Linux Kernel Network Subsystem Patching

[ted leslie]

Not familiar with Git patch, but I guess it is just "patch".
follow this

https://groups.google.com/forum/#!topic/msysgit/6NNvXeyd5xM

-tl


On Fri, Jan 10, 2014 at 8:42 PM, Aruna Hewapathirane <
aruna.hewapathirane-Re5JQEeQqe8AvxtiuMwx3w at public.gmane.org> wrote:

> Ted believe me I have tried and tried and finally decided to ask for help.
> The patch errors are below:
>
> aruna at aruna-desktop:/media/kernel-dev/next-net/net-next$ git format-patch
> --subject-prefix="PATCH v1" HEAD^
> 0001-Network-Subsystem-v2.patch
>
> aruna at aruna-desktop:/media/kernel-dev/next-net/net-next$ git apply
> --check 0001-Network-Subsystem-v2.patch
>
> error: patch failed: net/core/neighbour.c:117
>
> error: net/core/neighbour.c: patch does not apply
>
> It makes no sense to me because all the other files compile clean nil
> errors :(
>
> On Fri, Jan 10, 2014 at 8:36 PM, ted leslie <ted.leslie-Re5JQEeQqe8AvxtiuMwx3w at public.gmane.org> wrote:
>
>> What patch error are you getting?
>>
>> You can see what the patch is doing, you can just go put the changed two
>> lines in with editor.
>> -tl
>>
>>
>>
>>
>> On Fri, Jan 10, 2014 at 8:05 PM, Aruna Hewapathirane <
>> aruna.hewapathirane-Re5JQEeQqe8AvxtiuMwx3w at public.gmane.org> wrote:
>>
>>> Hello Everybody,
>>>
>>> I am a total newbie to kernel hacking and I have a question. Am trying
>>> to patch the Network Subsystem and changing the net_random() to direct
>>> calls of prandom_u32. Why ? Because it is
>>> harder to audit / grep and new commits actually use prandom_u32 only am
>>> told.
>>>
>>> u32 prandom_u32(void);    /* defined in include/linux/random.h: */
>>>
>>> #define net_random()        prandom_u32() /* defined in
>>> include/linux/net.h */
>>>
>>> Everything works as expected until I get to net/core/neighbour.c and
>>> then patch fails to apply.
>>>
>>> The diff is here:
>>>
>>>
>>> -------------------------------------------------------------------------------------------------------------------------------
>>> diff --git a/net/core/neighbour.c b/net/core/neighbour.c
>>> index 47d834e..ea97361 100644
>>> --- a/net/core/neighbour.c
>>> +++ b/net/core/neighbour.c
>>> @@ -117,7 +117,7 @@ static void neigh_cleanup_and_release(struct
>>> neighbour *neig
>>>
>>>  unsigned long neigh_rand_reach_time(unsigned long base)
>>>  {
>>> -       return base ? (prandom_u32() % base) + (base >> 1) : 0;
>>> +       return base ? (net_random() % base) + (base >> 1) : 0;
>>>  }
>>>  EXPORT_SYMBOL(neigh_rand_reach_time);
>>>
>>> @@ -1415,7 +1415,7 @@ void pneigh_enqueue(struct neigh_table *tbl,
>>> struct neigh_
>>>                     struct sk_buff *skb)
>>>  {
>>>         unsigned long now = jiffies;
>>> -       unsigned long sched_next = now + (prandom_u32() %
>>> +       unsigned long sched_next = now + (net_random() %
>>>                                           NEIGH_VAR(p, PROXY_DELAY));
>>>
>>>         if (tbl->proxy_queue.qlen > NEIGH_VAR(p, PROXY_QLEN)) {
>>>
>>> -------------------------------------------------------------------------------------------------------------------------------
>>>
>>> I can use some help and guidance please, what am I doing wrong ? Or not
>>> seeing ? It's just this neighbor.c file that is doing this and I am
>>> stumped.
>>>
>>> Thank you.
>>> Aruna
>>>
>>> <https://sites.google.com/site/arunahewapathirane/home/business-card/buisness-card.png?attredirects=0>
>>>
>>
>>
>
>
> --
> *Aruna Hewapathirane*
> Consultant/Trainer
> Phone : 647-709-9269
> Website: <http://goog_1768911931>Open Source Solutions<http://sahanaya.net/aruna/>
>
>
>
>
> <https://sites.google.com/site/arunahewapathirane/home/business-card/buisness-card.png?attredirects=0>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gtalug.org/pipermail/legacy/attachments/20140110/ba87dd92/attachment.html>