dynamic shared library .

dynamic shared library .

[ratheesh k]

when  dll is loaded  .

 at the start of execution of elf ?

     OR

 when the  reference is made into library function or data ?

-Ratheesh
--
To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: dynamic shared library .

[Michal Nazarewicz]

On Thu, 03 Feb 2011 15:50:27 +0100, ratheesh k <ratheesh.ksz@gmail.com>  
wrote:

> when  dll is loaded  .
>
>  at the start of execution of elf ?
>
>      OR
>
>  when the  reference is made into library function or data ?

Depends what you mean by "loaded".  At start of execution, dynamic loader
reads shared object's header files, maps the library to memory accordingly
and probably does some more magic with symbol resolution, etc. (I'm not
an expert and what dynamic loader does is not that trivial).

However, the whole file is not read into memory and instead, needed
portions are read as code references them.

-- 
Best regards,                                       _     _
.o. | Liege of Serenly Enlightened Majesty of     o' \,=./ `o
..o | Computer Science,  Michal "mina86" Nazarewicz  (o o)
ooo +-<email/jid: mnazarewicz@google.com>--------ooO--(_)--Ooo--

UDP data

[Randi]

Hi All,

	Since UDP socket is not reliable, what the common way in user level to
check UDP communication's data for error (eg. malformed, corrupted,
etc.)? is the checksum field in the UDP header (struct udphdr) intended
to this problem?

	Randi

Re: UDP data

[Chris Donovan]

Yes the checksum is for the header and data.
Keep in mind that the checksum value is optional in ipv4, but required in ipv6.

Chris-

On Thu, Feb 10, 2011 at 2:41 PM, Randi <nightdecoder@gmail.com> wrote:
> Hi All,
>
>        Since UDP socket is not reliable, what the common way in user level to
> check UDP communication's data for error (eg. malformed, corrupted,
> etc.)? is the checksum field in the UDP header (struct udphdr) intended
> to this problem?
>
>        Randi
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: UDP data

[Randi]

Chris, optional means it can be enable or not, how to enable it and get
the header to verify the checksum.


On Thu, 2011-02-10 at 15:14 +1100, Chris Donovan wrote:
> Yes the checksum is for the header and data.
> Keep in mind that the checksum value is optional in ipv4, but required in ipv6.
> 
> Chris-
> 
> On Thu, Feb 10, 2011 at 2:41 PM, Randi <nightdecoder@gmail.com> wrote:
> > Hi All,
> >
> >        Since UDP socket is not reliable, what the common way in user level to
> > check UDP communication's data for error (eg. malformed, corrupted,
> > etc.)? is the checksum field in the UDP header (struct udphdr) intended
> > to this problem?
> >
> >        Randi

Re: UDP data

[Glynn Clements]

Randi wrote:

> > Yes the checksum is for the header and data.
> > Keep in mind that the checksum value is optional in ipv4, but required in ipv6.
> 
> Chris, optional means it can be enable or not, how to enable it and get
> the header to verify the checksum.

Checksums are enabled by default. You can disable checksums when
sending using the SO_NO_CHECK option. On receipt, the kernel will
verify the checksum if it's present, and discard the packet if there
is an error.

-- 
Glynn Clements <glynn@gclements.plus.com>

Re: UDP data

[David Astua]

On Thu, Feb 10, 2011 at 9:41 AM, Glynn Clements
<glynn@gclements.plus.com> wrote:
>
> Randi wrote:
>
>> > Yes the checksum is for the header and data.
>> > Keep in mind that the checksum value is optional in ipv4, but required in ipv6.
>>
>> Chris, optional means it can be enable or not, how to enable it and get
>> the headehttps://mail.google.com/mail/?shva=1#inbox/12e0da8d716c914er to verify the checksum.
>
> Checksums are enabled by default. You can disable checksums when
> sending using the SO_NO_CHECK option. On receipt, the kernel will
> verify the checksum if it's present, and discard the packet if there
> is an error.
>
> --
> Glynn Clements <glynn@gclements.plus.com>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Also keep in mind that the check is for individual frames, even
checking that there's no certainty that all frames will arrive.

  -- David A.
--
To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: UDP data

[Chris Donovan]

I don't know libc, so I don't know what functions libc has to compute
(if any) the checksum.

I did find this page, that has some sample code though...
http://stackoverflow.com/questions/3737612/raw-socket-sendto-failed-using-c-on-linux

Chris-

On Thu, Feb 10, 2011 at 4:14 PM, Randi <nightdecoder@gmail.com> wrote:
> Chris, optional means it can be enable or not, how to enable it and get
> the header to verify the checksum.
>
>
> On Thu, 2011-02-10 at 15:14 +1100, Chris Donovan wrote:
>> Yes the checksum is for the header and data.
>> Keep in mind that the checksum value is optional in ipv4, but required in ipv6.
>>
>> Chris-
>>
>> On Thu, Feb 10, 2011 at 2:41 PM, Randi <nightdecoder@gmail.com> wrote:
>> > Hi All,
>> >
>> >        Since UDP socket is not reliable, what the common way in user level to
>> > check UDP communication's data for error (eg. malformed, corrupted,
>> > etc.)? is the checksum field in the UDP header (struct udphdr) intended
>> > to this problem?
>> >
>> >        Randi
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: dynamic shared library .

[Hendrik Visage]

On Thu, Feb 3, 2011 at 4:59 PM, Michal Nazarewicz <mina86@mina86.com> wrote:
> On Thu, 03 Feb 2011 15:50:27 +0100, ratheesh k <ratheesh.ksz@gmail.com>
> wrote:
>
>> when  dll is loaded  .
>>
>>  at the start of execution of elf ?
>>
>>     OR
>>
>>  when the  reference is made into library function or data ?
>
> Depends what you mean by "loaded".  At start of execution, dynamic loader
> reads shared object's header files, maps the library to memory accordingly
> and probably does some more magic with symbol resolution, etc. (I'm not
> an expert and what dynamic loader does is not that trivial).

That would be "normal" libraries that at compile time have been
referenced as needed to be dynamically linked. Ie. those libraries
that you have called/referenced in the compiled code and that the
linker have resolved to be in that dynamoc library. You can find that
list by using "ldd /path/to/<executable>" which would also show you
where the run time linker found those libraries (based on the ld.so
configured paths and the LD_LIBRARY_PATH setting)

Then the "fun" part are those dynamic executables that is loadedby the
program "on demand". An example systemwide is the name service stuff
refered to by the /etc/nsswitch.conf file. Another example is Apache's
loadable modules like the php, perl modules that is not refered to at
startup time by the ELF run time loader, but rather loaded as the
program executes and reads the config files that requests those
modules to be loaded.

>
> However, the whole file is not read into memory and instead, needed
> portions are read as code references them.

Typically the library/dll/.so is mmap(2)ed into the executable address
space, and only read as that part are referenced.

> --
> Best regards,                                       _     _
> .o. | Liege of Serenly Enlightened Majesty of     o' \,=./ `o
> ..o | Computer Science,  Michal "mina86" Nazarewicz  (o o)
> ooo +-<email/jid: mnazarewicz@google.com>--------ooO--(_)--Ooo--
> --
> To unsubscribe from this list: send the line "unsubscribe
> linux-c-programming" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-c-programming" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html