[Crash-utility] [PATCH v1 01/10] Add LoongArch64 framework code support

[Ming Wang]

Hi, Lianbo

On 10/8/23 20:00, lijiang wrote:
> Hi, Ming
> So sorry for the late reply.
> On Thu, Sep 7, 2023 at 9:56 AM <crash-utility-request at redhat.com> wrote:
>
>     Date: Thu,  7 Sep 2023 09:46:42 +0800
>     From: Ming Wang <wangming01 at loongson.cn>
>     To: crash-utility at redhat.com
>     Cc: yangtiezhu at loongson.cn,     lixuefeng at loongson.cn,
>             chenhuacai at kernel.org
>     Subject: [Crash-utility] [PATCH v1 01/10] Add LoongArch64 framework
>             code support
>     Message-ID: <20230907014651.167784-2-wangming01 at loongson.cn>
>     Content-Type: text/plain; charset=UTF-8
>
>  
> The gdb patches have their own format(style), for more details, please refer to the current gdb-10.2.patch.
>
I tested the diff /git diff and git format-patch methods to generate patches. The patch format generated by these methods seems to be different from gdb-10.2.patch.

I also read the man manuals of these methods and found no relevant parameters.

If I want to update gdb-10.2.patch, what command sequence should I use to generate a patch file in the following format:


--- gdb-10.2/Makefile.in.orig
 +++ gdb-10.2/Makefile.in
 @@ -340,6 +340,9 @@ AR_FOR_BUILD = @AR_FOR_BUILD@
  AS_FOR_BUILD = @AS_FOR_BUILD@
  CC_FOR_BUILD = @CC_FOR_BUILD@
  CFLAGS_FOR_BUILD = @CFLAGS_FOR_BUILD@
 +ifeq (${CRASH_TARGET}, PPC64)
 +CFLAGS_FOR_BUILD += -m64 -fPIC
 +endif


What confuses me is how the .orig in the patch is generated.

If I integrate loongarch's gdb modifications into the gdb-10.2.patch file, what method is recommended?