona do well trying to change that. The new driver name is "syscon" not "syscons", there is no name space conflict. Again, we will move it to sys/dev/extres, now can we stop whining about small details like that and do some actual work please ? -- Emmanuel Vadot ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
> On Tue, 12 Dec 2017 06:18:36 + > Alexey Dokuchaevwrote: > > > On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: > > > I think this name might confuse people looking for "syscons". Can it be > > > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > > > > +1 for rename, it's very confusing now. ++1 rename it or expect irrate users > > > > ./danfe > > It's not confusing. The spec is clear it's named "syscon" so our > driver must be named syscon, naming it otherwise would be confusing. > What's confusing is the directory name for "sc" named syscons, it's > config options are named SC_* and the device is named sc. The spec is wrong for dictating what any implementation must call this, the spec does not own the implementation name space. It is called called syscons and uses the variables and driver names sc beacuse that is what we did back in the 90's, drivers typically had 2 character names. > The only valuable remark in this thread is bde's one saying it's not > documented etc ... and yes we do sucks in this area in the arm world. Thank you for your dismissal of input from the community, does that mean your gong to ignore them? > Anyhow, I caught a discussion between mmel@ and kevans@ on IRC saying > that we will move the driver into extres after adding some standard > stuff common to extres framework. FYI why is this so bad an idea to call it syscons: man syscons How do you plan to resolve that name space conflict? AND this man page has existed since 1.0, So 25 years of finger memory, your just not gona do well trying to change that. -- Rod Grimes rgri...@freebsd.org ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Tue, 12 Dec 2017 06:18:36 + Alexey Dokuchaevwrote: > On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: > > I think this name might confuse people looking for "syscons". Can it be > > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > > +1 for rename, it's very confusing now. > > ./danfe It's not confusing. The spec is clear it's named "syscon" so our driver must be named syscon, naming it otherwise would be confusing. What's confusing is the directory name for "sc" named syscons, it's config options are named SC_* and the device is named sc. The only valuable remark in this thread is bde's one saying it's not documented etc ... and yes we do sucks in this area in the arm world. Anyhow, I caught a discussion between mmel@ and kevans@ on IRC saying that we will move the driver into extres after adding some standard stuff common to extres framework. -- Emmanuel Vadot ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
Bezüglich Alexey Dokuchaev's Nachricht vom 12.12.2017 07:18 (localtime): > On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: >> I think this name might confuse people looking for "syscons". Can it be >> renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > +1 for rename, it's very confusing now. Not that mine would have much weight, but also +1 from here. Stumbled at any commit log yet – mmdevcon corecon[f] chpcon[f] soccon[f] – just to throw in seomthing… -harry ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Mon, Dec 11, 2017 at 10:14:04AM -0800, Nathan Whitehorn wrote: > I think this name might confuse people looking for "syscons". Can it be > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? +1 for rename, it's very confusing now. ./danfe ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Mon, 11 Dec 2017, Nathan Whitehorn wrote: I think this name might confuse people looking for "syscons". Can it be renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? The wouldn't find it since it is not in any man page, GENERIC, or even in NOTES. It is not as bad as some drivers like drm2 that are also not in conf/files* so they are not puttable in GENERIC or NOTES or even in the user's config file. [.. Context lost to top posting] Bruce ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
On Mon, 2017-12-11 at 10:14 -0800, Nathan Whitehorn wrote: > I think this name might confuse people looking for "syscons". Can it > be > renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? > -Nathan > > On 12/11/17 10:04, Kyle Evans wrote: > > > > Author: kevans > > Date: Mon Dec 11 18:04:04 2017 > > New Revision: 326773 > > URL: https://svnweb.freebsd.org/changeset/base/326773 > > > > Log: > > Add generic 'syscon' driver > > It's definitely not arm-specific. It's the modern linux/fdt way of letting drivers access MMIO spaces they don't own (in terms of those ranges being included in a reg property that would cause us to set up rman resources for the device). I'm not sure I would have made a new directory in dev/ for it though... it's such a trivial little thing I would think it belongs with the dev/extres stuff. -- Ian ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r326773 - in head/sys: conf dev/syscon
I think this name might confuse people looking for "syscons". Can it be renamed? Also, if it is ARM-specific, maybe it belongs in sys/arm? -Nathan On 12/11/17 10:04, Kyle Evans wrote: Author: kevans Date: Mon Dec 11 18:04:04 2017 New Revision: 326773 URL: https://svnweb.freebsd.org/changeset/base/326773 Log: Add generic 'syscon' driver Upstream dts for allwinner will require a syscon driver, since the emac node coming in 4.15 will be using xref to /soc/syscon for configuring the emac clock. Add a generic syscon driver to attach to /soc/syscon for use by if_awg, providing basic read/write functionality to consumers. syscon driver will also be used by arm64 at least for A64+H5 emac/if_awg. Written by: mmel Reviewed by: manu Differential Revision: https://reviews.freebsd.org/D13295 Added: head/sys/dev/syscon/ head/sys/dev/syscon/syscon.c (contents, props changed) head/sys/dev/syscon/syscon_if.m (contents, props changed) Modified: head/sys/conf/files Modified: head/sys/conf/files == --- head/sys/conf/files Mon Dec 11 16:18:05 2017(r326772) +++ head/sys/conf/files Mon Dec 11 18:04:04 2017(r326773) @@ -3109,6 +3109,8 @@ dev/stg/tmc18c30_subr.c optional stg dev/stge/if_stge.coptional stge dev/sym/sym_hipd.coptional sym\ dependency "$S/dev/sym/sym_{conf,defs}.h" +dev/syscon/syscon.coptional fdt syscon +dev/syscon/syscon_if.m optional fdt syscon dev/syscons/blank/blank_saver.c optional blank_saver dev/syscons/daemon/daemon_saver.c optional daemon_saver dev/syscons/dragon/dragon_saver.c optional dragon_saver Added: head/sys/dev/syscon/syscon.c == --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ head/sys/dev/syscon/syscon.cMon Dec 11 18:04:04 2017 (r326773) @@ -0,0 +1,185 @@ +/*- + * Copyright (c) 2015 Michal Meloun + * 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 above copyright + *notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + *notice, this list of conditions and the following disclaimer in the + *documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +/* + * This is a generic syscon driver, whose purpose is to provide access to + * various unrelated bits packed in a single register space. It is usually used + * as a fallback to more specific driver, but works well enough for simple + * access. + */ + +#include +__FBSDID("$FreeBSD$"); +#include +#include +#include +#include +#include +#include + +#include + +#include +#include + +#include "syscon_if.h" + +#define SYSCON_LOCK(_sc) mtx_lock(&(_sc)->mtx) +#defineSYSCON_UNLOCK(_sc) mtx_unlock(&(_sc)->mtx) +#define SYSCON_LOCK_INIT(_sc) mtx_init(&(_sc)->mtx,\ + device_get_nameunit((_sc)->dev), "syscon", MTX_DEF) +#define SYSCON_LOCK_DESTROY(_sc) mtx_destroy(&(_sc)->mtx); +#define SYSCON_ASSERT_LOCKED(_sc) mtx_assert(&(_sc)->mtx, MA_OWNED); +#define SYSCON_ASSERT_UNLOCKED(_sc)mtx_assert(&(_sc)->mtx, MA_NOTOWNED); + +struct syscon_softc { + device_tdev; + struct resource *mem_res; + struct mtx mtx; +}; + +static struct ofw_compat_data compat_data[] = { + {"syscon",1}, + {NULL, 0} +}; + +static uint32_t +syscon_read_4(device_t dev, device_t consumer, bus_size_t offset) +{ + struct syscon_softc *sc; + uint32_t val; + + sc = device_get_softc(dev); + + SYSCON_LOCK(sc); + val = bus_read_4(sc->mem_res, offset); + SYSCON_UNLOCK(sc); + return (val); +} + +static void +syscon_write_4(device_t dev, device_t