From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pf1-f182.google.com (mail-pf1-f182.google.com [209.85.210.182]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C102735E1B5 for ; Mon, 13 Jul 2026 21:57:21 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.182 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1783979844; cv=none; b=jvF2omiAUfGMBUN+YU619l9CZzeUW+xDrEmE+VZWZiK31Ek74zGXXdwioPFiSZXrJBNI4/FPsa2q+a4w0WuFAjF0eSBr+JHERp0c9BCys7ITTto/Eb19aV4SI7BoIRRFwEhL8xXBxENybpi5ikwc/OyUFm3Ek57a/VyDoHgHT8E= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1783979844; c=relaxed/simple; bh=coTO6PwuDxPx0SSZlTNrbu8MbSA08SuMKb4lakXTOBw=; h=Mime-Version:Content-Type:Date:Message-Id:From:To:Cc:Subject: References:In-Reply-To; b=YUgzHRpEPE4qdCx2m8xlKWHphy6midAawtxMMzUibQpUmS6uklTD6zN+X5SSeZJ5/KCZZgpwVzyQngRnvKkyU9sWjH11NuNEAhdIo8UE9TPhnxUNIDJdMRYHSbcCcybGtqf/ybdDJKnhqphCzWH7o2WIlm5VelB0b2FAN8/lfTE= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=etsalapatis.com; spf=pass smtp.mailfrom=etsalapatis.com; dkim=pass (2048-bit key) header.d=etsalapatis-com.20251104.gappssmtp.com header.i=@etsalapatis-com.20251104.gappssmtp.com header.b=bb0Ru0yS; arc=none smtp.client-ip=209.85.210.182 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=etsalapatis.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=etsalapatis.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=etsalapatis-com.20251104.gappssmtp.com header.i=@etsalapatis-com.20251104.gappssmtp.com header.b="bb0Ru0yS" Received: by mail-pf1-f182.google.com with SMTP id d2e1a72fcca58-8487088510aso3460309b3a.0 for ; Mon, 13 Jul 2026 14:57:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=etsalapatis-com.20251104.gappssmtp.com; s=20251104; t=1783979841; x=1784584641; darn=vger.kernel.org; h=in-reply-to:references:subject:cc:to:from:message-id:date :content-type:content-transfer-encoding:mime-version:from:to:cc :subject:date:message-id:reply-to:content-type; bh=UrqBoQyrxvBcos5JbUmwXjWeYzr52NLOmMrsaM2Ub/k=; b=bb0Ru0ySsnkiOLTtWS5HDN9DQ1SqhJ1n4BYvijkz8/uPRIIfhcvcvjIe9t3LkeesoH TAK0Qbheuv+7gAgIHKKMSgfCtiWu+aBqtFyO3+V1eFFuI3HvFbgeKxC4RkYz1a7D0PBb NgxoxachohereFVr38xS/NCgypn8lfXqNaaBUHV+pL9Dby0+exIj+HEZQpDkg/lJ9xpt NglAYV+WKWg2oNKuMRObWZWZ/Z5ISNT0akjdY8xz8oHoZFpQVrVrnZSuCSTErfqp3rE+ ULJHzPbmMqF/sbNTZtigAinBwA8/Wlr662VWMLPylskLUvo9jkVcGYxFDXksSXR1YF4g m29Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1783979841; x=1784584641; h=in-reply-to:references:subject:cc:to:from:message-id:date :content-type:content-transfer-encoding:mime-version:x-gm-gg :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to :content-type; bh=UrqBoQyrxvBcos5JbUmwXjWeYzr52NLOmMrsaM2Ub/k=; b=gmqHfFAPFqjJ19fEgZJgDNjNe1RMp2A8RIItG1xIhKMo/8W5YV74QQEmOMBDyv1eaY 6kN0Le1k9LKusdF/AD8Xm5Y+0pCUCbcxea13EzlHjSUNyqXwuv5qbyHa20Ayul2ZzZ9h UiPh9zNNwyALGwR+2KJGknu4rpx89IEM5BK76PtXe1W8zXqvVDfm6NDpBVpf+UfxoU0u mcRTzDYCaFuaEaSJeqEZ2gWiRRqpFyZrv9AOWFwMYXnSlYTCdV/b5rRQWOVXt6Yr4xTu ym2qHfFe6D15hgiT+eczmpqi0vhS75I7jsJ0mgxCM+aXmTOHWCsyyahfEXu8kk514AxD aBAA== X-Gm-Message-State: AOJu0YxH8HC0fE4dJHXKQPU810hXrsgx2LP1E3Oku6rRzrRVahmlTZAn VVf//CYD7pEivA3OKA0YQkeaXoChEkXHK+Lwum4j6oLjGxLHiSSVHe4lWAtAZaPDxA4= X-Gm-Gg: AfdE7cnlAQ7LHE0qHxZp3+CeQjj1C9VWVmfGLLPn1VcqRn0BISva4r+S8xbjGDIM02g zAFCRaFRgvoiS1Jx13VPG/7aIh+ENmnS9/dk4d10ZVQgcagwZsbzisDkWYxjsSx9HGyGl+3Fnib x6FBQoNDyw2d7EvdTOinviZ9fo783gczLJiH3dEtnMzS3K4eZZF5NLkzMCN0RkLSbDawK+FWSlD y4kJqgo1wwemDzpUcDAW23NxeFoD/e+znLpVPsqrMdLXcA56unkdJyyromzt/Il291+E4kNnvCt lti257IF+YMY1+pSLfdYq1UaOH9YbJxWJSl9zDT/DSgPEgHaZRyECpiXsJzumwn3P14sg1+GlDV IiP3xsiJCBdg2nL3YuHz0U+wqJB5rfscPUCyfSdi6YU1fB7C8w5XZvr89I4a2rhsAVUenVgkQzJ NaiH0Ti2rKmoDXQAyihTroVMwMhE5ZBMlfbbnKfDNESQ== X-Received: by 2002:a05:6a00:983:b0:848:79aa:5d37 with SMTP id d2e1a72fcca58-84a5162f3admr637079b3a.55.1783979840901; Mon, 13 Jul 2026 14:57:20 -0700 (PDT) Received: from localhost (107-190-31-17.cpe.teksavvy.com. [107.190.31.17]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-ca5b3643cd6sm9131766a12.26.2026.07.13.14.57.20 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Mon, 13 Jul 2026 14:57:20 -0700 (PDT) Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=UTF-8 Date: Mon, 13 Jul 2026 17:57:19 -0400 Message-Id: From: "Emil Tsalapatis" To: "Amery Hung" , Cc: , , , , , , , , , , , Subject: Re: [PATCH bpf-next v3 08/15] bpf: Add a few bpf_cgroup_array_* helper functions X-Mailer: aerc 0.21.0-0-g5549850facc2 References: <20260706171918.317102-1-ameryhung@gmail.com> <20260706171918.317102-9-ameryhung@gmail.com> In-Reply-To: <20260706171918.317102-9-ameryhung@gmail.com> On Mon Jul 6, 2026 at 1:19 PM EDT, Amery Hung wrote: > From: Martin KaFai Lau > > In the upcoming patch, the array can store a struct_ops map. > The array could have a cfi_stubs acting as a dummy instead of > the dummy_bpf_prog. The array logic will need to skip the cfi_stubs > also in order to support storing struct_ops map in the array. > > bpf_cgroup_array_length(), bpf_cgroup_array_copy_to_user(), and > bpf_cgroup_array_delete_safe_at() are added as a preparation work > to allow skipping the cfi_stubs in the upcoming patch. This patch > only skips the dummy_bpf_prog which is the same as the existing behavior. > The current bpf_prog_array_*() callers are changed to call the new > bpf_cgroup_array_*(). This is a no-op change. > > Unlike bpf_prog_array_copy_to_user(), bpf_cgroup_array_copy_to_user() > does not need a temporary buffer. The cgroup caller already holds > cgroup_mutex and dereferences the effective array with > rcu_dereference_protected(), so it does not copy to userspace > from an RCU read-side critical section. Details in commit 0911287ce32b. > > Another addition is the bpf_cgroup_array_free(). This prepares > the array to have a different rcu gp for the struct_ops use case, > for example, a struct_ops could have mix of sleepable ops and > non-sleepable ops. In this patch, bpf_cgroup_array_free() only > goes through the regular rcu gp. This is a no-op change also. > > bpf_prog_dummy() is also added to return the global dummy_bpf_prog. > > bpf_cgroup_array_dummy() is added to decide the sentinel based on atype. > It now always returns bpf_prog_dummy(). In the upcoming patch, > it can return a cfi_stubs if the atype belongs to a struct_ops. > > Signed-off-by: Martin KaFai Lau > Signed-off-by: Amery Hung Reviewed-by: Emil Tsalapatis One style issue below. > --- > include/linux/bpf.h | 1 + > kernel/bpf/cgroup.c | 79 +++++++++++++++++++++++++++++++++++++++------ > kernel/bpf/core.c | 5 +++ > 3 files changed, 76 insertions(+), 9 deletions(-) > > diff --git a/include/linux/bpf.h b/include/linux/bpf.h > index 047ffc029666..e371a4733135 100644 > --- a/include/linux/bpf.h > +++ b/include/linux/bpf.h > @@ -2561,6 +2561,7 @@ int bpf_prog_array_copy(struct bpf_prog_array *old_= array, > struct bpf_prog *include_prog, > u64 bpf_cookie, > struct bpf_prog_array **new_array); > +struct bpf_prog *bpf_prog_dummy(void); > =20 > struct bpf_run_ctx {}; > =20 > diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c > index 7abbe12e108f..081d81de1816 100644 > --- a/kernel/bpf/cgroup.c > +++ b/kernel/bpf/cgroup.c > @@ -319,6 +319,67 @@ static void bpf_cgroup_link_auto_detach(struct bpf_c= group_link *link) > link->cgroup =3D NULL; > } > =20 > +static void bpf_cgroup_array_free(struct bpf_prog_array *array) > +{ > + if (!array || array =3D=3D &bpf_empty_prog_array) > + return; > + kfree_rcu(array, rcu); > +} > + > +static void *bpf_cgroup_array_dummy(enum cgroup_bpf_attach_type atype) > +{ > + return bpf_prog_dummy(); > +} > + > +static int bpf_cgroup_array_length(struct bpf_prog_array *array, > + enum cgroup_bpf_attach_type atype) > +{ > + struct bpf_prog_array_item *item; > + int cnt =3D 0; > + > + for (item =3D array->items; item->prog; item++) > + if (item->prog !=3D bpf_cgroup_array_dummy(atype)) > + cnt++; The outer parenthesis needs braces since its body is an if-statement. > + > + return cnt; > +} > + > +static int bpf_cgroup_array_copy_to_user(struct bpf_prog_array *array, > + __u32 __user *prog_ids, int cnt, > + enum cgroup_bpf_attach_type atype) > +{ > + struct bpf_prog_array_item *item; > + int i =3D 0; > + u32 id; > + > + for (item =3D array->items; item->prog && i < cnt; item++) { > + if (item->prog =3D=3D bpf_cgroup_array_dummy(atype)) > + continue; > + id =3D item->prog->aux->id; > + if (copy_to_user(prog_ids + i, &id, sizeof(id))) > + return -EFAULT; > + i++; > + } > + return item->prog ? -ENOSPC : 0; > +} > + > +static int bpf_cgroup_array_delete_safe_at(struct bpf_prog_array *array, > + int index, enum cgroup_bpf_attach_type atype) > +{ > + struct bpf_prog_array_item *item; > + > + for (item =3D array->items; item->prog; item++) { > + if (item->prog =3D=3D bpf_cgroup_array_dummy(atype)) > + continue; > + if (!index) { > + WRITE_ONCE(item->prog, bpf_cgroup_array_dummy(atype)); > + return 0; > + } > + index--; > + } > + return -ENOENT; > +} > + > /** > * cgroup_bpf_release() - put references of all bpf programs and > * release all cgroup bpf data > @@ -356,7 +417,7 @@ static void cgroup_bpf_release(struct work_struct *wo= rk) > old_array =3D rcu_dereference_protected( > cgrp->bpf.effective[atype], > lockdep_is_held(&cgroup_mutex)); > - bpf_prog_array_free(old_array); > + bpf_cgroup_array_free(old_array); > } > =20 > list_for_each_entry_safe(storage, stmp, storages, list_cg) { > @@ -530,7 +591,7 @@ static void activate_effective_progs(struct cgroup *c= grp, > /* free prog array after grace period, since __cgroup_bpf_run_*() > * might be still walking the array > */ > - bpf_prog_array_free(old_array); > + bpf_cgroup_array_free(old_array); > } > =20 > /** > @@ -570,7 +631,7 @@ static int cgroup_bpf_inherit(struct cgroup *cgrp) > return 0; > cleanup: > for (i =3D 0; i < NR; i++) > - bpf_prog_array_free(arrays[i]); > + bpf_cgroup_array_free(arrays[i]); > =20 > for (p =3D cgroup_parent(cgrp); p; p =3D cgroup_parent(p)) > cgroup_bpf_put(p); > @@ -625,7 +686,7 @@ static int update_effective_progs(struct cgroup *cgrp= , > =20 > if (percpu_ref_is_zero(&desc->bpf.refcnt)) { > if (unlikely(desc->bpf.inactive)) { > - bpf_prog_array_free(desc->bpf.inactive); > + bpf_cgroup_array_free(desc->bpf.inactive); > desc->bpf.inactive =3D NULL; > } > continue; > @@ -644,7 +705,7 @@ static int update_effective_progs(struct cgroup *cgrp= , > css_for_each_descendant_pre(css, &cgrp->self) { > struct cgroup *desc =3D container_of(css, struct cgroup, self); > =20 > - bpf_prog_array_free(desc->bpf.inactive); > + bpf_cgroup_array_free(desc->bpf.inactive); > desc->bpf.inactive =3D NULL; > } > =20 > @@ -1166,7 +1227,7 @@ static void purge_effective_progs(struct cgroup *cg= rp, struct bpf_prog_list *pl, > lockdep_is_held(&cgroup_mutex)); > =20 > /* Remove the program from the array */ > - WARN_ONCE(bpf_prog_array_delete_safe_at(progs, pos), > + WARN_ONCE(bpf_cgroup_array_delete_safe_at(progs, pos, atype), > "Failed to purge a prog from array at index %d", pos); > } > } > @@ -1296,7 +1357,7 @@ static int __cgroup_bpf_query(struct cgroup *cgrp, = const union bpf_attr *attr, > if (effective_query) { > effective =3D rcu_dereference_protected(cgrp->bpf.effective[atype], > lockdep_is_held(&cgroup_mutex)); > - total_cnt +=3D bpf_prog_array_length(effective); > + total_cnt +=3D bpf_cgroup_array_length(effective, atype); > } else { > total_cnt +=3D prog_list_length(&cgrp->bpf.progs[atype], NULL); > } > @@ -1326,8 +1387,8 @@ static int __cgroup_bpf_query(struct cgroup *cgrp, = const union bpf_attr *attr, > if (effective_query) { > effective =3D rcu_dereference_protected(cgrp->bpf.effective[atype], > lockdep_is_held(&cgroup_mutex)); > - cnt =3D min_t(int, bpf_prog_array_length(effective), total_cnt); > - ret =3D bpf_prog_array_copy_to_user(effective, prog_ids, cnt); > + cnt =3D min_t(int, bpf_cgroup_array_length(effective, atype), total_c= nt); > + ret =3D bpf_cgroup_array_copy_to_user(effective, prog_ids, cnt, atype= ); > } else { > struct hlist_head *progs; > struct bpf_prog_list *pl; > diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c > index 649cce41e13f..1837bb7bb4e9 100644 > --- a/kernel/bpf/core.c > +++ b/kernel/bpf/core.c > @@ -2740,6 +2740,11 @@ void bpf_prog_array_free_sleepable(struct bpf_prog= _array *progs) > call_rcu_tasks_trace(&progs->rcu, __bpf_prog_array_free_sleepable_cb); > } > =20 > +struct bpf_prog *bpf_prog_dummy(void) > +{ > + return &dummy_bpf_prog.prog; > +} > + > int bpf_prog_array_length(struct bpf_prog_array *array) > { > struct bpf_prog_array_item *item;