Sashiko change proposal

Sashiko change proposal

[David Hildenbrand (Arm)]

(CCing all people attending the mm-next call)

Hi,

After talking to Roman at MS, he just informed me that sashiko now has a new
feature combination:
- "embargo_hours": hours before review is public + emails sent
- Mechanism for signed-in maintainers to lift embargo immediately

That solves the key issue I had with sashiko: people/agents re-spinning stuff
too fast before a human at least had the chance to jump in first.

I suggest the following sashiko change for linux-mm:

diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
index a9198de..ba123da 100644
--- a/projects/linux/mailing_lists.toml
+++ b/projects/linux/mailing_lists.toml
@@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
 ["linux-integrity@vger.kernel.org"]
 ["linux-kernel@vger.kernel.org"]
 ["linux-kselftest@vger.kernel.org"]
-["linux-mm@kvack.org"]
 ["linux-nvme@lists.infradead.org"]
 ["linux-omap@vger.kernel.org"]
 ["linux-pm@vger.kernel.org"]
@@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]

 ["fsverity@lists.linux.dev"]
 reply_to = ["author", "list", "recipients"]
+
+["linux-mm@kvack.org"]
+reply_to = ["author", "list", "recipients"]
+embargo_hours = 24


Which means:
- Embargo 24h
- Mails sent to everybody (author, list, CC list)
- No positive-review ("looks great") mails (default)
- No other static email addresses that receive a copy (default)
- No additional ignored author email addresses (default)

I'd say we start with 24h and see how it goes.

Thoughts?

-- 
Cheers,

David

Re: Sashiko change proposal

[Lorenzo Stoakes (ARM)]

On Thu, Oct 15, 2026 at 10:38:14AM +0200, David Hildenbrand (Arm) wrote:
> (CCing all people attending the mm-next call)
>
> Hi,
>
> After talking to Roman at MS, he just informed me that sashiko now has a new
> feature combination:
> - "embargo_hours": hours before review is public + emails sent
> - Mechanism for signed-in maintainers to lift embargo immediately
>
> That solves the key issue I had with sashiko: people/agents re-spinning stuff
> too fast before a human at least had the chance to jump in first.
>
> I suggest the following sashiko change for linux-mm:
>
> diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
> index a9198de..ba123da 100644
> --- a/projects/linux/mailing_lists.toml
> +++ b/projects/linux/mailing_lists.toml
> @@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
>  ["linux-integrity@vger.kernel.org"]
>  ["linux-kernel@vger.kernel.org"]
>  ["linux-kselftest@vger.kernel.org"]
> -["linux-mm@kvack.org"]
>  ["linux-nvme@lists.infradead.org"]
>  ["linux-omap@vger.kernel.org"]
>  ["linux-pm@vger.kernel.org"]
> @@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]
>
>  ["fsverity@lists.linux.dev"]
>  reply_to = ["author", "list", "recipients"]
> +
> +["linux-mm@kvack.org"]
> +reply_to = ["author", "list", "recipients"]
> +embargo_hours = 24
>
>
> Which means:
> - Embargo 24h
> - Mails sent to everybody (author, list, CC list)
> - No positive-review ("looks great") mails (default)
> - No other static email addresses that receive a copy (default)
> - No additional ignored author email addresses (default)
>
> I'd say we start with 24h and see how it goes.
>
> Thoughts?

Yes, no notes :)

Reviewed-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>

--
Cheers, Lorenzo

Re: Sashiko change proposal

[Mike Rapoport]

On Thu, Oct 15, 2026 at 10:38:14AM +0200, David Hildenbrand (Arm) wrote:
> (CCing all people attending the mm-next call)
> 
> Hi,
> 
> After talking to Roman at MS, he just informed me that sashiko now has a new
> feature combination:
> - "embargo_hours": hours before review is public + emails sent
> - Mechanism for signed-in maintainers to lift embargo immediately
> 
> That solves the key issue I had with sashiko: people/agents re-spinning stuff
> too fast before a human at least had the chance to jump in first.
> 
> I suggest the following sashiko change for linux-mm:
> 
> diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
> index a9198de..ba123da 100644
> --- a/projects/linux/mailing_lists.toml
> +++ b/projects/linux/mailing_lists.toml
> @@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
>  ["linux-integrity@vger.kernel.org"]
>  ["linux-kernel@vger.kernel.org"]
>  ["linux-kselftest@vger.kernel.org"]
> -["linux-mm@kvack.org"]
>  ["linux-nvme@lists.infradead.org"]
>  ["linux-omap@vger.kernel.org"]
>  ["linux-pm@vger.kernel.org"]
> @@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]
> 
>  ["fsverity@lists.linux.dev"]
>  reply_to = ["author", "list", "recipients"]
> +
> +["linux-mm@kvack.org"]
> +reply_to = ["author", "list", "recipients"]
> +embargo_hours = 24
> 
> 
> Which means:
> - Embargo 24h
> - Mails sent to everybody (author, list, CC list)
> - No positive-review ("looks great") mails (default)
> - No other static email addresses that receive a copy (default)
> - No additional ignored author email addresses (default)
> 
> I'd say we start with 24h and see how it goes.

Sounds good!
 
> Thoughts?
> 
> -- 
> Cheers,
> 
> David
> 

-- 
Sincerely yours,
Mike.

Re: Sashiko change proposal

[Vlastimil Babka (SUSE)]

On 10/15/26 10:38, David Hildenbrand (Arm) wrote:
> (CCing all people attending the mm-next call)
> 
> Hi,
> 
> After talking to Roman at MS, he just informed me that sashiko now has a new
> feature combination:
> - "embargo_hours": hours before review is public + emails sent
> - Mechanism for signed-in maintainers to lift embargo immediately
> 
> That solves the key issue I had with sashiko: people/agents re-spinning stuff
> too fast before a human at least had the chance to jump in first.
> 
> I suggest the following sashiko change for linux-mm:
> 
> diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
> index a9198de..ba123da 100644
> --- a/projects/linux/mailing_lists.toml
> +++ b/projects/linux/mailing_lists.toml
> @@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
>  ["linux-integrity@vger.kernel.org"]
>  ["linux-kernel@vger.kernel.org"]
>  ["linux-kselftest@vger.kernel.org"]
> -["linux-mm@kvack.org"]
>  ["linux-nvme@lists.infradead.org"]
>  ["linux-omap@vger.kernel.org"]
>  ["linux-pm@vger.kernel.org"]
> @@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]
> 
>  ["fsverity@lists.linux.dev"]
>  reply_to = ["author", "list", "recipients"]
> +
> +["linux-mm@kvack.org"]
> +reply_to = ["author", "list", "recipients"]
> +embargo_hours = 24
> 
> 
> Which means:
> - Embargo 24h
> - Mails sent to everybody (author, list, CC list)
> - No positive-review ("looks great") mails (default)
> - No other static email addresses that receive a copy (default)
> - No additional ignored author email addresses (default)
> 
> I'd say we start with 24h and see how it goes.
> 
> Thoughts?

LGTM!

Re: Sashiko change proposal

[Pratyush Yadav]

On Thu, Oct 15 2026, David Hildenbrand (Arm) wrote:

> (CCing all people attending the mm-next call)
>
> Hi,
>
> After talking to Roman at MS, he just informed me that sashiko now has a new
> feature combination:
> - "embargo_hours": hours before review is public + emails sent
> - Mechanism for signed-in maintainers to lift embargo immediately
>
> That solves the key issue I had with sashiko: people/agents re-spinning stuff
> too fast before a human at least had the chance to jump in first.

If a sender re-spins stuff too fast after a Sashiko review, that's a
pretty clear signal that it is an agent. For me, that makes it an easy
decision to ignore the sender. I'm curious, why would you want to lose
that signal? Is the goal to preserve Sashiko bandwidth and reduce
mailing list spam?

>
> I suggest the following sashiko change for linux-mm:
>
> diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
> index a9198de..ba123da 100644
> --- a/projects/linux/mailing_lists.toml
> +++ b/projects/linux/mailing_lists.toml
> @@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
>  ["linux-integrity@vger.kernel.org"]
>  ["linux-kernel@vger.kernel.org"]
>  ["linux-kselftest@vger.kernel.org"]
> -["linux-mm@kvack.org"]
>  ["linux-nvme@lists.infradead.org"]
>  ["linux-omap@vger.kernel.org"]
>  ["linux-pm@vger.kernel.org"]
> @@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]
>
>  ["fsverity@lists.linux.dev"]
>  reply_to = ["author", "list", "recipients"]
> +
> +["linux-mm@kvack.org"]
> +reply_to = ["author", "list", "recipients"]
> +embargo_hours = 24
>
>
> Which means:
> - Embargo 24h
> - Mails sent to everybody (author, list, CC list)
> - No positive-review ("looks great") mails (default)
> - No other static email addresses that receive a copy (default)
> - No additional ignored author email addresses (default)
>
> I'd say we start with 24h and see how it goes.
>
> Thoughts?

-- 
Regards,
Pratyush Yadav

Re: Sashiko change proposal

[David Hildenbrand (Arm)]

On 10/15/26 13:23, Pratyush Yadav wrote:
> On Thu, Oct 15 2026, David Hildenbrand (Arm) wrote:
> 
>> (CCing all people attending the mm-next call)
>>
>> Hi,
>>
>> After talking to Roman at MS, he just informed me that sashiko now has a new
>> feature combination:
>> - "embargo_hours": hours before review is public + emails sent
>> - Mechanism for signed-in maintainers to lift embargo immediately
>>
>> That solves the key issue I had with sashiko: people/agents re-spinning stuff
>> too fast before a human at least had the chance to jump in first.
> 
> If a sender re-spins stuff too fast after a Sashiko review, that's a
> pretty clear signal that it is an agent. 

And I know for sure that that's not always the case.

-- 
Cheers,

David

Re: Sashiko change proposal

[Harry Yoo]

On Thu, Oct 15, 2026 at 10:38:14AM +0200, David Hildenbrand (Arm) wrote:
> (CCing all people attending the mm-next call)
> 
> Hi,
> 
> After talking to Roman at MS, he just informed me that sashiko now has a new
> feature combination:
> - "embargo_hours": hours before review is public + emails sent
> - Mechanism for signed-in maintainers to lift embargo immediately
>
> That solves the key issue I had with sashiko: people/agents re-spinning stuff
> too fast before a human at least had the chance to jump in first.
> 
> I suggest the following sashiko change for linux-mm:
> 
> diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
> index a9198de..ba123da 100644
> --- a/projects/linux/mailing_lists.toml
> +++ b/projects/linux/mailing_lists.toml
> @@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
>  ["linux-integrity@vger.kernel.org"]
>  ["linux-kernel@vger.kernel.org"]
>  ["linux-kselftest@vger.kernel.org"]
> -["linux-mm@kvack.org"]
>  ["linux-nvme@lists.infradead.org"]
>  ["linux-omap@vger.kernel.org"]
>  ["linux-pm@vger.kernel.org"]
> @@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]
> 
>  ["fsverity@lists.linux.dev"]
>  reply_to = ["author", "list", "recipients"]
> +
> +["linux-mm@kvack.org"]
> +reply_to = ["author", "list", "recipients"]
> +embargo_hours = 24
> 
> 
> Which means:
> - Embargo 24h
> - Mails sent to everybody (author, list, CC list)
> - No positive-review ("looks great") mails (default)
> - No other static email addresses that receive a copy (default)
> - No additional ignored author email addresses (default)
> 
> I'd say we start with 24h and see how it goes.
> 
> Thoughts?

Acked-by: Harry Yoo (Meta) <harry@kernel.org>

Quick question: when are maintainers supposed to lift embargo?

-- 
Cheers,
Harry / Hyeonggon

Re: Sashiko change proposal

[Roman Gushchin]

> On Oct 15, 2026, at 5:08 AM, Harry Yoo <harry@kernel.org> wrote:
> 
> On Thu, Oct 15, 2026 at 10:38:14AM +0200, David Hildenbrand (Arm) wrote:
>> (CCing all people attending the mm-next call)
>> 
>> Hi,
>> 
>> After talking to Roman at MS, he just informed me that sashiko now has a new
>> feature combination:
>> - "embargo_hours": hours before review is public + emails sent
>> - Mechanism for signed-in maintainers to lift embargo immediately
>> 
>> That solves the key issue I had with sashiko: people/agents re-spinning stuff
>> too fast before a human at least had the chance to jump in first.
>> 
>> I suggest the following sashiko change for linux-mm:
>> 
>> diff --git a/projects/linux/mailing_lists.toml b/projects/linux/mailing_lists.toml
>> index a9198de..ba123da 100644
>> --- a/projects/linux/mailing_lists.toml
>> +++ b/projects/linux/mailing_lists.toml
>> @@ -46,7 +46,6 @@ ignored_emails = ["patchwork@emeril.freedesktop.org"]
>> ["linux-integrity@vger.kernel.org"]
>> ["linux-kernel@vger.kernel.org"]
>> ["linux-kselftest@vger.kernel.org"]
>> -["linux-mm@kvack.org"]
>> ["linux-nvme@lists.infradead.org"]
>> ["linux-omap@vger.kernel.org"]
>> ["linux-pm@vger.kernel.org"]
>> @@ -315,3 +314,7 @@ reply_to = ["author", "list", "recipients"]
>> 
>> ["fsverity@lists.linux.dev"]
>> reply_to = ["author", "list", "recipients"]
>> +
>> +["linux-mm@kvack.org"]
>> +reply_to = ["author", "list", "recipients"]
>> +embargo_hours = 24
>> 
>> 
>> Which means:
>> - Embargo 24h
>> - Mails sent to everybody (author, list, CC list)
>> - No positive-review ("looks great") mails (default)
>> - No other static email addresses that receive a copy (default)
>> - No additional ignored author email addresses (default)
>> 
>> I'd say we start with 24h and see how it goes.
>> 
>> Thoughts?
> 
> Acked-by: Harry Yoo (Meta) <harry@kernel.org>
> 
> Quick question: when are maintainers supposed to lift embargo?

They don’t have to: it will be lifted automatically after 24h (or whatever is configured).
But if you as a maintainer has an immediate comment to Sashiko’s review or you’re ready to pick the change, you can lift it immediately.
And of course maintainers can see review results immediately.


Thanks