Skip to contentSkip to conversation

bpf: Add bpf_copy_from_user_task_str kfunc

PATCH In review Jordan Rome 4 participants 9 messages Jan 7, 2025 – Jan 17, 2025

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/2 bpf: Add bpf_copy_from_user_task_str kfunc

This new kfunc will be able to copy a string from another process's/task's address space. This is similar to bpf_copy_from_user_str but accepts a struct task_struct* argument.

This required adding an additional function in memory.c, namely copy_str_from_process_vm, which works similar to access_process_vm but utilizes the strncpy_from_user helper and only supports reading/copying and not writing.

---
 include/linux/mm.h   |   3 ++
 kernel/bpf/helpers.c |  46 ++++++++++++++++++++
 mm/memory.c          | 101 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 150 insertions(+)
include/linux/mm.h +3 -0
index c39c4945946c..52b304b20630 100644
@@ -2484,6 +2484,9 @@ extern int access_process_vm(struct task_struct *tsk, unsigned long addr,
 extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
 		void *buf, int len, unsigned int gup_flags);
 
+extern int copy_str_from_process_vm(struct task_struct *tsk, unsigned long addr,
+		void *buf, int len, unsigned int gup_flags);
+
 long get_user_pages_remote(struct mm_struct *mm,
 			   unsigned long start, unsigned long nr_pages,
 			   unsigned int gup_flags, struct page **pages,
kernel/bpf/helpers.c +46 -0

58 lines not on this page · raw message

mm/memory.c +101 -0

113 lines not on this page · raw message

--
2.43.5

#1 lore raw

PATCH v2 2/2 selftests/bpf: Add tests for bpf_copy_from_user_task_str

This adds tests for both the happy path and the error path (with and without the BPF_F_PAD_ZEROS flag).

---
 .../selftests/bpf/prog_tests/bpf_iter.c       |  7 +++
 .../selftests/bpf/prog_tests/read_vsyscall.c  |  1 +
 .../selftests/bpf/progs/bpf_iter_tasks.c      | 55 +++++++++++++++++++
 .../selftests/bpf/progs/read_vsyscall.c       |  6 +-
 4 files changed, 67 insertions(+), 2 deletions(-)
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +7 -0
index 6f1bfacd7375..8ed864793bd1 100644
@@ -34,6 +34,8 @@
 #include "bpf_iter_ksym.skel.h"
 #include "bpf_iter_sockmap.skel.h"
 
+static char test_data[] = "test_data";
+
 static void test_btf_id_or_null(void)
 {
 	struct bpf_iter_test_kern3 *skel;
@@ -328,12 +330,17 @@ static void test_task_sleepable(void)
 	if (!ASSERT_OK_PTR(skel, "bpf_iter_tasks__open_and_load"))
 		return;
 
+	skel->bss->user_ptr = test_data;
 	do_dummy_read(skel->progs.dump_task_sleepable);
 
 	ASSERT_GT(skel->bss->num_expected_failure_copy_from_user_task, 0,
 		  "num_expected_failure_copy_from_user_task");
 	ASSERT_GT(skel->bss->num_success_copy_from_user_task, 0,
 		  "num_success_copy_from_user_task");
+	ASSERT_GT(skel->bss->num_expected_failure_copy_from_user_task_str, 0,
+		  "num_expected_failure_copy_from_user_task_str");
+	ASSERT_GT(skel->bss->num_success_copy_from_user_task_str, 0,
+		  "num_success_copy_from_user_task_str");
 
 	bpf_iter_tasks__destroy(skel);
 }
tools/testing/selftests/bpf/prog_tests/read_vsyscall.c +1 -0
index c7b9ba8b1d06..a8d1eaa67020 100644
@@ -24,6 +24,7 @@ struct read_ret_desc {
 	{ .name = "copy_from_user", .ret = -EFAULT },
 	{ .name = "copy_from_user_task", .ret = -EFAULT },
 	{ .name = "copy_from_user_str", .ret = -EFAULT },
+	{ .name = "copy_from_user_task_str", .ret = -EFAULT },
 };
 
 void test_read_vsyscall(void)
tools/testing/selftests/bpf/progs/bpf_iter_tasks.c +55 -0

82 lines not on this page · raw message

tools/testing/selftests/bpf/progs/read_vsyscall.c +4 -2
index 39ebef430059..623c1c5bd2d0 100644
@@ -8,14 +8,15 @@
 
 int target_pid = 0;
 void *user_ptr = 0;
-int read_ret[9];
+int read_ret[10];
 
 char _license[] SEC("license") = "GPL";
 
 /*
- * This is the only kfunc, the others are helpers
+ * These are the kfuncs, the others are helpers
  */
 int bpf_copy_from_user_str(void *dst, u32, const void *, u64) __weak __ksym;
+int bpf_copy_from_user_task_str(void *dst, u32, const void *, struct task_struct *, u64) __weak __ksym;
 
 SEC("fentry/" SYS_PREFIX "sys_nanosleep")
 int do_probe_read(void *ctx)
@@ -47,6 +48,7 @@ int do_copy_from_user(void *ctx)
 	read_ret[7] = bpf_copy_from_user_task(buf, sizeof(buf), user_ptr,
 					      bpf_get_current_task_btf(), 0);
 	read_ret[8] = bpf_copy_from_user_str((char *)buf, sizeof(buf), user_ptr, 0);
+	read_ret[9] = bpf_copy_from_user_task_str((char *)buf, sizeof(buf), user_ptr, bpf_get_current_task_btf(), 0);
 
 	return 0;
 }

--
2.43.5

#2 lore raw

Re: [bpf-next v2 2/2] selftests/bpf: Add tests for bpf_copy_from_user_task_str

Quoted 126 lines · Jordan Rome

This adds tests for both the happy path and the error path (with and without the BPF_F_PAD_ZEROS flag).

---
 .../selftests/bpf/prog_tests/bpf_iter.c       |  7 +++
 .../selftests/bpf/prog_tests/read_vsyscall.c  |  1 +
 .../selftests/bpf/progs/bpf_iter_tasks.c      | 55 +++++++++++++++++++
 .../selftests/bpf/progs/read_vsyscall.c       |  6 +-
 4 files changed, 67 insertions(+), 2 deletions(-)
tools/testing/selftests/bpf/prog_tests/bpf_iter.c +7 -0
index 6f1bfacd7375..8ed864793bd1 100644
@@ -34,6 +34,8 @@
 #include "bpf_iter_ksym.skel.h"
 #include "bpf_iter_sockmap.skel.h"
 
+static char test_data[] = "test_data";
+
 static void test_btf_id_or_null(void)
 {
        struct bpf_iter_test_kern3 *skel;
@@ -328,12 +330,17 @@ static void test_task_sleepable(void)
        if (!ASSERT_OK_PTR(skel, "bpf_iter_tasks__open_and_load"))
                return;
 
+       skel->bss->user_ptr = test_data;
        do_dummy_read(skel->progs.dump_task_sleepable);
 
        ASSERT_GT(skel->bss->num_expected_failure_copy_from_user_task, 0,
                  "num_expected_failure_copy_from_user_task");
        ASSERT_GT(skel->bss->num_success_copy_from_user_task, 0,
                  "num_success_copy_from_user_task");
+       ASSERT_GT(skel->bss->num_expected_failure_copy_from_user_task_str, 0,
+                 "num_expected_failure_copy_from_user_task_str");
+       ASSERT_GT(skel->bss->num_success_copy_from_user_task_str, 0,
+                 "num_success_copy_from_user_task_str");
 
        bpf_iter_tasks__destroy(skel);
 }
tools/testing/selftests/bpf/prog_tests/read_vsyscall.c +1 -0
index c7b9ba8b1d06..a8d1eaa67020 100644
@@ -24,6 +24,7 @@ struct read_ret_desc {
        { .name = "copy_from_user", .ret = -EFAULT },
        { .name = "copy_from_user_task", .ret = -EFAULT },
        { .name = "copy_from_user_str", .ret = -EFAULT },
+       { .name = "copy_from_user_task_str", .ret = -EFAULT },
 };
 
 void test_read_vsyscall(void)
tools/testing/selftests/bpf/progs/bpf_iter_tasks.c +39 -0
index bc10c4e4b4fa..90691e34b915 100644
@@ -9,6 +9,7 @@ char _license[] SEC("license") = "GPL";
 uint32_t tid = 0;
 int num_unknown_tid = 0;
 int num_known_tid = 0;
+void *user_ptr = 0;
 
 SEC("iter/task")
 int dump_task(struct bpf_iter__task *ctx)
@@ -35,7 +36,9 @@ int dump_task(struct bpf_iter__task *ctx)
 }
 
 int num_expected_failure_copy_from_user_task = 0;
+int num_expected_failure_copy_from_user_task_str = 0;
 int num_success_copy_from_user_task = 0;
+int num_success_copy_from_user_task_str = 0;
 
 SEC("iter.s/task")
 int dump_task_sleepable(struct bpf_iter__task *ctx)
@@ -44,6 +47,9 @@ int dump_task_sleepable(struct bpf_iter__task *ctx)
        struct task_struct *task = ctx->task;
        static const char info[] = "    === END ===";
        struct pt_regs *regs;
+       char task_str1[10] = "aaaaaaaaaa";
+       char task_str2[10], task_str3[10];
+       char task_str4[20] = "aaaaaaaaaaaaaaaaaaaa";
        void *ptr;
        uint32_t user_data = 0;
        int ret;
@@ -78,8 +84,57 @@ int dump_task_sleepable(struct bpf_iter__task *ctx)
                BPF_SEQ_PRINTF(seq, "%s\n", info);
                return 0;
        }
+
        ++num_success_copy_from_user_task;
 
+       /* Read an invalid pointer and ensure we get an error */
+       ptr = NULL;
+       ret = bpf_copy_from_user_task_str((char *)task_str1, sizeof(task_str1), ptr, task, 0);
+       if (ret >= 0 || task_str1[9] != 'a') {
+               BPF_SEQ_PRINTF(seq, "%s\n", info);
+               return 0;
+       }
+
+       /* Read an invalid pointer and ensure we get error with pad zeros flag */
+       ptr = NULL;
+       ret = bpf_copy_from_user_task_str((char *)task_str1, sizeof(task_str1), ptr, task, BPF_F_PAD_ZEROS);
+       if (ret >= 0 || task_str1[9] != '\0') {
+               BPF_SEQ_PRINTF(seq, "%s\n", info);
+               return 0;
+       }
+
+       ++num_expected_failure_copy_from_user_task_str;
+
+       /* Same length as the string */
+       ret = bpf_copy_from_user_task_str((char *)task_str2, 10, user_ptr, task, 0);
+       if (bpf_strncmp(task_str2, 10, "test_data\0") != 0 || ret != 10) {
+               BPF_SEQ_PRINTF(seq, "%s\n", info);
+               return 0;
+       }
+
+       /* Shorter length than the string */
+       ret = bpf_copy_from_user_task_str((char *)task_str3, 9, user_ptr, task, 0);
+       if (bpf_strncmp(task_str3, 9, "test_dat\0") != 0 || ret != 9) {
+               BPF_SEQ_PRINTF(seq, "%s\n", info);
+               return 0;
+       }
+

tools/testing/selftests/bpf/progs/bpf_iter_tasks.c int dump_task_sleepable(struct bpf_iter__task *ctx) · line 131

Quoted 10 lines
tools/testing/selftests/bpf/progs/bpf_iter_tasks.c +10 -0
+       /* Longer length than the string */
+       ret = bpf_copy_from_user_task_str((char *)task_str4, 20, user_ptr, task, 0);
+       if (bpf_strncmp(task_str4, 10, "test_data\0") != 0 || ret != 10 || task_str4[sizeof(task_str4) - 1] != 'a') {
+               BPF_SEQ_PRINTF(seq, "%s\n", info);
+               return 0;
+       }
+
+       /* Longer length than the string with pad zeros flag */
+       ret = bpf_copy_from_user_task_str((char *)task_str4, 20, user_ptr, task, BPF_F_PAD_ZEROS);
+       if (bpf_strncmp(task_str4, 10, "test_data\0") != 0 || ret != 10 || task_str4[sizeof(task_str4) - 1] != '\0') {

looks like a long string, please check 100 character limit

tools/testing/selftests/bpf/progs/bpf_iter_tasks.c int dump_task_sleepable(struct bpf_iter__task *ctx) · line 133

Quoted 2 lines
tools/testing/selftests/bpf/progs/bpf_iter_tasks.c +2 -0
+               BPF_SEQ_PRINTF(seq, "%s\n", info);
+               return 0;

nit: this BPF_SEQ_PRINTF() + return 0 repetition is... repetitive :) goto drop_out; maybe?

Quoted 16 lines
tools/testing/selftests/bpf/progs/bpf_iter_tasks.c +4 -0
+       }
+
+       ++num_success_copy_from_user_task_str;
+
        if (ctx->meta->seq_num == 0)
                BPF_SEQ_PRINTF(seq, "    tgid      gid     data\n");
 
tools/testing/selftests/bpf/progs/read_vsyscall.c +0 -1
index 39ebef430059..623c1c5bd2d0 100644
@@ -8,14 +8,15 @@
 
 int target_pid = 0;
 void *user_ptr = 0;
-int read_ret[9];

tools/testing/selftests/bpf/progs/read_vsyscall.c line 19

Quoted 10 lines
tools/testing/selftests/bpf/progs/read_vsyscall.c +3 -1
+int read_ret[10];
 
 char _license[] SEC("license") = "GPL";
 
 /*
- * This is the only kfunc, the others are helpers
+ * These are the kfuncs, the others are helpers
  */
 int bpf_copy_from_user_str(void *dst, u32, const void *, u64) __weak __ksym;
+int bpf_copy_from_user_task_str(void *dst, u32, const void *, struct task_struct *, u64) __weak __ksym;

these definitions should be coming from vmlinux.h, no need to add them manually anymore

tools/testing/selftests/bpf/progs/read_vsyscall.c int do_copy_from_user(void *ctx) · line 51

Quoted 8 lines
tools/testing/selftests/bpf/progs/read_vsyscall.c +1 -0
 SEC("fentry/" SYS_PREFIX "sys_nanosleep")
 int do_probe_read(void *ctx)
@@ -47,6 +48,7 @@ int do_copy_from_user(void *ctx)
        read_ret[7] = bpf_copy_from_user_task(buf, sizeof(buf), user_ptr,
                                              bpf_get_current_task_btf(), 0);
        read_ret[8] = bpf_copy_from_user_str((char *)buf, sizeof(buf), user_ptr, 0);
+       read_ret[9] = bpf_copy_from_user_task_str((char *)buf, sizeof(buf), user_ptr, bpf_get_current_task_btf(), 0);

please drop all those (char *) casts, in C any pointer is implicitly castable to void * and void * is (implicitly) castable to any other pointer. C++ is stricter, but in C it's canonical to not do casting

Quoted 6 lines
tools/testing/selftests/bpf/progs/read_vsyscall.c +0 -1
        return 0;
 }
--

2.43.5

#3 lore raw

Re: [bpf-next v2 2/2] selftests/bpf: Add tests for bpf_copy_from_user_task_str

In reply to Jordan Rome
kernel test robot noticed "WARNING:possible_recursive_locking_detected" on:

Hello,

kernel test robot noticed "WARNING:possible_recursive_locking_detected" on:

commit: 974e24f3e253a8e69418b73e486f74c6fa40e449 ("[bpf-next v2 2/2] selftests/bpf: Add tests for bpf_copy_from_user_task_str")
url: https://github.com/intel-lab-lkp/linux/commits/Jordan-Rome/selftests-bpf-Add-tests-for-bpf_copy_from_user_task_str/20250107-100850
base: https://git.kernel.org/cgit/linux/kernel/git/bpf/bpf-next.git master
patch link: https://lore.kernel.org/all/[email protected]/
patch subject: [bpf-next v2 2/2] selftests/bpf: Add tests for bpf_copy_from_user_task_str

in testcase: kernel-selftests-bpf
version:
with following parameters:

group: bpf

config: x86_64-rhel-9.4-bpf
compiler: gcc-12
test machine: 8 threads 1 sockets Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz (Kaby Lake) with 32G memory

(please refer to attached dmesg/kmsg for entire log/backtrace)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-lkp/[email protected]

[ 1645.762718][T49812] WARNING: possible recursive locking detected
[ 1645.768742][T49812] 6.13.0-rc3-00084-g974e24f3e253 #1 Tainted: G           OE
[ 1645.776333][T49812] --------------------------------------------
[ 1645.782356][T49812] test_progs/49812 is trying to acquire lock:
[1645.788292][T49812] ffff88815d74c620 (&mm->mmap_lock){++++}-{4:4}, at: __might_fault (mm/memory.c:6852 mm/memory.c:6845) 
[ 1645.796945][T49812]
[ 1645.796945][T49812] but task is already holding lock:
[1645.804188][T49812] ffff88815d74c620 (&mm->mmap_lock){++++}-{4:4}, at: copy_str_from_process_vm (include/linux/mmap_lock.h:153 mm/memory.c:6686 mm/memory.c:6810) 
[ 1645.813879][T49812]
[ 1645.813879][T49812] other info that might help us debug this:
[ 1645.821816][T49812]  Possible unsafe locking scenario:
[ 1645.821816][T49812]
[ 1645.829146][T49812]        CPU0
[ 1645.832298][T49812]        ----
[ 1645.835450][T49812]   lock(&mm->mmap_lock);
[ 1645.839652][T49812]   lock(&mm->mmap_lock);
[ 1645.843865][T49812]
[ 1645.843865][T49812]  *** DEADLOCK ***
[ 1645.843865][T49812]
[ 1645.851888][T49812]  May be due to missing lock nesting notation
[ 1645.851888][T49812]
[ 1645.860086][T49812] 3 locks held by test_progs/49812:
[1645.865153][T49812] #0: ffff88835c5ab698 (&p->lock){+.+.}-{4:4}, at: bpf_seq_read (kernel/bpf/bpf_iter.c:105) 
[1645.873715][T49812] #1: ffffffff84ca6ec0 (rcu_read_lock_trace){....}-{0:0}, at: bpf_iter_run_prog (include/linux/rcupdate.h:337 include/linux/rcupdate_trace.h:58 kernel/bpf/bpf_iter.c:700) 
[1645.883680][T49812] #2: ffff88815d74c620 (&mm->mmap_lock){++++}-{4:4}, at: copy_str_from_process_vm (include/linux/mmap_lock.h:153 mm/memory.c:6686 mm/memory.c:6810) 
[ 1645.893817][T49812]
[ 1645.893817][T49812] stack backtrace:
[ 1645.899581][T49812] CPU: 3 UID: 0 PID: 49812 Comm: test_progs Tainted: G           OE      6.13.0-rc3-00084-g974e24f3e253 #1
[ 1645.910828][T49812] Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
[ 1645.916848][T49812] Hardware name: Dell Inc. OptiPlex 7050/062KRH, BIOS 1.2.0 12/22/2016
[ 1645.924972][T49812] Call Trace:
[ 1645.928128][T49812]  <TASK>
[1645.930938][T49812] dump_stack_lvl (lib/dump_stack.c:124) 
[1645.935320][T49812] print_deadlock_bug (kernel/locking/lockdep.c:3040) 
[1645.940224][T49812] validate_chain (kernel/locking/lockdep.c:3894) 
[1645.944784][T49812] ? __pfx_validate_chain (kernel/locking/lockdep.c:3860) 
[1645.949860][T49812] ? mark_lock (kernel/locking/lockdep.c:4727) 
[1645.954062][T49812] __lock_acquire (kernel/locking/lockdep.c:5226) 
[1645.958640][T49812] lock_acquire (kernel/locking/lockdep.c:467 kernel/locking/lockdep.c:5851 kernel/locking/lockdep.c:5814) 
[1645.963015][T49812] ? __might_fault (mm/memory.c:6852 mm/memory.c:6845) 
[1645.967479][T49812] ? __pfx_lock_acquire (kernel/locking/lockdep.c:5817) 
[1645.972377][T49812] ? __rcu_read_unlock (kernel/rcu/tree_plugin.h:440 (discriminator 2)) 
[1645.977187][T49812] ? mtree_load (lib/maple_tree.c:6337) 
[1645.981563][T49812] ? lock_is_held_type (kernel/locking/lockdep.c:5590 kernel/locking/lockdep.c:5921) 
[1645.986460][T49812] ? __might_fault (mm/memory.c:6852 mm/memory.c:6845) 
[1645.990923][T49812] __might_fault (mm/memory.c:6852 mm/memory.c:6845) 
[1645.995212][T49812] ? __might_fault (mm/memory.c:6852 mm/memory.c:6845) 
[1645.999699][T49812] strncpy_from_user (lib/strncpy_from_user.c:120) 
[1646.004425][T49812] copy_str_from_process_vm (include/linux/page-flags.h:242 include/linux/highmem.h:661 mm/memory.c:6722 mm/memory.c:6810) 
[1646.009845][T49812] ? __pfx_copy_str_from_process_vm (mm/memory.c:6802) 
[1646.015789][T49812] bpf_copy_from_user_task_str (kernel/bpf/helpers.c:3104) 
[1646.021297][T49812] bpf_prog_f57787fdd126684b_dump_task_sleepable+0x278/0x4d0 
[1646.028546][T49812] ? __might_fault (mm/memory.c:6852 mm/memory.c:6845) 
[1646.033011][T49812] bpf_iter_run_prog (include/linux/bpf.h:1290 include/linux/filter.h:701 include/linux/filter.h:708 kernel/bpf/bpf_iter.c:704) 
[1646.037821][T49812] ? __pfx_bpf_iter_run_prog (kernel/bpf/bpf_iter.c:695) 
[1646.043151][T49812] ? __pfx___lock_release+0x10/0x10 
[1646.048837][T49812] task_seq_show (kernel/bpf/task_iter.c:193) 
[1646.053124][T49812] ? __pfx_task_seq_show (kernel/bpf/task_iter.c:193) 
[1646.058111][T49812] bpf_seq_read (kernel/bpf/bpf_iter.c:138) 
[1646.062399][T49812] ? rw_verify_area (include/linux/fsnotify_backend.h:501 include/linux/fsnotify.h:24 include/linux/fsnotify.h:127 include/linux/fsnotify.h:153 fs/read_write.c:470) 
[1646.067035][T49812] vfs_read (fs/read_write.c:563) 
[1646.071062][T49812] ? __pfx___lock_release+0x10/0x10 
[1646.076740][T49812] ? __pfx_vfs_read (fs/read_write.c:546) 
[1646.081290][T49812] ? __pfx___lock_release+0x10/0x10 
[1646.086968][T49812] ? __mutex_unlock_slowpath (arch/x86/include/asm/atomic64_64.h:101 include/linux/atomic/atomic-arch-fallback.h:4329 include/linux/atomic/atomic-long.h:1506 include/linux/atomic/atomic-instrumented.h:4481 kernel/locking/mutex.c:913) 
[1646.092472][T49812] ? __fget_files (include/linux/rcupdate.h:347 include/linux/rcupdate.h:880 fs/file.c:1050) 
[1646.097022][T49812] ? __rcu_read_unlock (kernel/rcu/tree_plugin.h:440 (discriminator 2)) 
[1646.101831][T49812] ? __fget_files (fs/file.c:1053) 
[1646.106383][T49812] ksys_read (fs/read_write.c:709) 
[1646.110410][T49812] ? __pfx_ksys_read (fs/read_write.c:698) 
[1646.115042][T49812] ? fput (arch/x86/include/asm/atomic64_64.h:79 include/linux/atomic/atomic-arch-fallback.h:2913 include/linux/atomic/atomic-arch-fallback.h:3364 include/linux/atomic/atomic-long.h:698 include/linux/atomic/atomic-instrumented.h:3767 include/linux/file_ref.h:157 fs/file_table.c:501) 
[1646.118809][T49812] ? ksys_read (fs/read_write.c:698) 
[1646.123007][T49812] ? mark_held_locks (kernel/locking/lockdep.c:4309) 
[1646.127662][T49812] do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83) 
[1646.132033][T49812] ? lockdep_hardirqs_on_prepare (kernel/locking/lockdep.c:4347 kernel/locking/lockdep.c:4406) 
[1646.137887][T49812] ? syscall_exit_to_user_mode (arch/x86/include/asm/jump_label.h:36 include/linux/context_tracking_state.h:108 include/linux/context_tracking.h:41 include/linux/entry-common.h:364 kernel/entry/common.c:220) 
[1646.143481][T49812] ? do_syscall_64 (arch/x86/entry/common.c:102) 
[1646.148026][T49812] ? __pfx_vfs_read (fs/read_write.c:546) 
[1646.152574][T49812] ? fput (arch/x86/include/asm/atomic64_64.h:79 include/linux/atomic/atomic-arch-fallback.h:2913 include/linux/atomic/atomic-arch-fallback.h:3364 include/linux/atomic/atomic-long.h:698 include/linux/atomic/atomic-instrumented.h:3767 include/linux/file_ref.h:157 fs/file_table.c:501) 
[1646.156339][T49812] ? __fget_files (include/linux/rcupdate.h:347 include/linux/rcupdate.h:880 fs/file.c:1050) 
[1646.160891][T49812] ? __rcu_read_unlock (kernel/rcu/tree_plugin.h:440 (discriminator 2)) 
[1646.165712][T49812] ? __fget_files (fs/file.c:1053) 
[1646.170267][T49812] ? fput (arch/x86/include/asm/atomic64_64.h:79 include/linux/atomic/atomic-arch-fallback.h:2913 include/linux/atomic/atomic-arch-fallback.h:3364 include/linux/atomic/atomic-long.h:698 include/linux/atomic/atomic-instrumented.h:3767 include/linux/file_ref.h:157 fs/file_table.c:501) 
[1646.174039][T49812] ? ksys_read (fs/read_write.c:698) 
[1646.178243][T49812] ? __pfx_ksys_read (fs/read_write.c:698) 
[1646.182887][T49812] ? mark_held_locks (kernel/locking/lockdep.c:4309) 
[1646.187531][T49812] ? lockdep_hardirqs_on_prepare (kernel/locking/lockdep.c:4347 kernel/locking/lockdep.c:4406) 
[1646.193388][T49812] ? syscall_exit_to_user_mode (arch/x86/include/asm/jump_label.h:36 include/linux/context_tracking_state.h:108 include/linux/context_tracking.h:41 include/linux/entry-common.h:364 kernel/entry/common.c:220) 
[1646.198982][T49812] ? do_syscall_64 (arch/x86/entry/common.c:102) 
[1646.203529][T49812] ? syscall_exit_to_user_mode (arch/x86/include/asm/jump_label.h:36 include/linux/context_tracking_state.h:108 include/linux/context_tracking.h:41 include/linux/entry-common.h:364 kernel/entry/common.c:220) 
[1646.209122][T49812] ? do_syscall_64 (arch/x86/entry/common.c:102) 
[1646.213695][T49812] ? __fget_files (fs/file.c:1053) 
[1646.218244][T49812] ? __fget_files (include/linux/rcupdate.h:347 include/linux/rcupdate.h:880 fs/file.c:1050) 
[1646.222794][T49812] ? __rcu_read_unlock (kernel/rcu/tree_plugin.h:440 (discriminator 2)) 
[1646.227602][T49812] ? __fget_files (fs/file.c:1053) 
[1646.232151][T49812] ? fput (arch/x86/include/asm/atomic64_64.h:79 include/linux/atomic/atomic-arch-fallback.h:2913 include/linux/atomic/atomic-arch-fallback.h:3364 include/linux/atomic/atomic-long.h:698 include/linux/atomic/atomic-instrumented.h:3767 include/linux/file_ref.h:157 fs/file_table.c:501) 
[1646.235917][T49812] ? ksys_read (fs/read_write.c:698) 
[1646.240118][T49812] ? __pfx_ksys_read (fs/read_write.c:698) 
[1646.244752][T49812] ? mark_held_locks (kernel/locking/lockdep.c:4309) 
[1646.249391][T49812] ? lockdep_hardirqs_on_prepare (kernel/locking/lockdep.c:4347 kernel/locking/lockdep.c:4406) 
[1646.255245][T49812] ? syscall_exit_to_user_mode (arch/x86/include/asm/jump_label.h:36 include/linux/context_tracking_state.h:108 include/linux/context_tracking.h:41 include/linux/entry-common.h:364 kernel/entry/common.c:220) 
[1646.260837][T49812] ? do_syscall_64 (arch/x86/entry/common.c:102) 
[1646.265383][T49812] ? do_syscall_64 (arch/x86/entry/common.c:102) 
[1646.269930][T49812] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) 
[ 1646.275697][T49812] RIP: 0033:0x7fdb112fe25c
[ 1646.279982][T49812] Code: ec 28 48 89 54 24 18 48 89 74 24 10 89 7c 24 08 e8 d9 d5 f8 ff 48 8b 54 24 18 48 8b 74 24 10 41 89 c0 8b 7c 24 08 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 34 44 89 c7 48 89 44 24 08 e8 2f d6 f8 ff 48
All code
========
   0:	ec                   	in     (%dx),%al
   1:	28 48 89             	sub    %cl,-0x77(%rax)
   4:	54                   	push   %rsp
   5:	24 18                	and    $0x18,%al
   7:	48 89 74 24 10       	mov    %rsi,0x10(%rsp)
   c:	89 7c 24 08          	mov    %edi,0x8(%rsp)
  10:	e8 d9 d5 f8 ff       	call   0xfffffffffff8d5ee
  15:	48 8b 54 24 18       	mov    0x18(%rsp),%rdx
  1a:	48 8b 74 24 10       	mov    0x10(%rsp),%rsi
  1f:	41 89 c0             	mov    %eax,%r8d
  22:	8b 7c 24 08          	mov    0x8(%rsp),%edi
  26:	31 c0                	xor    %eax,%eax
  28:	0f 05                	syscall
  2a:*	48 3d 00 f0 ff ff    	cmp    $0xfffffffffffff000,%rax		<-- trapping instruction
  30:	77 34                	ja     0x66
  32:	44 89 c7             	mov    %r8d,%edi
  35:	48 89 44 24 08       	mov    %rax,0x8(%rsp)
  3a:	e8 2f d6 f8 ff       	call   0xfffffffffff8d66e
  3f:	48                   	rex.W

Code starting with the faulting instruction
===========================================

   0:	48 3d 00 f0 ff ff    	cmp    $0xfffffffffffff000,%rax
   6:	77 34                	ja     0x3c
   8:	44 89 c7             	mov    %r8d,%edi
   b:	48 89 44 24 08       	mov    %rax,0x8(%rsp)
  10:	e8 2f d6 f8 ff       	call   0xfffffffffff8d644
  15:	48                   	rex.W

The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20250116/[email protected]

Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
#4 lore raw
In reply to Jordan Rome
kernel test robot noticed the following build errors:

Hi Jordan,

kernel test robot noticed the following build errors:

[auto build test ERROR on bpf-next/master]

url:    https://github.com/intel-lab-lkp/linux/commits/Jordan-Rome/selftests-bpf-Add-tests-for-bpf_copy_from_user_task_str/20250107-100850
base:   https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link:    https://lore.kernel.org/r/20250107020632.170883-1-linux%40jordanrome.com
patch subject: [bpf-next v2 1/2] bpf: Add bpf_copy_from_user_task_str kfunc
config: arm-imxrt_defconfig (https://download.01.org/0day-ci/archive/20250109/[email protected]/config)
compiler: clang version 19.1.3 (https://github.com/llvm/llvm-project ab51eccf88f5321e7c60591c5546b254b6afab99)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250109/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/

All errors (new ones prefixed by >>):

>> ld.lld: error: undefined symbol: copy_str_from_process_vm
   >>> referenced by helpers.c
   >>>               kernel/bpf/helpers.o:(bpf_copy_from_user_task_str) in archive vmlinux.a
Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
#5 lore raw
In reply to Jordan Rome
kernel test robot noticed the following build errors:

Hi Jordan,

kernel test robot noticed the following build errors:

[auto build test ERROR on bpf-next/master]

url:    https://github.com/intel-lab-lkp/linux/commits/Jordan-Rome/selftests-bpf-Add-tests-for-bpf_copy_from_user_task_str/20250107-100850
base:   https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link:    https://lore.kernel.org/r/20250107020632.170883-1-linux%40jordanrome.com
patch subject: [bpf-next v2 1/2] bpf: Add bpf_copy_from_user_task_str kfunc
config: m68k-randconfig-r133-20250109 (https://download.01.org/0day-ci/archive/20250110/[email protected]/config)
compiler: m68k-linux-gcc (GCC) 14.2.0
reproduce: (https://download.01.org/0day-ci/archive/20250110/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/

All errors (new ones prefixed by >>):

m68k-linux-ld: kernel/bpf/helpers.o: in function `bpf_copy_from_user_task_str':

>> helpers.c:(.text+0x2a3a): undefined reference to `copy_str_from_process_vm'
Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
#6 lore raw
In reply to Jordan Rome
Quoted 3 lines · Jordan Rome

This new kfunc will be able to copy a string from another process's/task's address space.

nit: this is kernel code, task is unambiguous, so I'd drop the "process" reference here

Quoted 9 lines

This is similar to bpf_copy_from_user_str but accepts a struct task_struct* argument.

This required adding an additional function in memory.c, namely copy_str_from_process_vm, which works similar to access_process_vm but utilizes the strncpy_from_user helper and only supports reading/copying and not writing.

Quoted 7 lines
---
 include/linux/mm.h   |   3 ++
 kernel/bpf/helpers.c |  46 ++++++++++++++++++++
 mm/memory.c          | 101 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 150 insertions(+)

please check kernel test bot's complains as well

include/linux/mm.h extern int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 2488

Quoted 10 lines
include/linux/mm.h +2 -0
index c39c4945946c..52b304b20630 100644
@@ -2484,6 +2484,9 @@ extern int access_process_vm(struct task_struct *tsk, unsigned long addr,
 extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
                void *buf, int len, unsigned int gup_flags);
 
+extern int copy_str_from_process_vm(struct task_struct *tsk, unsigned long addr,
+               void *buf, int len, unsigned int gup_flags);

nit: curious what mm folks think about naming, I'd go with a slightly less verbose naming: "copy_remote_vm_str" (copy vs access, _str suffix for non fixed-sized semantics marking)

for the next revision, let's split out mm parts from helpers parts, I don't think we lose much as this new internal API is self-contained, and it will be easier for mm folks to review

Quoted 12 lines
include/linux/mm.h +1 -0
+
 long get_user_pages_remote(struct mm_struct *mm,
                           unsigned long start, unsigned long nr_pages,
                           unsigned int gup_flags, struct page **pages,
kernel/bpf/helpers.c +0 -0
index cd5f9884d85b..45d41b7a9906 100644
@@ -3072,6 +3072,51 @@ __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag)
        local_irq_restore(*flags__irq_flag);
 }
 

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3084

Quoted 10 lines
kernel/bpf/helpers.c +10 -0
+/**
+ * bpf_copy_from_user_task_str() - Copy a string from an task's address space
+ * @dst:             Destination address, in kernel space.  This buffer must be
+ *                   at least @dst__sz bytes long.
+ * @dst__sz:         Maximum number of bytes to copy, includes the trailing NUL.
+ * @unsafe_ptr__ign: Source address in the task's address space.
+ * @tsk:             The task whose address space will be used
+ * @flags:           The only supported flag is BPF_F_PAD_ZEROS
+ *
+ * Copies a NULL-terminated string from a task's address space to BPF space.

there is no "BPF space", really... maybe "copies string into *dst* buffer"

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3091

Quoted 7 lines
kernel/bpf/helpers.c +7 -0
+ * If user string is too long this will still ensure zero termination in the
+ * dst buffer unless buffer size is 0.
+ *
+ * If BPF_F_PAD_ZEROS flag is set, memset the tail of @dst to 0 on success and
+ * memset all of @dst on failure.
+ */
+__bpf_kfunc int bpf_copy_from_user_task_str(void *dst, u32 dst__sz, const void __user *unsafe_ptr__ign, struct task_struct *tsk, u64 flags)

this looks like a long line, does it fit under 100 characters?

Quoted 5 lines
kernel/bpf/helpers.c +5 -0
+{
+       int count = dst__sz - 1;
+       int ret = 0;
+
+       if (unlikely(flags & ~BPF_F_PAD_ZEROS))

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3106

Quoted 10 lines
kernel/bpf/helpers.c +10 -0
+               return -EINVAL;
+
+       if (unlikely(!dst__sz))
+               return 0;
+
+       ret = copy_str_from_process_vm(tsk, (unsigned long)unsafe_ptr__ign, dst, count, 0);
+
+       if (ret <= 0) {
+               if (flags & BPF_F_PAD_ZEROS)
+                       memset((char *)dst, 0, dst__sz);

nit: no need for (char *) cast? memset takes void *, I think

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3107

Quoted 1 line
kernel/bpf/helpers.c +1 -0
+               return ret;

if ret == 0, is that an error? If so, return ret ?: -EINVAL; or something along those lines?

pw-bot: cr

Quoted 28 lines
kernel/bpf/helpers.c +13 -0
+       }
+
+       if (ret < count) {
+               if (flags & BPF_F_PAD_ZEROS)
+                       memset((char *)dst + ret, 0, dst__sz - ret);
+       } else {
+               ((char *)dst)[count] = '\0';
+       }
+
+       return ret + 1;
+}
+
 __bpf_kfunc_end_defs();
 
 BTF_KFUNCS_START(generic_btf_ids)
@@ -3164,6 +3209,7 @@ BTF_ID_FLAGS(func, bpf_iter_bits_new, KF_ITER_NEW)
 BTF_ID_FLAGS(func, bpf_iter_bits_next, KF_ITER_NEXT | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_bits_destroy, KF_ITER_DESTROY)
 BTF_ID_FLAGS(func, bpf_copy_from_user_str, KF_SLEEPABLE)
+BTF_ID_FLAGS(func, bpf_copy_from_user_task_str, KF_SLEEPABLE)
 BTF_ID_FLAGS(func, bpf_get_kmem_cache)
 BTF_ID_FLAGS(func, bpf_iter_kmem_cache_new, KF_ITER_NEW | KF_SLEEPABLE)
 BTF_ID_FLAGS(func, bpf_iter_kmem_cache_next, KF_ITER_NEXT | KF_RET_NULL | KF_SLEEPABLE)
mm/memory.c +0 -0
index 75c2dfd04f72..514490bd7d6d 100644

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6678

Quoted 6 lines
mm/memory.c +3 -0
@@ -6673,6 +6673,75 @@ static int __access_remote_vm(struct mm_struct *mm, unsigned long addr,
        return buf - old_buf;
 }
 
+/*
+ * Copy a string from another process's address space as given in mm.
+ * Don't return partial results. If there is any error return -EFAULT.

What does "don't return partial results" mean? What happens if we read part of a string and then fail to read the rest?

Quoted 7 lines
mm/memory.c +7 -0
+ */
+static int __copy_str_from_remote_vm(struct mm_struct *mm, unsigned long addr,
+                             void *buf, int len, unsigned int gup_flags)
+{
+       void *old_buf = buf;
+       int err = 0;
+

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6695

Quoted 10 lines
mm/memory.c +10 -0
+       if (mmap_read_lock_killable(mm))
+               return -EFAULT;
+
+       /* Untag the address before looking up the VMA */
+       addr = untagged_addr_remote(mm, addr);
+
+       /* Avoid triggering the temporary warning in __get_user_pages */
+       if (!vma_lookup(mm, addr)) {
+               mmap_read_unlock(mm);
+               return -EFAULT;

maybe let's do (so that we do mmap_read_unlock in just one place)

err = -EFAULT;
goto err_out;

and then see below

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6704

Quoted 9 lines
mm/memory.c +9 -0
+       }
+
+       while (len) {
+               int bytes, offset, retval;
+               void *maddr;
+               struct vm_area_struct *vma = NULL;
+               struct page *page = get_user_page_vma_remote(mm, addr,
+                                                            gup_flags, &vma);
+

nit: I'd split page declaration and assignment and kept get_user_page_vma_remote() invocation on a single line

Quoted 7 lines
mm/memory.c +7 -0
+               if (IS_ERR(page)) {
+                       /*
+                        * Treat as a total failure for now until we decide how
+                        * to handle the CONFIG_HAVE_IOREMAP_PROT case and
+                        * stack expansion.
+                        */
+                       err = -EFAULT;

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6721

Quoted 10 lines
mm/memory.c +10 -0
+                       break;
+               }
+
+               bytes = len;
+               offset = addr & (PAGE_SIZE - 1);
+               if (bytes > PAGE_SIZE - offset)
+                       bytes = PAGE_SIZE - offset;
+
+               maddr = kmap_local_page(page);
+               retval = strncpy_from_user(buf, (const char __user *)addr, bytes);

you are not using maddr... that seems wrong (even if it works due to how kmap_local_page is currently implemented)

Quoted 5 lines
mm/memory.c +5 -0
+               unmap_and_put_page(page, maddr);
+
+               if (retval < 0) {
+                       err = retval;
+                       break;

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6736

Quoted 10 lines
mm/memory.c +10 -0
+               }
+
+               len -= retval;
+               buf += retval;
+               addr += retval;
+
+               /* Found the end of the string */
+               if (retval < bytes)
+                       break;
+       }

err_out: here

Quoted 12 lines
mm/memory.c +8 -0
+       mmap_read_unlock(mm);
+
+       if (err)
+               return err;
+
+       return buf - old_buf;
+}
+
 /**
  * access_remote_vm - access another process' address space
  * @mm:                the mm_struct of the target address space

mm/memory.c int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 6789

Quoted 7 lines
mm/memory.c +4 -0
@@ -6714,6 +6783,38 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr,
 }
 EXPORT_SYMBOL_GPL(access_process_vm);
 
+/**
+ * copy_str_from_process_vm - copy a string from another process's address space.
+ * @tsk:       the task of the target address space
+ * @addr:      start address to access

access -> read from

mm/memory.c int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 6790

Quoted 1 line
mm/memory.c +1 -0
+ * @buf:       source or destination buffer

for this api it's always the destination, right?

mm/memory.c int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 6797

Quoted 7 lines
mm/memory.c +7 -0
+ * @len:       number of bytes to transfer
+ * @gup_flags: flags modifying lookup behaviour
+ *
+ * The caller must hold a reference on @mm.
+ *
+ * Return: number of bytes copied from source to destination. If the string
+ * is shorter than @len then return the length of the string.

and if the string is longer than @len, then what happens? we should either specify or drop the "if string is shorter bit" and make it unambiguous whether terminating zero is included or not

Quoted 26 lines
mm/memory.c +20 -1
+ * On any error, return -EFAULT.
+ */
+int copy_str_from_process_vm(struct task_struct *tsk, unsigned long addr,
+               void *buf, int len, unsigned int gup_flags)
+{
+       struct mm_struct *mm;
+       int ret;
+
+       mm = get_task_mm(tsk);
+       if (!mm)
+               return -EFAULT;
+
+       ret = __copy_str_from_remote_vm(mm, addr, buf, len, gup_flags);
+
+       mmput(mm);
+
+       return ret;
+}
+EXPORT_SYMBOL_GPL(copy_str_from_process_vm);
+
 /*
  * Print the name of a VMA.
  */
--

2.43.5

#7 lore raw
Quoted 26 lines · Andrii Nakryiko
Quoted 3 lines · Jordan Rome

This new kfunc will be able to copy a string from another process's/task's address space.

nit: this is kernel code, task is unambiguous, so I'd drop the "process" reference here

Quoted 9 lines

This is similar to bpf_copy_from_user_str but accepts a struct task_struct* argument.

This required adding an additional function in memory.c, namely copy_str_from_process_vm, which works similar to access_process_vm but utilizes the strncpy_from_user helper and only supports reading/copying and not writing.

Quoted 7 lines
---
 include/linux/mm.h   |   3 ++
 kernel/bpf/helpers.c |  46 ++++++++++++++++++++
 mm/memory.c          | 101 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 150 insertions(+)
Quoted 1 line

please check kernel test bot's complains as well

Maybe I need an entry in nommu.c as well for 'copy_remote_vm_str'?

Quoted 9 lines
Quoted 9 lines
include/linux/mm.h +0 -0
index c39c4945946c..52b304b20630 100644
@@ -2484,6 +2484,9 @@ extern int access_process_vm(struct task_struct *tsk, unsigned long addr,
 extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
                void *buf, int len, unsigned int gup_flags);
 
Quoted 6 lines

include/linux/mm.h extern int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 2488

Quoted 2 lines
include/linux/mm.h +2 -0
+extern int copy_str_from_process_vm(struct task_struct *tsk, unsigned long addr,
+               void *buf, int len, unsigned int gup_flags);

nit: curious what mm folks think about naming, I'd go with a slightly less verbose naming: "copy_remote_vm_str" (copy vs access, _str suffix for non fixed-sized semantics marking)

Ack.

Quoted 4 lines

for the next revision, let's split out mm parts from helpers parts, I don't think we lose much as this new internal API is self-contained, and it will be easier for mm folks to review

Ack.

Quoted 13 lines
Quoted 13 lines
include/linux/mm.h +1 -0
+
 long get_user_pages_remote(struct mm_struct *mm,
                           unsigned long start, unsigned long nr_pages,
                           unsigned int gup_flags, struct page **pages,
kernel/bpf/helpers.c +0 -0
index cd5f9884d85b..45d41b7a9906 100644
@@ -3072,6 +3072,51 @@ __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag)
        local_irq_restore(*flags__irq_flag);
 }
 
Quoted 12 lines

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3084

Quoted 10 lines
kernel/bpf/helpers.c +10 -0
+/**
+ * bpf_copy_from_user_task_str() - Copy a string from an task's address space
+ * @dst:             Destination address, in kernel space.  This buffer must be
+ *                   at least @dst__sz bytes long.
+ * @dst__sz:         Maximum number of bytes to copy, includes the trailing NUL.
+ * @unsafe_ptr__ign: Source address in the task's address space.
+ * @tsk:             The task whose address space will be used
+ * @flags:           The only supported flag is BPF_F_PAD_ZEROS
+ *
+ * Copies a NULL-terminated string from a task's address space to BPF space.

there is no "BPF space", really... maybe "copies string into *dst* buffer"

Ack.

Quoted 10 lines

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3091

Quoted 8 lines
kernel/bpf/helpers.c +7 -0
+ * If user string is too long this will still ensure zero termination in the
+ * dst buffer unless buffer size is 0.
+ *
+ * If BPF_F_PAD_ZEROS flag is set, memset the tail of @dst to 0 on success and
+ * memset all of @dst on failure.
+ */
+__bpf_kfunc int bpf_copy_from_user_task_str(void *dst, u32 dst__sz, const void __user *unsafe_ptr__ign, struct task_struct *tsk, u64 flags)

this looks like a long line, does it fit under 100 characters?

I'll fix it in the next revision.

Quoted 19 lines
Quoted 6 lines
kernel/bpf/helpers.c +5 -0
+{
+       int count = dst__sz - 1;
+       int ret = 0;
+
+       if (unlikely(flags & ~BPF_F_PAD_ZEROS))

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3106

Quoted 10 lines
kernel/bpf/helpers.c +10 -0
+               return -EINVAL;
+
+       if (unlikely(!dst__sz))
+               return 0;
+
+       ret = copy_str_from_process_vm(tsk, (unsigned long)unsafe_ptr__ign, dst, count, 0);
+
+       if (ret <= 0) {
+               if (flags & BPF_F_PAD_ZEROS)
+                       memset((char *)dst, 0, dst__sz);

nit: no need for (char *) cast? memset takes void *, I think

Quoted 4 lines

kernel/bpf/helpers.c __bpf_kfunc void bpf_local_irq_restore(unsigned long *flags__irq_flag) · line 3107

Quoted 1 line
kernel/bpf/helpers.c +1 -0
+               return ret;

if ret == 0, is that an error? If so, return ret ?: -EINVAL; or something along those lines?

Good catch.

Quoted 34 lines

pw-bot: cr

Quoted 31 lines
kernel/bpf/helpers.c +13 -0
+       }
+
+       if (ret < count) {
+               if (flags & BPF_F_PAD_ZEROS)
+                       memset((char *)dst + ret, 0, dst__sz - ret);
+       } else {
+               ((char *)dst)[count] = '\0';
+       }
+
+       return ret + 1;
+}
+
 __bpf_kfunc_end_defs();
 
 BTF_KFUNCS_START(generic_btf_ids)
@@ -3164,6 +3209,7 @@ BTF_ID_FLAGS(func, bpf_iter_bits_new, KF_ITER_NEW)
 BTF_ID_FLAGS(func, bpf_iter_bits_next, KF_ITER_NEXT | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_bits_destroy, KF_ITER_DESTROY)
 BTF_ID_FLAGS(func, bpf_copy_from_user_str, KF_SLEEPABLE)
+BTF_ID_FLAGS(func, bpf_copy_from_user_task_str, KF_SLEEPABLE)
 BTF_ID_FLAGS(func, bpf_get_kmem_cache)
 BTF_ID_FLAGS(func, bpf_iter_kmem_cache_new, KF_ITER_NEW | KF_SLEEPABLE)
 BTF_ID_FLAGS(func, bpf_iter_kmem_cache_next, KF_ITER_NEXT | KF_RET_NULL | KF_SLEEPABLE)
mm/memory.c +0 -0
index 75c2dfd04f72..514490bd7d6d 100644
@@ -6673,6 +6673,75 @@ static int __access_remote_vm(struct mm_struct *mm, unsigned long addr,
        return buf - old_buf;
 }
 
Quoted 6 lines

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6678

Quoted 3 lines
mm/memory.c +3 -0
+/*
+ * Copy a string from another process's address space as given in mm.
+ * Don't return partial results. If there is any error return -EFAULT.

What does "don't return partial results" mean? What happens if we read part of a string and then fail to read the rest?

As per the last sentence, if we fail to read the rest of the string then an -EFAULT is returned.

Quoted 12 lines
Quoted 12 lines
mm/memory.c +11 -0
+ */
+static int __copy_str_from_remote_vm(struct mm_struct *mm, unsigned long addr,
+                             void *buf, int len, unsigned int gup_flags)
+{
+       void *old_buf = buf;
+       int err = 0;
+
+       if (mmap_read_lock_killable(mm))
+               return -EFAULT;
+
+       /* Untag the address before looking up the VMA */
Quoted 8 lines

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6695

Quoted 6 lines
mm/memory.c +6 -0
+       addr = untagged_addr_remote(mm, addr);
+
+       /* Avoid triggering the temporary warning in __get_user_pages */
+       if (!vma_lookup(mm, addr)) {
+               mmap_read_unlock(mm);
+               return -EFAULT;

maybe let's do (so that we do mmap_read_unlock in just one place)

Ack.

Quoted 6 lines

err = -EFAULT;
goto err_out;

and then see below

Quoted 12 lines

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6704

Quoted 9 lines
mm/memory.c +9 -0
+       }
+
+       while (len) {
+               int bytes, offset, retval;
+               void *maddr;
+               struct vm_area_struct *vma = NULL;
+               struct page *page = get_user_page_vma_remote(mm, addr,
+                                                            gup_flags, &vma);
+

nit: I'd split page declaration and assignment and kept get_user_page_vma_remote() invocation on a single line

Ack.

Quoted 13 lines
Quoted 13 lines
mm/memory.c +12 -0
+               if (IS_ERR(page)) {
+                       /*
+                        * Treat as a total failure for now until we decide how
+                        * to handle the CONFIG_HAVE_IOREMAP_PROT case and
+                        * stack expansion.
+                        */
+                       err = -EFAULT;
+                       break;
+               }
+
+               bytes = len;
+               offset = addr & (PAGE_SIZE - 1);
Quoted 8 lines

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6721

Quoted 5 lines
mm/memory.c +5 -0
+               if (bytes > PAGE_SIZE - offset)
+                       bytes = PAGE_SIZE - offset;
+
+               maddr = kmap_local_page(page);
+               retval = strncpy_from_user(buf, (const char __user *)addr, bytes);

you are not using maddr... that seems wrong (even if it works due to how kmap_local_page is currently implemented)

How do you think we should handle it then?

Quoted 34 lines
Quoted 6 lines
mm/memory.c +5 -0
+               unmap_and_put_page(page, maddr);
+
+               if (retval < 0) {
+                       err = retval;
+                       break;

mm/memory.c static int __access_remote_vm(struct mm_struct *mm, unsigned long addr, · line 6736

Quoted 10 lines
mm/memory.c +10 -0
+               }
+
+               len -= retval;
+               buf += retval;
+               addr += retval;
+
+               /* Found the end of the string */
+               if (retval < bytes)
+                       break;
+       }

err_out: here

Quoted 15 lines
mm/memory.c +8 -0
+       mmap_read_unlock(mm);
+
+       if (err)
+               return err;
+
+       return buf - old_buf;
+}
+
 /**
  * access_remote_vm - access another process' address space
  * @mm:                the mm_struct of the target address space
@@ -6714,6 +6783,38 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr,
 }
 EXPORT_SYMBOL_GPL(access_process_vm);
 
Quoted 6 lines

mm/memory.c int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 6789

Quoted 4 lines
mm/memory.c +4 -0
+/**
+ * copy_str_from_process_vm - copy a string from another process's address space.
+ * @tsk:       the task of the target address space
+ * @addr:      start address to access

access -> read from

Ack.

Quoted 4 lines

mm/memory.c int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 6790

Quoted 2 lines
mm/memory.c +1 -0
+ * @buf:       source or destination buffer

for this api it's always the destination, right?

Right. Will fix.

Quoted 9 lines

mm/memory.c int access_process_vm(struct task_struct *tsk, unsigned long addr, · line 6797

Quoted 8 lines
mm/memory.c +7 -0
+ * @len:       number of bytes to transfer
+ * @gup_flags: flags modifying lookup behaviour
+ *
+ * The caller must hold a reference on @mm.
+ *
+ * Return: number of bytes copied from source to destination. If the string
+ * is shorter than @len then return the length of the string.
Quoted 3 lines

and if the string is longer than @len, then what happens? we should either specify or drop the "if string is shorter bit" and make it unambiguous whether terminating zero is included or not

I'll make it clearer.

Quoted 27 lines
Quoted 27 lines
mm/memory.c +20 -1
+ * On any error, return -EFAULT.
+ */
+int copy_str_from_process_vm(struct task_struct *tsk, unsigned long addr,
+               void *buf, int len, unsigned int gup_flags)
+{
+       struct mm_struct *mm;
+       int ret;
+
+       mm = get_task_mm(tsk);
+       if (!mm)
+               return -EFAULT;
+
+       ret = __copy_str_from_remote_vm(mm, addr, buf, len, gup_flags);
+
+       mmput(mm);
+
+       return ret;
+}
+EXPORT_SYMBOL_GPL(copy_str_from_process_vm);
+
 /*
  * Print the name of a VMA.
  */
--

2.43.5

#8 lore raw
Quoted 29 lines · Jordan Rome
Quoted 26 lines · Andrii Nakryiko
Quoted 3 lines · Jordan Rome

This new kfunc will be able to copy a string from another process's/task's address space.

nit: this is kernel code, task is unambiguous, so I'd drop the "process" reference here

Quoted 16 lines

This is similar to bpf_copy_from_user_str but accepts a struct task_struct* argument.

This required adding an additional function in memory.c, namely copy_str_from_process_vm, which works similar to access_process_vm but utilizes the strncpy_from_user helper and only supports reading/copying and not writing.

---
 include/linux/mm.h   |   3 ++
 kernel/bpf/helpers.c |  46 ++++++++++++++++++++
 mm/memory.c          | 101 +++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 150 insertions(+)
Quoted 3 lines
Quoted 1 line

please check kernel test bot's complains as well

Maybe I need an entry in nommu.c as well for 'copy_remote_vm_str'?

yep, probably, I see that access_remote_vm has two implementations as well

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
include/linux/mm.h +0 -0
index c39c4945946c..52b304b20630 100644

[...]

Quoted 9 lines
Quoted 6 lines
Quoted 3 lines

+/*
+ * Copy a string from another process's address space as given in mm.
+ * Don't return partial results. If there is any error return -EFAULT.

What does "don't return partial results" mean? What happens if we read part of a string and then fail to read the rest?

As per the last sentence, if we fail to read the rest of the string then an -EFAULT is returned.

This is confusing because the buffer will contain the partially read string contents even with -EFAULT. And this "Don't return partial results" can be interpreted as this API sanitizing the buffer (by zeroing or something). So I'd drop the "Don't return partial results" part, as it just creates confusion.

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

+ */

[...]

Quoted 22 lines
Quoted 13 lines
Quoted 13 lines
include/linux/mm.h +12 -0
+               if (IS_ERR(page)) {
+                       /*
+                        * Treat as a total failure for now until we decide how
+                        * to handle the CONFIG_HAVE_IOREMAP_PROT case and
+                        * stack expansion.
+                        */
+                       err = -EFAULT;
+                       break;
+               }
+
+               bytes = len;
+               offset = addr & (PAGE_SIZE - 1);
Quoted 8 lines

include/linux/mm.h

Quoted 5 lines
include/linux/mm.h +5 -0
+               if (bytes > PAGE_SIZE - offset)
+                       bytes = PAGE_SIZE - offset;
+
+               maddr = kmap_local_page(page);
+               retval = strncpy_from_user(buf, (const char __user *)addr, bytes);

you are not using maddr... that seems wrong (even if it works due to how kmap_local_page is currently implemented)

Quoted 1 line

How do you think we should handle it then?

Use (maddr + offset_within_the_page) instead of addr itself?

Quoted 18 lines
Quoted 18 lines
Quoted 17 lines
include/linux/mm.h +15 -0
+               unmap_and_put_page(page, maddr);
+
+               if (retval < 0) {
+                       err = retval;
+                       break;
+               }
+
+               len -= retval;
+               buf += retval;
+               addr += retval;
+
+               /* Found the end of the string */
+               if (retval < bytes)
+                       break;
+       }

[...]