All of lore.kernel.org
 help / color / mirror / Atom feed
From: Petr Lautrbach <lautrbach@redhat.com>
To: James Carter <jwcart2@gmail.com>, selinux@vger.kernel.org
Cc: James Carter <jwcart2@gmail.com>
Subject: Re: [PATCH 2/3] libsepol: When resolving names check if a block is abstract
Date: Mon, 27 Apr 2026 20:17:08 +0200	[thread overview]
Message-ID: <87340gtgij.fsf@redhat.com> (raw)
In-Reply-To: <20260414191120.29067-2-jwcart2@gmail.com>

James Carter <jwcart2@gmail.com> writes:

> Nothing in an abstract block in CIL is instantiated until the block
> is inherited. No declartion, macro, or optional block within an
> abstract block should ever be referred to from outside of the block.
>
> Check for abstract blocks when resolving names and return an error
> if one is found.
>
> This patch is based on a report from the security firm Trail of Bits.


Since this commit, it's not possible to build Fedora selinux-policy:

# semodule -B -n
Failed to resolve pcp_var_lib_t at /var/lib/selinux/targeted/tmp/modules/100/abrt/cil:1241
Failed to resolve pcp_var_lib_t at /var/lib/selinux/targeted/tmp/modules/100/abrt/cil:1244
Failed to resolve pcp_var_lib_t at /var/lib/selinux/targeted/tmp/modules/100/abrt/cil:1245
Failed to resolve dirsrvadmin_config_t at /var/lib/selinux/targeted/tmp/modules/100/apache/cil:2384
Failed to resolve dirsrvadmin_tmp_t at /var/lib/selinux/targeted/tmp/modules/100/apache/cil:2385
Failed to resolve dirsrvadmin_unconfined_script_t at /var/lib/selinux/targeted/tmp/modules/100/apache/cil:2386
...



>
> Signed-off-by: James Carter <jwcart2@gmail.com>
> ---
>  libsepol/cil/src/cil_resolve_ast.c | 15 +++++++++++----
>  1 file changed, 11 insertions(+), 4 deletions(-)
>
> diff --git a/libsepol/cil/src/cil_resolve_ast.c b/libsepol/cil/src/cil_resolve_ast.c
> index bcac4026..b0965f1d 100644
> --- a/libsepol/cil/src/cil_resolve_ast.c
> +++ b/libsepol/cil/src/cil_resolve_ast.c
> @@ -4355,7 +4355,7 @@ int cil_resolve_name_keep_aliases(struct cil_tree_node *ast_node, char *name, en
>  		node = ast_node;
>  		if (*name == '.') {
>  			/* Leading '.' */
> -			symtab = &((struct cil_root *)db->ast->root->data)->symtab[CIL_SYM_BLOCKS];
> +			symtab = ((struct cil_root *)db->ast->root->data)->symtab;
>  		} else {
>  			rc = __cil_resolve_name_helper(db, node->parent, current, CIL_SYM_BLOCKS, datum);
>  			if (rc != SEPOL_OK) {
> @@ -4366,14 +4366,20 @@ int cil_resolve_name_keep_aliases(struct cil_tree_node *ast_node, char *name, en
>  		}
>  		/* Keep looking up blocks by name until only last part of name remains */
>  		while (next != NULL) {
> -			rc = cil_symtab_get_datum(symtab, current, datum);
> +			rc = cil_symtab_get_datum(&(symtab[CIL_SYM_BLOCKS]), current, datum);
>  			if (rc != SEPOL_OK) {
>  				free(name_dup);
>  				goto exit;
>  			}
>  			node = NODE(*datum);
>  			if (node->flavor == CIL_BLOCK) {
> -				symtab = &((struct cil_block*)node->data)->symtab[CIL_SYM_BLOCKS];
> +				if (((struct cil_block *)node->data)->is_abstract) {
> +					cil_log(CIL_WARN, "Found %s which is an abstract block and invalid for name resolution\n", current);
> +					free(name_dup);
> +					rc = SEPOL_ERR;
> +					goto exit;
> +				}
> +				symtab = ((struct cil_block*)node->data)->symtab;
>  			} else {
>  				if (ast_node->flavor != CIL_IN) {
>  					cil_log(CIL_WARN, "Can only use %s name for name resolution in \"in\" blocks\n", cil_node_to_string(node));
> @@ -4383,7 +4389,7 @@ int cil_resolve_name_keep_aliases(struct cil_tree_node *ast_node, char *name, en
>  				}
>  				if (node->flavor == CIL_MACRO) {
>  					struct cil_macro *macro = node->data;
> -					symtab = &macro->symtab[sym_index];
> +					symtab = macro->symtab;
>  				}
>  			}
>  			current = next;
> @@ -4401,6 +4407,7 @@ int cil_resolve_name_keep_aliases(struct cil_tree_node *ast_node, char *name, en
>  
>  exit:
>  	if (rc != SEPOL_OK) {
> +		cil_tree_log(ast_node, CIL_ERR, "Failed to resolve %s", name);
>  		*datum = NULL;
>  	}
>  
> -- 
> 2.53.0


  reply	other threads:[~2026-04-27 18:17 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-14 19:11 [PATCH 1/3] libsepol: Fix out-of-bounds memory write in discard_tunbables() James Carter
2026-04-14 19:11 ` [PATCH 2/3] libsepol: When resolving names check if a block is abstract James Carter
2026-04-27 18:17   ` Petr Lautrbach [this message]
2026-04-28 15:49     ` James Carter
2026-04-14 19:11 ` [PATCH 3/3] libsepol: Validate datum array entries for avrule blocks James Carter
2026-04-22 19:07 ` [PATCH 1/3] libsepol: Fix out-of-bounds memory write in discard_tunbables() James Carter

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87340gtgij.fsf@redhat.com \
    --to=lautrbach@redhat.com \
    --cc=jwcart2@gmail.com \
    --cc=selinux@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.