All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Hao Luo <haoluo@google.com>,
	Alexei Starovoitov <ast@kernel.org>
Subject: [PATCH 5.15 08/33] bpf: Make per_cpu_ptr return rdonly PTR_TO_MEM.
Date: Fri, 29 Apr 2022 12:41:55 +0200	[thread overview]
Message-ID: <20220429104052.585495115@linuxfoundation.org> (raw)
In-Reply-To: <20220429104052.345760505@linuxfoundation.org>

From: Hao Luo <haoluo@google.com>

commit 34d3a78c681e8e7844b43d1a2f4671a04249c821 upstream.

Tag the return type of {per, this}_cpu_ptr with RDONLY_MEM. The
returned value of this pair of helpers is kernel object, which
can not be updated by bpf programs. Previously these two helpers
return PTR_OT_MEM for kernel objects of scalar type, which allows
one to directly modify the memory. Now with RDONLY_MEM tagging,
the verifier will reject programs that write into RDONLY_MEM.

Fixes: 63d9b80dcf2c ("bpf: Introducte bpf_this_cpu_ptr()")
Fixes: eaa6bcb71ef6 ("bpf: Introduce bpf_per_cpu_ptr()")
Fixes: 4976b718c355 ("bpf: Introduce pseudo_btf_id")
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-8-haoluo@google.com
Cc: stable@vger.kernel.org # 5.15.x
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 kernel/bpf/helpers.c  |    4 ++--
 kernel/bpf/verifier.c |   30 ++++++++++++++++++++++++++----
 2 files changed, 28 insertions(+), 6 deletions(-)

--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -667,7 +667,7 @@ BPF_CALL_2(bpf_per_cpu_ptr, const void *
 const struct bpf_func_proto bpf_per_cpu_ptr_proto = {
 	.func		= bpf_per_cpu_ptr,
 	.gpl_only	= false,
-	.ret_type	= RET_PTR_TO_MEM_OR_BTF_ID | PTR_MAYBE_NULL,
+	.ret_type	= RET_PTR_TO_MEM_OR_BTF_ID | PTR_MAYBE_NULL | MEM_RDONLY,
 	.arg1_type	= ARG_PTR_TO_PERCPU_BTF_ID,
 	.arg2_type	= ARG_ANYTHING,
 };
@@ -680,7 +680,7 @@ BPF_CALL_1(bpf_this_cpu_ptr, const void
 const struct bpf_func_proto bpf_this_cpu_ptr_proto = {
 	.func		= bpf_this_cpu_ptr,
 	.gpl_only	= false,
-	.ret_type	= RET_PTR_TO_MEM_OR_BTF_ID,
+	.ret_type	= RET_PTR_TO_MEM_OR_BTF_ID | MEM_RDONLY,
 	.arg1_type	= ARG_PTR_TO_PERCPU_BTF_ID,
 };
 
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4166,15 +4166,30 @@ static int check_mem_access(struct bpf_v
 				mark_reg_unknown(env, regs, value_regno);
 			}
 		}
-	} else if (reg->type == PTR_TO_MEM) {
+	} else if (base_type(reg->type) == PTR_TO_MEM) {
+		bool rdonly_mem = type_is_rdonly_mem(reg->type);
+
+		if (type_may_be_null(reg->type)) {
+			verbose(env, "R%d invalid mem access '%s'\n", regno,
+				reg_type_str(env, reg->type));
+			return -EACCES;
+		}
+
+		if (t == BPF_WRITE && rdonly_mem) {
+			verbose(env, "R%d cannot write into %s\n",
+				regno, reg_type_str(env, reg->type));
+			return -EACCES;
+		}
+
 		if (t == BPF_WRITE && value_regno >= 0 &&
 		    is_pointer_value(env, value_regno)) {
 			verbose(env, "R%d leaks addr into mem\n", value_regno);
 			return -EACCES;
 		}
+
 		err = check_mem_region_access(env, regno, off, size,
 					      reg->mem_size, false);
-		if (!err && t == BPF_READ && value_regno >= 0)
+		if (!err && value_regno >= 0 && (t == BPF_READ || rdonly_mem))
 			mark_reg_unknown(env, regs, value_regno);
 	} else if (reg->type == PTR_TO_CTX) {
 		enum bpf_reg_type reg_type = SCALAR_VALUE;
@@ -6370,6 +6385,13 @@ static int check_helper_call(struct bpf_
 			regs[BPF_REG_0].type = PTR_TO_MEM | ret_flag;
 			regs[BPF_REG_0].mem_size = tsize;
 		} else {
+			/* MEM_RDONLY may be carried from ret_flag, but it
+			 * doesn't apply on PTR_TO_BTF_ID. Fold it, otherwise
+			 * it will confuse the check of PTR_TO_BTF_ID in
+			 * check_mem_access().
+			 */
+			ret_flag &= ~MEM_RDONLY;
+
 			regs[BPF_REG_0].type = PTR_TO_BTF_ID | ret_flag;
 			regs[BPF_REG_0].btf = meta.ret_btf;
 			regs[BPF_REG_0].btf_id = meta.ret_btf_id;
@@ -9172,7 +9194,7 @@ static int check_ld_imm(struct bpf_verif
 
 	if (insn->src_reg == BPF_PSEUDO_BTF_ID) {
 		dst_reg->type = aux->btf_var.reg_type;
-		switch (dst_reg->type) {
+		switch (base_type(dst_reg->type)) {
 		case PTR_TO_MEM:
 			dst_reg->mem_size = aux->btf_var.mem_size;
 			break;
@@ -11313,7 +11335,7 @@ static int check_pseudo_btf_id(struct bp
 			err = -EINVAL;
 			goto err_put;
 		}
-		aux->btf_var.reg_type = PTR_TO_MEM;
+		aux->btf_var.reg_type = PTR_TO_MEM | MEM_RDONLY;
 		aux->btf_var.mem_size = tsize;
 	} else {
 		aux->btf_var.reg_type = PTR_TO_BTF_ID;



  parent reply	other threads:[~2022-04-29 10:46 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-29 10:41 [PATCH 5.15 00/33] 5.15.37-rc1 review Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 01/33] floppy: disable FDRAWCMD by default Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 02/33] bpf: Introduce composable reg, ret and arg types Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 03/33] bpf: Replace ARG_XXX_OR_NULL with ARG_XXX | PTR_MAYBE_NULL Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 04/33] bpf: Replace RET_XXX_OR_NULL with RET_XXX " Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 05/33] bpf: Replace PTR_TO_XXX_OR_NULL with PTR_TO_XXX " Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 06/33] bpf: Introduce MEM_RDONLY flag Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 07/33] bpf: Convert PTR_TO_MEM_OR_NULL to composable types Greg Kroah-Hartman
2022-04-29 10:41 ` Greg Kroah-Hartman [this message]
2022-04-29 10:41 ` [PATCH 5.15 09/33] bpf: Add MEM_RDONLY for helper args that are pointers to rdonly mem Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 10/33] bpf/selftests: Test PTR_TO_RDONLY_MEM Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 11/33] bpf: Fix crash due to out of bounds access into reg2btf_ids Greg Kroah-Hartman
2022-04-29 10:41 ` [PATCH 5.15 12/33] spi: cadence-quadspi: fix write completion support Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 13/33] ARM: dts: socfpga: change qspi to "intel,socfpga-qspi" Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 14/33] mm: kfence: fix objcgs vector allocation Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 15/33] gup: Turn fault_in_pages_{readable,writeable} into fault_in_{readable,writeable} Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 16/33] iov_iter: Turn iov_iter_fault_in_readable into fault_in_iov_iter_readable Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 17/33] iov_iter: Introduce fault_in_iov_iter_writeable Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 18/33] gfs2: Add wrapper for iomap_file_buffered_write Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 19/33] gfs2: Clean up function may_grant Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 20/33] gfs2: Introduce flag for glock holder auto-demotion Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 21/33] gfs2: Move the inode glock locking to gfs2_file_buffered_write Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 22/33] gfs2: Eliminate ip->i_gh Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 23/33] gfs2: Fix mmap + page fault deadlocks for buffered I/O Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 24/33] iomap: Fix iomap_dio_rw return value for user copies Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 25/33] iomap: Support partial direct I/O on user copy failures Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 26/33] iomap: Add done_before argument to iomap_dio_rw Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 27/33] gup: Introduce FOLL_NOFAULT flag to disable page faults Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 28/33] iov_iter: Introduce nofault " Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 29/33] gfs2: Fix mmap + page fault deadlocks for direct I/O Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 30/33] btrfs: fix deadlock due to page faults during direct IO reads and writes Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 31/33] btrfs: fallback to blocking mode when doing async dio over multiple extents Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 32/33] mm: gup: make fault_in_safe_writeable() use fixup_user_fault() Greg Kroah-Hartman
2022-04-29 10:42 ` [PATCH 5.15 33/33] selftests/bpf: Add test for reg2btf_ids out of bounds access Greg Kroah-Hartman
2022-06-24 10:33   ` Po-Hsu Lin
2022-06-24 11:09     ` Greg Kroah-Hartman
2022-07-01 12:51       ` Po-Hsu Lin
2022-04-29 16:39 ` [PATCH 5.15 00/33] 5.15.37-rc1 review Florian Fainelli
2022-04-29 17:15 ` Jon Hunter
2022-04-29 18:36 ` Shuah Khan
2022-04-29 21:14 ` Naresh Kamboju
2022-04-29 23:47 ` Guenter Roeck
2022-04-29 23:54 ` Ron Economos
2022-04-30 10:17 ` Sudip Mukherjee

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=20220429104052.585495115@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=ast@kernel.org \
    --cc=haoluo@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@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.