hi,
> Module Name: src
> Committed By: christos
> Date: Wed Jan 25 00:28:36 UTC 2012
>
> Modified Files:
> src/lib/libc/sys: dup.2 fcntl.2 getsockopt.2 kqueue.2 open.2 pipe.2
> socket.2
> src/sys/kern: kern_descrip.c kern_event.c sys_descrip.c sys_generic.c
>
hi,
> On Jan 24, 3:20am, y...@mwd.biglobe.ne.jp (YAMAMOTO Takashi) wrote:
> -- Subject: Re: Adding an option to avoid SIGPIPE for all file descriptors
>
> | please don't forget compat_netbsd32 copy.
>
> I don't see anything to do there? What should I do?
i meant dofilewritev32.
YAMAMOTO Takas
On Tue 24 Jan 2012 at 21:33:44 +, Michael van Elst wrote:
> Unfortunately noone did ever think about a SIGNOSPACE and that's
> why every Unix system now requires an infinite disk.
Well, that does sound like a resource limit being exceeded, so why
not send
25SIGXFSZ terminate pr
On Tue, Jan 24, 2012 at 11:30:54PM +0100, Joerg Sonnenberger wrote:
> > > > There is also a related PR but which is for threads stack alignment:
> > > > lib/39465
> > >
> > > That bug is wrong. NetBSD uses SYSV ABI and that mandates 4 Bytes
> > > alignment for the stack. GCC versions bef
On Tue, Jan 24, 2012 at 10:13:39PM +, David Holland wrote:
> On Tue, Jan 24, 2012 at 09:55:30PM +0100, Joerg Sonnenberger wrote:
> > > There is also a related PR but which is for threads stack alignment:
> > > lib/39465
> >
> > That bug is wrong. NetBSD uses SYSV ABI and that mandates 4 By
On Tue, Jan 24, 2012 at 09:55:30PM +0100, Joerg Sonnenberger wrote:
> > There is also a related PR but which is for threads stack alignment:
> > lib/39465
>
> That bug is wrong. NetBSD uses SYSV ABI and that mandates 4 Bytes
> alignment for the stack. GCC versions before ~4.5 or so are just
On Tue, Jan 24, 2012 at 04:53:53PM -0500, Matthew Mondor wrote:
> It's in the x86-64 case that stack frames are 16-byte aligned, with
> arrays larger than 16 bytes also needing to be 16-byte aligned
> (possibly including the stack)...
Depending on when you look, the stack frame is either 8 mod 16
On Tue, 24 Jan 2012 21:55:30 +0100
Joerg Sonnenberger wrote:
> On Tue, Jan 24, 2012 at 03:30:37PM -0500, Matthew Mondor wrote:
> > There is also a related PR but which is for threads stack alignment:
> > lib/39465
>
> That bug is wrong. NetBSD uses SYSV ABI and that mandates 4 Bytes
> alignment
t...@panix.com (Thor Lancelot Simon) writes:
>On Tue, Jan 24, 2012 at 03:07:45AM +, David Holland wrote:
>> On Mon, Jan 23, 2012 at 09:28:48PM -0500, Christos Zoulas wrote:
>> > [stuff]
>>
>> One thing jumps out at me: if it's a file-level flag, shouldn't it be
>> sufficient to use F_GETFL /
On 2012-01-24, at 12:01 PM, Martin Husemann wrote:
> Can anyone explain why arm would need 8 byte alignment?
The ARM EABI requires 8-byte stack alignment at function entry. I don't know if
we're using the EABI, but this bug and its fix might indicate as much.
Steve
In article ,
Paul Fleischer wrote:
>Hello there,
>
>I am currently working on a NetBSD port for the FriendlyARM MINI2440,
>and have run into a situation where the arguments to user space
>programs is garbled.
>I've noticed it for the init-process and for the getty process.
>Booting with rc_config
On Tue, Jan 24, 2012 at 03:30:37PM -0500, Matthew Mondor wrote:
> There is also a related PR but which is for threads stack alignment:
> lib/39465
That bug is wrong. NetBSD uses SYSV ABI and that mandates 4 Bytes
alignment for the stack. GCC versions before ~4.5 or so are just
completely broken in
On Tue, Jan 24, 2012 at 03:30:37PM -0500, Matthew Mondor wrote:
> Would this be considered wasteful? Of course, x86-64 MD code could
> also be used...
I would prefer the x86 MD code, using the same technique as the (fixed)
arm code.
Maybe it could even depend on actual CPU type (i.e. SSE2 availa
On Tue, 24 Jan 2012 21:01:49 +0100
Martin Husemann wrote:
> On Tue, Jan 24, 2012 at 08:21:42PM +0100, Paul Fleischer wrote:
> > Is the usage of STACKALIGN indeed incorrect in this situation, or am I
> > missing the big picture?
>
> I stumbled across this when revamping execve1 for posix_spawn re
On Tue, Jan 24, 2012 at 12:16:49PM -0800, Stephen M. Rumble wrote:
> The ARM EABI requires 8-byte stack alignment at function entry. I don't know
> if we're using the EABI, but this bug and its fix might indicate as much.
I don't think we do yet, but we should ;-)
Anyway, let's make a up-roundin
On Tue, Jan 24, 2012 at 08:21:42PM +0100, Paul Fleischer wrote:
> Is the usage of STACKALIGN indeed incorrect in this situation, or am I
> missing the big picture?
I stumbled across this when revamping execve1 for posix_spawn recently.
The intention seems to be to align the stack on a 8 byte boun
On Tue, Jan 24, 2012 at 11:16:00AM -0600, David Young wrote:
> Does NetBSD run on any processor architectures where it is difficult
> or impossible in the kernel[1] to provide non-interlocked atomic
> operations? That is, operations that are atomic with respect to other
> operations *on the same p
Hello there,
I am currently working on a NetBSD port for the FriendlyARM MINI2440,
and have run into a situation where the arguments to user space
programs is garbled.
I've noticed it for the init-process and for the getty process.
Booting with rc_configured=NO in /etc/rc.conf, mounting procfs, an
On Jan 24, 5:19pm, dholland-t...@netbsd.org (David Holland) wrote:
-- Subject: Re: Adding an option to avoid SIGPIPE for all file descriptors
| On Mon, Jan 23, 2012 at 10:43:38PM -0500, Christos Zoulas wrote:
| > | On Mon, Jan 23, 2012 at 09:28:48PM -0500, Christos Zoulas wrote:
| > | > [stuff
On Mon, Jan 23, 2012 at 10:43:38PM -0500, Christos Zoulas wrote:
> | On Mon, Jan 23, 2012 at 09:28:48PM -0500, Christos Zoulas wrote:
> | > [stuff]
> |
> | One thing jumps out at me: if it's a file-level flag, shouldn't it be
> | sufficient to use F_GETFL / F_SETFL to manipulate it, rather t
Does NetBSD run on any processor architectures where it is difficult
or impossible in the kernel[1] to provide non-interlocked atomic
operations? That is, operations that are atomic with respect to other
operations *on the same processor*, but possibly divisible by operations
on other processors?
On Jan 24, 3:20am, y...@mwd.biglobe.ne.jp (YAMAMOTO Takashi) wrote:
-- Subject: Re: Adding an option to avoid SIGPIPE for all file descriptors
| please don't forget compat_netbsd32 copy.
I don't see anything to do there? What should I do?
Thanks,
christos
I wrote:
> So, are there any recommendations for current Opteron motherboards with good
> NetBSD support?
Since I didn't receive any specific suggestions, may I ask whether anyone is
running
NetBSD on either a Supermicro H8SCM-F or a Tyan S8010?
On Mon, Jan 23, 2012 at 10:58:59PM -0500, Steven Bellovin wrote:
> > I also wonder whether we should also have a note that disabled SIGPIPE.
> > similar to what paxctl does.
> >
> You mean a system-wide flag? That would worry me; I think it would have
> bad effects, since anything that did
>
>
24 matches
Mail list logo
