remove kern.arand from sysctl.8 and rnd.c (comment)

[Rob Pierce]

Index: sysctl.8
===
RCS file: /cvs/src/sbin/sysctl/sysctl.8,v
retrieving revision 1.205
diff -u -p -r1.205 sysctl.8
--- sysctl.87 Sep 2016 17:30:12 -   1.205
+++ sysctl.823 Sep 2016 02:16:14 -
@@ -144,7 +144,6 @@ and a few require a kernel compiled with
 .It kern.sysvmsg Ta integer Ta no
 .It kern.sysvsem Ta integer Ta no
 .It kern.sysvshm Ta integer Ta no
-.It kern.arandom Ta u_int Ta no
 .It kern.msgbufsize Ta integer Ta no
 .It kern.malloc.buckets Ta string Ta no
 .It kern.malloc.bucket. Ta string Ta no

Index: rnd.c
===
RCS file: /cvs/src/sys/dev/rnd.c,v
retrieving revision 1.186
diff -u -p -r1.186 rnd.c
--- rnd.c   22 Sep 2016 22:04:02 -  1.186
+++ rnd.c   23 Sep 2016 02:25:06 -
@@ -85,9 +85,8 @@
  * The output of this single ChaCha20 engine is then shared amongst many
  * consumers in the kernel and userland via a few interfaces:
  * arc4random_buf(), arc4random(), arc4random_uniform(), randomread()
- * for the set of /dev/random nodes, the sysctl kern.arandom, and the
- * system call getentropy(), which provides seeds for process-context
- * pseudorandom generators.
+ * for the set of /dev/random nodes, and the system call getentropy(),
+ * which provides seeds for process-context pseudorandom generators.
  *
  * Acknowledgements:
  * =

Two more arm aeabi functions for the kernel

[Mark Kettenis]

These functions are necessary to build our kernel with clang.  I
picked these from NetBSD as I didn't quite like the way FreeBSD
implemented them.  I stripped all the unecessary bits from the NetBSD
implementation (which is shared with userland).

Do the file names with the leading underscores bother people?  I could
rename the files ldivmod.S and uldivmod.S.

These seem to be the last two missing symbols.

ok?


Index: lib/libkern/arch/arm/__aeabi_ldivmod.S
===
RCS file: lib/libkern/arch/arm/__aeabi_ldivmod.S
diff -N lib/libkern/arch/arm/__aeabi_ldivmod.S
--- /dev/null   1 Jan 1970 00:00:00 -
+++ lib/libkern/arch/arm/__aeabi_ldivmod.S  22 Sep 2016 20:03:42 -
@@ -0,0 +1,133 @@
+/*-
+ * Copyright (c) 2012 The NetBSD Foundation, Inc.
+ * All rights reserved.
+ *
+ * This code is derived from software contributed to The NetBSD Foundation
+ * by Matt Thomas of 3am Software Foundry.
+ *
+ * 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 NETBSD FOUNDATION, INC. 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 FOUNDATION 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.
+ */
+
+#include 
+
+#ifdef __ARMEB__
+#defineALO r1  /* incoming numerator, outgoing quotient */
+#defineAHI r0  /* incoming numerator, outgoing quotient */
+#defineBLO r3  /* incoming denominator, outgoing remainder */
+#defineBHI r2  /* incoming denominator, outgoing remainder */
+#else
+#defineALO r0  /* incoming numerator, outgoing quotient */
+#defineAHI r1  /* incoming numerator, outgoing quotient */
+#defineBLO r2  /* incoming denominator, outgoing remainder */
+#defineBHI r3  /* incoming denominator, outgoing remainder */
+#endif
+
+ENTRY(__aeabi_ldivmod)
+   push{r4-r6, lr}
+#defineNEG r5
+   movsNEG, #0
+
+   cmp BHI, #0
+   bge 2f
+   movsNEG, #1 /* flip quotient sign */
+   bl  .Lnegate_b
+   bcs .Lmaxdenom
+
+2:
+   cmp AHI, #0
+   eorlt   NEG, NEG, #3/* flip quotient sign, flip remainder sign */
+   bllt.Lnegate_a
+
+   /*
+* Arguments are setup, allocate some stack for the remainder
+* and call __qdivrem for the heavy lifting.
+*/
+   sub sp, sp, #16
+   addsr4, sp, #8
+   str r4, [sp]
+   bl  __qdivrem
+   add sp, sp, #8
+
+   /*
+* The quotient is already in the right place and neither value
+* needs its sign flipped.
+*/
+   cmp NEG, #0 /* any signs to flip? */
+   beq .Lnegate_neither
+
+   cmp NEG, #2 /* does remainder need to be negative? */
+   beq .Lnegate_b_only /* 2 means b only */
+   bgt .Lnegate_both   /* 3 means both */
+.Lnegate_a_only:
+   bl  .Lnegate_a  /* 1 means a only */
+.Lnegate_neither:
+   pop {r2-r6, pc} /* grab b from stack */
+.Lnegate_both:
+   bl  .Lnegate_a
+.Lnegate_b_only:
+   pop {r2-r3} /* get remainder */
+   bl  .Lnegate_b  /* negate it */
+   pop {r4-r6, pc}
+
+   .align  0
+.Lnegate_a:
+   negsALO, ALO
+   rsc AHI, AHI, #0
+   mov pc, lr
+
+   .align  0
+.Lnegate_b:
+   negsBLO, BLO
+   rsc BHI, BHI, #0
+   mov pc, lr
+
+   .align  0
+.Lmaxdenom:
+   /*
+* We had a carry so the denominator must have INT64_MIN
+* Also BLO and BHI never changed values so we can use
+* them to see if the numerator has the same value.  We
+* don't have to worry about sign.
+*/
+   cmp BHI, AHI
+   cmpeq   BLO, ALO
+   bne 1f
+
+   /*
+* They were 

Re: aeabi stuff in libkern

[Philip Guenther]

On Thu, Sep 22, 2016 at 8:10 AM, Mark Kettenis  wrote:
> clang generates these calls using the official aeabi names instead of
> the gcc names.  Just add an alias for them.  FreeBSD does something
> similar.
>
> ok?

ok guenther@

Migrated Link in faq6.html

[wuy13]

The link to "run netstart" is incorrect. Corrected link to the man page.

Index: faq6.html
===
RCS file: /cvs/www/faq/faq6.html,v
retrieving revision 1.416
diff -u -p -r1.416 faq6.html
--- faq6.html   18 Sep 2016 11:22:37 -  1.416
+++ faq6.html   22 Sep 2016 12:33:03 -
@@ -325,7 +325,7 @@ inet alias 192.168.0.3 255.255.255.255
 inet alias 192.168.0.4 255.255.255.255
 

-Once you've created this file, run 
netstart or
+Once you've created this file, href="http://man.openbsd.org/netstart;>run netstart or

 reboot.
 To view all aliases, use $ ifconfig -A

aeabi stuff in libkern

[Mark Kettenis]

clang generates these calls using the official aeabi names instead of
the gcc names.  Just add an alias for them.  FreeBSD does something
similar.

ok?


Index: lib/libkern/arch/arm/divsi3.S
===
RCS file: /cvs/src/sys/lib/libkern/arch/arm/divsi3.S,v
retrieving revision 1.3
diff -u -p -r1.3 divsi3.S
--- lib/libkern/arch/arm/divsi3.S   30 Dec 2014 08:12:52 -  1.3
+++ lib/libkern/arch/arm/divsi3.S   22 Sep 2016 15:07:07 -
@@ -385,3 +385,6 @@ L_udivide_l1:
addhs   r3, r3, r2
mov r0, r3
mov pc, lr
+
+STRONG_ALIAS(__aeabi_idiv, __divsi3)
+STRONG_ALIAS(__aeabi_uidiv, __udivsi3)

Re: Check terminal width in ps(1)

[Theo Buehler]

On Thu, Sep 22, 2016 at 04:23:08AM -0600, Anthony J. Bentley wrote:
> Hi,
> 
> Unlike all the other software in the tree that checks terminal sizes, 
> ps(1) used to check the stdin and stderr terminal sizes as well as stdout.
> When I unified the terminal size checks in the tree in March, I missed
> this, and now "ps aux | blah" gets wrapped to 79 characters no matter
> how big the terminal is.
> 
> The below diff brings back the old behavior: if the output is a pipe,
> ps will check if either stderr or stdin is a terminal and if so will use
> its width.
> 
> ok?

ok.

It's nicer this way than the negated checks that were there in r1.69.

> 
> Index: ps.c
> ===
> RCS file: /cvs/src/bin/ps/ps.c,v
> retrieving revision 1.70
> diff -u -p -r1.70 ps.c
> --- ps.c  17 Mar 2016 05:27:10 -  1.70
> +++ ps.c  22 Sep 2016 10:14:10 -
> @@ -105,7 +105,10 @@ main(int argc, char *argv[])
>   termwidth = 0;
>   if ((cols = getenv("COLUMNS")) != NULL)
>   termwidth = strtonum(cols, 1, INT_MAX, NULL);
> - if (termwidth == 0 && ioctl(STDOUT_FILENO, TIOCGWINSZ, ) == 0 &&
> + if (termwidth == 0 &&
> + (ioctl(STDOUT_FILENO, TIOCGWINSZ, ) == 0 ||
> + ioctl(STDERR_FILENO, TIOCGWINSZ, ) == 0 ||
> + ioctl(STDIN_FILENO, TIOCGWINSZ, ) == 0) &&
>   ws.ws_col > 0)
>   termwidth = ws.ws_col - 1;
>   if (termwidth == 0)
> Index: ps.1
> ===
> RCS file: /cvs/src/bin/ps/ps.1,v
> retrieving revision 1.108
> diff -u -p -r1.108 ps.1
> --- ps.1  8 Sep 2016 15:54:36 -   1.108
> +++ ps.1  22 Sep 2016 10:14:10 -
> @@ -543,8 +543,13 @@ If set to a positive integer,
>  output is formatted to the given width in columns.
>  Otherwise,
>  .Nm
> -defaults to the terminal width \(mi 1, or 79 columns if the output is not a
> -terminal.
> +defaults to the terminal width \(mi 1,
> +or 79 columns if none of
> +.Dv stdout ,
> +.Dv stderr
> +and
> +.Dv stdin
> +are a terminal.
>  .It Ev TZ
>  The time zone to use when displaying dates.
>  See
> 

Check terminal width in ps(1)

[Anthony J. Bentley]

Hi,

Unlike all the other software in the tree that checks terminal sizes, 
ps(1) used to check the stdin and stderr terminal sizes as well as stdout.
When I unified the terminal size checks in the tree in March, I missed
this, and now "ps aux | blah" gets wrapped to 79 characters no matter
how big the terminal is.

The below diff brings back the old behavior: if the output is a pipe,
ps will check if either stderr or stdin is a terminal and if so will use
its width.

ok?

Index: ps.c
===
RCS file: /cvs/src/bin/ps/ps.c,v
retrieving revision 1.70
diff -u -p -r1.70 ps.c
--- ps.c17 Mar 2016 05:27:10 -  1.70
+++ ps.c22 Sep 2016 10:14:10 -
@@ -105,7 +105,10 @@ main(int argc, char *argv[])
termwidth = 0;
if ((cols = getenv("COLUMNS")) != NULL)
termwidth = strtonum(cols, 1, INT_MAX, NULL);
-   if (termwidth == 0 && ioctl(STDOUT_FILENO, TIOCGWINSZ, ) == 0 &&
+   if (termwidth == 0 &&
+   (ioctl(STDOUT_FILENO, TIOCGWINSZ, ) == 0 ||
+   ioctl(STDERR_FILENO, TIOCGWINSZ, ) == 0 ||
+   ioctl(STDIN_FILENO, TIOCGWINSZ, ) == 0) &&
ws.ws_col > 0)
termwidth = ws.ws_col - 1;
if (termwidth == 0)
Index: ps.1
===
RCS file: /cvs/src/bin/ps/ps.1,v
retrieving revision 1.108
diff -u -p -r1.108 ps.1
--- ps.18 Sep 2016 15:54:36 -   1.108
+++ ps.122 Sep 2016 10:14:10 -
@@ -543,8 +543,13 @@ If set to a positive integer,
 output is formatted to the given width in columns.
 Otherwise,
 .Nm
-defaults to the terminal width \(mi 1, or 79 columns if the output is not a
-terminal.
+defaults to the terminal width \(mi 1,
+or 79 columns if none of
+.Dv stdout ,
+.Dv stderr
+and
+.Dv stdin
+are a terminal.
 .It Ev TZ
 The time zone to use when displaying dates.
 See

Re: Minor improvement to socket(2) man page

[Philip Guenther]

On Wed, Sep 21, 2016 at 6:23 PM, Michael McConville  wrote:
> The current version is somewhat awkward and forgets to mention that
> errno is set. I adapted the paragraph found in most other system call
> man pages.

There are six syscalls that return an fd on success...and their
manpages all have totally different wording for the return value.
Whee.

accept
 The call returns -1 on error.  If it succeeds, it returns a non-negative
 integer that is a descriptor for the accepted socket.

fhopen
 Upon successful completion, fhopen() returns the file descriptor for the
 opened file, while fhstat() and fhstatfs() return 0.  Otherwise, -1 is
 returned and errno is set to indicate the error.

kqueue
 kqueue() creates a new kernel event queue and returns a file descriptor.
 If there was an error creating the kernel event queue, a value of -1 is
 returned and errno set.

open
 If successful, open() returns a non-negative integer, termed a file
 descriptor.  Otherwise, a value of -1 is returned and errno is set to
 indicate the error.

openat: same manpage as open but totally unmentioned in the RETURN
VALUES section

socket
 A -1 is returned if an error occurs, otherwise the return value is a
 descriptor referencing the socket.

  Some consistency that doesn't sacrifice clarity would be nice...

[PATCH] Remove deprecated newaliases NOTE

[Consus]

Since smtpd now uses file table instead of db this NOTE is not accurate.
---
 etc/mail/aliases | 4 
 1 file changed, 4 deletions(-)

diff --git a/etc/mail/aliases b/etc/mail/aliases
index fe08bf1..38f7691 100644
--- a/etc/mail/aliases
+++ b/etc/mail/aliases
@@ -4,10 +4,6 @@
 #  Aliases in this file will NOT be expanded in the header from
 #  Mail, but WILL be visible over networks or from /usr/libexec/mail.local.
 #
-#  >>  The program "newaliases" must be run after
-#  >> NOTE >>  this file is updated for any changes to
-#  >>  show through to smtpd.
-#
 
 # Basic system aliases -- these MUST be present
 MAILER-DAEMON: postmaster
-- 
2.9.3