Send connman mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.01.org/mailman/listinfo/connman
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of connman digest..."
Today's Topics:
1. storage handling (Nakamura, Yusuke (ADITJ/SWG))
2. Re: storage handling (Patrik Flykt)
3. Re: [PATCH 1/3] Use AC_USE_SYSTEM_EXTENSIONS (Marcel Holtmann)
4. Re: storage handling (wangfe-nestlabs)
----------------------------------------------------------------------
Message: 1
Date: Wed, 10 Aug 2016 02:52:47 +0000
From: "Nakamura, Yusuke (ADITJ/SWG)" <[email protected]>
To: "[email protected]" <[email protected]>
Cc: "Ishikawa, Tetsuri (ADITJ/SWG)" <[email protected]>,
"Hoyer, Marko (ADITG/SW2)" <[email protected]>
Subject: storage handling
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="us-ascii"
Hi
I'm working on Connman integrated in an automotive system.
I'd like to reduce the frequency of writing data to the file system as much as
possible since it requires to support long term lifecycle and limited storage
capability.
Connman always saves a keyfile to the file system
(/var/lib/connman/service_name) when it connects to an access point
successfully.
If the user connects to the usual access point, the key to be updated in the
keyfile is usually only "Modified".
As far as I can see no one in the Connman seems to rely on this key. Therefore
I'd like to make Connman not to save the keyfile to the file system if updated
key is only "Modified".
What do you think?
Thanks,
Yusuke.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.01.org/pipermail/connman/attachments/20160810/4ac2bbec/attachment-0001.html>
------------------------------
Message: 2
Date: Wed, 10 Aug 2016 13:29:32 +0300
From: Patrik Flykt <[email protected]>
To: "Nakamura, Yusuke (ADITJ/SWG)" <[email protected]>,
"[email protected]" <[email protected]>
Cc: "Hoyer, Marko (ADITG/SW2)" <[email protected]>, "Ishikawa,
Tetsuri (ADITJ/SWG)" <[email protected]>
Subject: Re: storage handling
Message-ID: <[email protected]>
Content-Type: text/plain; charset="UTF-8"
On Wed, 2016-08-10 at 02:52 +0000, Nakamura, Yusuke (ADITJ/SWG) wrote:
> Hi
> ?
> I?m working on Connman integrated in an automotive system.
> I?d like to reduce the frequency of writing data to the file system
> as much as possible since it requires to support long term lifecycle
> and limited storage capability.
> Connman always saves a keyfile to the file system
> (/var/lib/connman/service_name) when it connects to an access point
> successfully.
Yes, please check/audit the code so that only relevant writes go to the
file. If ConnMan does other excessive writes to the file, please fix.
> If the user connects to the usual access point, the key to be updated
> in the keyfile is usually only ?Modified?.
> As far as I can see no one in the Connman seems to rely on this key.
> Therefore I?d like to make Connman not to save the keyfile to the
> file system if updated key is only ?Modified?.
I have used the Modified value in a ConnMan specific shell script
sorting saved services according to last used time and deleting the not
recently used ones. As such, the Modified value is useful for
accounting/statistics purposes and I'd like to keep it. That does not
prevent you from creating a nice patch not saving the file due to the
Modified value being updated and even sending it as an RFC to the
mailing list for anyone else needing a similar solution.
> What do you think?
I'd keep saving of Modified value in upstream for now, what do others
think?
Cheers,
Patrik
------------------------------
Message: 3
Date: Wed, 10 Aug 2016 12:30:57 +0200
From: Marcel Holtmann <[email protected]>
To: Ross Burton <[email protected]>
Cc: [email protected]
Subject: Re: [PATCH 1/3] Use AC_USE_SYSTEM_EXTENSIONS
Message-ID: <[email protected]>
Content-Type: text/plain; charset=us-ascii
Hi Ross,
> Instead of using #define _GNU_SOURCE in some source files which causes
> problems
> when building with musl as more files need the define, simply use
> AC_USE_SYSTEM_EXTENSIONS in configure.ac to get it defined globally.
so my preference was to define _GNU_SOURCE manually in each *.c file as
described by the manpages. For example:
SYNOPSIS
#define _GNU_SOURCE /* See feature_test_macros(7) */
#include <stdio.h>
int asprintf(char **strp, const char *fmt, ...);
And if some of the musl vs glibc descriptions are broken, then lets get these
fixed instead of just globally declaring it and hand waiving issues away.
I know that musl is more strict than glibc and that is a good thing. We want to
support musl all the way. However they might not get everything correct either.
So lets rather update manpages as well to have this correct in both places.
Regards
Marcel
------------------------------
Message: 4
Date: Wed, 10 Aug 2016 10:34:09 -0700
From: wangfe-nestlabs <[email protected]>
To: Patrik Flykt <[email protected]>
Cc: "Nakamura, Yusuke (ADITJ/SWG)" <[email protected]>,
"[email protected]" <[email protected]>, "Hoyer, Marko
(ADITG/SW2)" <[email protected]>, "Ishikawa, Tetsuri (ADITJ/SWG)"
<[email protected]>
Subject: Re: storage handling
Message-ID: <[email protected]>
Content-Type: text/plain; charset=utf-8
Hi,
Just a reminder, there is statistic file ?data? in the service_name directory
is updated.
And I would like add a compile flag to disable it.
Patrick, what is your opinion?
Thanks,
Feng
> On Aug 10, 2016, at 3:29 AM, Patrik Flykt <[email protected]>
> wrote:
>
> On Wed, 2016-08-10 at 02:52 +0000, Nakamura, Yusuke (ADITJ/SWG) wrote:
>> Hi
>>
>> I?m working on Connman integrated in an automotive system.
>> I?d like to reduce the frequency of writing data to the file system
>> as much as possible since it requires to support long term lifecycle
>> and limited storage capability.
>> Connman always saves a keyfile to the file system
>> (/var/lib/connman/service_name) when it connects to an access point
>> successfully.
>
> Yes, please check/audit the code so that only relevant writes go to the
> file. If ConnMan does other excessive writes to the file, please fix.
>
>> If the user connects to the usual access point, the key to be updated
>> in the keyfile is usually only ?Modified?.
>> As far as I can see no one in the Connman seems to rely on this key.
>> Therefore I?d like to make Connman not to save the keyfile to the
>> file system if updated key is only ?Modified?.
>
> I have used the Modified value in a ConnMan specific shell script
> sorting saved services according to last used time and deleting the not
> recently used ones. As such, the Modified value is useful for
> accounting/statistics purposes and I'd like to keep it. That does not
> prevent you from creating a nice patch not saving the file due to the
> Modified value being updated and even sending it as an RFC to the
> mailing list for anyone else needing a similar solution.
>
>> What do you think?
>
> I'd keep saving of Modified value in upstream for now, what do others
> think?
>
> Cheers,
>
> Patrik
> _______________________________________________
> connman mailing list
> [email protected]
> https://lists.01.org/mailman/listinfo/connman
------------------------------
Subject: Digest Footer
_______________________________________________
connman mailing list
[email protected]
https://lists.01.org/mailman/listinfo/connman
------------------------------
End of connman Digest, Vol 10, Issue 6
**************************************
