diff mbox

Issue with patch in a reply to a patch

Message ID 20170405140603.GA5608@linux-mips.org
State Not Applicable
Headers show

Commit Message

Ralf Baechle April 5, 2017, 2:06 p.m. UTC
Hi,

since a recent upgrade of patchwork in February I've observed patchwork
misshandling patches sent as reply to another patch.  An example is

  https://patchwork.linux-mips.org/patch/15887/

where James' reply (first reply) was submitted like:

[...]
So I think something like this would be more correct. Does that fix your
problem?


Thanks
James
[...]

But then mangled into:

[...]
Thanks
Jamesdiff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
^^^^^^^^^
[...]

James' archived original mail in mbox format is available at:

https://www.linux-mips.org/cgi-bin/extract-mesg.cgi?a=linux-mips&m=2017-04&i=20170404203219.GK31606%40jhogan-linux.le.imgtec.org

the whole thread at

https://www.linux-mips.org/archives/linux-mips/2017-04/msg00035.html

  Ralf

Comments

Stephen Finucane April 9, 2017, 4:13 p.m. UTC | #1
On Wed, 2017-04-05 at 16:06 +0200, Ralf Baechle wrote:
> Hi,
> 
> since a recent upgrade of patchwork in February I've observed
> patchwork
> misshandling patches sent as reply to another patch.  An example is
> 
>   https://patchwork.linux-mips.org/patch/15887/
> 
> where James' reply (first reply) was submitted like:
> 
> [...]
> So I think something like this would be more correct. Does that fix
> your
> problem?
> 
> diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-
> r4k.c
> index 804d2a2a19fe..dd6a18bc10ab 100644
> --- a/arch/mips/kernel/cevt-r4k.c
> +++ b/arch/mips/kernel/cevt-r4k.c
> @@ -80,7 +80,7 @@ static unsigned int calculate_min_delta(void)
>                 }
> 
>                 /* Sorted insert of 75th percentile into buf2 */
> -               for (k = 0; k < i; ++k) {
> +               for (k = 0; k < i && k < ARRAY_SIZE(buf2); ++k) {
>                         if (buf1[ARRAY_SIZE(buf1) - 1] < buf2[k]) {
>                                 l = min_t(unsigned int,
>                                           i, ARRAY_SIZE(buf2) - 1);
> 
> Thanks
> James
> [...]
> 
> But then mangled into:
> 
> [...]
> Thanks
> Jamesdiff --git a/arch/mips/kernel/cevt-r4k.c
> b/arch/mips/kernel/cevt-r4k.c
> ^^^^^^^^^
> [...]
> 
> James' archived original mail in mbox format is available at:
> 
> https://www.linux-mips.org/cgi-bin/extract-mesg.cgi?a=linux-mips&m=20
> 17-04&i=20170404203219.GK31606%40jhogan-linux.le.imgtec.org
> 
> the whole thread at
> 
> https://www.linux-mips.org/archives/linux-mips/2017-04/msg00035.html
> 
>   Ralf

Aye, this looks like a bug introduced with some recent changes around
parsing replies. I've a fix up [1] and would appreciate your reviews :)

Cheers,
Stephen

[1] https://patchwork.ozlabs.org/patch/748756/
diff mbox

Patch

diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 804d2a2a19fe..dd6a18bc10ab 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -80,7 +80,7 @@  static unsigned int calculate_min_delta(void)
                }

                /* Sorted insert of 75th percentile into buf2 */
-               for (k = 0; k < i; ++k) {
+               for (k = 0; k < i && k < ARRAY_SIZE(buf2); ++k) {
                        if (buf1[ARRAY_SIZE(buf1) - 1] < buf2[k]) {
                                l = min_t(unsigned int,
                                          i, ARRAY_SIZE(buf2) - 1);