On Thu, Feb 18, 2010 at 03:00:13PM +0100, Lucas Nussbaum wrote:
> On 18/02/10 at 14:31 +0100, Bastian Blank wrote:
> > > I's only a few hundreds of megabytes per kernel image.
> >
> > Its around half a GiB per image.
>
> Isn't that a correct instance of "a few hundreds of megabytes"? :)
>
> > >
On 2010-02-18, Ben Hutchings wrote:
>
> --=-krd2Pvts3nDRsMxpCHn/
> Content-Type: text/plain; charset="UTF-8"
> Content-Transfer-Encoding: quoted-printable
>
> On Thu, 2010-02-18 at 15:00 +0100, Lucas Nussbaum wrote:
>> On 18/02/10 at 14:31 +0100, Bastian Blank wrote:
>> > > I's only a few hundreds
Troy Heber writes:
> On 02/16/10 13:54, dann frazier wrote:
>> The kernel team discussed this at our face to face meeting (due to
>> some prompting from the 'crash' maintainer).
> I didn't see this explicitly mentioned in the rest of the discussion,
> but debug kernel images are useful for more
On 2010-02-18, Ben Hutchings wrote:
> Currently I can upload a ~230 MB i386 build in 2 hours or so. Add debug
> info to this for all 6 images and I'm looking at more than a day, by
> which point I believe the incoming daemon would have given up waiting
> for a complete upload. Even if I switched
On Thu, 2010-02-18 at 15:00 +0100, Lucas Nussbaum wrote:
> On 18/02/10 at 14:31 +0100, Bastian Blank wrote:
> > > I's only a few hundreds of megabytes per kernel image.
> >
> > Its around half a GiB per image.
>
> Isn't that a correct instance of "a few hundreds of megabytes"? :)
[...]
I would o
On 02/16/10 13:54, dann frazier wrote:
>
> The kernel team discussed this at our face to face meeting (due to
> some prompting from the 'crash' maintainer).
I didn't see this explicitly mentioned in the rest of the discussion,
but debug kernel images are useful for more than just systemtap. It
w
On 18/02/10 at 14:31 +0100, Bastian Blank wrote:
> > I's only a few hundreds of megabytes per kernel image.
>
> Its around half a GiB per image.
Isn't that a correct instance of "a few hundreds of megabytes"? :)
> >Some games or
> > debug p
On Thu, Feb 18, 2010 at 02:01:19PM +0100, Lucas Nussbaum wrote:
> On 18/02/10 at 00:18 +0100, Frank Lin PIAT wrote:
> > On Tue, 2010-02-16 at 22:47 +0100, Lucas Nussbaum wrote:
> > > Why couldn't it be done using a normal deb?
It is just a matter of taste.
> I's only a few hundreds of megabytes p
On 18/02/10 at 00:18 +0100, Frank Lin PIAT wrote:
> On Tue, 2010-02-16 at 22:47 +0100, Lucas Nussbaum wrote:
> > On 16/02/10 at 22:27 +0100, Bastian Blank wrote:
> > > On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> > > > - disk space on buildds: at least 2 GiB are required to bui
On Tue, 2010-02-16 at 22:47 +0100, Lucas Nussbaum wrote:
> On 16/02/10 at 22:27 +0100, Bastian Blank wrote:
> > On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> > > - disk space on buildds: at least 2 GiB are required to build a kernel
> > > with debuginfo. (that doesn't sound to
On Wed, Feb 17, 2010 at 09:58:47AM +0100, Bastian Blank
wrote:
> This is a misinterpretation of my data. The effective
> space requirement is raised by factor 10 with debugging
> information.
And is that a problem for the buildds of the relevant
architectures? If so, what is the effective upper li
On Tue, Feb 16, 2010 at 10:17:27PM +0100, Kurt Roeckx wrote:
> On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> >
> > - disk space on buildds: at least 2 GiB are required to build a kernel
> > with debuginfo. (that doesn't sound too hard to satisfy)
> There currently are package
On 16/02/10 at 22:17 +0100, Kurt Roeckx wrote:
> On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> >
> > - disk space on buildds: at least 2 GiB are required to build a kernel
> > with debuginfo. (that doesn't sound too hard to satisfy)
>
> There currently are packages that requ
On 16/02/10 at 22:27 +0100, Bastian Blank wrote:
> On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> > - disk space on buildds: at least 2 GiB are required to build a kernel
> > with debuginfo. (that doesn't sound too hard to satisfy)
>
> A typical build includes between 2 and 10
On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> - disk space on buildds: at least 2 GiB are required to build a kernel
> with debuginfo. (that doesn't sound too hard to satisfy)
A typical build includes between 2 and 10 of them.
> - mirror space: each debug .deb would use ~ 45
On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
>
> - disk space on buildds: at least 2 GiB are required to build a kernel
> with debuginfo. (that doesn't sound too hard to satisfy)
There currently are packages that require more diskspace than
that, for instance the linux-2.6 pa
On Tue, Feb 16, 2010 at 09:13:06PM +0100, Lucas Nussbaum wrote:
> [ M-F-T set to debian-devel@ ]
>
> Hi,
>
> Systemtap[1] is a tool allowing to dynamically insert probes in the
> Linux kernel, similarly to what is possible with DTrace on Solaris.
> [1] http://sourceware.org/systemtap/
>
> The st
On 16/02/10 at 20:35 +, Philipp Kern wrote:
> On 2010-02-16, Lucas Nussbaum wrote:
> > Systemtap[1] is a tool allowing to dynamically insert probes in the
> > Linux kernel, similarly to what is possible with DTrace on Solaris.
> > [1] http://sourceware.org/systemtap/
> >
> > The state of syste
Hello,
16 лютого 2010 о 20:35 + Philipp Kern написав(-ла):
> Is that supported on all architectures or is it architecture-specific?
Systemtap now supports only i386/amd64/ia64/armel/powerpc/s390.
> Would it be sufficient to provide that infrastructure on the fast
> arches? I'm concerned how
On 2010-02-16, Lucas Nussbaum wrote:
> Systemtap[1] is a tool allowing to dynamically insert probes in the
> Linux kernel, similarly to what is possible with DTrace on Solaris.
> [1] http://sourceware.org/systemtap/
>
> The state of systemtap in Debian is currently worrying. First, the
> package h
[ M-F-T set to debian-devel@ ]
Hi,
Systemtap[1] is a tool allowing to dynamically insert probes in the
Linux kernel, similarly to what is possible with DTrace on Solaris.
[1] http://sourceware.org/systemtap/
The state of systemtap in Debian is currently worrying. First, the
package has been orph
21 matches
Mail list logo
