[oe] [PATCH 2/3] acpid: Make 1.0.10 compile.

[Christian Betz]

Søren and Paul:

I've officially signed off on my patch in the other thread, and have
also sent an email upstream to ted felix (maintainer of acpid-netlink
from which the patch in oe is derived). I'll let you know what I hear
back.

Of course my next question is why debian and oe are not using
acpid-netlink, but instead cherry pick certain files and then patch
vanilla acpid. I'll see if I can figure that out.

Christian

On Mon, May 16, 2011 at 9:08 AM, Paul Menzel
<paulepanter at users.sourceforge.net> wrote:
> Dear Søren,
>
>
> Am Montag, den 16.05.2011, 12:21 +0200 schrieb Søren Holm:
>> Patch is tested on i686.
>> Compilation failed with the following error :
>>
>> | i686-angstrom-linux-gcc -march=pentiumpro --sysroot=/home/sgh/oe-new-oe/build-vmc16/tmp/sysroots/i686-angstrom-linux -W -Wall -Werror -Wundef -Wshadow -D_GNU_SOURCE -O2 -g -DVERSION="\"1.0.10\""   -c -o
>> sock.o sock.c
>> | cc1: warnings being treated as errors
>> | libnetlink.c: In function 'addraw_l':
>> | libnetlink.c:521:56: error: comparison between signed and unsigned integer expressions
>
> Christian already submitted a patch for this issue [1][2], but did not
> provide a Signed-off-by line, which you also forgot in this patch by the
> way.
>
>> ---
>>  recipes/acpid/acpid-1.0.10/libnetlink.diff |   15 +++++++++++++++
>>  recipes/acpid/acpid_1.0.10.bb              |    1 +
>>  2 files changed, 16 insertions(+), 0 deletions(-)
>>  create mode 100644 recipes/acpid/acpid-1.0.10/libnetlink.diff
>>
>> diff --git a/recipes/acpid/acpid-1.0.10/libnetlink.diff b/recipes/acpid/acpid-1.0.10/libnetlink.diff
>> new file mode 100644
>> index 0000000..59ab951
>> --- /dev/null
>> +++ b/recipes/acpid/acpid-1.0.10/libnetlink.diff
>> @@ -0,0 +1,15 @@
>> +diff --git a/libnetlink.c b/libnetlink.c
>> +index 6f1ceea..d34dfe4 100644
>> +--- a/libnetlink.c
>> ++++ b/libnetlink.c
>> +@@ -518,8 +518,8 @@ int addattr_l(struct nlmsghdr *n, int maxlen, int type, const void *data,
>> +
>> + int addraw_l(struct nlmsghdr *n, int maxlen, const void *data, int len)
>> + {
>> +-    if ((int)NLMSG_ALIGN(n->nlmsg_len) + NLMSG_ALIGN(len) > maxlen) {
>> +-            fprintf(stderr, "addraw_l ERROR: message exceeded bound of %d\n",maxlen);
>> ++    if ((int)NLMSG_ALIGN(n->nlmsg_len) + (int)NLMSG_ALIGN(len) > maxlen) {
>> ++    fprintf(stderr, "addraw_l ERROR: message exceeded bound of %d\n",maxlen);
>
> This indentation change is not needed.
>
>> +             return -1;
>> +     }
>> +
>> diff --git a/recipes/acpid/acpid_1.0.10.bb b/recipes/acpid/acpid_1.0.10.bb
>> index 9d3170d..f0a6049 100644
>> --- a/recipes/acpid/acpid_1.0.10.bb
>> +++ b/recipes/acpid/acpid_1.0.10.bb
>> @@ -2,6 +2,7 @@ require acpid.inc
>>
>>  SRC_URI_append = " file://event.c.diff \
>>                     file://netlink.diff \
>> +                   file://libnetlink.diff \
>>                     file://gcc44.diff \
>>                     file://acpid-sys-stat.patch \
>>                   "
>
> I would say to wait for Christian’s reply to his thread and hopefully
> will get this in as soon as possible.
>
>
> Thanks,
>
> Paul
>
>
> [1] http://patches.openembedded.org/patch/4023/
> [2] http://patches.openembedded.org/patch/4061/
>



-- 
For there are moments when one can neither think nor feel.  And if one
can neither think nor feel, she thought, where is one?
 -- Virginia Woolf, "To the Lighthouse"

[Quoted in "VMS Internals and Data Structures", V4.4, when referring
to powerfail recovery.]