Re: [LTP] sched_setattr01: Convert to new API

[Andrea Cervesato via ltp <ltp@lists.linux.it>]

> Hi Andrea,
> 
> On Sep 4, 2026, Andrea Cervesato wrote:
> > sched_setattr01: Convert to new API
> 
> --- [PATCH 1/5] ---
> 
> > Rewrite the test to use the modern LTP API (tst_test.h) with a
> > struct tcase array and TST_EXP_* macros.
> 
> The body only restates how the test was converted. Please explain why the
> conversion is needed.
> 
> --- [PATCH 2/5] ---
> 
> > Rewrite the test to use the modern LTP API (tst_test.h) with
> > TST_EXP_* macros and cleanup callback.
> 
> The body only restates how the test was converted. Please explain why the
> conversion is needed.
> 
> --- [PATCH 3/5] ---
> 
> > Add a safe variant of sched_setattr() that breaks the test with TBROK
> > on failure and use it in sched_setattr01 and sched_getattr01.
> 
> The body describes the wrapper and its callers, but not why the shared
> wrapper is needed. Please add the motivation.
> 
> > +	if (ret == -1) {
> > +		tst_brk_(file, lineno, TBROK | TERRNO,
> > +			"sched_setattr(%i) failed", pid);
> > +	}
> 
> This body contains one statement. Please remove the braces as required by
> LTP C style.
> 
> --- [PATCH 5/5] ---
> 
> > +	TST_CHECKPOINT_WAKE(CP_SPRAYED);
> > +
> > +	while (!tst_atomic_load(&stop_spray)) {
> > +		prctl(PR_SET_MM, PR_SET_MM_MAP, (unsigned long)&mm_map,
> > +		      sizeof(mm_map), 0);
> > +	}
> 
> > +	TST_CHECKPOINT_WAIT2(CP_SPRAYED, 18000);
> > +
> > +	SAFE_SCHED_SETATTR(waiter_tid, &attr, 0);
> 
> CP_SPRAYED is signaled before the waiter enters PR_SET_MM_MAP, so nothing
> ensures that sched_setattr() overlaps the controlled kernel-stack buffer.
> Use the PoC's page-boundary and hole-punch mechanism, or an equivalent
> mechanism that guarantees a usable overlap; otherwise a vulnerable kernel
> can complete all attempts and incorrectly report TPASS.
> 
> Verdict - Needs revision
> 
> ---
> Note:
> 
> The agent can sometimes produce false positives although often its
> findings are genuine. If you find issues with the review, please
> comment this email or ignore the suggestions.
> 
> Regards,
> LTP AI Reviewer
> 
> -- 
> Mailing list info: https://lists.linux.it/listinfo/ltp

The whole AI review can be ignored

--
Andrea Cervesato
SUSE QE Automation Engineer Linux
andrea.cervesato@suse.com

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp