[9fans] request for testers of next beta release of tvx

[9fans] request for testers of next beta release of tvx

[EBo]

Sorry for the delay.  While working on the next release of tvx, TCL's
maintainers have been away for a little while and expect to get back to
addressing uploading the newest version of tvx (including the proposed name
changes) after Wednesday.

Is anybody interested and willing to do some testing between now and then?
If so, please upload the tarball from
ftp://ftp.swcp.com/pub/tmp/tvx_tcz_files.tgz and uncompress them into TCL's
/mnt/sd*1/tce/optional/.  I am hoping with a bit of testing I can release
version 1.0 of tvx, and start moving onto other things.

As a note, this version includes:

  * name change from Tvx-* to tvx* (while this has been tested it has not
completed TCL review/acceptance)

  * updates to tvx (including removal of -O3 flag, and cp's H/L or sudo
option)

  * fixes to tvxroot (including missing mkfile.proto, etc.)

  * addition of Ron's 9pm, ratrace, and live hg pull instructions to
tvxroot

  * minor patches to 9pm to avoid name clashes and use /tmp insead of
current directory for downloaded iso's

  * establishing a tvxroot repository on bitbucket.org

  * add date tags to tvxroot repository to help manage automated builds.

Thanks and best regards,

  EBo --

Re: [9fans] request for testers of next beta release of tvx

[ron minnich]

btw 9pm is what we named the package manager.

"It's 9pm. Do you know where your packages are?"

ron

Re: [9fans] request for testers of next beta release of tvx

[Skip Tavakkolian]

> btw 9pm is what we named the package manager.

this might cause a little confusion; 9pm is also the name of the plan9
emulation environment for windows.  it is still the only environment
available.

Re: [9fans] request for testers of next beta release of tvx

[EBo]

>> btw 9pm is what we named the package manager.
>
> this might cause a little confusion; 9pm is also the name of the plan9
> emulation environment for windows.  it is still the only environment
> available.

ouch...  Ron, what should we do about this?

  EBo --

Re: [9fans] request for testers of next beta release of tvx

[Lucio De Re]

On Tue, Jun 08, 2010 at 09:22:33AM -0600, EBo wrote:
>
> >> btw 9pm is what we named the package manager.
> >
> > this might cause a little confusion; 9pm is also the name of the plan9
> > emulation environment for windows.  it is still the only environment
> > available.
>
> ouch...  Ron, what should we do about this?
>
Different namespaces.  I think we'll live...

++L

Re: [9fans] request for testers of next beta release of tvx

[ron minnich]

On Tue, Jun 8, 2010 at 9:39 AM, Lucio De Re <lucio@proxima.alt.za> wrote:

> Different namespaces.  I think we'll live...

That's the point of private name spaces right?

I doubt a linux distro is that much concerned with anything related to windows.

ron

Re: [9fans] request for testers of next beta release of tvx

[Lucio De Re]

On Tue, Jun 08, 2010 at 09:10:33PM -0700, ron minnich wrote:
> On Tue, Jun 8, 2010 at 9:39 AM, Lucio De Re <lucio@proxima.alt.za> wrote:
> 
> > Different namespaces.  I think we'll live...
> 
> That's the point of private name spaces right?
> 
> I doubt a linux distro is that much concerned with anything related to windows.
> 
And I especially doubt that Windows will concern itself with a Linux
distro.  My point exactly.

++L

[9fans] Sed label handling inconsistency

[yard-ape]

P9P sed gives me the error "Label too long" on:

% echo foo | sed 's/./&/;ta;b;:a;s//*/'

but not on:

% echo foo | sed 's/./&/
ta
b
:a
s//*/'

As though the label commands require immediate newlines.

Anyone else experience this?

Re: [9fans] Sed label handling inconsistency

[erik quanstrom]

On Tue Jun  8 14:14:17 EDT 2010, yard-ape@telus.net wrote:
> P9P sed gives me the error "Label too long" on:
>
> % echo foo | sed 's/./&/;ta;b;:a;s//*/'
>
> but not on:
>
> % echo foo | sed 's/./&/
> ta
> b
> :a
> s//*/'
>
> As though the label commands require immediate newlines.
>
> Anyone else experience this?

without having access to a unix machine right now, i'm guessing that
the code at jtcommon should replace
	if(*cp == '\0')
with
	if(*cp == '\0' || *cp == ';')

at lines 428.  similar change at line 328.

- erik

Re: [9fans] Sed label handling inconsistency

[erik quanstrom]

> without having access to a unix machine right now, i'm guessing that
> the code at jtcommon should replace
> 	if(*cp == '\0')
> with
> 	if(*cp == '\0' || *cp == ';')
>
> at lines 428.  similar change at line 328.

sorry, fat fingers.  428 and 438.

- erik

Re: [9fans] Sed label handling inconsistency

[yard-ape]

erik quanstrom <quanstro@quanstro.net> wrote:

> > without having access to a unix machine right now, i'm guessing that
> > the code at jtcommon should replace
> > 	if(*cp == '\0')
> > with
> > 	if(*cp == '\0' || *cp == ';')
> >
> > at lines 428.  similar change at line 328.
>
> sorry, fat fingers.  428 and 438.
>
> - erik
Actually 328 is what I've got.  But no change in behaviour after recompile.

Re: [9fans] Sed label handling inconsistency

[erik quanstrom]

> erik quanstrom <quanstro@quanstro.net> wrote:
>
> > > without having access to a unix machine right now, i'm guessing that
> > > the code at jtcommon should replace
> > > 	if(*cp == '\0')
> > > with
> > > 	if(*cp == '\0' || *cp == ';')
> > >
> > > at lines 428.  similar change at line 328.
> >
> > sorry, fat fingers.  428 and 438.
> >
> > - erik
> Actually 328 is what I've got.  But no change in behaviour after recompile.

the plan 9 version works as you expect it should, and it's
not a 32-bit rune problem because that works on my system.
by looking at the differences between the p9p and plan 9
version, you should be able to find the problem rather
rapidly.  there's not much code involved.

- erik