svn commit: r303212 - in head: etc/mtree tools/build/mk usr.sbin/chown/tests

[Bryan Drewery]

Author: bdrewery
Date: Sat Jul 23 05:49:18 2016
New Revision: 303212
URL: https://svnweb.freebsd.org/changeset/base/303212

Log:
  Move chown tests to proper path
  
  Sponsored by: EMC / Isilon Storage Division

Modified:
  head/etc/mtree/BSD.tests.dist
  head/tools/build/mk/OptionalObsoleteFiles.inc
  head/usr.sbin/chown/tests/Makefile

Modified: head/etc/mtree/BSD.tests.dist
==
--- head/etc/mtree/BSD.tests.dist   Sat Jul 23 05:30:55 2016
(r303211)
+++ head/etc/mtree/BSD.tests.dist   Sat Jul 23 05:49:18 2016
(r303212)
@@ -8,8 +8,6 @@
 bin
 cat
 ..
-chown
-..
 date
 ..
 dd
@@ -652,6 +650,8 @@
 ..
 ..
 usr.sbin
+chown
+..
 etcupdate
 ..
 extattr

Modified: head/tools/build/mk/OptionalObsoleteFiles.inc
==
--- head/tools/build/mk/OptionalObsoleteFiles.inc   Sat Jul 23 05:30:55 
2016(r303211)
+++ head/tools/build/mk/OptionalObsoleteFiles.inc   Sat Jul 23 05:49:18 
2016(r303212)
@@ -7419,6 +7419,9 @@ OLD_FILES+=usr/share/aclocal/atf-c.m4
 OLD_FILES+=usr/share/aclocal/atf-common.m4
 OLD_FILES+=usr/share/aclocal/atf-sh.m4
 OLD_DIRS+=usr/share/aclocal
+OLD_DIRS+=usr/tests/bin/chown
+OLD_FILES+=usr/tests/bin/chown/Kyuafile
+OLD_FILES+=usr/tests/bin/chown/chown-f_test
 OLD_FILES+=usr/tests/bin/chown/units_basics
 OLD_FILES+=usr/tests/bin/date/legacy_test
 OLD_FILES+=usr/tests/bin/sh/legacy_test

Modified: head/usr.sbin/chown/tests/Makefile
==
--- head/usr.sbin/chown/tests/Makefile  Sat Jul 23 05:30:55 2016
(r303211)
+++ head/usr.sbin/chown/tests/Makefile  Sat Jul 23 05:49:18 2016
(r303212)
@@ -2,7 +2,7 @@
 
 .include 
 
-TESTSDIR=  ${TESTSBASE}/bin/chown
+TESTSDIR=  ${TESTSBASE}/usr.sbin/chown
 
 TAP_TESTS_SH=  chown-f_test
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303211 - in head: share/man/man9 sys/kern sys/sys

[Konstantin Belousov]

Author: kib
Date: Sat Jul 23 05:30:55 2016
New Revision: 303211
URL: https://svnweb.freebsd.org/changeset/base/303211

Log:
  Implement mtx_trylock_spin(9).
  
  Discussed with:   bde
  Reviewed by:  jhb
  Sponsored by: The FreeBSD Foundation
  MFC after:1 week
  Differential revision:https://reviews.freebsd.org/D7192

Modified:
  head/share/man/man9/Makefile
  head/share/man/man9/mutex.9
  head/sys/kern/kern_mutex.c
  head/sys/sys/mutex.h

Modified: head/share/man/man9/Makefile
==
--- head/share/man/man9/MakefileSat Jul 23 03:49:02 2016
(r303210)
+++ head/share/man/man9/MakefileSat Jul 23 05:30:55 2016
(r303211)
@@ -1181,6 +1181,8 @@ MLINKS+=mutex.9 mtx_assert.9 \
mutex.9 MTX_SYSINIT.9 \
mutex.9 mtx_trylock.9 \
mutex.9 mtx_trylock_flags.9 \
+   mutex.9 mtx_trylock_spin.9 \
+   mutex.9 mtx_trylock_spin_flags.9 \
mutex.9 mtx_unlock.9 \
mutex.9 mtx_unlock_flags.9 \
mutex.9 mtx_unlock_spin.9 \

Modified: head/share/man/man9/mutex.9
==
--- head/share/man/man9/mutex.9 Sat Jul 23 03:49:02 2016(r303210)
+++ head/share/man/man9/mutex.9 Sat Jul 23 05:30:55 2016(r303211)
@@ -28,7 +28,7 @@
 .\"from BSDI $Id: mutex.4,v 1.1.2.3 1998/04/27 22:53:13 ewv Exp $
 .\" $FreeBSD$
 .\"
-.Dd December 13, 2014
+.Dd July 18, 2016
 .Dt MUTEX 9
 .Os
 .Sh NAME
@@ -41,6 +41,8 @@
 .Nm mtx_lock_spin_flags ,
 .Nm mtx_trylock ,
 .Nm mtx_trylock_flags ,
+.Nm mtx_trylock_spin ,
+.Nm mtx_trylock_spin_flags ,
 .Nm mtx_unlock ,
 .Nm mtx_unlock_spin ,
 .Nm mtx_unlock_flags ,
@@ -73,6 +75,10 @@
 .Ft int
 .Fn mtx_trylock_flags "struct mtx *mutex" "int flags"
 .Ft void
+.Fn mtx_trylock_spin "struct mtx *mutex"
+.Ft int
+.Fn mtx_trylock_spin_flags "struct mtx *mutex" "int flags"
+.Ft void
 .Fn mtx_unlock "struct mtx *mutex"
 .Ft void
 .Fn mtx_unlock_spin "struct mtx *mutex"
@@ -249,26 +255,33 @@ argument, then the mutex can be acquired
 .Pp
 The
 .Fn mtx_trylock
-attempts to acquire the
+and
+.Fn mtx_trylock_spin
+functions attempt to acquire a
 .Dv MTX_DEF
-mutex pointed to by
+or
+.Dv MTX_SPIN
+mutex, respectively, pointed to by
 .Fa mutex .
-If the mutex cannot be immediately acquired
-.Fn mtx_trylock
-will return 0,
-otherwise the mutex will be acquired
-and a non-zero value will be returned.
+If the mutex cannot be immediately acquired, the functions will return 0,
+otherwise the mutex will be acquired and a non-zero value will be returned.
 .Pp
 The
 .Fn mtx_trylock_flags
-function has the same behavior as
+and
+.Fn mtx_trylock_spin_flags
+functions have the same behavior as
 .Fn mtx_trylock
-but should be used when the caller desires to pass in a
+and
+.Fn mtx_trylock_spin
+respectively, but should be used when the caller desires to pass in a
 .Fa flags
 value.
 Presently, the only valid value in the
 .Fn mtx_trylock
-case is
+and
+.Fn mtx_trylock_spin
+cases is
 .Dv MTX_QUIET ,
 and its effects are identical to those described for
 .Fn mtx_lock
@@ -447,6 +460,13 @@ while any spin lock is held.
 .It Dv MTX_RECURSE
 Specifies that the initialized mutex is allowed to recurse.
 This bit must be present if the mutex is permitted to recurse.
+.Pp
+Note that neither
+.Fn mtx_trylock
+nor
+.Fn mtx_trylock_spin
+support recursion;
+that is, attempting to acquire an already-owned mutex fails.
 .It Dv MTX_QUIET
 Do not log any mutex operations for this lock.
 .It Dv MTX_NOWITNESS
@@ -534,3 +554,7 @@ functions appeared in
 .Bsx 4.1
 and
 .Fx 5.0 .
+The
+.Fn mtx_trylock_spin
+function was added in
+.Fx 12.0 .

Modified: head/sys/kern/kern_mutex.c
==
--- head/sys/kern/kern_mutex.c  Sat Jul 23 03:49:02 2016(r303210)
+++ head/sys/kern/kern_mutex.c  Sat Jul 23 05:30:55 2016(r303211)
@@ -281,6 +281,34 @@ __mtx_lock_spin_flags(volatile uintptr_t
WITNESS_LOCK(>lock_object, opts | LOP_EXCLUSIVE, file, line);
 }
 
+int
+__mtx_trylock_spin_flags(volatile uintptr_t *c, int opts, const char *file,
+int line)
+{
+   struct mtx *m;
+
+   if (SCHEDULER_STOPPED())
+   return (1);
+
+   m = mtxlock2mtx(c);
+
+   KASSERT(m->mtx_lock != MTX_DESTROYED,
+   ("mtx_trylock_spin() of destroyed mutex @ %s:%d", file, line));
+   KASSERT(LOCK_CLASS(>lock_object) == _class_mtx_spin,
+   ("mtx_trylock_spin() of sleep mutex %s @ %s:%d",
+   m->lock_object.lo_name, file, line));
+   KASSERT((opts & MTX_RECURSE) == 0,
+   ("mtx_trylock_spin: unsupp. opt MTX_RECURSE on mutex %s @ %s:%d\n",
+   m->lock_object.lo_name, file, line));
+   if (__mtx_trylock_spin(m, curthread, opts, file, line)) {
+   LOCK_LOG_TRY("LOCK", >lock_object, opts, 1, file, line);
+   WITNESS_LOCK(>lock_object, opts | LOP_EXCLUSIVE, file, 

svn commit: r303210 - head/lib/libc/gen

[Andrey A. Chernov]

Author: ache
Date: Sat Jul 23 03:49:02 2016
New Revision: 303210
URL: https://svnweb.freebsd.org/changeset/base/303210

Log:
  1) POSIX defines well when GLOB_NOMATCH or original pattern
  (instead) should be returned, so we can't return GLOB_NOMATCH blindly
  just because we dislike something in the pattern.
  
  2) Remove extra condition.

Modified:
  head/lib/libc/gen/glob.c

Modified: head/lib/libc/gen/glob.c
==
--- head/lib/libc/gen/glob.cSat Jul 23 02:27:42 2016(r303209)
+++ head/lib/libc/gen/glob.cSat Jul 23 03:49:02 2016(r303210)
@@ -217,7 +217,8 @@ glob(const char * __restrict pattern, in
while (bufnext <= bufend) {
clen = mbrtowc(, patnext, MB_LEN_MAX, );
if (clen == (size_t)-1 || clen == (size_t)-2)
-   return (GLOB_NOMATCH);
+   return (globfinal(pglob, ,
+   pglob->gl_pathc, pattern));
else if (clen == 0) {
too_long = 0;
break;
@@ -239,7 +240,8 @@ glob(const char * __restrict pattern, in
prot = 0;
clen = mbrtowc(, patnext, MB_LEN_MAX, );
if (clen == (size_t)-1 || clen == (size_t)-2)
-   return (GLOB_NOMATCH);
+   return (globfinal(pglob, ,
+   pglob->gl_pathc, pattern));
else if (clen == 0) {
too_long = 0;
break;
@@ -249,7 +251,7 @@ glob(const char * __restrict pattern, in
}
}
if (too_long)
-   return (GLOB_NOMATCH);
+   return (globfinal(pglob, , pglob->gl_pathc, pattern));
*bufnext = EOS;
 
if (flags & GLOB_BRACE)
@@ -613,9 +615,9 @@ globfinal(glob_t *pglob, struct glob_lim
 * GLOB_NOMAGIC is there just for compatibility with csh.
 */
if (pglob->gl_pathc == oldpathc) {
-   if (origpat != NULL && ((pglob->gl_flags & GLOB_NOCHECK) ||
+   if ((pglob->gl_flags & GLOB_NOCHECK) ||
((pglob->gl_flags & GLOB_NOMAGIC) &&
-   !(pglob->gl_flags & GLOB_MAGCHAR
+   !(pglob->gl_flags & GLOB_MAGCHAR)))
return (globextend(NULL, pglob, limit, origpat));
else
return (GLOB_NOMATCH);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303209 - head/sys/powerpc/booke

[Bruce Evans]

On Sat, 23 Jul 2016, Justin Hibbits wrote:


Log:
 Use label math instead of hard-coding offsets for return addresses.

 Though the chances of the code in these sections changing are low, future-proof
 the sections and use label math.


Not with numeric labels.


 Renumber the surrounding areas to avoid duplicate label numbers.


New numeric labels tend to move the addresses of old labels.  Renumbering
to give unique numeric labels more than defeats their reason for existence --
you have to manage them on every change to a label anywhere in the file
(and perhaps in macros and included files).  You never get a error for
a duplicated numeric label.  Non-unique numeric labels are also a good
obfuscation.  Given a branch to 1f or 1b, grepping the file for 1: may
find many labels 1:.  You have to examine all the labels named 1: or
search only near the jump to see where 1 is.  Its name is supposed to
not matter, but perhaps it does.

Numeric labels should only be used in macros, and then only when the
macro processor is too feeble to support generating unique non-numeric
labels.  Standard cpp is too feeble for this.

I managed to remove all numeric labels in i386/exception.s (except in
macros and included files), and none came back.  amd64/exception.S has
19 of them, 7 with the additional style bug of being on a line with
a statement or comment.

Bruce
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303209 - head/sys/powerpc/booke

[Justin Hibbits]

Author: jhibbits
Date: Sat Jul 23 02:27:42 2016
New Revision: 303209
URL: https://svnweb.freebsd.org/changeset/base/303209

Log:
  Use label math instead of hard-coding offsets for return addresses.
  
  Though the chances of the code in these sections changing are low, 
future-proof
  the sections and use label math.
  
  Renumber the surrounding areas to avoid duplicate label numbers.

Modified:
  head/sys/powerpc/booke/locore.S

Modified: head/sys/powerpc/booke/locore.S
==
--- head/sys/powerpc/booke/locore.S Sat Jul 23 01:21:58 2016
(r303208)
+++ head/sys/powerpc/booke/locore.S Sat Jul 23 02:27:42 2016
(r303209)
@@ -171,7 +171,7 @@ __start:
ori %r3, %r3, (PSL_IS | PSL_DS)
bl  2f
 2: mflr%r4
-   addi%r4, %r4, 20
+   addi%r4, %r4, (3f - 2b)
mtspr   SPR_SRR0, %r4
mtspr   SPR_SRR1, %r3
rfi /* Switch context */
@@ -179,6 +179,7 @@ __start:
 /*
  * Invalidate initial entry
  */
+3:
mr  %r3, %r29
bl  tlb1_inval_entry
 
@@ -224,7 +225,7 @@ __start:
rlwinm  %r4, %r4, 0, 8, 31  /* Current offset from kernel load 
address */
rlwinm  %r3, %r3, 0, 0, 19
add %r4, %r4, %r3   /* Convert to kernel virtual address */
-   addi%r4, %r4, 36
+   addi%r4, %r4, (5f - 4b)
li  %r3, PSL_DE /* Note AS=0 */
mtspr   SPR_SRR0, %r4
mtspr   SPR_SRR1, %r3
@@ -233,6 +234,7 @@ __start:
 /*
  * Invalidate temp mapping
  */
+5:
mr  %r3, %r28
bl  tlb1_inval_entry
 
@@ -362,7 +364,7 @@ bp_kernload:
ori %r3, %r3, (PSL_IS | PSL_DS)
bl  3f
 3: mflr%r4
-   addi%r4, %r4, 20
+   addi%r4, %r4, (4f - 3b)
mtspr   SPR_SRR0, %r4
mtspr   SPR_SRR1, %r3
rfi /* Switch context */
@@ -370,6 +372,7 @@ bp_kernload:
 /*
  * Invalidate initial entry
  */
+4:
mr  %r3, %r29
bl  tlb1_inval_entry
 
@@ -395,10 +398,10 @@ bp_kernload:
isync
 
/* Retrieve kernel load [physical] address from bp_kernload */
-   bl  4f
+   bl  5f
.long   bp_kernload
.long   __boot_page
-4: mflr%r3
+5: mflr%r3
lwz %r4, 0(%r3)
lwz %r5, 4(%r3)
rlwinm  %r3, %r3, 0, 0, 19
@@ -414,15 +417,16 @@ bp_kernload:
msync
 
/* Switch to the final mapping */
-   bl  5f
-5: mflr%r3
+   bl  6f
+6: mflr%r3
rlwinm  %r3, %r3, 0, 0xfff  /* Offset from boot page start */
add %r3, %r3, %r5   /* Make this virtual address */
-   addi%r3, %r3, 32
+   addi%r3, %r3, (7f - 6b)
li  %r4, 0  /* Note AS=0 */
mtspr   SPR_SRR0, %r3
mtspr   SPR_SRR1, %r4
rfi
+7:
 
 /*
  * At this point we're running at virtual addresses KERNBASE and beyond so
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303188 - head/share/mk

[Bryan Drewery]

On 7/22/2016 5:25 PM, Bruce Evans wrote:
> On Fri, 22 Jul 2016, Ngie Cooper wrote:
> 
>> On Fri, Jul 22, 2016 at 2:57 PM, Bruce Evans 
>> wrote:
>> ...
> +# GCC 6.1.0
> +.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
> +CWARNFLAGS+=   -Wno-error=unused-const-variable=
> -Wno-error=nonnull-compare -Wno-error=shift-negative-value
> -Wno-error=misleading-indentation -Wno-error=tautological-compare
> +.endif


 Extra trailing = in the first one as well?
>>>
>>> I don't see one.  I see only unreadably long lines containing unreadably
>>> unordered flags.  Perhaps your mail client mangled the line splitting,
>>> but the above quite displays OK in mutt and vi here (except mutt makes
>>> the line wrapping especially ugly by highlighting it).
>>
>> Look for "-Wno-error=unused-const-variable=".
> 
> It is not trailing :-).  It is only trailing in the above quote and
> apparently in the mail as seen by jhb because some mail programs mangle
> the formatting worse than others.  In the above quote as seen by me
> in EDITOR=vi, the long line is split and each part is quoted.  But in
> the previous quote, the line wasn't split.  Since it was unreadable
> by me, I used the editor search function to check if it had a trailing
> '=' using a regexp, and none was found.

It's not a mail issue. It is in the actual file. See
https://svnweb.freebsd.org/base/head/share/mk/bsd.sys.mk?r1=303188=303187=303188
and
https://svnweb.freebsd.org/base/head/share/mk/bsd.sys.mk?revision=303188=markup=303188#l119

Specifically "-Wno-error=unused-const-variable=", the first value.

-- 
Regards,
Bryan Drewery



signature.asc
Description: OpenPGP digital signature

svn commit: r303208 - head/lib/libc/gen

[Andrey A. Chernov]

Author: ache
Date: Sat Jul 23 01:21:58 2016
New Revision: 303208
URL: https://svnweb.freebsd.org/changeset/base/303208

Log:
  1) We need the original pattern (in the next round of changes) not only in
  case it fully constructed, but for half-constructed too, so have no
  other choice to pass original pattern from glob() down to globextend()
  instead of attempt to reconstruct I implement previously.
  
  2) Instead of copy the same big enough code, make function for it:
  globfinal().

Modified:
  head/lib/libc/gen/glob.c

Modified: head/lib/libc/gen/glob.c
==
--- head/lib/libc/gen/glob.cFri Jul 22 23:13:25 2016(r303207)
+++ head/lib/libc/gen/glob.cSat Jul 23 01:21:58 2016(r303208)
@@ -145,11 +145,12 @@ typedef uint_fast64_t Char;
 #defineM_RNG   META(L'-')
 #defineM_SET   META(L'[')
 #defineismeta(c)   (((c)_QUOTE) != 0)
+#ifdef DEBUG
 #defineisprot(c)   (((c)_PROTECT) != 0)
-
+#endif
 
 static int  compare(const void *, const void *);
-static int  g_Ctoc(const Char *, char *, size_t, int);
+static int  g_Ctoc(const Char *, char *, size_t);
 static int  g_lstat(Char *, struct stat *, glob_t *);
 static DIR *g_opendir(Char *, glob_t *);
 static const Char *g_strchr(const Char *, wchar_t);
@@ -157,19 +158,24 @@ static const Char *g_strchr(const Char *
 static Char*g_strcat(Char *, const Char *);
 #endif
 static int  g_stat(Char *, struct stat *, glob_t *);
-static int  glob0(const Char *, glob_t *, struct glob_limit *, int);
+static int  glob0(const Char *, glob_t *, struct glob_limit *,
+const char *);
 static int  glob1(Char *, glob_t *, struct glob_limit *);
 static int  glob2(Char *, Char *, Char *, Char *, glob_t *,
 struct glob_limit *);
 static int  glob3(Char *, Char *, Char *, Char *, Char *, glob_t *,
 struct glob_limit *);
-static int  globextend(const Char *, glob_t *, struct glob_limit *, int);
+static int  globextend(const Char *, glob_t *, struct glob_limit *,
+const char *);
 static const Char *
 globtilde(const Char *, Char *, size_t, glob_t *);
-static int  globexp0(const Char *, glob_t *, struct glob_limit *);
+static int  globexp0(const Char *, glob_t *, struct glob_limit *,
+const char *);
 static int  globexp1(const Char *, glob_t *, struct glob_limit *);
 static int  globexp2(const Char *, const Char *, glob_t *,
 struct glob_limit *);
+static int  globfinal(glob_t *, struct glob_limit *, size_t,
+const char *);
 static int  match(Char *, Char *, Char *);
 #ifdef DEBUG
 static void qprintf(const char *, Char *);
@@ -247,14 +253,14 @@ glob(const char * __restrict pattern, in
*bufnext = EOS;
 
if (flags & GLOB_BRACE)
-   return (globexp0(patbuf, pglob, ));
+   return (globexp0(patbuf, pglob, , pattern));
else
-   return (glob0(patbuf, pglob, , 1));
+   return (glob0(patbuf, pglob, , pattern));
 }
 
 static int
-globexp0(const Char *pattern, glob_t *pglob, struct glob_limit *limit)
-{
+globexp0(const Char *pattern, glob_t *pglob, struct glob_limit *limit,
+const char *origpat) {
int rv;
size_t oldpathc;
 
@@ -265,31 +271,15 @@ globexp0(const Char *pattern, glob_t *pg
errno = 0;
return (GLOB_NOSPACE);
}
-   return (glob0(pattern, pglob, limit, 1));
+   return (glob0(pattern, pglob, limit, origpat));
}
 
oldpathc = pglob->gl_pathc;
 
if ((rv = globexp1(pattern, pglob, limit)) != 0)
return rv;
-   /*
-* If there was no match we are going to append the pattern
-* if GLOB_NOCHECK was specified or if GLOB_NOMAGIC was specified
-* and the pattern did not contain any magic characters
-* GLOB_NOMAGIC is there just for compatibility with csh.
-*/
-   if (pglob->gl_pathc == oldpathc) {
-   if (((pglob->gl_flags & GLOB_NOCHECK) ||
-   ((pglob->gl_flags & GLOB_NOMAGIC) &&
-   !(pglob->gl_flags & GLOB_MAGCHAR
-   return (globextend(pattern, pglob, limit, 1));
-   else
-   return (GLOB_NOMATCH);
-   }
-   if (!(pglob->gl_flags & GLOB_NOSORT))
-   qsort(pglob->gl_pathv + pglob->gl_offs + oldpathc,
-   pglob->gl_pathc - oldpathc, sizeof(char *), compare);
-   return (0);
+
+   return (globfinal(pglob, limit, oldpathc, origpat));
 }
 
 /*
@@ -311,7 +301,7 @@ globexp1(const Char *pattern, glob_t *pg
return (globexp2(ptr, pattern, pglob, limit));
}
 
-   return (glob0(pattern, pglob, limit, 0));
+   return (glob0(pattern, pglob, limit, NULL));
 }
 
 
@@ -359,7 +349,7 @@ globexp2(const Char *ptr, const 

Re: svn commit: r303188 - head/share/mk

[Bruce Evans]

On Fri, 22 Jul 2016, Ngie Cooper wrote:


On Fri, Jul 22, 2016 at 2:57 PM, Bruce Evans  wrote:
...

+# GCC 6.1.0
+.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
+CWARNFLAGS+=   -Wno-error=unused-const-variable=
-Wno-error=nonnull-compare -Wno-error=shift-negative-value
-Wno-error=misleading-indentation -Wno-error=tautological-compare
+.endif



Extra trailing = in the first one as well?


I don't see one.  I see only unreadably long lines containing unreadably
unordered flags.  Perhaps your mail client mangled the line splitting,
but the above quite displays OK in mutt and vi here (except mutt makes
the line wrapping especially ugly by highlighting it).


Look for "-Wno-error=unused-const-variable=".


It is not trailing :-).  It is only trailing in the above quote and
apparently in the mail as seen by jhb because some mail programs mangle
the formatting worse than others.  In the above quote as seen by me
in EDITOR=vi, the long line is split and each part is quoted.  But in
the previous quote, the line wasn't split.  Since it was unreadable
by me, I used the editor search function to check if it had a trailing
'=' using a regexp, and none was found.

The corresponding line for gcc 5.2.0 in bsd.sys.mk has the same quality.

The '='s in the correct part of the syntax are not very good either.  I
think they break our policy that all warnings are fatal unless NO_WERROR
is set.  Already clang allows too many things to compile without even
warnings.  I think this allows them to compile with gcc's stricter
warnings, but only with newer gcc's.  NO_WERROR is too global, but is
at least easy to understand.

Bruce
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303188 - head/share/mk

[Ruslan Bukin]

On Fri, Jul 22, 2016 at 03:20:51PM -0700, Ngie Cooper wrote:
> On Fri, Jul 22, 2016 at 2:57 PM, Bruce Evans  wrote:
> ...
> >>> +# GCC 6.1.0
> >>> +.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
> >>> +CWARNFLAGS+=   -Wno-error=unused-const-variable=
> >>> -Wno-error=nonnull-compare -Wno-error=shift-negative-value
> >>> -Wno-error=misleading-indentation -Wno-error=tautological-compare
> >>> +.endif
> >>
> >>
> >> Extra trailing = in the first one as well?
> >
> > I don't see one.  I see only unreadably long lines containing unreadably
> > unordered flags.  Perhaps your mail client mangled the line splitting,
> > but the above quite displays OK in mutt and vi here (except mutt makes
> > the line wrapping especially ugly by highlighting it).
> 
> Look for "-Wno-error=unused-const-variable=".

I copy-pasted this from compiler error message:

/home/rb743/dev/freebsd-riscv/lib/libc/isc/ev_timers.c:23:19: error: 'rcsid' 
defined but not used [-Werror=unused-const-variable=]
 static const char rcsid[] = "$Id: ev_timers.c,v 1.6 2005/04/27 04:56:36 sra 
Exp $";

I am not sure yet why compiler set extra trailing = for some errors only

Ruslan

___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303205 - in head/sys: conf dev/cxgbe modules/cxgbe/if_cxgbe

[John Baldwin]

On Friday, July 22, 2016 10:46:41 PM John Baldwin wrote:
> Author: jhb
> Date: Fri Jul 22 22:46:41 2016
> New Revision: 303205
> URL: https://svnweb.freebsd.org/changeset/base/303205
> 
> Log:
>   Add a driver to create VF devices on Chelsio T4/T5 NICs.
>   
>   Chelsio NICs are a bit unique compared to some other NICs in that they
>   expose different functionality on different physical functions.  In
>   particular, PF4 is used to manage the NIC interfaces ('t4nex' and 't5nex').
>   However, PF4 is not able to create VF devices.  Instead, VFs are only
>   supported by physical functions 0 through 3.  This commit adds 't4iov'
>   and 't5iov' drivers that attach to PF0-3.
>   
>   One extra wrinkle is that the iov devices cannot enable SR-IOV until the
>   firwmare has been initialized by the main PF4 driver.  To handle this
>   case, a new t4_if kobj interface has been added to permit cross-calls
>   between the PF drivers.  The PF4 driver notifies sibling drivers when it
>   is fully attached.  It also requests sibling drivers to detach before it
>   detaches.  Sibling drivers query the PF4 driver during their attach
>   routine to see if it is attached.  If not, the sibling drivers defer
>   their attach actions until the PF4 driver informs them it is attached.
>   
>   VF devices are associated with a single port on the NIC.  VF devices
>   created from PF0 are associated with the first port on the NIC, VFs
>   from PF1 are associated with the second port, etc.  VF devices can
>   only be created from a PF device that has an associated port.  Thus,
>   on a 2-port card, VFs are only supported on PF0 and PF1.
>   
>   Reviewed by:np (earlier versions)
>   MFC after:  1 month
>   Sponsored by:   Chelsio Communications

Note that this only permits the creation of VFs, it does not include a driver
for these VFs.  I have a mostly working driver, but it is against an older
HEAD and has to be rebased.  In addition, Linux VF drivers will not work
without some hacks to the FreeBSD cxgbe/cxl driver because the Linux VF driver
only wants page-sized buffers whereas the FreeBSD drivers use 2k buffers
(since that is an mbuf cluster).

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303205 - in head/sys: conf dev/cxgbe modules/cxgbe/if_cxgbe

[John Baldwin]

Author: jhb
Date: Fri Jul 22 22:46:41 2016
New Revision: 303205
URL: https://svnweb.freebsd.org/changeset/base/303205

Log:
  Add a driver to create VF devices on Chelsio T4/T5 NICs.
  
  Chelsio NICs are a bit unique compared to some other NICs in that they
  expose different functionality on different physical functions.  In
  particular, PF4 is used to manage the NIC interfaces ('t4nex' and 't5nex').
  However, PF4 is not able to create VF devices.  Instead, VFs are only
  supported by physical functions 0 through 3.  This commit adds 't4iov'
  and 't5iov' drivers that attach to PF0-3.
  
  One extra wrinkle is that the iov devices cannot enable SR-IOV until the
  firwmare has been initialized by the main PF4 driver.  To handle this
  case, a new t4_if kobj interface has been added to permit cross-calls
  between the PF drivers.  The PF4 driver notifies sibling drivers when it
  is fully attached.  It also requests sibling drivers to detach before it
  detaches.  Sibling drivers query the PF4 driver during their attach
  routine to see if it is attached.  If not, the sibling drivers defer
  their attach actions until the PF4 driver informs them it is attached.
  
  VF devices are associated with a single port on the NIC.  VF devices
  created from PF0 are associated with the first port on the NIC, VFs
  from PF1 are associated with the second port, etc.  VF devices can
  only be created from a PF device that has an associated port.  Thus,
  on a 2-port card, VFs are only supported on PF0 and PF1.
  
  Reviewed by:  np (earlier versions)
  MFC after:1 month
  Sponsored by: Chelsio Communications

Added:
  head/sys/dev/cxgbe/t4_if.m   (contents, props changed)
  head/sys/dev/cxgbe/t4_iov.c   (contents, props changed)
Modified:
  head/sys/conf/files
  head/sys/dev/cxgbe/t4_main.c
  head/sys/modules/cxgbe/if_cxgbe/Makefile

Modified: head/sys/conf/files
==
--- head/sys/conf/files Fri Jul 22 21:52:07 2016(r303204)
+++ head/sys/conf/files Fri Jul 22 22:46:41 2016(r303205)
@@ -1243,6 +1243,8 @@ dev/cxgb/sys/uipc_mvec.c  optional cxgb p
compile-with "${NORMAL_C} -I$S/dev/cxgb"
 dev/cxgb/cxgb_t3fw.c   optional cxgb cxgb_t3fw \
compile-with "${NORMAL_C} -I$S/dev/cxgb"
+dev/cxgbe/t4_iov.c optional cxgbe pci \
+   compile-with "${NORMAL_C} -I$S/dev/cxgbe"
 dev/cxgbe/t4_mp_ring.c optional cxgbe pci \
compile-with "${NORMAL_C} -I$S/dev/cxgbe"
 dev/cxgbe/t4_main.coptional cxgbe pci \

Added: head/sys/dev/cxgbe/t4_if.m
==
--- /dev/null   00:00:00 1970   (empty, because file is newly added)
+++ head/sys/dev/cxgbe/t4_if.m  Fri Jul 22 22:46:41 2016(r303205)
@@ -0,0 +1,65 @@
+#-
+# Copyright (c) 2015-2016 Chelsio Communications, Inc.
+# All rights reserved.
+# Written by: John Baldwin 
+#
+# 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.
+#
+# $FreeBSD$
+#
+
+#include 
+
+INTERFACE t4;
+
+# The "main" device of a T4/T5 NIC is the PF4 device.  Drivers for other
+# functions on the NIC need to wait for the main device to be initialized
+# before finishing attach.  These routines allow drivers for other devices
+# to coordinate with the main driver for the PF4.
+
+# Called by a driver during attach to determine if the PF4 driver is
+# initialized.  If the main driver is not ready, the driver should defer
+# further initialization until 'attach_child'.
+METHOD int is_main_ready {
+   device_tdev;
+};
+
+# Called by the PF4 driver on each sibling device when the PF4 driver is
+# initialized.
+METHOD int attach_child {
+   device_t

Re: svn commit: r303188 - head/share/mk

[Ngie Cooper]

On Fri, Jul 22, 2016 at 2:57 PM, Bruce Evans  wrote:
...
>>> +# GCC 6.1.0
>>> +.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
>>> +CWARNFLAGS+=   -Wno-error=unused-const-variable=
>>> -Wno-error=nonnull-compare -Wno-error=shift-negative-value
>>> -Wno-error=misleading-indentation -Wno-error=tautological-compare
>>> +.endif
>>
>>
>> Extra trailing = in the first one as well?
>
> I don't see one.  I see only unreadably long lines containing unreadably
> unordered flags.  Perhaps your mail client mangled the line splitting,
> but the above quite displays OK in mutt and vi here (except mutt makes
> the line wrapping especially ugly by highlighting it).

Look for "-Wno-error=unused-const-variable=".
Thanks,
-Ngie
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303188 - head/share/mk

[Bruce Evans]

On Fri, 22 Jul 2016, John Baldwin wrote:


On Friday, July 22, 2016 03:00:38 PM Ruslan Bukin wrote:


Log:
  Add warn flags for GCC 6.1 compiler.

  Sponsored by: DARPA, AFRL

Modified:
  head/share/mk/bsd.sys.mk

Modified: head/share/mk/bsd.sys.mk
==
--- head/share/mk/bsd.sys.mkFri Jul 22 14:57:26 2016(r303187)
+++ head/share/mk/bsd.sys.mkFri Jul 22 15:00:38 2016(r303188)
@@ -114,6 +114,11 @@ CWARNFLAGS+=   -Wno-format
 CWARNFLAGS+=   -Wno-error=unused-function -Wno-error=enum-compare 
-Wno-error=logical-not-parentheses -Wno-error=bool-compare 
-Wno-error=uninitialized -Wno-error=array-bounds -Wno-error=clobbered 
-Wno-error=cast-align -Wno-error=extra -Wno-error=attributes -Wno-error=inline 
-Wno-error=unused-but-set-variable -Wno-error=unused-value 
-Wno-error=strict-aliasing -Wno-error=address
 .endif

+# GCC 6.1.0
+.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
+CWARNFLAGS+=   -Wno-error=unused-const-variable= -Wno-error=nonnull-compare 
-Wno-error=shift-negative-value -Wno-error=misleading-indentation 
-Wno-error=tautological-compare
+.endif


Extra trailing = in the first one as well?


I don't see one.  I see only unreadably long lines containing unreadably
unordered flags.  Perhaps your mail client mangled the line splitting,
but the above quite displays OK in mutt and vi here (except mutt makes
the line wrapping especially ugly by highlighting it).

Bruce
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303204 - head/sys/dev/cxgbe

[John Baldwin]

Author: jhb
Date: Fri Jul 22 21:52:07 2016
New Revision: 303204
URL: https://svnweb.freebsd.org/changeset/base/303204

Log:
  Install a handler for firmware work request error messages.
  
  If a driver sends an malformed or disallowed work request, the firmware
  responds with a work request error.  Previously the driver treated this is
  as an unexpected message and panicked.  Now it decodes the error message
  to aid in debugging.
  
  Reviewed by:  np (older version)
  MFC after:1 month
  Sponsored by: Chelsio Communications
  Differential Revision:https://reviews.freebsd.org/D6950

Modified:
  head/sys/dev/cxgbe/t4_sge.c

Modified: head/sys/dev/cxgbe/t4_sge.c
==
--- head/sys/dev/cxgbe/t4_sge.c Fri Jul 22 21:49:41 2016(r303203)
+++ head/sys/dev/cxgbe/t4_sge.c Fri Jul 22 21:52:07 2016(r303204)
@@ -244,6 +244,7 @@ static int handle_sge_egr_update(struct 
 struct mbuf *);
 static int handle_fw_msg(struct sge_iq *, const struct rss_header *,
 struct mbuf *);
+static int t4_handle_wrerr_rpl(struct adapter *, const __be64 *);
 static void wrq_tx_drain(void *, int);
 static void drain_wrq_wr_list(struct adapter *, struct sge_wrq *);
 
@@ -402,6 +403,7 @@ t4_sge_modload(void)
t4_register_cpl_handler(CPL_SGE_EGR_UPDATE, handle_sge_egr_update);
t4_register_cpl_handler(CPL_RX_PKT, t4_eth_rx);
t4_register_fw_msg_handler(FW6_TYPE_CMD_RPL, t4_handle_fw_rpl);
+   t4_register_fw_msg_handler(FW6_TYPE_WRERR_RPL, t4_handle_wrerr_rpl);
 }
 
 void
@@ -4782,6 +4784,71 @@ handle_fw_msg(struct sge_iq *iq, const s
return (t4_fw_msg_handler[cpl->type](sc, >data[0]));
 }
 
+/**
+ * t4_handle_wrerr_rpl - process a FW work request error message
+ * @adap: the adapter
+ * @rpl: start of the FW message
+ */
+static int
+t4_handle_wrerr_rpl(struct adapter *adap, const __be64 *rpl)
+{
+   u8 opcode = *(const u8 *)rpl;
+   const struct fw_error_cmd *e = (const void *)rpl;
+   unsigned int i;
+
+   if (opcode != FW_ERROR_CMD) {
+   log(LOG_ERR,
+   "%s: Received WRERR_RPL message with opcode %#x\n",
+   device_get_nameunit(adap->dev), opcode);
+   return (EINVAL);
+   }
+   log(LOG_ERR, "%s: FW_ERROR (%s) ", device_get_nameunit(adap->dev),
+   G_FW_ERROR_CMD_FATAL(be32toh(e->op_to_type)) ? "fatal" :
+   "non-fatal");
+   switch (G_FW_ERROR_CMD_TYPE(be32toh(e->op_to_type))) {
+   case FW_ERROR_TYPE_EXCEPTION:
+   log(LOG_ERR, "exception info:\n");
+   for (i = 0; i < nitems(e->u.exception.info); i++)
+   log(LOG_ERR, "%s%08x", i == 0 ? "\t" : " ",
+   be32toh(e->u.exception.info[i]));
+   log(LOG_ERR, "\n");
+   break;
+   case FW_ERROR_TYPE_HWMODULE:
+   log(LOG_ERR, "HW module regaddr %08x regval %08x\n",
+   be32toh(e->u.hwmodule.regaddr),
+   be32toh(e->u.hwmodule.regval));
+   break;
+   case FW_ERROR_TYPE_WR:
+   log(LOG_ERR, "WR cidx %d PF %d VF %d eqid %d hdr:\n",
+   be16toh(e->u.wr.cidx),
+   G_FW_ERROR_CMD_PFN(be16toh(e->u.wr.pfn_vfn)),
+   G_FW_ERROR_CMD_VFN(be16toh(e->u.wr.pfn_vfn)),
+   be32toh(e->u.wr.eqid));
+   for (i = 0; i < nitems(e->u.wr.wrhdr); i++)
+   log(LOG_ERR, "%s%02x", i == 0 ? "\t" : " ",
+   e->u.wr.wrhdr[i]);
+   log(LOG_ERR, "\n");
+   break;
+   case FW_ERROR_TYPE_ACL:
+   log(LOG_ERR, "ACL cidx %d PF %d VF %d eqid %d %s",
+   be16toh(e->u.acl.cidx),
+   G_FW_ERROR_CMD_PFN(be16toh(e->u.acl.pfn_vfn)),
+   G_FW_ERROR_CMD_VFN(be16toh(e->u.acl.pfn_vfn)),
+   be32toh(e->u.acl.eqid),
+   G_FW_ERROR_CMD_MV(be16toh(e->u.acl.mv_pkd)) ? "vlanid" :
+   "MAC");
+   for (i = 0; i < nitems(e->u.acl.val); i++)
+   log(LOG_ERR, " %02x", e->u.acl.val[i]);
+   log(LOG_ERR, "\n");
+   break;
+   default:
+   log(LOG_ERR, "type %#x\n",
+   G_FW_ERROR_CMD_TYPE(be32toh(e->op_to_type)));
+   return (EINVAL);
+   }
+   return (0);
+}
+
 static int
 sysctl_uint16(SYSCTL_HANDLER_ARGS)
 {
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303203 - head/share/misc

[Michael Zhilin]

Author: mizhka
Date: Fri Jul 22 21:49:41 2016
New Revision: 303203
URL: https://svnweb.freebsd.org/changeset/base/303203

Log:
  [new-committer:mizhka] add committer into graph
  
  Approved by: adrian (mentor)
  Differential Revision: https://reviews.freebsd.org/D7288

Modified:
  head/share/misc/committers-src.dot

Modified: head/share/misc/committers-src.dot
==
--- head/share/misc/committers-src.dot  Fri Jul 22 20:52:40 2016
(r303202)
+++ head/share/misc/committers-src.dot  Fri Jul 22 21:49:41 2016
(r303203)
@@ -237,6 +237,7 @@ maxim [label="Maxim Konovalov\nmaxim@Fre
 mdf [label="Matthew Fleming\n...@freebsd.org\n2010/06/04"]
 mdodd [label="Matthew N. Dodd\nmd...@freebsd.org\n1999/07/27"]
 melifaro [label="Alexander V. Chernikov\nmelif...@freebsd.org\n2011/10/04"]
+mizhka [label="Michael Zhilin\nmiz...@freebsd.org\n2016/07/19"]
 mjacob [label="Matt Jacob\nmja...@freebsd.org\n1997/08/13"]
 mjg [label="Mateusz Guzik\n...@freebsd.org\n2012/06/04"]
 mlaier [label="Max Laier\nmla...@freebsd.org\n2004/02/10"]
@@ -353,6 +354,7 @@ adrian -> jmcneill
 adrian -> landonf
 adrian -> lidl
 adrian -> loos
+adrian -> mizhka
 adrian -> monthadar
 adrian -> ray
 adrian -> rmh
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303076 - head/sys/dev/fb

[John Baldwin]

On Wednesday, July 20, 2016 09:29:39 AM Roger Pau Monné wrote:
> Author: royger
> Date: Wed Jul 20 09:29:39 2016
> New Revision: 303076
> URL: https://svnweb.freebsd.org/changeset/base/303076
> 
> Log:
>   vesa: fix panic on suspend
>   
>   Fix the following panic seen when migrating a FreeBSD guest on Xen:
>   
>   panic: mtx_lock() of destroyed mutex @ /usr/src/sys/dev/fb/vesa.c:541
>   cpuid = 0
>   KDB: stack backtrace:
>   db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 
> 0xfe001d2fa4f0
>   vpanic() at vpanic+0x182/frame 0xfe001d2fa570
>   kassert_panic() at kassert_panic+0x126/frame 0xfe001d2fa5e0
>   __mtx_lock_flags() at __mtx_lock_flags+0x15b/frame 0xfe001d2fa630
>   vesa_bios_save_restore() at vesa_bios_save_restore+0x78/frame 
> 0xfe001d2fa680
>   vga_suspend() at vga_suspend+0xa3/frame 0xfe001d2fa6b0
>   isavga_suspend() at isavga_suspend+0x1d/frame 0xfe001d2fa6d0
>   bus_generic_suspend_child() at bus_generic_suspend_child+0x44/frame
>   [...]
>   
>   This is caused because vga_sub_configure (which is called if the VGA adapter
>   is attached after VESA tried to initialize), points to vesa_configure, which
>   doesn't initialize the VESA mutex. In order to fix it, make sure
>   vga_sub_configure points to vesa_load, so that all the needed vesa
>   components are properly initialized.

This panics a box without VESA (a dumb VGA) since vesa_init_done doesn't get set
to true if VESA fails to find anything during its probe, so the mutex can be
initialized twice.  I think it is better to just revert this change and use
MTX_SYSINIT to initialize the mutex instead.

vga0:  at port 0x3d0-0x3db iomem 0xb8000-0xb on isa0
panic: lock "VESA lock" 0x81c64af0 already initialized
cpuid = 0
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0x824ec7a0
kdb_backtrace() at kdb_backtrace+0x39/frame 0x824ec850
vpanic() at vpanic+0x181/frame 0x824ec8d0
kassert_panic() at kassert_panic+0x17d/frame 0x824ec960
lock_init() at lock_init+0x186/frame 0x824ec9a0
_mtx_init() at _mtx_init+0x90/frame 0x824ec9e0
vesa_late_load() at vesa_late_load+0x2d/frame 0x824eca00
vga_init() at vga_init+0x65/frame 0x824eca20
vga_attach_unit() at vga_attach_unit+0x4b/frame 0x824eca50
isavga_attach() at isavga_attach+0x6c/frame 0x824eca90
DEVICE_ATTACH() at DEVICE_ATTACH+0x44/frame 0x824ecab0

https://reviews.freebsd.org/D7290 is a suggested patch that works in my
test case.

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r301453 - in head/sys: arm/arm arm64/arm64 dev/fdt dev/gpio dev/iicbus dev/ofw dev/pci dev/vnic kern mips/mips sys

[Nathan Whitehorn]

On 07/22/16 13:53, Andrew Turner wrote:

On Fri, 22 Jul 2016 13:19:32 -0700
John Baldwin  wrote:


On Thursday, July 21, 2016 10:51:20 PM Nathan Whitehorn wrote:

On 07/21/16 14:35, John Baldwin wrote:

On Thursday, July 21, 2016 01:37:42 PM Andrew Turner wrote:

On Wed, 20 Jul 2016 13:28:53 +0200
Michal Meloun  wrote:

Dne 19.07.2016 v 17:06 Nathan Whitehorn napsal(a):

2. It partially duplicates the functionality of
OFW_BUS_MAP_INTR(), but is both problematically more general
and less flexible (it has requirements on timing of PIC
attachment vs. driver resource allocation)

OFW_BUS_MAP_INTR()  can parse only OFW  based data and expect
that parsed data are magicaly stored within the call.
The new method, bus_map_intr(),  can parse data from multiple
sources (OFW, UEFI / ACPI, synthetic[gpio device + pin
number]).  It also returns parsed data back to caller.
And no, it  doesn't  add any additional timing requirements .

I've been looking at ACPI on arm64. So far I have not found the
need for this with ACPI as we don't need to send the data to the
interrupt controller driver to be parsed in the way OFW/FDT
needs to.

ACPI though has a gross hack where we call BUS_CONFIG_INTR on the
IRQ in bus_alloc_resource().

I hadn't realized that. It looks like you could do essentially the
same thing we do on PowerPC to clean this up by explicitly mapping
the ACPI interrupt domains to different PICs with varying default
interrupt properties.
   

What I had advocated in the discussions
leading up to this was to have some sort of opaque structure
containing a set of properties (the sort of thing
bus_map_resource and make_dev_s use) that was passed up at
bus_setup_intr() time.
   

I think it should now
be passed up at bus_alloc_resource() time instead, but it would
allow bus drivers to "decorate" a SYS_RES_IRQ request as it goes
up the device tree with properties that the interrupt controller
can then associate with the IRQ cookie it allocates in its own
code.


We used to do this on PPC and MIPS, and the current code still
supports it, but it turned out not to be useful in the end for
IRQs. The hierarchy for IRQs rarely (read: almost never) follows
the bus hierarchy and often is enumerated in a different order. I
have hardware, for example, where the children of a single parent
bus are all wired to different interrupt controllers and sometimes
to a mixture of interrupt controllers. Those controllers are
cascaded in ways that cross the newbus tree laterally and, on some
of them, the parent device from the bus topology has interrupts
handled by its own (bus) children. Trying to make the newbus
parents do something sensible with all of this would be crazy and,
in the case where parents depend on resources provided by their own
children, impossible.

This is all to say that, since you want the interrupts to be
decorated along a path that usually has nothing to do with the
newbus hierarchy, it doesn't add much to add extra features to
resource allocation. ofw_bus_map_intr() is a newbus method to
support this kind of thing but, on all supported platforms, it is
implemented only in nexus and no cases have appeared where anyone
ever wanted anything at the intermediate layers.

Mmm.  Another idea that has been bandied about is to create a separate
"plane" in new-bus for an interrupt hierarchy and allow devices to
have "interrupt" parents that are not the same as the "bus" parent.
(Additional planes for power and clocks might also make sense.)  The
idea is borrowed from IOKit on Darwin which has multiple planes.  The
"bus" plane is always fully populated, but the other planes (Darwin
has one for power for example) can be sparse.  ACPI has methods that
effectively describe the power plane on x86.

For some planes the structure would look more like a directed
graph. Devices can have multiple clock parents, e.g. one for the
register interface, and another to drive the external bus.

I can also imagine the case where a device could have multiple
interrupt parents, an MMC/SD controller comes to mind where it may have
an interrupt on a GPIO line to detect the card being inserted, with
another for command completion, etc. In this case one parent would be
the standard interrupt controller, with the other being the GPIO
controller.

Andrew



That's a good point for the generalized system (this already works for 
the specific case of interrupts, of course). It's more the resources 
that have a non-bus parent than the device. There are also devices with 
multiple bus parents, but those are more oddball kinds of things (Apple 
made some on-board PCI devices that are also connected by a second 
non-PCI path).

-Nathan
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r301453 - in head/sys: arm/arm arm64/arm64 dev/fdt dev/gpio dev/iicbus dev/ofw dev/pci dev/vnic kern mips/mips sys

[Andrew Turner]

On Fri, 22 Jul 2016 13:19:32 -0700
John Baldwin  wrote:

> On Thursday, July 21, 2016 10:51:20 PM Nathan Whitehorn wrote:
> > 
> > On 07/21/16 14:35, John Baldwin wrote:  
> > > On Thursday, July 21, 2016 01:37:42 PM Andrew Turner wrote:  
> > >> On Wed, 20 Jul 2016 13:28:53 +0200
> > >> Michal Meloun  wrote:  
> > >>> Dne 19.07.2016 v 17:06 Nathan Whitehorn napsal(a):  
> >  2. It partially duplicates the functionality of
> >  OFW_BUS_MAP_INTR(), but is both problematically more general
> >  and less flexible (it has requirements on timing of PIC
> >  attachment vs. driver resource allocation)  
> > >>> OFW_BUS_MAP_INTR()  can parse only OFW  based data and expect
> > >>> that parsed data are magicaly stored within the call.
> > >>> The new method, bus_map_intr(),  can parse data from multiple
> > >>> sources (OFW, UEFI / ACPI, synthetic[gpio device + pin
> > >>> number]).  It also returns parsed data back to caller.
> > >>> And no, it  doesn't  add any additional timing requirements .  
> > >> I've been looking at ACPI on arm64. So far I have not found the
> > >> need for this with ACPI as we don't need to send the data to the
> > >> interrupt controller driver to be parsed in the way OFW/FDT
> > >> needs to.  
> > > ACPI though has a gross hack where we call BUS_CONFIG_INTR on the
> > > IRQ in bus_alloc_resource().  
> > 
> > I hadn't realized that. It looks like you could do essentially the
> > same thing we do on PowerPC to clean this up by explicitly mapping
> > the ACPI interrupt domains to different PICs with varying default
> > interrupt properties.
> >   
> > > What I had advocated in the discussions
> > > leading up to this was to have some sort of opaque structure
> > > containing a set of properties (the sort of thing
> > > bus_map_resource and make_dev_s use) that was passed up at
> > > bus_setup_intr() time.  
> >   
> > > I think it should now
> > > be passed up at bus_alloc_resource() time instead, but it would
> > > allow bus drivers to "decorate" a SYS_RES_IRQ request as it goes
> > > up the device tree with properties that the interrupt controller
> > > can then associate with the IRQ cookie it allocates in its own
> > > code.  
> > 
> > 
> > We used to do this on PPC and MIPS, and the current code still
> > supports it, but it turned out not to be useful in the end for
> > IRQs. The hierarchy for IRQs rarely (read: almost never) follows
> > the bus hierarchy and often is enumerated in a different order. I
> > have hardware, for example, where the children of a single parent
> > bus are all wired to different interrupt controllers and sometimes
> > to a mixture of interrupt controllers. Those controllers are
> > cascaded in ways that cross the newbus tree laterally and, on some
> > of them, the parent device from the bus topology has interrupts
> > handled by its own (bus) children. Trying to make the newbus
> > parents do something sensible with all of this would be crazy and,
> > in the case where parents depend on resources provided by their own
> > children, impossible.
> > 
> > This is all to say that, since you want the interrupts to be
> > decorated along a path that usually has nothing to do with the
> > newbus hierarchy, it doesn't add much to add extra features to
> > resource allocation. ofw_bus_map_intr() is a newbus method to
> > support this kind of thing but, on all supported platforms, it is
> > implemented only in nexus and no cases have appeared where anyone
> > ever wanted anything at the intermediate layers.  
> 
> Mmm.  Another idea that has been bandied about is to create a separate
> "plane" in new-bus for an interrupt hierarchy and allow devices to
> have "interrupt" parents that are not the same as the "bus" parent.
> (Additional planes for power and clocks might also make sense.)  The
> idea is borrowed from IOKit on Darwin which has multiple planes.  The
> "bus" plane is always fully populated, but the other planes (Darwin
> has one for power for example) can be sparse.  ACPI has methods that
> effectively describe the power plane on x86.

For some planes the structure would look more like a directed
graph. Devices can have multiple clock parents, e.g. one for the
register interface, and another to drive the external bus.

I can also imagine the case where a device could have multiple
interrupt parents, an MMC/SD controller comes to mind where it may have
an interrupt on a GPIO line to detect the card being inserted, with
another for command completion, etc. In this case one parent would be
the standard interrupt controller, with the other being the GPIO
controller.

Andrew
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303202 - head/share/man/man4

[John Baldwin]

Author: jhb
Date: Fri Jul 22 20:52:40 2016
New Revision: 303202
URL: https://svnweb.freebsd.org/changeset/base/303202

Log:
  Update after r303154 to note that operations on local files are safe.

Modified:
  head/share/man/man4/aio.4

Modified: head/share/man/man4/aio.4
==
--- head/share/man/man4/aio.4   Fri Jul 22 19:36:11 2016(r303201)
+++ head/share/man/man4/aio.4   Fri Jul 22 20:52:40 2016(r303202)
@@ -54,7 +54,9 @@ the
 .Va vfs.aio.enable_unsafe
 sysctl node to a non-zero value.
 .Pp
-Asynchronous I/O operations on sockets and raw disk devices do not block
+Asynchronous I/O operations on sockets,
+raw disk devices,
+and regular files on local filesystems do not block
 indefinitely and are always enabled.
 .Pp
 The
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r301453 - in head/sys: arm/arm arm64/arm64 dev/fdt dev/gpio dev/iicbus dev/ofw dev/pci dev/vnic kern mips/mips sys

[Nathan Whitehorn]

On 07/22/16 13:19, John Baldwin wrote:

On Thursday, July 21, 2016 10:51:20 PM Nathan Whitehorn wrote:

On 07/21/16 14:35, John Baldwin wrote:

On Thursday, July 21, 2016 01:37:42 PM Andrew Turner wrote:

On Wed, 20 Jul 2016 13:28:53 +0200
Michal Meloun  wrote:

Dne 19.07.2016 v 17:06 Nathan Whitehorn napsal(a):

2. It partially duplicates the functionality of OFW_BUS_MAP_INTR(),
but is both problematically more general and less flexible (it has
requirements on timing of PIC attachment vs. driver resource
allocation)

OFW_BUS_MAP_INTR()  can parse only OFW  based data and expect that
parsed data are magicaly stored within the call.
The new method, bus_map_intr(),  can parse data from multiple sources
(OFW, UEFI / ACPI, synthetic[gpio device + pin number]).  It also
returns parsed data back to caller.
And no, it  doesn't  add any additional timing requirements .

I've been looking at ACPI on arm64. So far I have not found the need
for this with ACPI as we don't need to send the data to the interrupt
controller driver to be parsed in the way OFW/FDT needs to.

ACPI though has a gross hack where we call BUS_CONFIG_INTR on the IRQ
in bus_alloc_resource().

I hadn't realized that. It looks like you could do essentially the same
thing we do on PowerPC to clean this up by explicitly mapping the ACPI
interrupt domains to different PICs with varying default interrupt
properties.


What I had advocated in the discussions
leading up to this was to have some sort of opaque structure containing
a set of properties (the sort of thing bus_map_resource and make_dev_s
use) that was passed up at bus_setup_intr() time.
I think it should now
be passed up at bus_alloc_resource() time instead, but it would allow bus
drivers to "decorate" a SYS_RES_IRQ request as it goes up the device tree
with properties that the interrupt controller can then associate with
the IRQ cookie it allocates in its own code.


We used to do this on PPC and MIPS, and the current code still supports
it, but it turned out not to be useful in the end for IRQs. The
hierarchy for IRQs rarely (read: almost never) follows the bus hierarchy
and often is enumerated in a different order. I have hardware, for
example, where the children of a single parent bus are all wired to
different interrupt controllers and sometimes to a mixture of interrupt
controllers. Those controllers are cascaded in ways that cross the
newbus tree laterally and, on some of them, the parent device from the
bus topology has interrupts handled by its own (bus) children. Trying to
make the newbus parents do something sensible with all of this would be
crazy and, in the case where parents depend on resources provided by
their own children, impossible.

This is all to say that, since you want the interrupts to be decorated
along a path that usually has nothing to do with the newbus hierarchy,
it doesn't add much to add extra features to resource allocation.
ofw_bus_map_intr() is a newbus method to support this kind of thing but,
on all supported platforms, it is implemented only in nexus and no cases
have appeared where anyone ever wanted anything at the intermediate layers.

Mmm.  Another idea that has been bandied about is to create a separate
"plane" in new-bus for an interrupt hierarchy and allow devices to have
"interrupt" parents that are not the same as the "bus" parent.  (Additional
planes for power and clocks might also make sense.)  The idea is borrowed
from IOKit on Darwin which has multiple planes.  The "bus" plane is always
fully populated, but the other planes (Darwin has one for power for example)
can be sparse.  ACPI has methods that effectively describe the power plane
on x86.



That's basically what the virtual IRQ code does: it implements a 
separate IRQ "plane" that is lazily interconnected during bus attachment 
and then strongly interconnected at configure_final() (i.e. it panics at 
that point if the topology hasn't converged yet). A generalized version 
of this concept would be useful for a number of other resource types 
(power, clocks, GPIOs, etc.) and I would be very happy to see it as a 
standard part of the OS.

-Nathan
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r301453 - in head/sys: arm/arm arm64/arm64 dev/fdt dev/gpio dev/iicbus dev/ofw dev/pci dev/vnic kern mips/mips sys

[John Baldwin]

On Thursday, July 21, 2016 10:51:20 PM Nathan Whitehorn wrote:
> 
> On 07/21/16 14:35, John Baldwin wrote:
> > On Thursday, July 21, 2016 01:37:42 PM Andrew Turner wrote:
> >> On Wed, 20 Jul 2016 13:28:53 +0200
> >> Michal Meloun  wrote:
> >>> Dne 19.07.2016 v 17:06 Nathan Whitehorn napsal(a):
>  2. It partially duplicates the functionality of OFW_BUS_MAP_INTR(),
>  but is both problematically more general and less flexible (it has
>  requirements on timing of PIC attachment vs. driver resource
>  allocation)
> >>> OFW_BUS_MAP_INTR()  can parse only OFW  based data and expect that
> >>> parsed data are magicaly stored within the call.
> >>> The new method, bus_map_intr(),  can parse data from multiple sources
> >>> (OFW, UEFI / ACPI, synthetic[gpio device + pin number]).  It also
> >>> returns parsed data back to caller.
> >>> And no, it  doesn't  add any additional timing requirements .
> >> I've been looking at ACPI on arm64. So far I have not found the need
> >> for this with ACPI as we don't need to send the data to the interrupt
> >> controller driver to be parsed in the way OFW/FDT needs to.
> > ACPI though has a gross hack where we call BUS_CONFIG_INTR on the IRQ
> > in bus_alloc_resource().
> 
> I hadn't realized that. It looks like you could do essentially the same 
> thing we do on PowerPC to clean this up by explicitly mapping the ACPI 
> interrupt domains to different PICs with varying default interrupt 
> properties.
> 
> > What I had advocated in the discussions
> > leading up to this was to have some sort of opaque structure containing
> > a set of properties (the sort of thing bus_map_resource and make_dev_s
> > use) that was passed up at bus_setup_intr() time.
> 
> > I think it should now
> > be passed up at bus_alloc_resource() time instead, but it would allow bus
> > drivers to "decorate" a SYS_RES_IRQ request as it goes up the device tree
> > with properties that the interrupt controller can then associate with
> > the IRQ cookie it allocates in its own code.
> 
> 
> We used to do this on PPC and MIPS, and the current code still supports 
> it, but it turned out not to be useful in the end for IRQs. The 
> hierarchy for IRQs rarely (read: almost never) follows the bus hierarchy 
> and often is enumerated in a different order. I have hardware, for 
> example, where the children of a single parent bus are all wired to 
> different interrupt controllers and sometimes to a mixture of interrupt 
> controllers. Those controllers are cascaded in ways that cross the 
> newbus tree laterally and, on some of them, the parent device from the 
> bus topology has interrupts handled by its own (bus) children. Trying to 
> make the newbus parents do something sensible with all of this would be 
> crazy and, in the case where parents depend on resources provided by 
> their own children, impossible.
> 
> This is all to say that, since you want the interrupts to be decorated 
> along a path that usually has nothing to do with the newbus hierarchy, 
> it doesn't add much to add extra features to resource allocation. 
> ofw_bus_map_intr() is a newbus method to support this kind of thing but, 
> on all supported platforms, it is implemented only in nexus and no cases 
> have appeared where anyone ever wanted anything at the intermediate layers.

Mmm.  Another idea that has been bandied about is to create a separate
"plane" in new-bus for an interrupt hierarchy and allow devices to have
"interrupt" parents that are not the same as the "bus" parent.  (Additional
planes for power and clocks might also make sense.)  The idea is borrowed
from IOKit on Darwin which has multiple planes.  The "bus" plane is always
fully populated, but the other planes (Darwin has one for power for example)
can be sparse.  ACPI has methods that effectively describe the power plane
on x86.

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303188 - head/share/mk

[John Baldwin]

On Friday, July 22, 2016 03:00:38 PM Ruslan Bukin wrote:
> Author: br
> Date: Fri Jul 22 15:00:38 2016
> New Revision: 303188
> URL: https://svnweb.freebsd.org/changeset/base/303188
> 
> Log:
>   Add warn flags for GCC 6.1 compiler.
>   
>   Sponsored by:   DARPA, AFRL
> 
> Modified:
>   head/share/mk/bsd.sys.mk
> 
> Modified: head/share/mk/bsd.sys.mk
> ==
> --- head/share/mk/bsd.sys.mk  Fri Jul 22 14:57:26 2016(r303187)
> +++ head/share/mk/bsd.sys.mk  Fri Jul 22 15:00:38 2016(r303188)
> @@ -114,6 +114,11 @@ CWARNFLAGS+= -Wno-format
>  CWARNFLAGS+= -Wno-error=unused-function -Wno-error=enum-compare 
> -Wno-error=logical-not-parentheses -Wno-error=bool-compare 
> -Wno-error=uninitialized -Wno-error=array-bounds -Wno-error=clobbered 
> -Wno-error=cast-align -Wno-error=extra -Wno-error=attributes 
> -Wno-error=inline -Wno-error=unused-but-set-variable -Wno-error=unused-value 
> -Wno-error=strict-aliasing -Wno-error=address
>  .endif
>  
> +# GCC 6.1.0
> +.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
> +CWARNFLAGS+= -Wno-error=unused-const-variable= -Wno-error=nonnull-compare 
> -Wno-error=shift-negative-value -Wno-error=misleading-indentation 
> -Wno-error=tautological-compare
> +.endif

Extra trailing = in the first one as well?

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303190 - head/sys/conf

[John Baldwin]

On Friday, July 22, 2016 04:15:35 PM Ruslan Bukin wrote:
> Author: br
> Date: Fri Jul 22 16:15:35 2016
> New Revision: 303190
> URL: https://svnweb.freebsd.org/changeset/base/303190
> 
> Log:
>   Add GCC 6.1 warn flags for kernel as well.
>   
>   Sponsored by:   DARPA, AFRL
> 
> Modified:
>   head/sys/conf/kern.mk
> 
> Modified: head/sys/conf/kern.mk
> ==
> --- head/sys/conf/kern.mk Fri Jul 22 15:22:49 2016(r303189)
> +++ head/sys/conf/kern.mk Fri Jul 22 16:15:35 2016(r303190)
> @@ -50,6 +50,9 @@ CWARNEXTRA?=-Wno-error=inline -Wno-erro
>   -Wno-error=array-bounds -Wno-error=address \
>   -Wno-error=cast-qual -Wno-error=sequence-point 
> -Wno-error=attributes \
>   -Wno-error=strict-overflow -Wno-error=overflow
> +.if ${COMPILER_VERSION} >= 60100
> +CWARNEXTRA+= -Wno-error=nonnull-compare -Wno-error=shift-overflow=

Extra trailing =?

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303201 - stable/10/usr.bin/tr

[Andrey A. Chernov]

Author: ache
Date: Fri Jul 22 19:36:11 2016
New Revision: 303201
URL: https://svnweb.freebsd.org/changeset/base/303201

Log:
  MFC: r302827
  
  Optimize [Cc]flag case: don't repeatedly add the last character of
  string2 to squeeze cset when string2 reach its EOS state.

Modified:
  stable/10/usr.bin/tr/tr.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/usr.bin/tr/tr.c
==
--- stable/10/usr.bin/tr/tr.c   Fri Jul 22 19:34:43 2016(r303200)
+++ stable/10/usr.bin/tr/tr.c   Fri Jul 22 19:36:11 2016(r303201)
@@ -272,10 +272,11 @@ endloop:
if (Cflag && !iswrune(cnt))
continue;
if (cmap_lookup(map, cnt) == OOBCH) {
-   if (next())
+   if (next()) {
cmap_add(map, cnt, s2.lastch);
-   if (sflag)
-   cset_add(squeeze, s2.lastch);
+   if (sflag)
+   cset_add(squeeze, s2.lastch);
+   }
} else
cmap_add(map, cnt, cnt);
if ((s2.state == EOS || s2.state == INFINITE) &&
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303200 - stable/10/usr.bin/tr

[Andrey A. Chernov]

Author: ache
Date: Fri Jul 22 19:34:43 2016
New Revision: 303200
URL: https://svnweb.freebsd.org/changeset/base/303200

Log:
  MFC: r302826
  
  Document incomplete support of [=equiv=] and collation for ranges.

Modified:
  stable/10/usr.bin/tr/tr.1
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/usr.bin/tr/tr.1
==
--- stable/10/usr.bin/tr/tr.1   Fri Jul 22 19:28:23 2016(r303199)
+++ stable/10/usr.bin/tr/tr.1   Fri Jul 22 19:34:43 2016(r303200)
@@ -334,6 +334,10 @@ should be used instead of explicit chara
 and
 .Dq Li A-Z .
 .Pp
+.Dq Li [=equiv=]
+expression and collation for ranges
+are implemented for single byte locales only.
+.Pp
 System V has historically implemented character ranges using the syntax
 .Dq Li [c-c]
 instead of the
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303199 - head/usr.sbin/ctld

[Navdeep Parhar]

Author: np
Date: Fri Jul 22 19:28:23 2016
New Revision: 303199
URL: https://svnweb.freebsd.org/changeset/base/303199

Log:
  ctld(8): Fix MaxBurstLength negotiation.
  
  The target must reply with the selected value of MaxBurstSize instead of
  just echoing back the initiator's offered value.
  
  Reviewed by:  mav@
  Sponsored by: Chelsio Communications
  Differential Revision:https://reviews.freebsd.org/D7278

Modified:
  head/usr.sbin/ctld/login.c

Modified: head/usr.sbin/ctld/login.c
==
--- head/usr.sbin/ctld/login.c  Fri Jul 22 17:36:40 2016(r303198)
+++ head/usr.sbin/ctld/login.c  Fri Jul 22 19:28:23 2016(r303199)
@@ -569,7 +569,7 @@ login_negotiate_key(struct pdu *request,
tmp = MAX_BURST_LENGTH;
}
conn->conn_max_burst_length = tmp;
-   keys_add(response_keys, name, value);
+   keys_add_int(response_keys, name, tmp);
} else if (strcmp(name, "FirstBurstLength") == 0) {
tmp = strtoul(value, NULL, 10);
if (tmp <= 0) {
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303198 - stable/10/sys/kern

[Konstantin Belousov]

Author: kib
Date: Fri Jul 22 17:36:40 2016
New Revision: 303198
URL: https://svnweb.freebsd.org/changeset/base/303198

Log:
  MFC r302919:
  In ptrace_vm_entry(), do not call vmspace_free() while owning a vm
  object lock.

Modified:
  stable/10/sys/kern/sys_process.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/sys/kern/sys_process.c
==
--- stable/10/sys/kern/sys_process.cFri Jul 22 17:34:58 2016
(r303197)
+++ stable/10/sys/kern/sys_process.cFri Jul 22 17:36:40 2016
(r303198)
@@ -389,7 +389,6 @@ ptrace_vm_entry(struct thread *td, struc
} while (0);
 
vm_map_unlock_read(map);
-   vmspace_free(vm);
 
pve->pve_fsid = VNOVAL;
pve->pve_fileid = VNOVAL;
@@ -434,6 +433,7 @@ ptrace_vm_entry(struct thread *td, struc
free(freepath, M_TEMP);
}
}
+   vmspace_free(vm);
if (error == 0)
CTR3(KTR_PTRACE, "PT_VM_ENTRY: pid %d, entry %d, start %p",
p->p_pid, pve->pve_entry, pve->pve_start);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303197 - stable/11/sys/kern

[Konstantin Belousov]

Author: kib
Date: Fri Jul 22 17:34:58 2016
New Revision: 303197
URL: https://svnweb.freebsd.org/changeset/base/303197

Log:
  MFC r302919:
  In ptrace_vm_entry(), do not call vmspace_free() while owning a vm
  object lock.
  
  Approved by:  re (gjb)

Modified:
  stable/11/sys/kern/sys_process.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/sys/kern/sys_process.c
==
--- stable/11/sys/kern/sys_process.cFri Jul 22 17:34:28 2016
(r303196)
+++ stable/11/sys/kern/sys_process.cFri Jul 22 17:34:58 2016
(r303197)
@@ -433,7 +433,6 @@ ptrace_vm_entry(struct thread *td, struc
} while (0);
 
vm_map_unlock_read(map);
-   vmspace_free(vm);
 
pve->pve_fsid = VNOVAL;
pve->pve_fileid = VNOVAL;
@@ -478,6 +477,7 @@ ptrace_vm_entry(struct thread *td, struc
free(freepath, M_TEMP);
}
}
+   vmspace_free(vm);
if (error == 0)
CTR3(KTR_PTRACE, "PT_VM_ENTRY: pid %d, entry %d, start %p",
p->p_pid, pve->pve_entry, pve->pve_start);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303196 - in stable/11: contrib/llvm/projects/libunwind/include contrib/llvm/projects/libunwind/src gnu/lib/libgcc

[Ed Maste]

Author: emaste
Date: Fri Jul 22 17:34:28 2016
New Revision: 303196
URL: https://svnweb.freebsd.org/changeset/base/303196

Log:
  MFC libunwind improvements
  
  r302450: libunwind: update to upstream snapshot r272680
  
  The key improvement is that it may be built without cross-unwinding
  support, which significantly reduces the stack space requirement.
  
  r302456: libunwind: enable only the native unwinder by default
  
  This significantly reduces stack space requirements, and runtimes
  require only native unwinding.
  
  r302475: libunwind: limit stack usage in unwind cursor
  
  This may be reworked upstream but in the interim should address the
  stack usage issue reported in the PR.
  
  r303016: llvm-libunwind: use conventional (non-Darwin) X86 register numbers
  
  For historical reasons Darwin/i386 has ebp and esp swapped in the
  eh_frame register numbering.  That is:
  
Darwin  Other
  Reg #eh_frameeh_frameDWARF
  ==
4ebp espesp
5esp ebpebp
  
  Although the UNW_X86_* constants are not supposed to be coupled to
  DWARF / eh_frame numbering they are currently conflated in LLVM
  libunwind, and thus we require the non-Darwin numbering.
  
  PR:   206384
  Approved by:  re (kib)
  Sponsored by: The FreeBSD Foundation

Modified:
  stable/11/contrib/llvm/projects/libunwind/include/__libunwind_config.h
  stable/11/contrib/llvm/projects/libunwind/include/libunwind.h
  stable/11/contrib/llvm/projects/libunwind/src/AddressSpace.hpp
  stable/11/contrib/llvm/projects/libunwind/src/CompactUnwinder.hpp
  stable/11/contrib/llvm/projects/libunwind/src/Registers.hpp
  stable/11/contrib/llvm/projects/libunwind/src/Unwind-EHABI.cpp
  stable/11/contrib/llvm/projects/libunwind/src/UnwindCursor.hpp
  stable/11/contrib/llvm/projects/libunwind/src/UnwindLevel1.c
  stable/11/contrib/llvm/projects/libunwind/src/UnwindRegistersRestore.S
  stable/11/contrib/llvm/projects/libunwind/src/UnwindRegistersSave.S
  stable/11/contrib/llvm/projects/libunwind/src/config.h
  stable/11/contrib/llvm/projects/libunwind/src/libunwind.cpp
  stable/11/gnu/lib/libgcc/Makefile
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/contrib/llvm/projects/libunwind/include/__libunwind_config.h
==
--- stable/11/contrib/llvm/projects/libunwind/include/__libunwind_config.h  
Fri Jul 22 17:31:14 2016(r303195)
+++ stable/11/contrib/llvm/projects/libunwind/include/__libunwind_config.h  
Fri Jul 22 17:34:28 2016(r303196)
@@ -17,4 +17,47 @@
 #define _LIBUNWIND_ARM_EHABI 0
 #endif
 
+#if defined(_LIBUNWIND_IS_NATIVE_ONLY)
+# if defined(__i386__)
+#  define _LIBUNWIND_TARGET_I386 1
+#  define _LIBUNWIND_CONTEXT_SIZE 8
+#  define _LIBUNWIND_CURSOR_SIZE 19
+# elif defined(__x86_64__)
+#  define _LIBUNWIND_TARGET_X86_64 1
+#  define _LIBUNWIND_CONTEXT_SIZE 21
+#  define _LIBUNWIND_CURSOR_SIZE 33
+# elif defined(__ppc__)
+#  define _LIBUNWIND_TARGET_PPC 1
+#  define _LIBUNWIND_CONTEXT_SIZE 117
+#  define _LIBUNWIND_CURSOR_SIZE 128
+# elif defined(__aarch64__)
+#  define _LIBUNWIND_TARGET_AARCH64 1
+#  define _LIBUNWIND_CONTEXT_SIZE 66
+#  define _LIBUNWIND_CURSOR_SIZE 78
+# elif defined(__arm__)
+#  define _LIBUNWIND_TARGET_ARM 1
+#  define _LIBUNWIND_CONTEXT_SIZE 60
+#  define _LIBUNWIND_CURSOR_SIZE 67
+# elif defined(__or1k__)
+#  define _LIBUNWIND_TARGET_OR1K 1
+#  define _LIBUNWIND_CONTEXT_SIZE 16
+#  define _LIBUNWIND_CURSOR_SIZE 28
+# elif defined(__riscv__)
+#  define _LIBUNWIND_TARGET_RISCV 1
+#  define _LIBUNWIND_CONTEXT_SIZE 128 /* XXX */
+#  define _LIBUNWIND_CURSOR_SIZE 140 /* XXX */
+# else
+#  error "Unsupported architecture."
+# endif
+#else // !_LIBUNWIND_IS_NATIVE_ONLY
+# define _LIBUNWIND_TARGET_I386 1
+# define _LIBUNWIND_TARGET_X86_64 1
+# define _LIBUNWIND_TARGET_PPC 1
+# define _LIBUNWIND_TARGET_AARCH64 1
+# define _LIBUNWIND_TARGET_ARM 1
+# define _LIBUNWIND_TARGET_OR1K 1
+# define _LIBUNWIND_CONTEXT_SIZE 128
+# define _LIBUNWIND_CURSOR_SIZE 140
+#endif // _LIBUNWIND_IS_NATIVE_ONLY
+
 #endif // LIBUNWIND_CONFIG_H__

Modified: stable/11/contrib/llvm/projects/libunwind/include/libunwind.h
==
--- stable/11/contrib/llvm/projects/libunwind/include/libunwind.h   Fri Jul 
22 17:31:14 2016(r303195)
+++ stable/11/contrib/llvm/projects/libunwind/include/libunwind.h   Fri Jul 
22 17:34:28 2016(r303196)
@@ -46,12 +46,12 @@ enum {
 };
 
 struct unw_context_t {
-  uint64_t data[128];
+  uint64_t data[_LIBUNWIND_CONTEXT_SIZE];
 };
 typedef struct unw_context_t unw_context_t;
 
 struct unw_cursor_t {
-  uint64_t data[140];
+  uint64_t data[_LIBUNWIND_CURSOR_SIZE];
 };
 typedef struct unw_cursor_t unw_cursor_t;
 
@@ -151,8 +151,8 @@ enum {
   UNW_X86_ECX = 1,
   UNW_X86_EDX = 2,
   

svn commit: r303195 - head/sys/conf

[Bryan Drewery]

Author: bdrewery
Date: Fri Jul 22 17:31:14 2016
New Revision: 303195
URL: https://svnweb.freebsd.org/changeset/base/303195

Log:
  Don't run find(1) for __MPATH with NO_MODULES set.
  
  It's a waste of time when it won't be used.
  
  Submitted by: bde
  MFC after:3 days

Modified:
  head/sys/conf/config.mk
  head/sys/conf/kern.pre.mk

Modified: head/sys/conf/config.mk
==
--- head/sys/conf/config.mk Fri Jul 22 17:25:28 2016(r303194)
+++ head/sys/conf/config.mk Fri Jul 22 17:31:14 2016(r303195)
@@ -54,7 +54,7 @@ KERN_OPTS!=cat ${KERNBUILDDIR}/opt*.h | 
 .export KERN_OPTS
 .endif
 
-.if !defined(__MPATH)
+.if !defined(NO_MODULES) && !defined(__MPATH)
 __MPATH!=find ${SYSDIR:tA}/ -name \*_if.m
 .export __MPATH
 .endif

Modified: head/sys/conf/kern.pre.mk
==
--- head/sys/conf/kern.pre.mk   Fri Jul 22 17:25:28 2016(r303194)
+++ head/sys/conf/kern.pre.mk   Fri Jul 22 17:31:14 2016(r303195)
@@ -192,7 +192,7 @@ SYSTEM_LD_TAIL= @${OBJCOPY} --strip-symb
 SYSTEM_DEP+= ${LDSCRIPT}
 
 # Calculate path for .m files early, if needed.
-.if !defined(__MPATH)
+.if !defined(NO_MODULES) && !defined(__MPATH)
 __MPATH!=find ${S:tA}/ -name \*_if.m
 .endif
 
@@ -212,7 +212,9 @@ MKMODULESENV+=  MODULES_OVERRIDE="${MODUL
 .if defined(DEBUG)
 MKMODULESENV+= DEBUG_FLAGS="${DEBUG}"
 .endif
+.if !defined(NO_MODULES)
 MKMODULESENV+= __MPATH="${__MPATH}"
+.endif
 
 # Architecture and output format arguments for objdump to convert image to
 # object file
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303194 - stable/10/sys/kern

[Konstantin Belousov]

Author: kib
Date: Fri Jul 22 17:25:28 2016
New Revision: 303194
URL: https://svnweb.freebsd.org/changeset/base/303194

Log:
  MFC r302893:
  Do not allow creation of char or block special nodes with VNOVAL dev_t.

Modified:
  stable/10/sys/kern/vfs_syscalls.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/sys/kern/vfs_syscalls.c
==
--- stable/10/sys/kern/vfs_syscalls.c   Fri Jul 22 17:24:36 2016
(r303193)
+++ stable/10/sys/kern/vfs_syscalls.c   Fri Jul 22 17:25:28 2016
(r303194)
@@ -1269,6 +1269,8 @@ kern_mknodat(struct thread *td, int fd, 
case S_IFCHR:
case S_IFBLK:
error = priv_check(td, PRIV_VFS_MKNOD_DEV);
+   if (error == 0 && dev == VNOVAL)
+   error = EINVAL;
break;
case S_IFMT:
error = priv_check(td, PRIV_VFS_MKNOD_BAD);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303193 - stable/11/libexec/rtld-elf

[Bryan Drewery]

Author: bdrewery
Date: Fri Jul 22 17:24:36 2016
New Revision: 303193
URL: https://svnweb.freebsd.org/changeset/base/303193

Log:
  MFC r302908:
  
Fix dlsym(RTLD_NEXT) handling to only return the next library in last 
library
cases.
  
  Approved by:  re (gjb)

Modified:
  stable/11/libexec/rtld-elf/rtld.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/libexec/rtld-elf/rtld.c
==
--- stable/11/libexec/rtld-elf/rtld.c   Fri Jul 22 17:22:37 2016
(r303192)
+++ stable/11/libexec/rtld-elf/rtld.c   Fri Jul 22 17:24:36 2016
(r303193)
@@ -3291,7 +3291,7 @@ do_dlsym(void *handle, const char *name,
   handle == RTLD_SELF) { /* ... caller included */
if (handle == RTLD_NEXT)
obj = globallist_next(obj);
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
res = symlook_obj(, obj);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r300349 - head

[Bryan Drewery]

On 5/20/16 6:32 PM, Bryan Drewery wrote:
> Author: bdrewery
> Date: Sat May 21 01:32:07 2016
> New Revision: 300349
> URL: https://svnweb.freebsd.org/changeset/base/300349
> 
> Log:
>   Enable and utilize WITHOUT_CROSS_COMPILER logic for external CC.
>   
>   This is a NOP.
>   

It turns out it isn't a NOP!

>   Reviewed by:brooks, bapt
>   Sponsored by:   EMC / Isilon Storage Division
>   Differential Revision:  https://reviews.freebsd.org/D6354
> 
> Modified:
>   head/Makefile.inc1
> 
> Modified: head/Makefile.inc1
> ==
> --- head/Makefile.inc1Sat May 21 01:32:04 2016(r300348)
> +++ head/Makefile.inc1Sat May 21 01:32:07 2016(r300349)
> @@ -91,6 +91,12 @@ X${BINUTIL}?=  ${CROSS_BINUTILS_PREFIX}${
>  X${BINUTIL}?=${${BINUTIL}}
>  .endif
>  .endfor
> +# If a full path to an external cross compiler is given, don't build
> +# a cross compiler.
> +.if ${XCC:N${CCACHE_BIN}:M/*}
> +MK_CROSS_COMPILER=   no
> +.endif

Now MK_CROSS_COMPILER=no is set.

> +
>  .include# don't depend on src.opts.mk doing it
>  .include "share/mk/src.opts.mk"  
>  

src.opts.mk has:

.if ${MK_CROSS_COMPILER} == "no"
MK_BINUTILS_BOOTSTRAP:= no
MK_CLANG_BOOTSTRAP:= no
MK_ELFTOOLCHAIN_BOOTSTRAP:= no
MK_GCC_BOOTSTRAP:= no
.endif

So binutils and elftoolchain don't get built.  In my SYSTEM_COMPILER
commit r300354 I had intended that binutils and elftoolchain continue to
build as they lack version checks like were added for the compiler.

I'll get some sort of fix in today for this.


> @@ -454,7 +460,7 @@ BFLAGS+=  -B${CROSS_BINUTILS_PREFIX}
>  .endif
>  
>  # External compiler needs sysroot and target flags.
> -.if ${XCC:N${CCACHE_BIN}:M/*} || ${MK_CROSS_COMPILER} == "no"
> +.if ${MK_CROSS_COMPILER} == "no"
>  .if !defined(CROSS_BINUTILS_PREFIX) || !exists(${CROSS_BINUTILS_PREFIX})
>  BFLAGS+= -B${WORLDTMP}/usr/bin
>  .endif
> @@ -478,7 +484,7 @@ XCFLAGS+= -target ${TARGET_TRIPLE}
>  .endif
>  XCFLAGS+=--sysroot=${WORLDTMP}
>  .else
> -.endif # ${XCC:M/*} || ${MK_CROSS_COMPILER} == "no"
> +.endif # ${MK_CROSS_COMPILER} == "no"
>  
>  .if !empty(BFLAGS)
>  XCFLAGS+=${BFLAGS}
> @@ -1714,9 +1720,7 @@ _elftctools=lib/libelftc \
>   usr.bin/elfcopy
>  .endif
>  
> -# If an full path to an external cross compiler is given, don't build
> -# a cross compiler.
> -.if ${XCC:N${CCACHE_BIN}:M/*} == "" && ${MK_CROSS_COMPILER} != "no"
> +.if ${MK_CROSS_COMPILER} != "no"
>  .if ${MK_CLANG_BOOTSTRAP} != "no"
>  _clang=  usr.bin/clang
>  _clang_libs= lib/clang
> 


-- 
Regards,
Bryan Drewery
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303192 - stable/11/sys/kern

[Konstantin Belousov]

Author: kib
Date: Fri Jul 22 17:22:37 2016
New Revision: 303192
URL: https://svnweb.freebsd.org/changeset/base/303192

Log:
  MFC r302893:
  Do not allow creation of char or block special nodes with VNOVAL dev_t.
  
  Approved by:  re (gjb)

Modified:
  stable/11/sys/kern/vfs_syscalls.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/sys/kern/vfs_syscalls.c
==
--- stable/11/sys/kern/vfs_syscalls.c   Fri Jul 22 16:48:39 2016
(r303191)
+++ stable/11/sys/kern/vfs_syscalls.c   Fri Jul 22 17:22:37 2016
(r303192)
@@ -1167,6 +1167,8 @@ kern_mknodat(struct thread *td, int fd, 
case S_IFCHR:
case S_IFBLK:
error = priv_check(td, PRIV_VFS_MKNOD_DEV);
+   if (error == 0 && dev == VNOVAL)
+   error = EINVAL;
break;
case S_IFMT:
error = priv_check(td, PRIV_VFS_MKNOD_BAD);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303191 - stable/10/lib/libc/locale

[Andrey A. Chernov]

Author: ache
Date: Fri Jul 22 16:48:39 2016
New Revision: 303191
URL: https://svnweb.freebsd.org/changeset/base/303191

Log:
  Forget to merge this changes in prev. MFC

Modified:
  stable/10/lib/libc/locale/collcmp.c

Modified: stable/10/lib/libc/locale/collcmp.c
==
--- stable/10/lib/libc/locale/collcmp.c Fri Jul 22 16:15:35 2016
(r303190)
+++ stable/10/lib/libc/locale/collcmp.c Fri Jul 22 16:48:39 2016
(r303191)
@@ -42,10 +42,12 @@ __FBSDID("$FreeBSD$");
 
 int __collate_range_cmp(char c1, char c2)
 {
-   static char s1[2], s2[2];
+   char s1[2], s2[2];
 
s1[0] = c1;
+   s1[1] = '\0';
s2[0] = c2;
+   s2[1] = '\0';
return (strcoll(s1, s2));
 }
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303190 - head/sys/conf

[Ruslan Bukin]

Author: br
Date: Fri Jul 22 16:15:35 2016
New Revision: 303190
URL: https://svnweb.freebsd.org/changeset/base/303190

Log:
  Add GCC 6.1 warn flags for kernel as well.
  
  Sponsored by: DARPA, AFRL

Modified:
  head/sys/conf/kern.mk

Modified: head/sys/conf/kern.mk
==
--- head/sys/conf/kern.mk   Fri Jul 22 15:22:49 2016(r303189)
+++ head/sys/conf/kern.mk   Fri Jul 22 16:15:35 2016(r303190)
@@ -50,6 +50,9 @@ CWARNEXTRA?=  -Wno-error=inline -Wno-erro
-Wno-error=array-bounds -Wno-error=address \
-Wno-error=cast-qual -Wno-error=sequence-point 
-Wno-error=attributes \
-Wno-error=strict-overflow -Wno-error=overflow
+.if ${COMPILER_VERSION} >= 60100
+CWARNEXTRA+=   -Wno-error=nonnull-compare -Wno-error=shift-overflow=
+.endif
 .else
 # For gcc 4.2, eliminate the too-often-wrong warnings about uninitialized vars.
 CWARNEXTRA?=   -Wno-uninitialized
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303189 - head/share/mk

[Ruslan Bukin]

Author: br
Date: Fri Jul 22 15:22:49 2016
New Revision: 303189
URL: https://svnweb.freebsd.org/changeset/base/303189

Log:
  Set the soft-float flag for assembly code as well.
  This fixes compilation with GCC 6.1.
  
  Sponsored by: DARPA, AFRL

Modified:
  head/share/mk/bsd.cpu.mk

Modified: head/share/mk/bsd.cpu.mk
==
--- head/share/mk/bsd.cpu.mkFri Jul 22 15:00:38 2016(r303188)
+++ head/share/mk/bsd.cpu.mkFri Jul 22 15:22:49 2016(r303189)
@@ -329,6 +329,7 @@ CFLAGS += -mfloat-abi=softfp
 
 .if ${MACHINE_CPUARCH} == "riscv"
 CFLAGS += -msoft-float
+ACFLAGS += -msoft-float
 .endif
 
 # NB: COPTFLAGS is handled in /usr/src/sys/conf/kern.pre.mk
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303187 - head/contrib/llvm/projects/libunwind/include

[Ruslan Bukin]

Author: br
Date: Fri Jul 22 14:57:26 2016
New Revision: 303187
URL: https://svnweb.freebsd.org/changeset/base/303187

Log:
  Set real values for context/cursor sizes for RISC-V to prevent static
  assertions.
  
  Reviewed by:  emaste
  Sponsored by: DARPA, AFRL

Modified:
  head/contrib/llvm/projects/libunwind/include/__libunwind_config.h

Modified: head/contrib/llvm/projects/libunwind/include/__libunwind_config.h
==
--- head/contrib/llvm/projects/libunwind/include/__libunwind_config.h   Fri Jul 
22 14:39:55 2016(r303186)
+++ head/contrib/llvm/projects/libunwind/include/__libunwind_config.h   Fri Jul 
22 14:57:26 2016(r303187)
@@ -50,8 +50,8 @@
 #  define _LIBUNWIND_MAX_REGISTER 32
 # elif defined(__riscv__)
 #  define _LIBUNWIND_TARGET_RISCV 1
-#  define _LIBUNWIND_CONTEXT_SIZE 128 /* XXX */
-#  define _LIBUNWIND_CURSOR_SIZE 140 /* XXX */
+#  define _LIBUNWIND_CONTEXT_SIZE 64
+#  define _LIBUNWIND_CURSOR_SIZE 76
 #  define _LIBUNWIND_MAX_REGISTER 96
 # else
 #  error "Unsupported architecture."
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303188 - head/share/mk

[Ruslan Bukin]

Author: br
Date: Fri Jul 22 15:00:38 2016
New Revision: 303188
URL: https://svnweb.freebsd.org/changeset/base/303188

Log:
  Add warn flags for GCC 6.1 compiler.
  
  Sponsored by: DARPA, AFRL

Modified:
  head/share/mk/bsd.sys.mk

Modified: head/share/mk/bsd.sys.mk
==
--- head/share/mk/bsd.sys.mkFri Jul 22 14:57:26 2016(r303187)
+++ head/share/mk/bsd.sys.mkFri Jul 22 15:00:38 2016(r303188)
@@ -114,6 +114,11 @@ CWARNFLAGS+=   -Wno-format
 CWARNFLAGS+=   -Wno-error=unused-function -Wno-error=enum-compare 
-Wno-error=logical-not-parentheses -Wno-error=bool-compare 
-Wno-error=uninitialized -Wno-error=array-bounds -Wno-error=clobbered 
-Wno-error=cast-align -Wno-error=extra -Wno-error=attributes -Wno-error=inline 
-Wno-error=unused-but-set-variable -Wno-error=unused-value 
-Wno-error=strict-aliasing -Wno-error=address
 .endif
 
+# GCC 6.1.0
+.if ${COMPILER_TYPE} == "gcc" && ${COMPILER_VERSION} >= 60100
+CWARNFLAGS+=   -Wno-error=unused-const-variable= -Wno-error=nonnull-compare 
-Wno-error=shift-negative-value -Wno-error=misleading-indentation 
-Wno-error=tautological-compare
+.endif
+
 # How to handle FreeBSD custom printf format specifiers.
 .if ${COMPILER_TYPE} == "clang" && ${COMPILER_VERSION} >= 30600
 FORMAT_EXTENSIONS= -D__printf__=__freebsd_kprintf__
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303146 - head/usr.bin/sed

[Pedro Giffuni]

On 07/22/16 02:13, Bruce Evans wrote:

On Fri, 22 Jul 2016, Alexey Dokuchaev wrote:


Log:
  sed(1): Appease older GCC.


"Appease" actually seems to be the correct wording here since gcc's
detection
of a variable that might be used unitialized seems to report one that is
not used uninitialized.


Isn't it also being dictated by style(9) and common sense? :)


You missed that this combines a style fix in previous gcc appeasement
(or just excessive paranoia) in one variable with appeasement for
another variable, since copying the previous appeasement would copy
its style bug.  The 2 variables are used in exactly the same limited
way.



Yes, the first one is just a style fix while I was there.

The oldpsanl bogusness was breaking the build with gcc42. gcc48+, clang
and coverity all agree it was a false positive.
It was likely a side effect of raising the WARNS level to 5.


Modified:
  head/usr.bin/sed/process.c

@@ -97,11 +97,12 @@ process(void)
 {
 struct s_command *cp;
 SPACE tspace;
-size_t oldpsl = 0;
+size_t oldpsl;
 char *p;
 int oldpsanl;

 p = NULL;
+oldpsanl = oldpsl = 0;


Multiple assignments on a single line is not very good style and is
probably
not KNF.  Here it is further from being good style since the variables
have different types.  Since both types are integral and the value is 0
the implicit type conversions don't change the value.  However, compilers
should warn about down-converting a size_t to an int unless they do the
analysis that this is safe because the value in the size_t is known to
fit in the int.



Doing the multiple assignment seemed natural and readable as both are
(perhaps equally bogus) initializations.

I did notice the different types after committing. Perhaps swapping the
assignment would have been preferable? I suspect the compiler manages to 
optimize out the casting.


Pedro.
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303186 - head/sys/arm/allwinner/a20

[Emmanuel Vadot]

Author: manu
Date: Fri Jul 22 14:39:55 2016
New Revision: 303186
URL: https://svnweb.freebsd.org/changeset/base/303186

Log:
  PC5 doesn't have mmc2 function.

Modified:
  head/sys/arm/allwinner/a20/a20_padconf.c

Modified: head/sys/arm/allwinner/a20/a20_padconf.c
==
--- head/sys/arm/allwinner/a20/a20_padconf.cFri Jul 22 14:24:17 2016
(r303185)
+++ head/sys/arm/allwinner/a20/a20_padconf.cFri Jul 22 14:39:55 2016
(r303186)
@@ -87,7 +87,7 @@ const static struct allwinner_pins a20_p
{"PC2",  2, 2,  {"gpio_in", "gpio_out", "nand0", "spi0", NULL, NULL, 
NULL, NULL}},
{"PC3",  2, 3,  {"gpio_in", "gpio_out", "nand0", NULL, NULL, NULL, 
NULL, NULL}},
{"PC4",  2, 4,  {"gpio_in", "gpio_out", "nand0", NULL, NULL, NULL, 
NULL, NULL}},
-   {"PC5",  2, 5,  {"gpio_in", "gpio_out", "nand0", "mmc2", NULL, NULL, 
NULL, NULL}},
+   {"PC5",  2, 5,  {"gpio_in", "gpio_out", "nand0", NULL, NULL, NULL, 
NULL, NULL}},
{"PC6",  2, 6,  {"gpio_in", "gpio_out", "nand0", "mmc2", NULL, NULL, 
NULL, NULL}},
{"PC7",  2, 7,  {"gpio_in", "gpio_out", "nand0", "mmc2", NULL, NULL, 
NULL, NULL}},
{"PC8",  2, 8,  {"gpio_in", "gpio_out", "nand0", "mmc2", NULL, NULL, 
NULL, NULL}},
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303185 - in stable/10/lib/libc: gen locale regex stdio

[Andrey A. Chernov]

Author: ache
Date: Fri Jul 22 14:24:17 2016
New Revision: 303185
URL: https://svnweb.freebsd.org/changeset/base/303185

Log:
  MFC: r302824
  
  1) Eliminate possibility to call __*collate_range_cmp() with inclomplete
  locale (which cause core dump) by removing whole 'table' argument
  by which it passed.
  
  2) Restore __collate_range_cmp() in __sccl().
  
  3) Collating [a-z] range in regcomp() works for single byte locales only
  (we can't do it for other ones). In previous state only first 256
  wide chars are considered and all others are just silently dropped from the
  range.

Modified:
  stable/10/lib/libc/gen/fnmatch.c
  stable/10/lib/libc/gen/glob.c
  stable/10/lib/libc/locale/collate.h
  stable/10/lib/libc/locale/collcmp.c
  stable/10/lib/libc/regex/regcomp.c
  stable/10/lib/libc/stdio/vfscanf.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/lib/libc/gen/fnmatch.c
==
--- stable/10/lib/libc/gen/fnmatch.cFri Jul 22 12:55:31 2016
(r303184)
+++ stable/10/lib/libc/gen/fnmatch.cFri Jul 22 14:24:17 2016
(r303185)
@@ -304,8 +304,8 @@ rangematch(pattern, test, flags, newp, p
 
if (table->__collate_load_error ?
c <= test && test <= c2 :
-  __collate_range_cmp(table, c, test) <= 0
-   && __collate_range_cmp(table, test, c2) <= 0
+  __wcollate_range_cmp(c, test) <= 0
+   && __wcollate_range_cmp(test, c2) <= 0
   )
ok = 1;
} else if (c == test)

Modified: stable/10/lib/libc/gen/glob.c
==
--- stable/10/lib/libc/gen/glob.c   Fri Jul 22 12:55:31 2016
(r303184)
+++ stable/10/lib/libc/gen/glob.c   Fri Jul 22 14:24:17 2016
(r303185)
@@ -836,8 +836,8 @@ match(Char *name, Char *pat, Char *paten
if ((*pat & M_MASK) == M_RNG) {
if (table->__collate_load_error ?
CHAR(c) <= CHAR(k) && CHAR(k) <= 
CHAR(pat[1]) :
-  __collate_range_cmp(table, 
CHAR(c), CHAR(k)) <= 0
-   && __collate_range_cmp(table, 
CHAR(k), CHAR(pat[1])) <= 0
+  __wcollate_range_cmp(CHAR(c), 
CHAR(k)) <= 0
+   && __wcollate_range_cmp(CHAR(k), 
CHAR(pat[1])) <= 0
   )
ok = 1;
pat += 2;

Modified: stable/10/lib/libc/locale/collate.h
==
--- stable/10/lib/libc/locale/collate.h Fri Jul 22 12:55:31 2016
(r303184)
+++ stable/10/lib/libc/locale/collate.h Fri Jul 22 14:24:17 2016
(r303185)
@@ -72,7 +72,8 @@ u_char*__collate_strdup(u_char *);
 u_char *__collate_substitute(struct xlocale_collate *, const u_char *);
 int__collate_load_tables(const char *);
 void   __collate_lookup(struct xlocale_collate *, const u_char *, int *, int 
*, int *);
-int__collate_range_cmp(struct xlocale_collate *, int, int);
+int__collate_range_cmp(char, char);
+int__wcollate_range_cmp(wchar_t, wchar_t);
 #ifdef COLLATE_DEBUG
 void   __collate_print_tables(void);
 #endif

Modified: stable/10/lib/libc/locale/collcmp.c
==
--- stable/10/lib/libc/locale/collcmp.c Fri Jul 22 12:55:31 2016
(r303184)
+++ stable/10/lib/libc/locale/collcmp.c Fri Jul 22 14:24:17 2016
(r303185)
@@ -33,20 +33,29 @@
 __FBSDID("$FreeBSD$");
 
 #include 
-#include 
+#include 
 #include "collate.h"
 
 /*
  * Compare two characters using collate
  */
 
-int __collate_range_cmp(struct xlocale_collate *table, int c1, int c2)
+int __collate_range_cmp(char c1, char c2)
 {
static char s1[2], s2[2];
 
s1[0] = c1;
s2[0] = c2;
-   struct _xlocale l = {{0}};
-   l.components[XLC_COLLATE] = (struct xlocale_component *)table;
-   return (strcoll_l(s1, s2, ));
+   return (strcoll(s1, s2));
+}
+
+int __wcollate_range_cmp(wchar_t c1, wchar_t c2)
+{
+   wchar_t s1[2], s2[2];
+
+   s1[0] = c1;
+   s1[1] = L'\0';
+   s2[0] = c2;
+   s2[1] = L'\0';
+   return (wcscoll(s1, s2));
 }

Modified: stable/10/lib/libc/regex/regcomp.c
==
--- stable/10/lib/libc/regex/regcomp.c  Fri Jul 22 12:55:31 2016
(r303184)
+++ stable/10/lib/libc/regex/regcomp.c  Fri Jul 22 14:24:17 2016
(r303185)
@@ -51,7 +51,6 @@ __FBSDID("$FreeBSD$");
 #include 

svn commit: r303184 - stable/11/contrib/llvm/tools/clang/lib/Driver

[Ed Maste]

Author: emaste
Date: Fri Jul 22 12:55:31 2016
New Revision: 303184
URL: https://svnweb.freebsd.org/changeset/base/303184

Log:
  MFC r303031: clang++: Always use --eh-frame-hdr on FreeBSD, even for -static
  
  FreeBSD uses LLVM's libunwind on FreeBSD/arm64 today (and we expect to
  use it more widely in the future) and it requires the EH frame segment
  in static binaries.
  
  Approved by:  re (kib)

Modified:
  stable/11/contrib/llvm/tools/clang/lib/Driver/Tools.cpp
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/contrib/llvm/tools/clang/lib/Driver/Tools.cpp
==
--- stable/11/contrib/llvm/tools/clang/lib/Driver/Tools.cpp Fri Jul 22 
06:21:03 2016(r303183)
+++ stable/11/contrib/llvm/tools/clang/lib/Driver/Tools.cpp Fri Jul 22 
12:55:31 2016(r303184)
@@ -7922,12 +7922,12 @@ void freebsd::Linker::ConstructJob(Compi
   if (IsPIE)
 CmdArgs.push_back("-pie");
 
+  CmdArgs.push_back("--eh-frame-hdr");
   if (Args.hasArg(options::OPT_static)) {
 CmdArgs.push_back("-Bstatic");
   } else {
 if (Args.hasArg(options::OPT_rdynamic))
   CmdArgs.push_back("-export-dynamic");
-CmdArgs.push_back("--eh-frame-hdr");
 if (Args.hasArg(options::OPT_shared)) {
   CmdArgs.push_back("-Bshareable");
 } else {
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r303146 - head/usr.bin/sed

[Bruce Evans]

On Fri, 22 Jul 2016, Alexey Dokuchaev wrote:


Log:
  sed(1): Appease older GCC.


"Appease" actually seems to be the correct wording here since gcc's detection
of a variable that might be used unitialized seems to report one that is
not used uninitialized.


Isn't it also being dictated by style(9) and common sense? :)


You missed that this combines a style fix in previous gcc appeasement
(or just excessive paranoia) in one variable with appeasement for
another variable, since copying the previous appeasement would copy
its style bug.  The 2 variables are used in exactly the same limited
way.


Modified:
  head/usr.bin/sed/process.c

@@ -97,11 +97,12 @@ process(void)
 {
struct s_command *cp;
SPACE tspace;
-   size_t oldpsl = 0;
+   size_t oldpsl;
char *p;
int oldpsanl;

p = NULL;
+   oldpsanl = oldpsl = 0;


Multiple assignments on a single line is not very good style and is probably
not KNF.  Here it is further from being good style since the variables
have different types.  Since both types are integral and the value is 0
the implicit type conversions don't change the value.  However, compilers
should warn about down-converting a size_t to an int unless they do the
analysis that this is safe because the value in the size_t is known to
fit in the int.

Bruce

___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r303183 - head/sys/modules/cam

[Konstantin Belousov]

Author: kib
Date: Fri Jul 22 06:21:03 2016
New Revision: 303183
URL: https://svnweb.freebsd.org/changeset/base/303183

Log:
  Make cam.ko loadable.
  
  Sponsored by: The FreeBSD Foundation
  MFC after:1 week

Modified:
  head/sys/modules/cam/Makefile

Modified: head/sys/modules/cam/Makefile
==
--- head/sys/modules/cam/Makefile   Fri Jul 22 05:09:08 2016
(r303182)
+++ head/sys/modules/cam/Makefile   Fri Jul 22 06:21:03 2016
(r303183)
@@ -2,7 +2,7 @@
 
 S= ${.CURDIR}/../..
 
-.PATH: $S/cam $S/cam/scsi $S/cam/ata $S/${MACHINE}/${MACHINE}
+.PATH: $S/cam $S/cam/scsi $S/cam/ata $S/cam/nvme $S/${MACHINE}/${MACHINE}
 
 KMOD=  cam
 
@@ -41,6 +41,8 @@ SRCS+=ata_da.c
 SRCS+= ata_machdep.c
 .endif
 SRCS+= ata_pmp.c
+SRCS+= nvme_all.c
+SRCS+= nvme_xpt.c
 
 EXPORT_SYMS=   YES # XXX evaluate
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"