linux-kernel - Re: [PATCH v4 25/30] context_tracking,x86: Defer kernel text patching IPIs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Z4bTlZkqihaAyGb4@google.com>
Date: Tue, 14 Jan 2025 13:13:57 -0800
From: Sean Christopherson <seanjc@...gle.com>
To: Valentin Schneider <vschneid@...hat.com>
Cc: linux-kernel@...r.kernel.org, x86@...nel.org, 
	virtualization@...ts.linux.dev, linux-arm-kernel@...ts.infradead.org, 
	loongarch@...ts.linux.dev, linux-riscv@...ts.infradead.org, 
	linux-perf-users@...r.kernel.org, xen-devel@...ts.xenproject.org, 
	kvm@...r.kernel.org, linux-arch@...r.kernel.org, rcu@...r.kernel.org, 
	linux-hardening@...r.kernel.org, linux-mm@...ck.org, 
	linux-kselftest@...r.kernel.org, bpf@...r.kernel.org, 
	bcm-kernel-feedback-list@...adcom.com, Peter Zijlstra <peterz@...radead.org>, 
	Nicolas Saenz Julienne <nsaenzju@...hat.com>, Juergen Gross <jgross@...e.com>, 
	Ajay Kaher <ajay.kaher@...adcom.com>, Alexey Makhalov <alexey.amakhalov@...adcom.com>, 
	Russell King <linux@...linux.org.uk>, Catalin Marinas <catalin.marinas@....com>, 
	Will Deacon <will@...nel.org>, Huacai Chen <chenhuacai@...nel.org>, WANG Xuerui <kernel@...0n.name>, 
	Paul Walmsley <paul.walmsley@...ive.com>, Palmer Dabbelt <palmer@...belt.com>, 
	Albert Ou <aou@...s.berkeley.edu>, Thomas Gleixner <tglx@...utronix.de>, 
	Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>, 
	Dave Hansen <dave.hansen@...ux.intel.com>, "H. Peter Anvin" <hpa@...or.com>, 
	Arnaldo Carvalho de Melo <acme@...nel.org>, Namhyung Kim <namhyung@...nel.org>, 
	Mark Rutland <mark.rutland@....com>, 
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>, Jiri Olsa <jolsa@...nel.org>, 
	Ian Rogers <irogers@...gle.com>, Adrian Hunter <adrian.hunter@...el.com>, 
	Kan Liang <kan.liang@...ux.intel.com>, Boris Ostrovsky <boris.ostrovsky@...cle.com>, 
	Josh Poimboeuf <jpoimboe@...nel.org>, Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>, 
	Paolo Bonzini <pbonzini@...hat.com>, Andy Lutomirski <luto@...nel.org>, Arnd Bergmann <arnd@...db.de>, 
	Frederic Weisbecker <frederic@...nel.org>, "Paul E. McKenney" <paulmck@...nel.org>, 
	Jason Baron <jbaron@...mai.com>, Steven Rostedt <rostedt@...dmis.org>, 
	Ard Biesheuvel <ardb@...nel.org>, Neeraj Upadhyay <neeraj.upadhyay@...nel.org>, 
	Joel Fernandes <joel@...lfernandes.org>, Josh Triplett <josh@...htriplett.org>, 
	Boqun Feng <boqun.feng@...il.com>, Uladzislau Rezki <urezki@...il.com>, 
	Mathieu Desnoyers <mathieu.desnoyers@...icios.com>, Lai Jiangshan <jiangshanlai@...il.com>, 
	Zqiang <qiang.zhang1211@...il.com>, Juri Lelli <juri.lelli@...hat.com>, 
	Clark Williams <williams@...hat.com>, Yair Podemsky <ypodemsk@...hat.com>, 
	Tomas Glozar <tglozar@...hat.com>, Vincent Guittot <vincent.guittot@...aro.org>, 
	Dietmar Eggemann <dietmar.eggemann@....com>, Ben Segall <bsegall@...gle.com>, 
	Mel Gorman <mgorman@...e.de>, Kees Cook <kees@...nel.org>, 
	Andrew Morton <akpm@...ux-foundation.org>, Christoph Hellwig <hch@...radead.org>, 
	Shuah Khan <shuah@...nel.org>, Sami Tolvanen <samitolvanen@...gle.com>, 
	Miguel Ojeda <ojeda@...nel.org>, Alice Ryhl <aliceryhl@...gle.com>, 
	"Mike Rapoport (Microsoft)" <rppt@...nel.org>, Samuel Holland <samuel.holland@...ive.com>, Rong Xu <xur@...gle.com>, 
	Geert Uytterhoeven <geert@...ux-m68k.org>, Yosry Ahmed <yosryahmed@...gle.com>, 
	"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>, 
	"Masami Hiramatsu (Google)" <mhiramat@...nel.org>, Jinghao Jia <jinghao7@...inois.edu>, 
	Luis Chamberlain <mcgrof@...nel.org>, Randy Dunlap <rdunlap@...radead.org>, 
	Tiezhu Yang <yangtiezhu@...ngson.cn>
Subject: Re: [PATCH v4 25/30] context_tracking,x86: Defer kernel text patching IPIs

On Tue, Jan 14, 2025, Valentin Schneider wrote:
> text_poke_bp_batch() sends IPIs to all online CPUs to synchronize
> them vs the newly patched instruction. CPUs that are executing in userspace
> do not need this synchronization to happen immediately, and this is
> actually harmful interference for NOHZ_FULL CPUs.

...

> This leaves us with static keys and static calls.

...

> @@ -2317,11 +2334,20 @@ static void text_poke_bp_batch(struct text_poke_loc *tp, unsigned int nr_entries
>  	 * First step: add a int3 trap to the address that will be patched.
>  	 */
>  	for (i = 0; i < nr_entries; i++) {
> -		tp[i].old = *(u8 *)text_poke_addr(&tp[i]);
> -		text_poke(text_poke_addr(&tp[i]), &int3, INT3_INSN_SIZE);
> +		void *addr = text_poke_addr(&tp[i]);
> +
> +		/*
> +		 * There's no safe way to defer IPIs for patching text in
> +		 * .noinstr, record whether there is at least one such poke.
> +		 */
> +		if (is_kernel_noinstr_text((unsigned long)addr))
> +			cond = NULL;

Maybe pre-check "cond", especially if multiple ranges need to be checked?  I.e.

		if (cond && is_kernel_noinstr_text(...))
> +
> +		tp[i].old = *((u8 *)addr);
> +		text_poke(addr, &int3, INT3_INSN_SIZE);
>  	}
>  
> -	text_poke_sync();
> +	__text_poke_sync(cond);
>  
>  	/*
>  	 * Second step: update all but the first byte of the patched range.

...

> +/**
> + * is_kernel_noinstr_text - checks if the pointer address is located in the
> + *                    .noinstr section
> + *
> + * @addr: address to check
> + *
> + * Returns: true if the address is located in .noinstr, false otherwise.
> + */
> +static inline bool is_kernel_noinstr_text(unsigned long addr)
> +{
> +	return addr >= (unsigned long)__noinstr_text_start &&
> +	       addr < (unsigned long)__noinstr_text_end;
> +}

This doesn't do the right thing for modules, which matters because KVM can be
built as a module on x86, and because context tracking understands transitions
to GUEST mode, i.e. CPUs that are running in a KVM guest will be treated as not
being in the kernel, and thus will have IPIs deferred.  If KVM uses a static key
or branch between guest_state_enter_irqoff() and guest_state_exit_irqoff(), the
patching code won't wait for CPUs to exit guest mode, i.e. KVM could theoretically
use the wrong static path.

I don't expect this to ever cause problems in practice, because patching code in
KVM's VM-Enter/VM-Exit path that has *functional* implications, while CPUs are
actively running guest code, would be all kinds of crazy.  But I do think we
should plug the hole.

If this issue is unique to KVM, i.e. is not a generic problem for all modules (I
assume module code generally isn't allowed in the entry path, even via NMI?), one
idea would be to let KVM register its noinstr section for text poking.