From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 75E4EBA49 for ; Tue, 7 Mar 2023 18:33:12 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 63FF2C433EF; Tue, 7 Mar 2023 18:33:11 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1678213992; bh=1PUpIQRDHlg0QANfyUGpYpycg/Po5EgS5kF056alFB0=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=zQR/k/lBna3VyzTa2byS6DbogU4Iobw/uvyjIVezeNNVEMG4XUZM7Cai5k+LvkLfB GrOW1f2t0Jmq31h2bEfC+h4UOBj5XoP8XVCwyqPf+VrJ5mPFdKgwZvbsIJgBFM0soV UwExI9d9+yJHR9oK7I5wXgLSmTydXZZ6P6zVSWHA= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Heiko Carstens , Vasily Gorbik Subject: [PATCH 6.1 668/885] s390/kprobes: fix current_kprobe never cleared after kprobes reenter Date: Tue, 7 Mar 2023 18:00:02 +0100 Message-Id: <20230307170031.183653226@linuxfoundation.org> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20230307170001.594919529@linuxfoundation.org> References: <20230307170001.594919529@linuxfoundation.org> User-Agent: quilt/0.67 Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From: Vasily Gorbik commit cd57953936f2213dfaccce10d20f396956222c7d upstream. Recent test_kprobe_missed kprobes kunit test uncovers the following problem. Once kprobe is triggered from another kprobe (kprobe reenter), all future kprobes on this cpu are considered as kprobe reenter, thus pre_handler and post_handler are not being called and kprobes are counted as "missed". Commit b9599798f953 ("[S390] kprobes: activation and deactivation") introduced a simpler scheme for kprobes (de)activation and status tracking by using push_kprobe/pop_kprobe, which supposed to work for both initial kprobe entry as well as kprobe reentry and helps to avoid handling those two cases differently. The problem is that a sequence of calls in case of kprobes reenter: push_kprobe() <- NULL (current_kprobe) push_kprobe() <- kprobe1 (current_kprobe) pop_kprobe() -> kprobe1 (current_kprobe) pop_kprobe() -> kprobe1 (current_kprobe) leaves "kprobe1" as "current_kprobe" on this cpu, instead of setting it to NULL. In fact push_kprobe/pop_kprobe can only store a single state (there is just one prev_kprobe in kprobe_ctlblk). Which is a hack but sufficient, there is no need to have another prev_kprobe just to store NULL. To make a simple and backportable fix simply reset "prev_kprobe" when kprobe is poped from this "stack". No need to worry about "kprobe_status" in this case, because its value is only checked when current_kprobe != NULL. Cc: stable@vger.kernel.org Fixes: b9599798f953 ("[S390] kprobes: activation and deactivation") Reviewed-by: Heiko Carstens Signed-off-by: Vasily Gorbik Signed-off-by: Heiko Carstens Signed-off-by: Greg Kroah-Hartman --- arch/s390/kernel/kprobes.c | 1 + 1 file changed, 1 insertion(+) --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -279,6 +279,7 @@ static void pop_kprobe(struct kprobe_ctl { __this_cpu_write(current_kprobe, kcb->prev_kprobe.kp); kcb->kprobe_status = kcb->prev_kprobe.status; + kcb->prev_kprobe.kp = NULL; } NOKPROBE_SYMBOL(pop_kprobe);