e a
reboot.
--apb (Alan Barrett)
in
discussion as soon as you learn that the issue was controversial.
Continuing without discussion is not appropriate.
I think that NetBSD's acronyms file should be for acronyms that
ordinary people are likely to encounter in ordinary situations;
not for acronyms used in some small subculture.
--apb (Alan Barrett)
On Fri, 24 Apr 2015, Taylor R Campbell wrote:
Modified Files:
src/share/misc: acronyms acronyms-o.real
Log Message:
Per discussion with board@, remove inappropriate, hostile acronyms.
Thank you.
--apb (Alan Barrett)
u could write a test that uses nm|awk to check that all
externally visible symbols have some desired prefix.
--apb (Alan Barrett)
distinct from %Z) already accepts "GMT"
and "UT". If we are going to accept "UTC" as well, which I think
is a good idea, then please allow both %z and %Z to accept all
three of UT, UTC, and GMT.
--apb (Alan Barrett)
file whose name
contains a space. Whether or not it's supported is separate from
whether or not the service(8) script does strange things when it
happens.
--apb (Alan Barrett)
[ -f "$_f" -a -x "$_f" ] && echo "$_f"
done
fi
--apb (Alan Barrett)
uot; is that it leaves people wondering whether
or not particular hardware is old enough to be classified as
"legacy". If the test is "Does it have an ISA bus without a PCI
bus?" then I'd suggest a name like ISA_NO_PCI.
This is another case where prior discussion wou
was made u_int to match previous art by FreeBSD. Can you please
explain why it needs compat32 handling?
Sorry, it doesn't need compat32 handling, because all existing
NetBSD platforms have 32-bit int, and all exiting NetBSD platforms
have 64-bit off_t.
--apb (Alan Barrett)
On Mon, 29 Dec 2014, Michael van Elst wrote:
Log Message:
Implement DIOCGMEDIASIZE and DIOCGSECTORSIZE from FreeBSD.
This needs compat32 handling, at least for the u_int arg to
DIOCGSECTORSIZE.
Why not make it a fixed size, like uint32_t, so compat32 handling will
not be needed?
--apb (Alan
On Mon, 17 Nov 2014, Masao Uebayashi wrote:
How about adding a new make variable CONFIGOPTS, which is passed to config(1)?
This name is similar to COPTS and works similarly too.
Yes, that could work.
--apb (Alan Barrett)
the environment or in the kernel configuration?
Was this discussed anywhere?
--apb (Alan Barrett)
r than incrementally processing dependencies as
the parser makes progress.
That sounds as though it would do what I want.
--apb (Alan Barrett)
"options FOO" in line 3 should put it back.
In the cases that I care about, the "options" and "no options"
lines will be in different files, referenced via "include"
directives.
--apb (Alan Barrett)
anded value.
Oh, you'd better fix it in NetBSD's bsd.own.mk then, which was where I took
the definition above from.
It probably works because .OBJDIR is magic, and is unlikely to need
recursive expansion.
--apb (Alan Barrett)
E} -V '${.OBJDIR}' to
get the recursively-expanded value of .OBJDIR. ${MAKE} -V .OBJDIR
would print the unexpanded value.
--apb (Alan Barrett)
ve to wait a few
days. Meanwhile, backing out the problematic revision should be
fine.
--apb (Alan Barrett)
think open has enough flags.
Yes, that would probably work, and it's easier than my ideas.
--apb (Alan Barrett)
First try open(O_EXCL), and if that fails then try
open(O_SPECIAL).
--apb (Alan Barrett)
{ if (xxx) yyy; } while (0)
a week or two ago. (That change was not pulled up to netbsd-7,
although other changes made around the same time were pulled up.)
Do we need to support any compilers that don't support __VA_ARGS__ ?
Even microsoft's compiler almost supports it.
I don't know.
--apb (Alan Barrett)
On Sun, 05 Oct 2014, Alistair Crooks wrote:
On Sun, Oct 05, 2014 at 02:13:15PM +, Alan Barrett wrote:
#ifdef PUFFSDEBUG
extern int puffsdebug; /* puffs_subr.c */
-#define DPRINTF(x) if (puffsdebug > 0) printf x
-#define DPRINTF_VERBOSE(x) if (puffsdebug > 1) printf x
+#define DPR
-modern toolchain I can think of without C99 support is MSVC and we
don't support that anyway.
Perhaps it does make sense to allow C99 in tools, but I think we
have tried to avoid it until now.
--apb (Alan Barrett)
.)
Can you say that again, please?
I think he means: Please add links from
<http://www.netbsd.org/developers/releng/pullups.html>
to <https://releng.netbsd.org/index-7.html>,
<https://releng.netbsd.org/index-6.html>, and
<https://releng.netbsd.org/index-5.html>.
--apb (Alan Barrett)
$ in the
shell escaping case, which I guess is what you meant by the "don't
encode this" list.
--apb (Alan Barrett)
correctly handle it.
unvis is not intended to reverse shell-style escapes. You can use
the shell's eval command for that.
--apb (Alan Barrett)
lready there
before the user adds their own restricted list of interface names.
--apb (Alan Barrett)
rtsol from the section
about things that may be removed in the future. Please also add
rtsol to the section about things that have been removed.
--apb (Alan Barrett)
lt to pull
up the "ifconfig -w" change and the rc.d/network changes to other
branches.
--apb (Alan Barrett)
to do
with changing defaults. Also, it probably needs a man page update.
--apb (Alan Barrett)
.
INITIALIZE(ktime);
+ ktime = 0; /* XXX: gcc */
The #define GNUC_or_lint and #ifdef GNUC_or_lint in time/private.h
should have made it so INITIALIZE(ktime) was enough.
--apb (Alan Barrett)
P_INACTIVE().
- Make ptyfs_used_get() private to ptyfs_subr.c
- Stop copying device attributes from traditional ptys on first allocation.
- Remove unneeded argument "lwp" from ptyfs_allocvp() and ptyfs_free_get().
Please could you update the mount_ptyfs(8) man page to explain the
new beh
mark them as "obsolete"
instead,
} #unless otherwise stated below.
^
Heh, thanks for the reminder. I'll fix it.
--apb (Alan Barrett)
non-zero inputs) but to round down (and to return 0 for
very small inputs) when hz > 100. Once the desired behaviour is
clarified, you could implement it by making adjustments to the
value before or after calling tvtohz.
--apb (Alan Barrett)
ement these ioctls in any compat code.
--apb (Alan Barrett)
if this doesn't get reverted for being rushed, then
please import it on a vendor branch as per the usual practice.
Did you mean moving it from -current to a specific branch?
No, I mean using "cvs import" the way we usually do it. I'll explain
in more detail in a private message.
--apb (Alan Barrett)
his doesn't get reverted for being rushed, then please
import it on a vendor branch as per the usual practice.
--apb (Alan Barrett)
de)
to produce different output streams from the same initial seed or
state.
--apb (Alan Barrett)
r new uses, please just write __RCSID(...);
without any #if/#endif wrapper.
--apb (Alan Barrett)
- net.inet6.multicast_kludge for temporarily removed multicast groups
Please document these in sysctl(7).
--apb (Alan Barrett)
On Sat, 24 May 2014, Masao Uebayashi wrote:
Log Message:
Unbreak syspkg by escaping '[' by vis(1) to match the new mtree(8) format.
I'd prefer to escape using TOOL_SED so we don't need to add a new
TOOL_VIS.
--apb (Alan Barrett)
On Sat, 17 May 2014, Christos Zoulas wrote:
Add tzdata2netbsd, a script to help import new versions of tzdata.
Why all the || exit $? instead of set -e
set -e would probably be better.
--apb (Alan Barrett)
o
ask for restricted namespaces...
src/tools/compat/compat_defs.h goes to a lot of trouble to ensure
that _NETBSD_SOURCE does not get defined. I don't understand why
that is done.
--apb (Alan Barrett)
and 7.21.2.1 "The memcpy function" does not give any explicit
permission for use of null pointers.
I don't object if the implementation wants to allow null pointers
with zero size as an extension, but it should be clear that this
is non-standard.
--apb (Alan Barrett)
time, which seems to match
the description in hier(7) perfectly.
--apb (Alan Barrett)
y */
# endif
# if defined(__HEADER_NAME_TOKEN)
# define __tmp_z2 __HEADER_NAME_TOKEN
# else
# define __tmp_z2 /* Empty */
# endif
# define __CTASSERT(x) _CTASSERT0(x, __ctassert, __tmp_z1, __tmp_z2, __LINE__)
# undef __tmp_z1
# undef __tmp_z2
#endif
--apb (Alan Barrett)
scussed
that meets all reasonable requirements for compatibility.
--apb (Alan Barrett)
actual size.
--apb (Alan Barrett)
than deleting them.
--apb (Alan Barrett)
) CTL_CREATE idiom?
--apb (Alan Barrett)
instead of "find . type
f -print", but I'd like the options to be compatible with other
operating systems.
--apb (Alan Barrett)
replacements), and
what are the replacements?
I see no mention of deprecation in the man page for a version of
atf-run from November 2013.
--apb (Alan Barrett)
On Wed, 29 Jan 2014, Matt Thomas wrote:
Modified Files:
src/sys/arch/arm/include: int_fmtio.h int_mwgwtypes.h
Log Message:
Make {,u}int{8,16,32} be of type int.
I think you mean:
Make {,u}int_fast{8,16,32} be of type int.
--apb (Alan Barrett)
ain.
Will this make malloc fail 0.5GB earlier than before? The data
size limits on i386 are already annoyingly small, and I would
prefer not to make them smaller. Please could you find a way to
penalise only a.out programs.
--apb (Alan Barrett)
o not see them.
--apb (Alan Barrett)
then please adjust bsd.README to say so.
Currently, they are documented as being independent (but there
seem to be some accidental yes/no inversions in the documentation).
--apb (Alan Barrett)
gus extra link (if it exists)
rm -f /dirname/destination/destination
"ln -sfh" is not portable, so we can't use that.
--apb (Alan Barrett)
#ifndef __dead
#define __dead /* nothing */
#endif
--apb (Alan Barrett)
On Mon, 06 Jan 2014, Alan Barrett wrote:
On Mon, 06 Jan 2014, Erik Fair wrote:
Unless I misunderstand NTP configuration semantics, your additional
"restrict" statements for the NTP pool names will do the wrong
thing, in that each reference to a given netbsd.pool.ntp.org name
return
at machines that
are syncing from them?
Yes, that's polite. It is suggested by the NTP peering
guidelines (and I know "server" and "peer" are different).
I am sure I have heard this advice in the past, but I can't find a
reference to it.
--apb (Alan Barrett)
t works. Perhaps a lot of the variable
definitions from bsd.sys.mk can be moved to bsd.own.mk, or a file
that's included by bsd.own.mk.
Also, if you think that Makefiles should not include bsd.sys.mk,
then please edit share/mk/bsd.README to document how things should
be done.
--apb (Alan Barrett)
f you have "restrict default nopeer noquery" (the uncommented
line in my commit), then time service will still work, but the
configured servers will be denied query permission.
If you use "restrict default ignore", then time service does not
work.
--apb (Alan Barrett)
also usable by the kernel.
* change lua_Number to int64_t for the kernel Lua implementation.
I would have preferred to see two separate commits. Even with only
one commit, the log message should have described both changes.
--apb (Alan Barrett)
_list.tqe_next;
... could be rewritten to use TAILQ_FOREACH.
--apb (Alan Barrett)
to do that without booting a different kernel for
every test.
- rmind has said he'll address this eventually,
OK.
--apb (Alan Barrett)
Message:
implement AT_RANDOM.
random() returns a predictable value based on a seed. Given the
supposed use of AT_RANDOM for security, I think that it would be more
appropriate to use a stronger source, such as cprng_strong32().
--apb (Alan Barrett)
On Thu, 31 Oct 2013, Marc Balmer wrote:
Modified Files:
src/sys/modules/lua: Makefile
Log Message:
fix build on arm
Please describe the actual change, not just the reason for it.
Fopr example:
Move "-include ${.CURDIR}/luaconf.h" from CPPFLAGS to CFLAGS.
--apb (Alan Barrett)
On Mon, 28 Oct 2013, Marc Balmer wrote:
Modified Files:
src/distrib/sets/lists/modules: mi
src/sys/modules: Makefile
Log Message:
linke pmf(9l) to the build
Messages like this would be more useful if they said "9lua" instyead
of "9l".
--apb (Alan Barrett)
tried to collect some of these to attribute(3), which might be worth
updating (or not).
Thank you! I didn't know about attribute(3).
--apb (Alan Barrett)
cern was with other compilers in the future. If the pointer
access is volatile, then I think it's fine.
--apb (Alan Barrett)
ink that the change from the use of volatile to the use of
__USE() is a change from reliance on the C standard's guarantee
that the memory location behind (cpi)->intreg_4m->pi_pend really
will be accessed, to a reliance on what a particular compiler
happens to do in this situation.
--apb (Alan Barrett)
. Please don't steamroll over public discussions like
this.
Just for the record, yes, I got core's approval.
Could you tell us approximately when you got that approval? It
wasn't in the past two years while I have been a member of core.
--apb (Alan Barrett)
oject
instead of the NetBSD project then I might care about that.
--apb (Alan Barrett)
On Fri, 18 Oct 2013, Christos Zoulas wrote:
The result from strlen() has type size_t, so print it with "%zd" format.
%zu is correct.
Fixed, thanks.
ch a case, "yes" means accept the result silently,
while "ask" means ask the user (the first time). If the DNS
answer was not secured by DNSSEC, then both "yes" and "ask" end up
asking the user.
By the way, I think that's a bug in ssh that the Host patterns are
case sensitive.
--apb (Alan Barrett)
, or the fact that it corresponds to this
import, will not be obvious to future users of "cvs log".
If the ChangeLog is small, then pasting it into the commit log is
useful. If the ChangeLog is large, then please extract only the
highlights.
--apb (Alan Barrett)
On Mon, 02 Sep 2013, Simon J. Gerraty wrote:
Modified Files:
src/usr.bin/make: compat.c
Log Message:
Do not apply shellErrFlag unless errCheck is true.
To generate a diff of this commit:
cvs rdiff -u -r1.92 -r1.93 src/usr.bin/make/compat.c
Will this fix PR 45356?
--apb (Alan
s? I'd be inclined to say
"Coverity CID " instead of just "CID " in these log
messages.
--apb (Alan Barrett)
time_t shall be an integer type. [Option End]"
(The "[CX][Option Start]...[Option End]" just means "Extension to he
ISO C standard".)
The tzcode implementation goes to some trouble to be portable to
environments with integer or floating point types for time_t.
--apb (Alan Barrett)
at code trying to do?
If time_t is a floating point type, then round to the nearest
second. If time_t is an integral type, then do nothing. There
really should be a comment to explain that.
--apb (Alan Barrett)
lt; _y ? _x : _y); })
--apb (Alan Barrett)
= {.membername = 0};
where membername is any valid member of the struct.
I wish gcc -Wmissing-field-initializers would recognise that {0} is
a special case that does nto deserve a warning, so we could write the
natural and well-defined code
typename varname = {0};
--apb (Alan Barrett)
both have 64-bit long long,
so it doesn't affect them, but other pairs of related platforms
(perhaps in the future) might be different.
--apb (Alan Barrett)
--apb (Alan Barrett)
hostname.
-I Display each IP address on all interfaces.
-i Display the IP address(es) for the hostname.
Not to go all Rob Pike on you (cf. "cat -v considered harmful"),
but what the heck is all this for?
And also: Where was this change discussed?
--apb (Alan Barrett)
On Sun, 07 Jul 2013, Erik Fair wrote:
Pullup request to netbsd-6 and netbsd-5?
Yes, I always request such pullups. I like to wait a day or two, but
it's occasionally taken me much longer than that.
--apb (Alan Barrett)
On Mon, 24 Jun 2013, Taylor R Campbell wrote:
Discuss on tech-userlevel?
Yes please.
--apb (Alan Barrett)
r do they
compare for sorting?" (0 for equal).
By the way, I have an implementation of consttime_memcmp()
at <ftp://ftp.netbsd.org/pub/NetBSD/misc/apb/consttime_memcmp.c>.
--apb (Alan Barrett)
sions of
some useful types without polluting the namespace, and then use
__intptr_t.
--apb (Alan Barrett)
doc equivalent for the
${TOOL_GROFF} command there, feel free to edit the Makefile.
--apb (Alan Barrett)
On Sat, 11 May 2013, Blue Rats wrote:
.Fn iconv_close ,
and
.Fn iconv
-conform to
+conforms to
.St -p1003.1-2001 .
"conform" was correct, and "conforms" is wrong. Please reinstate the
correct wording.
--apb (Alan Barrett)
ation phase. I am not sure why it's not
good enough for host-mkdep, but two ideas spring to mind: perhaps
host-mkdep is run before the dummy include file is created, or
perhaps host-mkdep is run without the -I flags that would allow it
to find the dummy include file.
--apb (Alan Barrett)
the underlying problem without adding a new special target,
such as using the newish "meta mode". Also, a better name than
".STALE" might have been suggested.
--apb (Alan Barrett)
On Sun, 03 Mar 2013, Alan Barrett wrote:
Module Name:src
Committed By: apb
Date: Sun Mar 3 20:07:09 UTC 2013
Modified Files:
src/distrib/sets/lists/base: mi
Log Message:
/tmp/cvs12835b
I have updated the log message to say:
Add zoneinfo files for new timezones from
about characters,
that's because it was written at a time when vis did not attempt
to deal with multibyte characters. I think that we should revert
to the original semantics of lengths being measured in bytes, and
adjust both the man pages and callers appropriately.
--apb (Alan Barrett)
I and II, 2 for
UltraSPARC-III and up
Does this need to use #define CPU_VIS, or could it use a new-style
dynamic MIB entry, passing CTL_CREATE to sysctl_createv(9)?
--apb (Alan Barrett)
ad of bar".
--apb (Alan Barrett)
;s a well-defined kind of wraparound, not the undefined kind
of overflow. I'd expect an error like EINVAL rather than an error like
ENOMEM for this case.
--apb (Alan Barrett)
tappit-hen as an example.
Good idea.
--apb (Alan Barrett)
ile-name=crti.o
_GCC_CRTN!= ${CC} --print-file-name=crtn.o
.for v in _GCC_CRTBEGINS _GCC_CRTENDS _GCC_CRTI _GCC_CRTN
# If the value does not contain '/' then replace it with the empty string
${v} := ${${v}:C,^[^/]*$,,}
.endfor
--apb (Alan Barrett)
the implementation si still incomplete, then exposing it when
_NETBSD_SOURCE is defined seems wrong. I don't see any third scenario
in which it makes sense to test defined(_NETBSD_SOURCE) ||
defined(_INCOMPLETE_
--apb (Alan Barrett)
From: Alan Barrett
To: source-changes-d@NetBSD.org
Cc:
1 - 100 of 237 matches
Mail list logo
