On Tue, Jul 28, 2026, at 9:55 PM, Jakub Kicinski wrote:
> On Tue, 28 Jul 2026 21:46:57 -0400 Chuck Lever wrote:
>> On Tue, Jul 28, 2026, at 9:43 PM, Jakub Kicinski wrote:
>> > On Mon, 20 Jul 2026 10:27:54 -0400 Chuck Lever wrote:
>> >> - The no-data cap (patch 1) is a prerequisite, not a stand-alone fix.
>> >> Once control records reach read_sock, a record carrying no payload
>> >> stops advancing the caller's read descriptor, so a peer streaming
>> >> such records would pin the socket lock and the kernel receive
>> >> context for as long as the flood lasts. Bounding consecutive
>> >> no-data records supplies the return boundary a system call would
>> >> otherwise provide. The cap is scoped to tls_sw_read_sock() alone:
>> >> splice and recvmsg run in the caller's own context, reschedule, and
>> >> drop the lock on return, so they need nothing.
>> >
>> > Is it just me or this is incomprehensible slop? Run on sentences
>> > full of terms no kernel developer would use?
>>
>> It makes sense to me. Which terms do you have trouble with?
>>
>> Patch 1 is moving to a pre-requisite series, so this can be dropped
>> from the cover letter.
>
> Did you write this? If not what model did you use?
I did not write this text. It was written with Opus 5 and reviewed by
gpt-5.6, sashiko (gemini), and myself, repeatedly.
I’m not aware of a requirement to add an AI-related tag to a series
cover letter, nor am I aware of any a priori prohibition against
using assistance when composing one.
I agree it needlessly duplicates content from the patch commit messages.
It is bland, but I don’t find it nonsensical, or I wouldn’t have posted
it for review.
--
Chuck Lever
