[PATCH 0/4] catch syscall -- try 3 -- Introduction

[PATCH 0/4] catch syscall -- try 3 -- Introduction

[Sérgio Durigan Júnior]

Hi, it's me again :-)

This is the third attempt to push this patch upstream :-). I've
basically reworked everything that you asked me, so I think things are
pretty much better now. Basically, the changes are:

- Incremented the documentation (as suggested by Eli)

- Fixed some bugs

- Now you can catch as many syscalls as you want in one single command
(thanks to Tom Tromey!)

- Create a "linux-tdep.[ch]", and put the common syscall functions there
(suggestion by Daniel Jacobowits)

- Put the syscall_state inside lwp_info (as requested by Pedro)

... and other minor changes. Unfortunately (only for Eli, apparently), I
decided to keep the mechanism the way it is, so you'll still find
syscalls being stored as numbers in this version of the patch. That's
not personal; I only thought that people in general were happy with the
way I did.

It's also good to mention that the support for x86_64 arch is almost
done (I have one little regression to solve). That's for you, Phil ;-).

Ok, I'm tired (3:30am here) so I won't write too much. I'd appreciate
your review :-).

Thanks,
 
-- 
Sérgio Durigan Júnior
Linux on Power Toolchain - Software Engineer
Linux Technology Center - LTC
IBM Brazil

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[Eli Zaretskii]

> From: =?ISO-8859-1?Q?S=E9rgio?= Durigan =?ISO-8859-1?Q?J=FAnior?= <sergiodj@linux.vnet.ibm.com>
> Date: Tue, 18 Nov 2008 03:45:10 -0200
> 
> ... and other minor changes. Unfortunately (only for Eli, apparently), I
> decided to keep the mechanism the way it is, so you'll still find
> syscalls being stored as numbers in this version of the patch. That's
> not personal; I only thought that people in general were happy with the
> way I did.

That's okay, I'm used to have my opinions overruled here.

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[Sérgio Durigan Júnior]

Hello guys,

Ping. :-)

[]s

On Tue, 2008-11-18 at 03:45 -0200, Sérgio Durigan Júnior wrote:
> Hi, it's me again :-)
> 
> This is the third attempt to push this patch upstream :-). I've
> basically reworked everything that you asked me, so I think things are
> pretty much better now. Basically, the changes are:
> 
> - Incremented the documentation (as suggested by Eli)
> 
> - Fixed some bugs
> 
> - Now you can catch as many syscalls as you want in one single command
> (thanks to Tom Tromey!)
> 
> - Create a "linux-tdep.[ch]", and put the common syscall functions there
> (suggestion by Daniel Jacobowits)
> 
> - Put the syscall_state inside lwp_info (as requested by Pedro)
> 
> ... and other minor changes. Unfortunately (only for Eli, apparently), I
> decided to keep the mechanism the way it is, so you'll still find
> syscalls being stored as numbers in this version of the patch. That's
> not personal; I only thought that people in general were happy with the
> way I did.
> 
> It's also good to mention that the support for x86_64 arch is almost
> done (I have one little regression to solve). That's for you, Phil ;-).
> 
> Ok, I'm tired (3:30am here) so I won't write too much. I'd appreciate
> your review :-).
> 
> Thanks,
> 
-- 
Sérgio Durigan Júnior
Linux on Power Toolchain - Software Engineer
Linux Technology Center - LTC
IBM Brazil

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[Sérgio Durigan Júnior]

Hi,

Ping again.

Thanks.

On Mon, 2008-11-24 at 21:16 -0200, Sérgio Durigan Júnior wrote:
> Hello guys,
> 
> Ping. :-)
> 
> []s
> 
> On Tue, 2008-11-18 at 03:45 -0200, Sérgio Durigan Júnior wrote:
> > Hi, it's me again :-)
> > 
> > This is the third attempt to push this patch upstream :-). I've
> > basically reworked everything that you asked me, so I think things are
> > pretty much better now. Basically, the changes are:
> > 
> > - Incremented the documentation (as suggested by Eli)
> > 
> > - Fixed some bugs
> > 
> > - Now you can catch as many syscalls as you want in one single command
> > (thanks to Tom Tromey!)
> > 
> > - Create a "linux-tdep.[ch]", and put the common syscall functions there
> > (suggestion by Daniel Jacobowits)
> > 
> > - Put the syscall_state inside lwp_info (as requested by Pedro)
> > 
> > ... and other minor changes. Unfortunately (only for Eli, apparently), I
> > decided to keep the mechanism the way it is, so you'll still find
> > syscalls being stored as numbers in this version of the patch. That's
> > not personal; I only thought that people in general were happy with the
> > way I did.
> > 
> > It's also good to mention that the support for x86_64 arch is almost
> > done (I have one little regression to solve). That's for you, Phil ;-).
> > 
> > Ok, I'm tired (3:30am here) so I won't write too much. I'd appreciate
> > your review :-).
> > 
> > Thanks,
> > 
-- 
Sérgio Durigan Júnior
Linux on Power Toolchain - Software Engineer
Linux Technology Center - LTC
IBM Brazil

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[Sérgio Durigan Júnior]

Let's try again...

sergio@miki # ping gdb-patches

:-)

On Tue, 2008-12-09 at 17:34 -0200, Sérgio Durigan Júnior wrote:
> Hi,
> 
> Ping again.
> 
> Thanks.
> 
> On Mon, 2008-11-24 at 21:16 -0200, Sérgio Durigan Júnior wrote:
> > Hello guys,
> > 
> > Ping. :-)
> > 
> > []s
> > 
> > On Tue, 2008-11-18 at 03:45 -0200, Sérgio Durigan Júnior wrote:
> > > Hi, it's me again :-)
> > > 
> > > This is the third attempt to push this patch upstream :-). I've
> > > basically reworked everything that you asked me, so I think things are
> > > pretty much better now. Basically, the changes are:
> > > 
> > > - Incremented the documentation (as suggested by Eli)
> > > 
> > > - Fixed some bugs
> > > 
> > > - Now you can catch as many syscalls as you want in one single command
> > > (thanks to Tom Tromey!)
> > > 
> > > - Create a "linux-tdep.[ch]", and put the common syscall functions there
> > > (suggestion by Daniel Jacobowits)
> > > 
> > > - Put the syscall_state inside lwp_info (as requested by Pedro)
> > > 
> > > ... and other minor changes. Unfortunately (only for Eli, apparently), I
> > > decided to keep the mechanism the way it is, so you'll still find
> > > syscalls being stored as numbers in this version of the patch. That's
> > > not personal; I only thought that people in general were happy with the
> > > way I did.
> > > 
> > > It's also good to mention that the support for x86_64 arch is almost
> > > done (I have one little regression to solve). That's for you, Phil ;-).
> > > 
> > > Ok, I'm tired (3:30am here) so I won't write too much. I'd appreciate
> > > your review :-).
> > > 
> > > Thanks,
> > > 
-- 
Sérgio Durigan Júnior
Linux on Power Toolchain - Software Engineer
Linux Technology Center - LTC
IBM Brazil

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[teawater]

This patch still not in?

On Tue, Dec 23, 2008 at 04:28, Sérgio Durigan Júnior
<sergiodj@linux.vnet.ibm.com> wrote:
> Let's try again...
>
> sergio@miki # ping gdb-patches
>
> :-)
>
> On Tue, 2008-12-09 at 17:34 -0200, Sérgio Durigan Júnior wrote:
>> Hi,
>>
>> Ping again.
>>
>> Thanks.
>>
>> On Mon, 2008-11-24 at 21:16 -0200, Sérgio Durigan Júnior wrote:
>> > Hello guys,
>> >
>> > Ping. :-)
>> >
>> > []s
>> >
>> > On Tue, 2008-11-18 at 03:45 -0200, Sérgio Durigan Júnior wrote:
>> > > Hi, it's me again :-)
>> > >
>> > > This is the third attempt to push this patch upstream :-). I've
>> > > basically reworked everything that you asked me, so I think things are
>> > > pretty much better now. Basically, the changes are:
>> > >
>> > > - Incremented the documentation (as suggested by Eli)
>> > >
>> > > - Fixed some bugs
>> > >
>> > > - Now you can catch as many syscalls as you want in one single command
>> > > (thanks to Tom Tromey!)
>> > >
>> > > - Create a "linux-tdep.[ch]", and put the common syscall functions there
>> > > (suggestion by Daniel Jacobowits)
>> > >
>> > > - Put the syscall_state inside lwp_info (as requested by Pedro)
>> > >
>> > > ... and other minor changes. Unfortunately (only for Eli, apparently), I
>> > > decided to keep the mechanism the way it is, so you'll still find
>> > > syscalls being stored as numbers in this version of the patch. That's
>> > > not personal; I only thought that people in general were happy with the
>> > > way I did.
>> > >
>> > > It's also good to mention that the support for x86_64 arch is almost
>> > > done (I have one little regression to solve). That's for you, Phil ;-).
>> > >
>> > > Ok, I'm tired (3:30am here) so I won't write too much. I'd appreciate
>> > > your review :-).
>> > >
>> > > Thanks,
>> > >
> --
> Sérgio Durigan Júnior
> Linux on Power Toolchain - Software Engineer
> Linux Technology Center - LTC
> IBM Brazil
>
>

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[Daniel Jacobowitz]

On Fri, Jan 16, 2009 at 10:25:46AM +0800, teawater wrote:
> This patch still not in?

If I followed correctly, Sergio got some useful feedback from Jan
K. on the archer list.  Sergio, once that's resolved, could you
resubmit?  I agree with Jan about getting rid of the entry breakpoint.

-- 
Daniel Jacobowitz
CodeSourcery

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[Sérgio Durigan Júnior]

On Fri, 2009-01-16 at 09:56 -0500, Daniel Jacobowitz wrote:

> If I followed correctly, Sergio got some useful feedback from Jan
> K. on the archer list.  Sergio, once that's resolved, could you
> resubmit?  I agree with Jan about getting rid of the entry breakpoint.

Hi Daniel,

You're right, Jan gave me useful feedback about the catch syscall patch.
I'm preparing the new version for re-submission; it shouldn't take too
long.

Thanks,

-- 
Sérgio Durigan Júnior
Linux on Power Toolchain - Software Engineer
Linux Technology Center - LTC
IBM Brazil

Re: [PATCH 0/4] catch syscall -- try 3 -- Introduction

[teawater]

That's great!  I can't wait to use it.  :)

On Sat, Jan 17, 2009 at 00:22, Sérgio Durigan Júnior
<sergiodj@linux.vnet.ibm.com> wrote:
> On Fri, 2009-01-16 at 09:56 -0500, Daniel Jacobowitz wrote:
>
>> If I followed correctly, Sergio got some useful feedback from Jan
>> K. on the archer list.  Sergio, once that's resolved, could you
>> resubmit?  I agree with Jan about getting rid of the entry breakpoint.
>
> Hi Daniel,
>
> You're right, Jan gave me useful feedback about the catch syscall patch.
> I'm preparing the new version for re-submission; it shouldn't take too
> long.
>
> Thanks,
>
> --
> Sérgio Durigan Júnior
> Linux on Power Toolchain - Software Engineer
> Linux Technology Center - LTC
> IBM Brazil
>
>