Any News on cut-and-paste bug?

Any News on cut-and-paste bug?

[Hart Larry]

Quite some months ago Bill Acker suggested I login as an
ssh localhost
on each console where I would want to cut-and-paste.  Well, actually unless 
there were a way for this anoyance to just ruin 1 tty instead of freeze an 
entire machine?  I also suppose having a script on bootup log us in a 
localhost.
This bug just comes so suddenly, no warning--and-best as we can tell all 
activity stops.
I realize-and-appreciate that we have an active community, many who are 
knowledgeable, ETC.  But it almost seems Speakup may join YASR as having gotten 
at a certain level-and-thats it.
I've been on this list since 2003, but now for `quite some time I still cannot 
move up past 2.632 as I would have no DecTalk speech.  I think John Heim wrote 
a patch to fix this, but I have no idea what steps will install?
And lastly, still about the DecTalk, if we can ever produce a log showing 
commands which Speakup is sending, James says he can assist.
Thanks so much in advance
Hart

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

On Tue, 30 Apr 2013, Hart Larry wrote:

> Quite some months ago Bill Acker suggested I login as an
> ssh localhost
> on each console where I would want to cut-and-paste.  Well, actually unless 
> there were a way for this anoyance to just ruin 1 tty instead of freeze an 
> entire machine?  I also suppose having a script on bootup log us in a 
> localhost.
> This bug just comes so suddenly, no warning--and-best as we can tell all 
> activity stops.

As far as I know, Chris Brannon submitted a patch to fix the
cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
patch ever made it into the kernel speakup version or not.

> I realize-and-appreciate that we have an active community, many who are 
> knowledgeable, ETC.  But it almost seems Speakup may join YASR as having 
> gotten at a certain level-and-thats it.

Without new blood interested in taking speakup further, you may very
well be correct.

> I've been on this list since 2003, but now for `quite some time I still 
> cannot move up past 2.632 as I would have no DecTalk speech.  I think John 
> Heim wrote a patch to fix this, but I have no idea what steps will install?
> And lastly, still about the DecTalk, if we can ever produce a log showing 
> commands which Speakup is sending, James says he can assist.

Unless someone decides to take on writing external drivers for USB and
RS232C synths, you will never see a DECTalk Express fix. The only
support over the past few years has been for the softsynth version of
speakup. There have been a few serial fixes but they have been more of
an aside than anything else. The serial synth substructure is terribly
out of date and nobody appears to be willing to rewrite it.


-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

Kirk Reiser, le Wed 01 May 2013 08:27:30 -0400, a écrit :
> an aside than anything else. The serial synth substructure is terribly
> out of date and nobody appears to be willing to rewrite it.

s/willing/having time/

Samuel

Re: Any News on cut-and-paste bug?

[covici]

I have the cut and paste patch, but it was eventually put in the
kernel.  There is a work around for using serial synths in newer
kernels, although there seems to be some problems in and around 3.7.  I
have wanted to figure this all out, but my time ...

Kirk Reiser <kirk@reisers.ca> wrote:

> On Tue, 30 Apr 2013, Hart Larry wrote:
> 
> > Quite some months ago Bill Acker suggested I login as an
> > ssh localhost
> > on each console where I would want to cut-and-paste.  Well, actually
> > unless there were a way for this anoyance to just ruin 1 tty instead
> > of freeze an entire machine?  I also suppose having a script on
> > bootup log us in a localhost.
> > This bug just comes so suddenly, no warning--and-best as we can tell
> > all activity stops.
> 
> As far as I know, Chris Brannon submitted a patch to fix the
> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
> patch ever made it into the kernel speakup version or not.
> 
> > I realize-and-appreciate that we have an active community, many who
> > are knowledgeable, ETC.  But it almost seems Speakup may join YASR
> > as having gotten at a certain level-and-thats it.
> 
> Without new blood interested in taking speakup further, you may very
> well be correct.
> 
> > I've been on this list since 2003, but now for `quite some time I
> > still cannot move up past 2.632 as I would have no DecTalk speech.
> > I think John Heim wrote a patch to fix this, but I have no idea what
> > steps will install?
> > And lastly, still about the DecTalk, if we can ever produce a log
> > showing commands which Speakup is sending, James says he can assist.
> 
> Unless someone decides to take on writing external drivers for USB and
> RS232C synths, you will never see a DECTalk Express fix. The only
> support over the past few years has been for the softsynth version of
> speakup. There have been a few serial fixes but they have been more of
> an aside than anything else. The serial synth substructure is terribly
> out of date and nobody appears to be willing to rewrite it.
> 
> 
> -- 
> Well that's it then, colour me gone!
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[John G. Heim]

A big part of the problem is that even if someone is willing to take on 
writing fixes for speakup, the kernel people won't cooperate. I tried to 
get some help/advice from the linux kernel  list on implementing the bug 
fix I had for serial synths. Their advice -- start over.

PS: I think that particular bug got fixed for real by someone who knows 
way more about the speakup code than I do. I'm running a 3.2 kernel from 
squeeze backports and the bbug is fixed.

On 05/01/13 07:27, Kirk Reiser wrote:
> On Tue, 30 Apr 2013, Hart Larry wrote:
>
>> Quite some months ago Bill Acker suggested I login as an
>> ssh localhost
>> on each console where I would want to cut-and-paste.  Well, actually
>> unless there were a way for this anoyance to just ruin 1 tty instead
>> of freeze an entire machine?  I also suppose having a script on bootup
>> log us in a localhost.
>> This bug just comes so suddenly, no warning--and-best as we can tell
>> all activity stops.
>
> As far as I know, Chris Brannon submitted a patch to fix the
> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
> patch ever made it into the kernel speakup version or not.
>
>> I realize-and-appreciate that we have an active community, many who
>> are knowledgeable, ETC.  But it almost seems Speakup may join YASR as
>> having gotten at a certain level-and-thats it.
>
> Without new blood interested in taking speakup further, you may very
> well be correct.
>
>> I've been on this list since 2003, but now for `quite some time I
>> still cannot move up past 2.632 as I would have no DecTalk speech.  I
>> think John Heim wrote a patch to fix this, but I have no idea what
>> steps will install?
>> And lastly, still about the DecTalk, if we can ever produce a log
>> showing commands which Speakup is sending, James says he can assist.
>
> Unless someone decides to take on writing external drivers for USB and
> RS232C synths, you will never see a DECTalk Express fix. The only
> support over the past few years has been for the softsynth version of
> speakup. There have been a few serial fixes but they have been more of
> an aside than anything else. The serial synth substructure is terribly
> out of date and nobody appears to be willing to rewrite it.
>
>

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

John G. Heim, le Wed 01 May 2013 08:45:54 -0500, a écrit :
> A big part of the problem is that even if someone is willing to take on
> writing fixes for speakup, the kernel people won't cooperate. I tried to get
> some help/advice from the linux kernel  list on implementing the bug fix I
> had for serial synths. Their advice -- start over.

Which isn't really not cooperating. The current code will keep having
such kind of issues until it gets rewritten a more integrated way.

Samuel

Re: Any News on cut-and-paste bug?

[covici]

I would like to have a conference call with those interested in
developing speakup, even if they don't have much time -- maybe we can at
least sort out what things should happen, etc.  I like that much better
than Email exchanges!

covici@ccs.covici.com wrote:

> I have the cut and paste patch, but it was eventually put in the
> kernel.  There is a work around for using serial synths in newer
> kernels, although there seems to be some problems in and around 3.7.  I
> have wanted to figure this all out, but my time ...
> 
> Kirk Reiser <kirk@reisers.ca> wrote:
> 
> > On Tue, 30 Apr 2013, Hart Larry wrote:
> > 
> > > Quite some months ago Bill Acker suggested I login as an
> > > ssh localhost
> > > on each console where I would want to cut-and-paste.  Well, actually
> > > unless there were a way for this anoyance to just ruin 1 tty instead
> > > of freeze an entire machine?  I also suppose having a script on
> > > bootup log us in a localhost.
> > > This bug just comes so suddenly, no warning--and-best as we can tell
> > > all activity stops.
> > 
> > As far as I know, Chris Brannon submitted a patch to fix the
> > cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
> > patch ever made it into the kernel speakup version or not.
> > 
> > > I realize-and-appreciate that we have an active community, many who
> > > are knowledgeable, ETC.  But it almost seems Speakup may join YASR
> > > as having gotten at a certain level-and-thats it.
> > 
> > Without new blood interested in taking speakup further, you may very
> > well be correct.
> > 
> > > I've been on this list since 2003, but now for `quite some time I
> > > still cannot move up past 2.632 as I would have no DecTalk speech.
> > > I think John Heim wrote a patch to fix this, but I have no idea what
> > > steps will install?
> > > And lastly, still about the DecTalk, if we can ever produce a log
> > > showing commands which Speakup is sending, James says he can assist.
> > 
> > Unless someone decides to take on writing external drivers for USB and
> > RS232C synths, you will never see a DECTalk Express fix. The only
> > support over the past few years has been for the softsynth version of
> > speakup. There have been a few serial fixes but they have been more of
> > an aside than anything else. The serial synth substructure is terribly
> > out of date and nobody appears to be willing to rewrite it.
> > 
> > 
> > -- 
> > Well that's it then, colour me gone!
> > _______________________________________________
> > Speakup mailing list
> > Speakup@linux-speakup.org
> > http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
> 
> -- 
> Your life is like a penny.  You're going to lose it.  The question is:
> How do
> you spend it?
> 
>          John Covici
>          covici@ccs.covici.com
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[John G. Heim]

On 05/01/13 08:50, Samuel Thibault wrote:
> John G. Heim, le Wed 01 May 2013 08:45:54 -0500, a écrit :
>> A big part of the problem is that even if someone is willing to take on
>> writing fixes for speakup, the kernel people won't cooperate. I tried to get
>> some help/advice from the linux kernel  list on implementing the bug fix I
>> had for serial synths. Their advice -- start over.
>
> Which isn't really not cooperating. The current code will keep having
> such kind of issues until it gets rewritten a more integrated way.

Well, the details of this particular bug are significant. The problem 
was that the speakup code was erroring out on what I believe was a 
totally meaningless error code. All I did was change the code to print a 
warning and keep going.

If you are a programmer, you may be saying to yourself, "Well, that's 
not a good way to solve a problem." But I am about as sure as I can be 
that the function call that returned the error code did nothing. The 
function was part of the kernel code, not speakup. So I went on the 
kernel list to ask what the function was supposed to do,   was it 
necessary to call it at all,  and how to call it correctly.

All I wanted to do was get rid of what I suspected was a call to a 
function that apparently did nothing and the subsequent erroring out. 
But nobody seemed to know what the function did or if they did, they 
weren't sharing. They did, however, take the time to criticize the 
speakup code itself.

So I was like, "Come on, people. Can't we just focus on this one 
problem? It seems like a small fix for a huge problem. I mean, I cannot 
use my hardware speech synth without patching the kernel code. This 
could cost me my job. It could cost a lot of blind systems admins their 
jobs."

No love.

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
> All I wanted to do was get rid of what I suspected was a call to a function
> that apparently did nothing and the subsequent erroring out. But nobody
> seemed to know what the function did or if they did, they weren't sharing.
> They did, however, take the time to criticize the speakup code itself.

Do you have a reference of the thread? (like the precise date when it
happened, or the subject, etc.)

Samuel

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
> They did, however, take the time to criticize the speakup code itself.

I'm afraid that's for a good reason: speakup works around the kernel,
that's not an approach that can work on the long term.

Samuel

Re: Any News on cut-and-paste bug?

[John G. Heim]

You mean skype?


On 05/01/13 09:10, covici@ccs.covici.com wrote:
> I would like to have a conference call with those interested in
> developing speakup, even if they don't have much time -- maybe we can at
> least sort out what things should happen, etc.  I like that much better
> than Email exchanges!
>
> covici@ccs.covici.com wrote:
>
>> I have the cut and paste patch, but it was eventually put in the
>> kernel.  There is a work around for using serial synths in newer
>> kernels, although there seems to be some problems in and around 3.7.  I
>> have wanted to figure this all out, but my time ...
>>
>> Kirk Reiser <kirk@reisers.ca> wrote:
>>
>>> On Tue, 30 Apr 2013, Hart Larry wrote:
>>>
>>>> Quite some months ago Bill Acker suggested I login as an
>>>> ssh localhost
>>>> on each console where I would want to cut-and-paste.  Well, actually
>>>> unless there were a way for this anoyance to just ruin 1 tty instead
>>>> of freeze an entire machine?  I also suppose having a script on
>>>> bootup log us in a localhost.
>>>> This bug just comes so suddenly, no warning--and-best as we can tell
>>>> all activity stops.
>>>
>>> As far as I know, Chris Brannon submitted a patch to fix the
>>> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
>>> patch ever made it into the kernel speakup version or not.
>>>
>>>> I realize-and-appreciate that we have an active community, many who
>>>> are knowledgeable, ETC.  But it almost seems Speakup may join YASR
>>>> as having gotten at a certain level-and-thats it.
>>>
>>> Without new blood interested in taking speakup further, you may very
>>> well be correct.
>>>
>>>> I've been on this list since 2003, but now for `quite some time I
>>>> still cannot move up past 2.632 as I would have no DecTalk speech.
>>>> I think John Heim wrote a patch to fix this, but I have no idea what
>>>> steps will install?
>>>> And lastly, still about the DecTalk, if we can ever produce a log
>>>> showing commands which Speakup is sending, James says he can assist.
>>>
>>> Unless someone decides to take on writing external drivers for USB and
>>> RS232C synths, you will never see a DECTalk Express fix. The only
>>> support over the past few years has been for the softsynth version of
>>> speakup. There have been a few serial fixes but they have been more of
>>> an aside than anything else. The serial synth substructure is terribly
>>> out of date and nobody appears to be willing to rewrite it.
>>>
>>>
>>> --
>>> Well that's it then, colour me gone!
>>> _______________________________________________
>>> Speakup mailing list
>>> Speakup@linux-speakup.org
>>> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>>
>> --
>> Your life is like a penny.  You're going to lose it.  The question is:
>> How do
>> you spend it?
>>
>>           John Covici
>>           covici@ccs.covici.com
>> _______________________________________________
>> Speakup mailing list
>> Speakup@linux-speakup.org
>> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>

Re: Any News on cut-and-paste bug?

[Kelly Prescott]

On Wed, 1 May 2013, covici@ccs.covici.com wrote:
I would be interested in helping, I am decent in C but have absolutely no 
kernel background, but I want to change that.

-- Kelly Prescott


> I would like to have a conference call with those interested in
> developing speakup, even if they don't have much time -- maybe we can at
> least sort out what things should happen, etc.  I like that much better
> than Email exchanges!

Re: Any News on cut-and-paste bug?

[covici]

I wonder if that error code is trying to tell us something -- I do know
if I try to use the soft synth with that patch installed, it crashes,
presumably because its trying to release a null pointer or something.

I also saw on the serial list, some kinf of call to platform register
(not exactly sure of the name), so I wonder if these are the right calls
anymore?


John G. Heim <jheim@math.wisc.edu> wrote:

> 
> 
> On 05/01/13 08:50, Samuel Thibault wrote:
> > John G. Heim, le Wed 01 May 2013 08:45:54 -0500, a écrit :
> >> A big part of the problem is that even if someone is willing to take on
> >> writing fixes for speakup, the kernel people won't cooperate. I tried to get
> >> some help/advice from the linux kernel  list on implementing the bug fix I
> >> had for serial synths. Their advice -- start over.
> >
> > Which isn't really not cooperating. The current code will keep having
> > such kind of issues until it gets rewritten a more integrated way.
> 
> Well, the details of this particular bug are significant. The problem
> was that the speakup code was erroring out on what I believe was a
> totally meaningless error code. All I did was change the code to print
> a warning and keep going.
> 
> If you are a programmer, you may be saying to yourself, "Well, that's
> not a good way to solve a problem." But I am about as sure as I can be
> that the function call that returned the error code did nothing. The
> function was part of the kernel code, not speakup. So I went on the
> kernel list to ask what the function was supposed to do,   was it
> necessary to call it at all,  and how to call it correctly.
> 
> All I wanted to do was get rid of what I suspected was a call to a
> function that apparently did nothing and the subsequent erroring
> out. But nobody seemed to know what the function did or if they did,
> they weren't sharing. They did, however, take the time to criticize
> the speakup code itself.
> 
> So I was like, "Come on, people. Can't we just focus on this one
> problem? It seems like a small fix for a huge problem. I mean, I
> cannot use my hardware speech synth without patching the kernel
> code. This could cost me my job. It could cost a lot of blind systems
> admins their jobs."
> 
> No love.
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[covici]

Or my freeswitch server, or Kirks.

John G. Heim <jheim@math.wisc.edu> wrote:

> You mean skype?
> 
> 
> On 05/01/13 09:10, covici@ccs.covici.com wrote:
> > I would like to have a conference call with those interested in
> > developing speakup, even if they don't have much time -- maybe we can at
> > least sort out what things should happen, etc.  I like that much better
> > than Email exchanges!
> >
> > covici@ccs.covici.com wrote:
> >
> >> I have the cut and paste patch, but it was eventually put in the
> >> kernel.  There is a work around for using serial synths in newer
> >> kernels, although there seems to be some problems in and around 3.7.  I
> >> have wanted to figure this all out, but my time ...
> >>
> >> Kirk Reiser <kirk@reisers.ca> wrote:
> >>
> >>> On Tue, 30 Apr 2013, Hart Larry wrote:
> >>>
> >>>> Quite some months ago Bill Acker suggested I login as an
> >>>> ssh localhost
> >>>> on each console where I would want to cut-and-paste.  Well, actually
> >>>> unless there were a way for this anoyance to just ruin 1 tty instead
> >>>> of freeze an entire machine?  I also suppose having a script on
> >>>> bootup log us in a localhost.
> >>>> This bug just comes so suddenly, no warning--and-best as we can tell
> >>>> all activity stops.
> >>>
> >>> As far as I know, Chris Brannon submitted a patch to fix the
> >>> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
> >>> patch ever made it into the kernel speakup version or not.
> >>>
> >>>> I realize-and-appreciate that we have an active community, many who
> >>>> are knowledgeable, ETC.  But it almost seems Speakup may join YASR
> >>>> as having gotten at a certain level-and-thats it.
> >>>
> >>> Without new blood interested in taking speakup further, you may very
> >>> well be correct.
> >>>
> >>>> I've been on this list since 2003, but now for `quite some time I
> >>>> still cannot move up past 2.632 as I would have no DecTalk speech.
> >>>> I think John Heim wrote a patch to fix this, but I have no idea what
> >>>> steps will install?
> >>>> And lastly, still about the DecTalk, if we can ever produce a log
> >>>> showing commands which Speakup is sending, James says he can assist.
> >>>
> >>> Unless someone decides to take on writing external drivers for USB and
> >>> RS232C synths, you will never see a DECTalk Express fix. The only
> >>> support over the past few years has been for the softsynth version of
> >>> speakup. There have been a few serial fixes but they have been more of
> >>> an aside than anything else. The serial synth substructure is terribly
> >>> out of date and nobody appears to be willing to rewrite it.
> >>>
> >>>
> >>> --
> >>> Well that's it then, colour me gone!
> >>> _______________________________________________
> >>> Speakup mailing list
> >>> Speakup@linux-speakup.org
> >>> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
> >>
> >> --
> >> Your life is like a penny.  You're going to lose it.  The question is:
> >> How do
> >> you spend it?
> >>
> >>           John Covici
> >>           covici@ccs.covici.com
> >> _______________________________________________
> >> Speakup mailing list
> >> Speakup@linux-speakup.org
> >> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
> >

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

On Wed, 1 May 2013, Samuel Thibault wrote:

> s/willing/having time/

Um, not having time translates to not finding time which translates to
not willing. You can't invalidate the point by playing samantics
games.

However, I don't think you need to feel guilty about not getting
around to this rewrite.  It is not your responsibility afterall. It
isn't anyones responsibility.

My point still stands that until someone is willing to take on the
task and do it, it ain't gonna happin'.

I will straight out say unless others are willing to get involved and
help do the work, I'm no longer interested.

-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[covici]

I am definitely willing, but I don't know how things work anymore, so I
would like to have a discussion to see what we should do.

Kirk Reiser <kirk@reisers.ca> wrote:

> On Wed, 1 May 2013, Samuel Thibault wrote:
> 
> > s/willing/having time/
> 
> Um, not having time translates to not finding time which translates to
> not willing. You can't invalidate the point by playing samantics
> games.
> 
> However, I don't think you need to feel guilty about not getting
> around to this rewrite.  It is not your responsibility afterall. It
> isn't anyones responsibility.
> 
> My point still stands that until someone is willing to take on the
> task and do it, it ain't gonna happin'.
> 
> I will straight out say unless others are willing to get involved and
> help do the work, I'm no longer interested.
> 
> -- 
> Well that's it then, colour me gone!
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

On Wed, 1 May 2013, covici@ccs.covici.com wrote:

> Or my freeswitch server, or Kirks.

I don't have a freeswitch server. Well, I do but my network is so
flooded it isn't usable.

So I would suggest that Brandon's be used at
sip://3000@bmcginty.hopto.org.

You will all need to agree on a time however and help them what wish
to attend but don't know about sip to find something that works and
help them learn.

I am not a Windows user so have no experience with sip clients for
that OS.

-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[John G. Heim]

On 05/01/13 10:45, Samuel Thibault wrote:
> John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
>> All I wanted to do was get rid of what I suspected was a call to a function
>> that apparently did nothing and the subsequent erroring out. But nobody
>> seemed to know what the function did or if they did, they weren't sharing.
>> They did, however, take the time to criticize the speakup code itself.
>
> Do you have a reference of the thread? (like the precise date when it
> happened, or the subject, etc.)

Here are 2 of the messages I posted. I haven't found an archive of the 
linux-kernel list. But I haven't looked too hard.  But these messages 
show the exact date and subject line of the 2 threads  on the list.

On 03/03/12 11:18, John G. Heim wrote:
 >Subject: speakup bug
 > I need help fixing a bug in the driver for serial hardware  speech
 > synths in the speakup screen reader. According to the comments in the
 > code, it is in a part of the code that is trying to "steal" the serial
 > port.
 >
 > First it calls request_region and when that fails (it always fails), it
 > calls         __release_region(&then it calls release_region again to
 > see if __release_region worked. But it never works because the region
 > being requested is already taken.
 >
 > The code in question is in 2 source code files in
 > drivers/staging/speakup. It starts in serialio.c on line 38. Here it
 > calls a function named synth_request_region which in turn, calls
 > request_region. On line 41 it calls __release_region, and on line 42 it
 > calls synth_request_region again. The function synth_request_region
 > (which calls request_region) is in a file named synth.c. But this code
 > always fails.  Here is a kind of simplified version of it...
 >
 > int error;
 > struct resource tmp;
 > tmp .name = "ltlk";
 > tmps.start = 0x3F8;
 > tmp.end = 0x3FF;
 > tmp.flags = IORESOURCE_BUSY;
 > error = request_resource (&ioport_resource, &tmp);
 >
 > The error returned is always -16. I looked at the code in
 > kernel/resource.c where the request_region function is  defined. It
 > builds a linked list of resources with start & end addresses. If you
 > request a region that is already within the start-end range of a
 > resource already in the list, it returns an error code. But it looks as
 > if the region for a serial port, 0x3f8 - 0x3ff,  in ioport_resource
 > cannot be reserved because the entire range from 0x000 through 0xcf7 is
 > already taken by something named "PCI Bus 0000:00".  Therefore calling
 > request_resource always fails and the driver for the speech synth errors
 > out.
 >
 > And therefore I can't use my hardware speech synth without modifying the
 > kernel code.  If you comment out the line that checks the return code
 > from request_region, it works. So you have to modify  the kernel code
 > and compile a custom kernel to use a hardware speech synth. That's not
 > such a problem for me but it is for a lot of people. Plus, the grml live
 > CD doesn't work with hardware speech.  That is a problem for me.
 >
 > Can anyone tell me how to fix this so it can be patched in the official
 > kernel code?
 >


On 05/11/12 10:36, John Heim wrote:
 >Subject: patch for speakup serial hardware synths
 > A few weeks ago I asked about a problem with the speakup screen reader
 > code. It does not work with serial hardware speech synthesizers. But I
 > managed to get it working. How can I submit my fix for integration into
 > the kernel code. The patch file can be downloaded here:
 > http://www.math.wisc.edu/~jheim/downloads/patch-2012-03-06.patch
 >
 > To install it you cd to the linux source directory and do this:
 > patch -i patch-2012-03-06.patch drivers/staging/speakup/serialio.c"
 >

Re: Any News on cut-and-paste bug?

[Brandon McGinty-Carroll]

I'll join you all.
I would like to get a handle on the code and what it's doing, perhaps figure out where speakup hooks the kernel, and disassociate the rest of the code from those points.
Further, having some documentation in the code (well, considerably more than that is there now), would be quite helpful, even if the current design stays.

Brandon McGinty-Carroll



On Wed, May 01, 2013 at 12:22:20PM -0400, Kirk Reiser wrote:
> On Wed, 1 May 2013, covici@ccs.covici.com wrote:
> 
> >Or my freeswitch server, or Kirks.
> 
> I don't have a freeswitch server. Well, I do but my network is so
> flooded it isn't usable.
> 
> So I would suggest that Brandon's be used at
> sip://3000@bmcginty.hopto.org.
> 
> You will all need to agree on a time however and help them what wish
> to attend but don't know about sip to find something that works and
> help them learn.
> 
> I am not a Windows user so have no experience with sip clients for
> that OS.
> 
> -- 
> Well that's it then, colour me gone!
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

Re: Any News on cut-and-paste bug?

[covici]

I meant his, not your actual one.  I have a client which is quite
accessible, although Randy had some problems with it.  And you don't
need sip if you want to call my box instead, I have numbers for that.

Kirk Reiser <kirk@reisers.ca> wrote:

> On Wed, 1 May 2013, covici@ccs.covici.com wrote:
> 
> > Or my freeswitch server, or Kirks.
> 
> I don't have a freeswitch server. Well, I do but my network is so
> flooded it isn't usable.
> 
> So I would suggest that Brandon's be used at
> sip://3000@bmcginty.hopto.org.
> 
> You will all need to agree on a time however and help them what wish
> to attend but don't know about sip to find something that works and
> help them learn.
> 
> I am not a Windows user so have no experience with sip clients for
> that OS.
> 
> -- 
> Well that's it then, colour me gone!
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[covici]

I would not accept that fix in the kernel, either -- nice workaround,
but its just hiding the real problem.

John G. Heim <jheim@math.wisc.edu> wrote:

> 
> 
> On 05/01/13 10:45, Samuel Thibault wrote:
> > John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
> >> All I wanted to do was get rid of what I suspected was a call to a function
> >> that apparently did nothing and the subsequent erroring out. But nobody
> >> seemed to know what the function did or if they did, they weren't sharing.
> >> They did, however, take the time to criticize the speakup code itself.
> >
> > Do you have a reference of the thread? (like the precise date when it
> > happened, or the subject, etc.)
> 
> Here are 2 of the messages I posted. I haven't found an archive of the
> linux-kernel list. But I haven't looked too hard.  But these messages
> show the exact date and subject line of the 2 threads  on the list.
> 
> On 03/03/12 11:18, John G. Heim wrote:
> >Subject: speakup bug
> > I need help fixing a bug in the driver for serial hardware  speech
> > synths in the speakup screen reader. According to the comments in the
> > code, it is in a part of the code that is trying to "steal" the serial
> > port.
> >
> > First it calls request_region and when that fails (it always fails), it
> > calls         __release_region(&then it calls release_region again to
> > see if __release_region worked. But it never works because the region
> > being requested is already taken.
> >
> > The code in question is in 2 source code files in
> > drivers/staging/speakup. It starts in serialio.c on line 38. Here it
> > calls a function named synth_request_region which in turn, calls
> > request_region. On line 41 it calls __release_region, and on line 42 it
> > calls synth_request_region again. The function synth_request_region
> > (which calls request_region) is in a file named synth.c. But this code
> > always fails.  Here is a kind of simplified version of it...
> >
> > int error;
> > struct resource tmp;
> > tmp .name = "ltlk";
> > tmps.start = 0x3F8;
> > tmp.end = 0x3FF;
> > tmp.flags = IORESOURCE_BUSY;
> > error = request_resource (&ioport_resource, &tmp);
> >
> > The error returned is always -16. I looked at the code in
> > kernel/resource.c where the request_region function is  defined. It
> > builds a linked list of resources with start & end addresses. If you
> > request a region that is already within the start-end range of a
> > resource already in the list, it returns an error code. But it looks as
> > if the region for a serial port, 0x3f8 - 0x3ff,  in ioport_resource
> > cannot be reserved because the entire range from 0x000 through 0xcf7 is
> > already taken by something named "PCI Bus 0000:00".  Therefore calling
> > request_resource always fails and the driver for the speech synth errors
> > out.
> >
> > And therefore I can't use my hardware speech synth without modifying the
> > kernel code.  If you comment out the line that checks the return code
> > from request_region, it works. So you have to modify  the kernel code
> > and compile a custom kernel to use a hardware speech synth. That's not
> > such a problem for me but it is for a lot of people. Plus, the grml live
> > CD doesn't work with hardware speech.  That is a problem for me.
> >
> > Can anyone tell me how to fix this so it can be patched in the official
> > kernel code?
> >
> 
> 
> On 05/11/12 10:36, John Heim wrote:
> >Subject: patch for speakup serial hardware synths
> > A few weeks ago I asked about a problem with the speakup screen reader
> > code. It does not work with serial hardware speech synthesizers. But I
> > managed to get it working. How can I submit my fix for integration into
> > the kernel code. The patch file can be downloaded here:
> > http://www.math.wisc.edu/~jheim/downloads/patch-2012-03-06.patch
> >
> > To install it you cd to the linux source directory and do this:
> > patch -i patch-2012-03-06.patch drivers/staging/speakup/serialio.c"
> >
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

That has always een the attitude of the general folks or a few loud
ones anyway on the kernel mailing list.  They have almost always been
more likely to throw stones than be helpful. I have a suspician it's
because the ones that post like that don't actually know how to be
helpful and can't stand the idea of speakup being in the kernel so
much being quiet isn't an option for them.

I think if you are looking for kernel help you are better off writing
the few folks that have been helpful already such as Samuel or Alan
Cox, Dan Drake, Greg? I believe they may be more tolerant helping you
learn what it is you wish to learn to make the modifications.

That's been my experience anyway.

On Wed, 1 May 2013, John G. Heim wrote:

>
>
> On 05/01/13 10:45, Samuel Thibault wrote:
>> John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
>>> All I wanted to do was get rid of what I suspected was a call to a 
>>> function
>>> that apparently did nothing and the subsequent erroring out. But nobody
>>> seemed to know what the function did or if they did, they weren't sharing.
>>> They did, however, take the time to criticize the speakup code itself.
>> 
>> Do you have a reference of the thread? (like the precise date when it
>> happened, or the subject, etc.)
>
> Here are 2 of the messages I posted. I haven't found an archive of the 
> linux-kernel list. But I haven't looked too hard.  But these messages show 
> the exact date and subject line of the 2 threads  on the list.
>
> On 03/03/12 11:18, John G. Heim wrote:
>> Subject: speakup bug
>> I need help fixing a bug in the driver for serial hardware  speech
>> synths in the speakup screen reader. According to the comments in the
>> code, it is in a part of the code that is trying to "steal" the serial
>> port.
>>
>> First it calls request_region and when that fails (it always fails), it
>> calls         __release_region(&then it calls release_region again to
>> see if __release_region worked. But it never works because the region
>> being requested is already taken.
>>
>> The code in question is in 2 source code files in
>> drivers/staging/speakup. It starts in serialio.c on line 38. Here it
>> calls a function named synth_request_region which in turn, calls
>> request_region. On line 41 it calls __release_region, and on line 42 it
>> calls synth_request_region again. The function synth_request_region
>> (which calls request_region) is in a file named synth.c. But this code
>> always fails.  Here is a kind of simplified version of it...
>>
>> int error;
>> struct resource tmp;
>> tmp .name = "ltlk";
>> tmps.start = 0x3F8;
>> tmp.end = 0x3FF;
>> tmp.flags = IORESOURCE_BUSY;
>> error = request_resource (&ioport_resource, &tmp);
>>
>> The error returned is always -16. I looked at the code in
>> kernel/resource.c where the request_region function is  defined. It
>> builds a linked list of resources with start & end addresses. If you
>> request a region that is already within the start-end range of a
>> resource already in the list, it returns an error code. But it looks as
>> if the region for a serial port, 0x3f8 - 0x3ff,  in ioport_resource
>> cannot be reserved because the entire range from 0x000 through 0xcf7 is
>> already taken by something named "PCI Bus 0000:00".  Therefore calling
>> request_resource always fails and the driver for the speech synth errors
>> out.
>>
>> And therefore I can't use my hardware speech synth without modifying the
>> kernel code.  If you comment out the line that checks the return code
>> from request_region, it works. So you have to modify  the kernel code
>> and compile a custom kernel to use a hardware speech synth. That's not
>> such a problem for me but it is for a lot of people. Plus, the grml live
>> CD doesn't work with hardware speech.  That is a problem for me.
>>
>> Can anyone tell me how to fix this so it can be patched in the official
>> kernel code?
>>
>
>
> On 05/11/12 10:36, John Heim wrote:
>> Subject: patch for speakup serial hardware synths
>> A few weeks ago I asked about a problem with the speakup screen reader
>> code. It does not work with serial hardware speech synthesizers. But I
>> managed to get it working. How can I submit my fix for integration into
>> the kernel code. The patch file can be downloaded here:
>> http://www.math.wisc.edu/~jheim/downloads/patch-2012-03-06.patch
>>
>> To install it you cd to the linux source directory and do this:
>> patch -i patch-2012-03-06.patch drivers/staging/speakup/serialio.c"
>>
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>

-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[John G. Heim]

But that's no answer.  This is the same problem I had with the responses 
  I got on the linux-kernel list. I asked extremely specific  questions 
below and just saying no isn't enough.  Specifically, what is wrong with 
my patch? Does speakup need to call request_reggion at all? If so, why,? 
  What does request_region do? It looks as if request_region always 
fails if you request a region at the address of the serial port even if 
the serial port is not in use? Why is that?  If this behaviour is by 
design, what is the speakup code supposed to do instead?

You can say nothing. Say you're too busy or whatever. But don't just say no.

On 05/01/13 11:30, covici@ccs.covici.com wrote:
> I would not accept that fix in the kernel, either -- nice workaround,
> but its just hiding the real problem.
>
> John G. Heim <jheim@math.wisc.edu> wrote:
>
>>
>>
>> On 05/01/13 10:45, Samuel Thibault wrote:
>>> John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
>>>> All I wanted to do was get rid of what I suspected was a call to a function
>>>> that apparently did nothing and the subsequent erroring out. But nobody
>>>> seemed to know what the function did or if they did, they weren't sharing.
>>>> They did, however, take the time to criticize the speakup code itself.
>>>
>>> Do you have a reference of the thread? (like the precise date when it
>>> happened, or the subject, etc.)
>>
>> Here are 2 of the messages I posted. I haven't found an archive of the
>> linux-kernel list. But I haven't looked too hard.  But these messages
>> show the exact date and subject line of the 2 threads  on the list.
>>
>> On 03/03/12 11:18, John G. Heim wrote:
>>> Subject: speakup bug
>>> I need help fixing a bug in the driver for serial hardware  speech
>>> synths in the speakup screen reader. According to the comments in the
>>> code, it is in a part of the code that is trying to "steal" the serial
>>> port.
>>>
>>> First it calls request_region and when that fails (it always fails), it
>>> calls         __release_region(&then it calls release_region again to
>>> see if __release_region worked. But it never works because the region
>>> being requested is already taken.
>>>
>>> The code in question is in 2 source code files in
>>> drivers/staging/speakup. It starts in serialio.c on line 38. Here it
>>> calls a function named synth_request_region which in turn, calls
>>> request_region. On line 41 it calls __release_region, and on line 42 it
>>> calls synth_request_region again. The function synth_request_region
>>> (which calls request_region) is in a file named synth.c. But this code
>>> always fails.  Here is a kind of simplified version of it...
>>>
>>> int error;
>>> struct resource tmp;
>>> tmp .name = "ltlk";
>>> tmps.start = 0x3F8;
>>> tmp.end = 0x3FF;
>>> tmp.flags = IORESOURCE_BUSY;
>>> error = request_resource (&ioport_resource, &tmp);
>>>
>>> The error returned is always -16. I looked at the code in
>>> kernel/resource.c where the request_region function is  defined. It
>>> builds a linked list of resources with start & end addresses. If you
>>> request a region that is already within the start-end range of a
>>> resource already in the list, it returns an error code. But it looks as
>>> if the region for a serial port, 0x3f8 - 0x3ff,  in ioport_resource
>>> cannot be reserved because the entire range from 0x000 through 0xcf7 is
>>> already taken by something named "PCI Bus 0000:00".  Therefore calling
>>> request_resource always fails and the driver for the speech synth errors
>>> out.
>>>
>>> And therefore I can't use my hardware speech synth without modifying the
>>> kernel code.  If you comment out the line that checks the return code
>>> from request_region, it works. So you have to modify  the kernel code
>>> and compile a custom kernel to use a hardware speech synth. That's not
>>> such a problem for me but it is for a lot of people. Plus, the grml live
>>> CD doesn't work with hardware speech.  That is a problem for me.
>>>
>>> Can anyone tell me how to fix this so it can be patched in the official
>>> kernel code?
>>>
>>
>>
>> On 05/11/12 10:36, John Heim wrote:
>>> Subject: patch for speakup serial hardware synths
>>> A few weeks ago I asked about a problem with the speakup screen reader
>>> code. It does not work with serial hardware speech synthesizers. But I
>>> managed to get it working. How can I submit my fix for integration into
>>> the kernel code. The patch file can be downloaded here:
>>> http://www.math.wisc.edu/~jheim/downloads/patch-2012-03-06.patch
>>>
>>> To install it you cd to the linux source directory and do this:
>>> patch -i patch-2012-03-06.patch drivers/staging/speakup/serialio.c"
>>>
>> _______________________________________________
>> Speakup mailing list
>> Speakup@linux-speakup.org
>> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

John G. Heim, le Wed 01 May 2013 13:05:23 -0500, a écrit :
> But that's no answer.  This is the same problem I had with the responses  I
> got on the linux-kernel list. I asked extremely specific  questions below
> and just saying no isn't enough.

That's an XYZ problem. People won't answer Y because the actual issue is
X, whose answer is Z.

> Specifically, what is wrong with my patch?

It's dealing with Y, not X.

> Does speakup need to call request_reggion at all?

It shouldn't have to.

> If so, why,?  What does request_region do?

Try to circumvent the kernel.

> It looks as if request_region always fails if you request
> a region at the address of the serial port even if the serial port is not in
> use? Why is that?  If this behaviour is by design, what is the speakup code
> supposed to do instead?

It is supposed to open the serial code a correct way.

Samuel

Re: Any News on cut-and-paste bug?

[covici]

What I mean is that the error code is there for some reason -- its
trying to warn you of something, so investigation is needed and a real
solution found.  I would look at the 8250 module and see how it does the
registration/release itself.  However, maybe we need to go to a
different way of doing this -- this is some of what I want to discuss on
the call.

John G. Heim <jheim@math.wisc.edu> wrote:

> But that's no answer.  This is the same problem I had with the
> responses I got on the linux-kernel list. I asked extremely specific
> questions below and just saying no isn't enough.  Specifically, what
> is wrong with my patch? Does speakup need to call request_reggion at
> all? If so, why,? What does request_region do? It looks as if
> request_region always fails if you request a region at the address of
> the serial port even if the serial port is not in use? Why is that?
> If this behaviour is by design, what is the speakup code supposed to
> do instead?
> 
> You can say nothing. Say you're too busy or whatever. But don't just say no.
> 
> On 05/01/13 11:30, covici@ccs.covici.com wrote:
> > I would not accept that fix in the kernel, either -- nice workaround,
> > but its just hiding the real problem.
> >
> > John G. Heim <jheim@math.wisc.edu> wrote:
> >
> >>
> >>
> >> On 05/01/13 10:45, Samuel Thibault wrote:
> >>> John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
> >>>> All I wanted to do was get rid of what I suspected was a call to a function
> >>>> that apparently did nothing and the subsequent erroring out. But nobody
> >>>> seemed to know what the function did or if they did, they weren't sharing.
> >>>> They did, however, take the time to criticize the speakup code itself.
> >>>
> >>> Do you have a reference of the thread? (like the precise date when it
> >>> happened, or the subject, etc.)
> >>
> >> Here are 2 of the messages I posted. I haven't found an archive of the
> >> linux-kernel list. But I haven't looked too hard.  But these messages
> >> show the exact date and subject line of the 2 threads  on the list.
> >>
> >> On 03/03/12 11:18, John G. Heim wrote:
> >>> Subject: speakup bug
> >>> I need help fixing a bug in the driver for serial hardware  speech
> >>> synths in the speakup screen reader. According to the comments in the
> >>> code, it is in a part of the code that is trying to "steal" the serial
> >>> port.
> >>>
> >>> First it calls request_region and when that fails (it always fails), it
> >>> calls         __release_region(&then it calls release_region again to
> >>> see if __release_region worked. But it never works because the region
> >>> being requested is already taken.
> >>>
> >>> The code in question is in 2 source code files in
> >>> drivers/staging/speakup. It starts in serialio.c on line 38. Here it
> >>> calls a function named synth_request_region which in turn, calls
> >>> request_region. On line 41 it calls __release_region, and on line 42 it
> >>> calls synth_request_region again. The function synth_request_region
> >>> (which calls request_region) is in a file named synth.c. But this code
> >>> always fails.  Here is a kind of simplified version of it...
> >>>
> >>> int error;
> >>> struct resource tmp;
> >>> tmp .name = "ltlk";
> >>> tmps.start = 0x3F8;
> >>> tmp.end = 0x3FF;
> >>> tmp.flags = IORESOURCE_BUSY;
> >>> error = request_resource (&ioport_resource, &tmp);
> >>>
> >>> The error returned is always -16. I looked at the code in
> >>> kernel/resource.c where the request_region function is  defined. It
> >>> builds a linked list of resources with start & end addresses. If you
> >>> request a region that is already within the start-end range of a
> >>> resource already in the list, it returns an error code. But it looks as
> >>> if the region for a serial port, 0x3f8 - 0x3ff,  in ioport_resource
> >>> cannot be reserved because the entire range from 0x000 through 0xcf7 is
> >>> already taken by something named "PCI Bus 0000:00".  Therefore calling
> >>> request_resource always fails and the driver for the speech synth errors
> >>> out.
> >>>
> >>> And therefore I can't use my hardware speech synth without modifying the
> >>> kernel code.  If you comment out the line that checks the return code
> >>> from request_region, it works. So you have to modify  the kernel code
> >>> and compile a custom kernel to use a hardware speech synth. That's not
> >>> such a problem for me but it is for a lot of people. Plus, the grml live
> >>> CD doesn't work with hardware speech.  That is a problem for me.
> >>>
> >>> Can anyone tell me how to fix this so it can be patched in the official
> >>> kernel code?
> >>>
> >>
> >>
> >> On 05/11/12 10:36, John Heim wrote:
> >>> Subject: patch for speakup serial hardware synths
> >>> A few weeks ago I asked about a problem with the speakup screen reader
> >>> code. It does not work with serial hardware speech synthesizers. But I
> >>> managed to get it working. How can I submit my fix for integration into
> >>> the kernel code. The patch file can be downloaded here:
> >>> http://www.math.wisc.edu/~jheim/downloads/patch-2012-03-06.patch
> >>>
> >>> To install it you cd to the linux source directory and do this:
> >>> patch -i patch-2012-03-06.patch drivers/staging/speakup/serialio.c"
> >>>
> >> _______________________________________________
> >> Speakup mailing list
> >> Speakup@linux-speakup.org
> >> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
> >
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: Any News on cut-and-paste bug?

[John G. Heim]

Are you saying the bug was fixed by modifying speakup to open the serial 
port the right way? That problem is gone?

On 05/01/13 13:13, Samuel Thibault wrote:
> John G. Heim, le Wed 01 May 2013 13:05:23 -0500, a écrit :
>> But that's no answer.  This is the same problem I had with the responses  I
>> got on the linux-kernel list. I asked extremely specific  questions below
>> and just saying no isn't enough.
>
> That's an XYZ problem. People won't answer Y because the actual issue is
> X, whose answer is Z.
>
>> Specifically, what is wrong with my patch?
>
> It's dealing with Y, not X.
>
>> Does speakup need to call request_reggion at all?
>
> It shouldn't have to.
>
>> If so, why,?  What does request_region do?
>
> Try to circumvent the kernel.
>
>> It looks as if request_region always fails if you request
>> a region at the address of the serial port even if the serial port is not in
>> use? Why is that?  If this behaviour is by design, what is the speakup code
>> supposed to do instead?
>
> It is supposed to open the serial code a correct way.
>
> Samuel
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>

Re: Any News on cut-and-paste bug?

[William Hubbs]

On Wed, May 01, 2013 at 05:46:28PM +0200, Samuel Thibault wrote:
> John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
> > They did, however, take the time to criticize the speakup code itself.
> 
> I'm afraid that's for a good reason: speakup works around the kernel,
> that's not an approach that can work on the long term.

Samuel is correct.

I haven't heard anything on lkml indicating that there is a vendetta
against speakup being in the kernel; it is just a very complex issue
to solve, and no one has found a solution yet.

Basically, staging/speakup/serialio.c needs to be completely rewritten,
or possibly thrown out. The problem is that it uses hard coded addresses
for the serial ports instead of communicating with them through the
kernel serial port driver. That is why we are having more and more
systems lately where speakup is having issues with serial ports. This is
also why Samuel correctly says that speakup is working around the
kernel.

The proper solution to this issue would be to have a way that speakup
can interface with the kernel serial port drivers.

It will take some rewriting on the kernel side, as well as in speakup,
to give us that functionality. It all comes down to time, and figuring
out how to make this happen.

William

Re: Any News on cut-and-paste bug?

[Al Sten-Clanton]

Is there a layman's explanation you can give or point to about why 
Speakup's working "around the kernel" is a bad approach?  I ask because 
I've understood one virtue of it to be that it's what allows a blind 
person using a hardware speech output device to get information almost 
as early as a sighted person can on boot-up.

I admit knowing very little about the kernel.  I've just begun stucying 
C in the hope that some day I might be able to help with problems like 
this.  (This was prompted by recent discussions about accessibility--or 
comparative lack of it--on Red Hat and Fedora.)

Incidentally, if somebody knowledgeable on the matter thinks I'd be 
superfluous or wasting my time, feel free to say so and explain why.

Thanks!

Al

On 05/01/2013 03:16 PM, William Hubbs wrote:
> On Wed, May 01, 2013 at 05:46:28PM +0200, Samuel Thibault wrote:
>> John G. Heim, le Wed 01 May 2013 10:42:39 -0500, a écrit :
>>> They did, however, take the time to criticize the speakup code itself.
>>
>> I'm afraid that's for a good reason: speakup works around the kernel,
>> that's not an approach that can work on the long term.
>
> Samuel is correct.
>
> I haven't heard anything on lkml indicating that there is a vendetta
> against speakup being in the kernel; it is just a very complex issue
> to solve, and no one has found a solution yet.
>
> Basically, staging/speakup/serialio.c needs to be completely rewritten,
> or possibly thrown out. The problem is that it uses hard coded addresses
> for the serial ports instead of communicating with them through the
> kernel serial port driver. That is why we are having more and more
> systems lately where speakup is having issues with serial ports. This is
> also why Samuel correctly says that speakup is working around the
> kernel.
>
> The proper solution to this issue would be to have a way that speakup
> can interface with the kernel serial port drivers.
>
> It will take some rewriting on the kernel side, as well as in speakup,
> to give us that functionality. It all comes down to time, and figuring
> out how to make this happen.
>
> William
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

On Wed, 1 May 2013, covici@ccs.covici.com wrote:

> I meant his, not your actual one.  I have a client which is quite
> accessible, although Randy had some problems with it.  And you don't
> need sip if you want to call my box instead, I have numbers for that.

Hahahah, what kind of numbers would those be? 'grin'

It doesn't really matter to me where it is held. If you can handle
incoming skype to make it easier on Windows users fine. Otherwise a
lot of folks in this community already are comfortable with Brandon's
server.

What I would suggest however is people write the list or myself with
times they would be available if they wish to participate. Their time
zones would also be necessary.

I also think folks should submit agenda points and issues they'd like
to see discussed so we could have a useful meeting. So we could at
least have a direction anyway.

In some ways, having the discussion here would be better in my
opinion, so that we have a permanent record of any
discussion. However, I understand that some folks are more comfortable
in conversation than they are writing down their thoughts. I suppose
there is no reason not to have the discussion here and on a conference
call as well.


-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[Brandon McGinty-Carroll]

I'll take notes or record (with permission) the call for future reference.
My time is my own for better or worse, so I'll work with whatever restrictions the group has.

Brandon

On Wed, May 01, 2013 at 04:05:33PM -0400, Kirk Reiser wrote:
> On Wed, 1 May 2013, covici@ccs.covici.com wrote:
> 
> >I meant his, not your actual one.  I have a client which is quite
> >accessible, although Randy had some problems with it.  And you don't
> >need sip if you want to call my box instead, I have numbers for that.
> 
> Hahahah, what kind of numbers would those be? 'grin'
> 
> It doesn't really matter to me where it is held. If you can handle
> incoming skype to make it easier on Windows users fine. Otherwise a
> lot of folks in this community already are comfortable with Brandon's
> server.
> 
> What I would suggest however is people write the list or myself with
> times they would be available if they wish to participate. Their time
> zones would also be necessary.
> 
> I also think folks should submit agenda points and issues they'd like
> to see discussed so we could have a useful meeting. So we could at
> least have a direction anyway.
> 
> In some ways, having the discussion here would be better in my
> opinion, so that we have a permanent record of any
> discussion. However, I understand that some folks are more comfortable
> in conversation than they are writing down their thoughts. I suppose
> there is no reason not to have the discussion here and on a conference
> call as well.
> 
> 
> -- 
> Well that's it then, colour me gone!
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

Re: Any News on cut-and-paste bug?

[Chris Brannon]

Kirk Reiser <kirk@reisers.ca> writes:

> As far as I know, Chris Brannon submitted a patch to fix the
> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
> patch ever made it into the kernel speakup version or not.

You're thinking of the "goto position" bug, which I fixed.  The system
would crash every time someone tried to use the "go to" function.

As for the cut and paste bug, I haven't been able to reproduce it
consistently, and I never tracked down the cause.

-- Chris

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

Woops! Sorry about that.

I guess that should be an item for the list of topics.

On Wed, 1 May 2013, Chris Brannon wrote:

> Kirk Reiser <kirk@reisers.ca> writes:
>
>> As far as I know, Chris Brannon submitted a patch to fix the
>> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
>> patch ever made it into the kernel speakup version or not.
>
> You're thinking of the "goto position" bug, which I fixed.  The system
> would crash every time someone tried to use the "go to" function.
>
> As for the cut and paste bug, I haven't been able to reproduce it
> consistently, and I never tracked down the cause.
>
> -- Chris
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
>

-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

John G. Heim, le Wed 01 May 2013 14:07:28 -0500, a écrit :
> Are you saying the bug was fixed by modifying speakup to open the serial
> port the right way?

I'm said the bug should be fixed that way.

> That problem is gone?

Re: Any News on cut-and-paste bug?

[Samuel Thibault]

Al Sten-Clanton, le Wed 01 May 2013 15:46:44 -0400, a écrit :
> Is there a layman's explanation you can give or point to about why Speakup's
> working "around the kernel" is a bad approach?

At least because that  can not work with PCI serial devices, USB devices,
etc.

Samuel

Re: Any News on cut-and-paste bug?

[Jason White]

Kirk Reiser  <speakup@linux-speakup.org> wrote:

>Without new blood interested in taking speakup further, you may very
>well be correct.

What is the required background for this sort of work?

Knowledge of C, algorithms and data structures, kernel design/programming,
and a lot of software development experience would seem to be the recipe.

I don't qualify but I am working to improve my background in order to
contribute to more accessibility-related projects.

With apologies if I'm taking the thread in an undesired direction.

Re: Any News on cut-and-paste bug?

[Tony Baechler]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

I haven't tried the 3.2 from Squeeze backports, but I tried 3.2 from
Wheezy and my system locked up as soon as I loaded the DEC Express module,
after first killing espeakup and removing speakup_soft.  Are you running a
32-bit kernel?  I got a report that serial synths work fine with 32-bit.
I only run 64-bit here.

On 5/1/2013 6:45 AM, John G. Heim wrote:
> 
> 
> A big part of the problem is that even if someone is willing to take
> on writing fixes for speakup, the kernel people won't cooperate. I
> tried to get some help/advice from the linux kernel  list on
> implementing the bug fix I had for serial synths. Their advice -- start
> over.
> 
> PS: I think that particular bug got fixed for real by someone who knows
> way more about the speakup code than I do. I'm running a 3.2 kernel
> from squeeze backports and the bbug is fixed.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (MingW32)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=4fwD
-----END PGP SIGNATURE-----

Re: Any News on cut-and-paste bug?

[Kirk Reiser]

On Thu, 2 May 2013, Jason White wrote:

> What is the required background for this sort of work?
>
> Knowledge of C, algorithms and data structures, kernel design/programming,
> and a lot of software development experience would seem to be the recipe.

Hi Jason: Actually there is lots of things you or anyone interested
can do with this latest set of suggestions William Hubbs has posted
from the kernel staging mailing list.

Many of the suggestions are simple clean-up changes which needs more
shear manpower than real technical expertise. It can also help one
become familiar with the speakup code base and may even shed a tad of
understanding.

We can always use more willing hands and heads.

-- 
Well that's it then, colour me gone!

Re: Any News on cut-and-paste bug?

[Janina Sajka]

Cut and paste hasn't been a problem for me on Fedora 18 kernels since
the patch went in.

Janina

Kirk Reiser writes:
> On Tue, 30 Apr 2013, Hart Larry wrote:
> 
> >Quite some months ago Bill Acker suggested I login as an
> >ssh localhost
> >on each console where I would want to cut-and-paste.  Well,
> >actually unless there were a way for this anoyance to just ruin 1
> >tty instead of freeze an entire machine?  I also suppose having a
> >script on bootup log us in a localhost.
> >This bug just comes so suddenly, no warning--and-best as we can
> >tell all activity stops.
> 
> As far as I know, Chris Brannon submitted a patch to fix the
> cut-and-paste lock-up bug somewhere around 3.2.x. I don't know if that
> patch ever made it into the kernel speakup version or not.
> 
> >I realize-and-appreciate that we have an active community, many
> >who are knowledgeable, ETC.  But it almost seems Speakup may join
> >YASR as having gotten at a certain level-and-thats it.
> 
> Without new blood interested in taking speakup further, you may very
> well be correct.
> 
> >I've been on this list since 2003, but now for `quite some time I
> >still cannot move up past 2.632 as I would have no DecTalk speech.
> >I think John Heim wrote a patch to fix this, but I have no idea
> >what steps will install?
> >And lastly, still about the DecTalk, if we can ever produce a log
> >showing commands which Speakup is sending, James says he can
> >assist.
> 
> Unless someone decides to take on writing external drivers for USB and
> RS232C synths, you will never see a DECTalk Express fix. The only
> support over the past few years has been for the softsynth version of
> speakup. There have been a few serial fixes but they have been more of
> an aside than anything else. The serial synth substructure is terribly
> out of date and nobody appears to be willing to rewrite it.
> 
> 
> -- 
> Well that's it then, colour me gone!
> _______________________________________________
> Speakup mailing list
> Speakup@linux-speakup.org
> http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

-- 

Janina Sajka,	Phone:	+1.443.300.2200
			sip:janina@asterisk.rednote.net
		Email:	janina@rednote.net

Linux Foundation Fellow
Executive Chair, Accessibility Workgroup:	http://a11y.org

The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
Chair,	Protocols & Formats	http://www.w3.org/wai/pf
	Indie UI			http://www.w3.org/WAI/IndieUI/