[Bug 223158] need driver it(4) from OpenBSD for ITE87xx

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223158

Bug ID: 223158
   Summary: need driver it(4) from OpenBSD for ITE87xx
   Product: Base System
   Version: 11.1-RELEASE
  Hardware: Any
OS: Any
Status: New
  Severity: Affects Only Me
  Priority: ---
 Component: kern
  Assignee: freebsd-bugs@FreeBSD.org
  Reporter: okolit...@gmail.com

need driver it(4) from OpenBSD for ITE87xx

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223165] Crash with pthread_condattr_destroy(NULL) on freebsd-11

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223165

Bug ID: 223165
   Summary: Crash with pthread_condattr_destroy(NULL) on
freebsd-11
   Product: Base System
   Version: 11.0-STABLE
  Hardware: amd64
OS: Any
Status: New
  Severity: Affects Many People
  Priority: ---
 Component: kern
  Assignee: freebsd-bugs@FreeBSD.org
  Reporter: rajendra...@gmail.com

Below program crashes on FreeBSD-11.

$ cat test.c
#include 
int main(int argc, char *argv[]) {
pthread_condattr_destroy(NULL);
return 0;
}
$ clang -o test test.c -lpthread
test.c:3:31: warning: null passed to a callee that requires a non-null argument
[-Wnonnull]
pthread_condattr_destroy(NULL);
 ^
1 warning generated.

syrajendra@[bm64-fbsd11] # ./test
Segmentation fault (core dumped)


The "pthread_condattr_destroy" has a argument which has _Nonnull quilifier but
the function implementation properly handles the NULL value by returning
"EINVAL" if argument is NULL.

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 192394] Base system bootstrapping broken with gcc; not clang

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=192394

Ngie Cooper  changed:

   What|Removed |Added

   Assignee|freebsd-bugs@FreeBSD.org|bdrew...@freebsd.org
 Resolution|--- |FIXED
 Status|New |Closed

--- Comment #1 from Ngie Cooper  ---
Bryan fixed this a year or two ago.

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223168] [patch] Support for iBook G4 German keyboard in syscons.

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223168

Bug ID: 223168
   Summary: [patch] Support for iBook G4 German keyboard in
syscons.
   Product: Base System
   Version: 9.3-RELEASE
  Hardware: powerpc
OS: Any
Status: New
  Keywords: patch
  Severity: Affects Only Me
  Priority: ---
 Component: misc
  Assignee: freebsd-bugs@FreeBSD.org
  Reporter: xelalex_ma...@web.de
  Keywords: patch

Created attachment 187368
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=187368&action=edit
Syscons-keymap for German iBook G4.

The current keymaps for generic German keyboard layouts don't work too well on
my old iBook G4 (12"/800MHz).

I modified german.iso.acc.kbd to give the following results:

- The degree (°) works,
- the caret works as a character as well as an accent,
- curly, square and angle brackets, as well as @ and | are located in the usual
positions for Apple-machines,
- Alt+u works as an umlaut-accent,
- Alt+a works as an overring-accent,
- Alt+n works as a tilde-accent,
- Alt+c works as a cedille-accent.

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"

[Bug 223165] Crash with pthread_condattr_destroy(NULL) on freebsd-11

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223165

Conrad Meyer  changed:

   What|Removed |Added

 CC||c...@freebsd.org

--- Comment #1 from Conrad Meyer  ---
Does not seem to affect CURRENT, FWIW.

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 178062] make depend broken on CURRENT with lib/libgpib for clang/gcc

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=178062

Ngie Cooper  changed:

   What|Removed |Added

 Status|In Progress |Closed
 Resolution|--- |Unable to Reproduce

--- Comment #1 from Ngie Cooper  ---
This happened because the headers weren't being installed and it wasn't looking
for them under ${SRCTOP}. Not a bug with FreeBSD: it was a bug with the
FreeBSD-fork (Isilon OneFS).

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 221337] -fsanitize=address (asan) fails on i386

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=221337

Dimitry Andric  changed:

   What|Removed |Added

   Assignee|freebsd-bugs@FreeBSD.org|d...@freebsd.org
 Status|Open|In Progress

--- Comment #10 from Dimitry Andric  ---
Created attachment 187379
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=187379&action=edit
Ensure alignment of jemalloc's TSD is 8 bytes on 32-bit platforms

Here is a possible way of fixing this issue: it ensures that the big tsd_t
struct in contrib/jemalloc/src/tsd.c is aligned to 8 bytes, on !LP64 platforms
(i.e. i386, but it could also apply to arm or mips, if AddressSanitizer ever
gets to work there).

Before r319971, on i386 the TLS items in libc.so.7 added up to 80 bytes of
storage (a multiple of 8 bytes), assuming __je_tsd_initialized is aligned at 4
bytes:

$ readelf -sW /usr/obj/head-r319970/usr/src/lib/libc/libc.so.7|grep -w TLS
  1139: 0058 4 TLS GLOBAL DEFAULT   17 _ThreadRuneLocale@@FBSD_1.3
   547: 0054 4 TLS LOCAL  DEFAULT   17 __thread_locale
   607: 64 TLS LOCAL  DEFAULT   16 __je_tsd_tls
   608: 004c 1 TLS LOCAL  DEFAULT   17 __je_tsd_initialized
  3088: 0058 4 TLS GLOBAL DEFAULT   17 _ThreadRuneLocale

With r319971, this increased quite a lot, to 2404 bytes (which is not a
multiple of 8 bytes anymore, unfortunately):

$ readelf -sW /usr/obj/head-r319971/usr/src/lib/libc/libc.so.7|grep -w TLS
  1139: 096c 4 TLS GLOBAL DEFAULT   17 _ThreadRuneLocale@@FBSD_1.3
   441: 0968 4 TLS LOCAL  DEFAULT   17 __thread_locale
   500:   2388 TLS LOCAL  DEFAULT   16 __je_tsd_tls
   502: 0960 1 TLS LOCAL  DEFAULT   17 __je_tsd_initialized
  2845: 096c 4 TLS GLOBAL DEFAULT   17 _ThreadRuneLocale

All the growth is in the __je_tsd_tls struct, as you can see.  If this struct
is forced to align to 8 bytes, the total amount of TLS data also becomes
aligned to 8 bytes.

An alternative would be to unconditionally align the struct at, say, 16 bytes,
and get rid of the #ifdef.

Yet another alternative would be to place some bogus padding 4 byte entity
somewhere else in libc.so to ensure the TLS data is a multiple of 8 bytes. 
Suggestions as to where are welcome. :)

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223151] [patch] provide a way to get initial interface name

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223151

Ed Maste  changed:

   What|Removed |Added

   Keywords||patch
Summary|provide a way to get|[patch] provide a way to
   |initial interface name  |get initial interface name
 CC||ema...@freebsd.org

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223177] ipfw table delete returns non-zero exit status even with -q option specified

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223177

Bug ID: 223177
   Summary: ipfw table delete returns non-zero exit status even
with -q option specified
   Product: Base System
   Version: 11.1-RELEASE
  Hardware: Any
OS: Any
Status: New
  Severity: Affects Some People
  Priority: ---
 Component: kern
  Assignee: freebsd-bugs@FreeBSD.org
  Reporter: ab...@f-t-w.wtf

Behaviour of 'ipfw -q table 1 delete 1.2.3.4' changed between 10.3-RELEASE and
11.1-RELEASE. In 11.1, the -q option is apparently ignored and an attempt to
delete a nonexistent entry from the table returns an error message and non-zero
exit status.  In 10.3, the -q option suppressed the error message and always
returned a zero exit status.   This change breaks makefiles and shell scripts
that depended on the exit status being zero and no error output always.

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223177] ipfw table entry delete returns non-zero exit status even with -q option specified

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223177

Brian  changed:

   What|Removed |Added

Summary|ipfw table delete returns   |ipfw table entry delete
   |non-zero exit status even   |returns non-zero exit
   |with -q option specified|status even with -q option
   ||specified

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223178] netinet/raw_ip.c undocumented change in byte order.

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223178

Bug ID: 223178
   Summary: netinet/raw_ip.c undocumented change in byte order.
   Product: Base System
   Version: 11.1-RELEASE
  Hardware: Any
OS: Any
Status: New
  Severity: Affects Some People
  Priority: ---
 Component: kern
  Assignee: freebsd-bugs@FreeBSD.org
  Reporter: ab...@f-t-w.wtf

netinet/raw_ip.c changed from host byte order to network byte order for ip_len
and ip_off from 10.x to 11.x breaking programs doing raw socket I/O.  This
change was not documented in release notes or man pages.

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223179] lib/libcompiler_rt fails to compile with gcc; some flags are being applied incorrectly

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223179

Bug ID: 223179
   Summary: lib/libcompiler_rt fails to compile with gcc; some
flags are being applied incorrectly
   Product: Base System
   Version: CURRENT
  Hardware: Any
OS: Any
Status: New
  Severity: Affects Some People
  Priority: ---
 Component: bin
  Assignee: freebsd-bugs@FreeBSD.org
  Reporter: n...@freebsd.org

Per gcc(1), the following clangs (minus -fno-sanitize=...) are c++ flags, and
are being applied to CFLAGS instead of CXXFLAGS.

===> lib/libclang_rt/profile (obj,all,install)
cc1: warning: command line option "-fno-rtti" is valid for C++/ObjC++ but not
for C
cc1: error: unrecognized command line option "-fno-sanitize=safe-stack"
cc1: warning: command line option "-fvisibility-inlines-hidden" is valid for
C++/ObjC++ but not for C

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223179] lib/libcompiler_rt fails to compile with gcc; some flags are being applied incorrectly

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223179

--- Comment #1 from commit-h...@freebsd.org ---
A commit references this bug:

Author: ngie
Date: Mon Oct 23 04:22:17 UTC 2017
New revision: 324894
URL: https://svnweb.freebsd.org/changeset/base/324894

Log:
  Support compiling lib/libclang_rt/profile with gcc

  Several of the flags were being treated as CFLAGS, when they were actually
  technically CXXFLAGS. Move them there.

  Also, only apply -fno-sanitize=safe-stack with clang.

  This is a draft diff.

  PR:   223179

Changes:
  projects/runtime-coverage/lib/libclang_rt/Makefile.inc

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"


[Bug 223179] lib/libcompiler_rt fails to compile with gcc; some flags are being applied incorrectly

2017-10-22 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223179

--- Comment #2 from commit-h...@freebsd.org ---
A commit references this bug:

Author: ngie
Date: Mon Oct 23 05:11:01 UTC 2017
New revision: 324899
URL: https://svnweb.freebsd.org/changeset/base/324899

Log:
  -fvisibility is a c++ thing, per gcc(1)

  Followup to r324894

  PR:   223179

Changes:
  projects/runtime-coverage/lib/libclang_rt/profile/Makefile

-- 
You are receiving this mail because:
You are the assignee for the bug.
___
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"