On Wed, Jan 18, 2006 at 01:19:21AM +0100, Paolo 'Blaisorblade' Giarrusso wrote:
> Done by Jeff around January for ubd only, later lost, then restored in his
> tree
> - however I'm merging it now since there's no reason to leave this here.
The original was dinged by hch for covering over a problem
> From: Jeff Dike <[EMAIL PROTECTED]>, Paolo 'Blaisorblade' Giarrusso <[EMAIL
> PROTECTED]>
>
> Define a release method for the ubd and network driver so that sysfs doesn't
> complain when one is removed via:
What? No. The kernel is complaining for a reason, don't try to
out-smart it.
>
> ho
"Paolo 'Blaisorblade' Giarrusso" <[EMAIL PROTECTED]> wrote:
>
> +int __cant_sleep(void) {
> + return in_atomic() || irqs_disabled() || in_interrupt();
aww, man, this is awful. Code is supposed to know what context it's
running in, not go fishing about in core internals trying to fix up its ow
"Paolo 'Blaisorblade' Giarrusso" <[EMAIL PROTECTED]> wrote:
>
>
> From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
>
> In this error path, when the interface has had a problem, we call dev_close(),
> which is disallowed for two reasons:
>
> *) takes again the UML internal spinlock, insid
From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
Avoid uninitialized data in the daemon_data structure. I used this transport
before doing proper setup before-hand, and I got some very nice SLAB corruption
due to freeing crap pointers. So just make sure to clear everything when
appropriat
*) mark as "EXPERIMENTAL" various items that either aren't very stable or that
are actively crashing the setup of users which don't really need them (i.e.
HIGHMEM and 3-level pagetables on x86 - nobody needs either, everybody reports
"I'm using it and getting trouble").
*) move net/Kconfig near t
When the user specifies both a COW file and its backing file, if the previous
backing file is not found, currently UML tries again to use it and fails.
This can be corrected by changing same_backing_files() return value in that
case, so that the caller will try to change the COW file to point to
From: Jeff Dike <[EMAIL PROTECTED]>, Paolo 'Blaisorblade' Giarrusso <[EMAIL
PROTECTED]>
Define a release method for the ubd and network driver so that sysfs doesn't
complain when one is removed via:
host $ uml_mconsole remove
Done by Jeff around January for ubd only, later lost, then restore
From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
Pre-clear transport-specific private structure before passing it down.
In fact, I just got a slab corruption and kernel panic on exit because kfree()
was called on a pointer which probably was never allocated, BUT hadn't been set
to NULL b
Correct a bit of whitespace problems while working here.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
---
arch/um/drivers/ubd_kern.c | 31 +--
1 files changed, 17 insertions(+), 14 deletions(-)
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um
From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
Ugly trick to help make malloc not sleeping - we can't do anything else. But
this is not yet optimal, since spinlock don't trigger in_atomic() when
preemption is disabled.
Also, even if ugly, this was already used in one place, and was eve
From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
In a previous patch I shifted an allocation to being atomic.
In this patch, a better but more intrusive solution is implemented, i.e. hold
the lock only when really needing it, especially not over pipe operations, nor
over the culprit allo
From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
In this error path, when the interface has had a problem, we call dev_close(),
which is disallowed for two reasons:
*) takes again the UML internal spinlock, inside the ->stop method of this
device
*) can be called in process context on
Merge various fixes, mainly for locking issues.
A bit of attention is to give to "uml: avoid malloc to sleep in atomic
sections". It's really an hack, but we already use this hack in a more buggy
way (see the patch).
In practice, when malloc() is called, it's translated in
a call to kmalloc(GFP_?
On Wed, Jan 18, 2006 at 12:26:27AM +0100, Paolo 'Blaisorblade' Giarrusso wrote:
>
> From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
>
> I added this line to share this file with UML, but now it's no longer shared
> so
> remove this useless leftover.
ACK - Thanks, I saw this and meant t
Clear-Text: http://bugs.gentoo.org/show_bug.cgi?id=111575
Secure: https://bugs.gentoo.org/show_bug.cgi?id=111575
[EMAIL PROTECTED] changed:
What|Removed |Added
CC|
From: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
I added this line to share this file with UML, but now it's no longer shared so
remove this useless leftover.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <[EMAIL PROTECTED]>
---
include/asm-i386/futex.h |2 +-
1 files changed, 1 in
On Tuesday 17 January 2006 20:52, Daniel Drake wrote:
> Blaisorblade wrote:
> > Vanilla can do, however -bs has one little additional feature
> > (mconsole-exec) and contains all bug fixes written after the release
> > (only few important ones go to -stable, they're very restrictive and some
> > -b
On Tue, Jan 17, Blaisorblade wrote:
> Disable CONFIG_SMP, it doesn't work well anyway, however thanks for the
> report.
Some of the libc headers my define it, its not enabled in the kernel
.config.
The patch gets me further, thanks. Next error is:
arch/um/os-Linux/sys-i386/registers.c: In func
Blaisorblade wrote:
Vanilla can do, however -bs has one little additional feature (mconsole-exec)
and contains all bug fixes written after the release (only few important ones
go to -stable, they're very restrictive and some -bs patches don't fix
critical "regressions"/crashes).
In fact, most
On Tuesday 17 January 2006 04:32, Jeff Dike wrote:
> On Tue, Jan 17, 2006 at 01:24:31AM +0100, Blaisorblade wrote:
> > ~25 %? Good! Which is delay vs. host?
>
> Delay vs a UML without the patch.
>
> > A curiosity - did you look at the similar code in Ingo Molnar's VCPU
> > patch? I never found the
On Tuesday 17 January 2006 04:45, Jeff Dike wrote:
> On Tue, Jan 17, 2006 at 02:22:59AM +0100, Blaisorblade wrote:
> > On Monday 16 January 2006 18:14, Olaf Hering wrote:
> > > It picks up the libc headers instead of the 2.6.15 ones.
> >
> > Yep, that's correct, the bug is that shouldn't be using C
On Tuesday 17 January 2006 04:41, Jeff Dike wrote:
> On Tue, Jan 17, 2006 at 01:26:12AM +0100, Blaisorblade wrote:
> > I ran by accident scripts/checkstack.pl and found that slip_init and
> > slirp_init are two big offenders.
> >
> > They are since of such lines:
> >
> > *spri = ((struct sl
On Tuesday 17 January 2006 04:37, Jeff Dike wrote:
> On Tue, Jan 17, 2006 at 02:33:59AM +0100, Blaisorblade wrote:
> > what
> > about load_TLS? I only got -22, I fixed it by removing tls-debug
>
> That's caused by referring to errno in kernel code. That gets you the
> value of kernel_errno, which
24 matches
Mail list logo
