Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 5 Jan 2021 14:39:28 +0100
From:   Jessica Yu <jeyu@kernel.org>
To:     Frank van der Linden <fllinden@amazon.com>
Cc:     linux-kernel@vger.kernel.org
Subject: Re: [PATCH] module: harden ELF info handling
Message-ID: <20210105133928.GA27293@linux-8ccs.fritz.box>
References: <20201221234921.31129-1-fllinden@amazon.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
In-Reply-To: <20201221234921.31129-1-fllinden@amazon.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Precedence: bulk

Hi Frank,

Sorry for the delay. I've just gotten back from vacation :-)

+++ Frank van der Linden [21/12/20 23:49 +0000]:
>5fdc7db644 ("module: setup load info before module_sig_check()")
>moved the ELF setup, so that it was done before the signature
>check. This made the module name available to signature error
>messages.
>
>However, the checks for ELF correctness in setup_load_info
>are not sufficient to prevent bad memory references due to
>corrupted offset fields, indices, etc.
>
>So, there's a regression in behavior here: a corrupt and unsigned
>(or badly signed) module, which might previously have been rejected
>immediately, can now cause an oops/crash.
>
>Harden ELF handling for module loading by doing the following:
>
>- Move the signature check back up so that it comes before ELF
>  initialization. It's best to do the signature check to see
>  if we can trust the module, before using the ELF structures
>  inside it. This also makes checks against info->len
>  more accurate again, as this field will be reduced by the
>  length of the signature in mod_check_sig().
>
>  The module name is now once again not available for error
>  messages during the signature check, but that seems like
>  a fair tradeoff.

I vaguely remember that I had made the module name available in
response to a one-off request, IIRC someone had wanted the module name
logged to be able to figure out which module(s) had failed signature
verification. But I do agree with your line of reasoning, that we
should probably not access internal module structures until we have
verified that we can trust the module. It is a chicken and egg problem
unfortunately. Although, it is probably worth it to trade ease of
debugging for a more hardened approach.

>- Check if sections have offset / size fields that at least don't
>  exceed the length of the module.
>
>- Check if sections have section name offsets that don't fall
>  outside the section name table.
>
>- Add a few other sanity checks against invalid section indices,
>  etc.
>
>This is not an exhaustive consistency check, but the idea is to
>at least get through the signature and blacklist checks without
>crashing because of corrupted ELF info, and to error out gracefully
>for most issues that would have caused problems later on.
>
>Fixes: 5fdc7db644 ("module: setup load info before module_sig_check()")
>Signed-off-by: Frank van der Linden <fllinden@amazon.com>
>---
> kernel/module.c           | 143 +++++++++++++++++++++++++++++++++-----
> kernel/module_signature.c |   2 +-
> kernel/module_signing.c   |   2 +-
> 3 files changed, 126 insertions(+), 21 deletions(-)
>
>diff --git a/kernel/module.c b/kernel/module.c
>index 4bf30e4b3eaa..ef7681a22a1a 100644
>--- a/kernel/module.c
>+++ b/kernel/module.c
>@@ -2964,7 +2964,7 @@ static int module_sig_check(struct load_info *info, int flags)
> 	}
>
> 	if (is_module_sig_enforced()) {
>-		pr_notice("%s: loading of %s is rejected\n", info->name, reason);
>+		pr_notice("loading of %s is rejected\n", reason);

Small nit: Let's start with a capital letter perhaps? Just to be
consistent with the other log messages that don't start with a prefix.
Same goes for the other pr_err()s below.

> 		return -EKEYREJECTED;
> 	}
>
>@@ -2977,9 +2977,33 @@ static int module_sig_check(struct load_info *info, int flags)
> }
> #endif /* !CONFIG_MODULE_SIG */
>
>-/* Sanity checks against invalid binaries, wrong arch, weird elf version. */
>-static int elf_header_check(struct load_info *info)
>+static int validate_section_offset(struct load_info *info, Elf_Shdr *shdr)
> {
>+	unsigned long secend;
>+
>+	/*
>+	 * Check for both overflow and offset/size being
>+	 * too large.
>+	 */
>+	secend = shdr->sh_offset + shdr->sh_size;
>+	if (secend < shdr->sh_offset || secend >= info->len)

Should this not be secend > info->len?

>+		return -ENOEXEC;
>+
>+	return 0;
>+}
>+
>+/*
>+ * Sanity checks against invalid binaries, wrong arch, weird elf version.
>+ *
>+ * Also do basic validity checks against section offsets and sizes, the
>+ * section name string table, and the indices used for it (sh_name).
>+ */
>+static int elf_validity_check(struct load_info *info)
>+{
>+	unsigned int i;
>+	Elf_Shdr *shdr, *strhdr;
>+	int err;
>+
> 	if (info->len < sizeof(*(info->hdr)))
> 		return -ENOEXEC;
>
>@@ -2989,11 +3013,78 @@ static int elf_header_check(struct load_info *info)
> 	    || info->hdr->e_shentsize != sizeof(Elf_Shdr))
> 		return -ENOEXEC;
>
>+	/*
>+	 * e_shnum is 16 bits, and sizeof(Elf_Shdr) is
>+	 * known and small. So e_shnum * sizeof(Elf_Shdr)
>+	 * will not overflow unsigned long on any platform.
>+	 */
> 	if (info->hdr->e_shoff >= info->len
> 	    || (info->hdr->e_shnum * sizeof(Elf_Shdr) >
> 		info->len - info->hdr->e_shoff))
> 		return -ENOEXEC;
>
>+	info->sechdrs = (void *)info->hdr + info->hdr->e_shoff;
>+
>+	/*
>+	 * Verify if the section name table index is valid.
>+	 */
>+	if (info->hdr->e_shstrndx == SHN_UNDEF
>+	    || info->hdr->e_shstrndx >= info->hdr->e_shnum)
>+		return -ENOEXEC;
>+
>+	strhdr = &info->sechdrs[info->hdr->e_shstrndx];
>+	err = validate_section_offset(info, strhdr);
>+	if (err < 0)
>+		return err;
>+
>+	/*
>+	 * The section name table must be NUL-terminated, as required
>+	 * by the spec. This makes strcmp and pr_* calls that access
>+	 * strings in the section safe.
>+	 */
>+	info->secstrings = (void *)info->hdr + strhdr->sh_offset;
>+	if (info->secstrings[strhdr->sh_size - 1] != '\0')
>+		return -ENOEXEC;
>+
>+	/*
>+	 * The code assumes that section 0 has a length of zero and
>+	 * an addr of zero, so check for it.
>+	 */
>+	if (info->sechdrs[0].sh_type != SHT_NULL
>+	    || info->sechdrs[0].sh_size != 0
>+	    || info->sechdrs[0].sh_addr != 0)
>+		return -ENOEXEC;
>+
>+	for (i = 1; i < info->hdr->e_shnum; i++) {
>+		shdr = &info->sechdrs[i];
>+		switch (shdr->sh_type) {
>+		case SHT_NULL:
>+		case SHT_NOBITS:
>+			continue;
>+		case SHT_SYMTAB:
>+			if (shdr->sh_link == SHN_UNDEF
>+			    || shdr->sh_link >= info->hdr->e_shnum)
>+				return -ENOEXEC;
>+			fallthrough;
>+		default:
>+			err = validate_section_offset(info, shdr);
>+			if (err < 0) {
>+				pr_err("invalid ELF section in module num %u type %u\n",
>+					i, shdr->sh_type);

Same as the first comment here. Also, this is personal preference but I
think the "in module num %u type %u" reads a bit awkwardly. Maybe
something like "Invalid ELF section in module (section ndx %u type %u)"?

>+				return err;
>+			}
>+
>+			if (shdr->sh_flags & SHF_ALLOC) {
>+				if (shdr->sh_name >= strhdr->sh_size) {
>+					pr_err("invalid ELF section name in module num %u type %u\n",
>+					       i, shdr->sh_type);

Same here.

Thanks!

Jessica