On Sun, Oct 13, 2024 at 3:45 AM Andrey Konovalov <andreyknvl@xxxxxxxxx> wrote: > > On Fri, Oct 11, 2024 at 5:52 AM Sabyrzhan Tasbolatov > <snovitoll@xxxxxxxxx> wrote: > > > > Instrument copy_from_kernel_nofault() with KMSAN for uninitialized kernel > > memory check and copy_to_kernel_nofault() with KASAN, KCSAN to detect > > the memory corruption. > > > > syzbot reported that bpf_probe_read_kernel() kernel helper triggered > > KASAN report via kasan_check_range() which is not the expected behaviour > > as copy_from_kernel_nofault() is meant to be a non-faulting helper. > > > > Solution is, suggested by Marco Elver, to replace KASAN, KCSAN check in > > copy_from_kernel_nofault() with KMSAN detection of copying uninitilaized > > kernel memory. In copy_to_kernel_nofault() we can retain > > instrument_write() explicitly for the memory corruption instrumentation. > > For future reference: please write commit messages in a way that is > readable standalone. I.e. without obscured references to the > discussions or problems in the previous versions of the patch. It's > fine to give such references in itself, but you need to give enough > context in the commit message to make it understandable without > looking up those discussions. > > > copy_to_kernel_nofault() is tested on x86_64 and arm64 with > > CONFIG_KASAN_SW_TAGS. On arm64 with CONFIG_KASAN_HW_TAGS, > > kunit test currently fails. Need more clarification on it. > > > > Link: https://lore.kernel.org/linux-mm/CANpmjNMAVFzqnCZhEity9cjiqQ9CVN1X7qeeeAp_6yKjwKo8iw@xxxxxxxxxxxxxx/ > > Reviewed-by: Marco Elver <elver@xxxxxxxxxx> > > Reported-by: syzbot+61123a5daeb9f7454599@xxxxxxxxxxxxxxxxxxxxxxxxx > > Closes: https://syzkaller.appspot.com/bug?extid=61123a5daeb9f7454599 > > Reported-by: Andrey Konovalov <andreyknvl@xxxxxxxxx> > > Closes: https://bugzilla.kernel.org/show_bug.cgi?id=210505 > > Signed-off-by: Sabyrzhan Tasbolatov <snovitoll@xxxxxxxxx> > > --- > > v2: > > - squashed previous submitted in -mm tree 2 patches based on Linus tree > > v3: > > - moved checks to *_nofault_loop macros per Marco's comments > > - edited the commit message > > v4: > > - replaced Suggested-by with Reviewed-by > > v5: > > - addressed Andrey's comment on deleting CONFIG_KASAN_HW_TAGS check in > > mm/kasan/kasan_test_c.c > > - added explanatory comment in kasan_test_c.c > > - added Suggested-by: Marco Elver back per Andrew's comment. > > v6: > > - deleted checks KASAN_TAG_MIN, KASAN_TAG_KERNEL per Andrey's comment. > > - added empty line before kfree. > > --- > > mm/kasan/kasan_test_c.c | 34 ++++++++++++++++++++++++++++++++++ > > mm/kmsan/kmsan_test.c | 17 +++++++++++++++++ > > mm/maccess.c | 10 ++++++++-- > > 3 files changed, 59 insertions(+), 2 deletions(-) > > > > diff --git a/mm/kasan/kasan_test_c.c b/mm/kasan/kasan_test_c.c > > index a181e4780d9d..716f2cac9708 100644 > > --- a/mm/kasan/kasan_test_c.c > > +++ b/mm/kasan/kasan_test_c.c > > @@ -1954,6 +1954,39 @@ static void rust_uaf(struct kunit *test) > > KUNIT_EXPECT_KASAN_FAIL(test, kasan_test_rust_uaf()); > > } > > > > +static void copy_to_kernel_nofault_oob(struct kunit *test) > > +{ > > + char *ptr; > > + char buf[128]; > > + size_t size = sizeof(buf); > > + > > + /* This test currently fails with the HW_TAGS mode. > > + * The reason is unknown and needs to be investigated. */ > > + KASAN_TEST_NEEDS_CONFIG_OFF(test, CONFIG_KASAN_HW_TAGS); > > + > > + ptr = kmalloc(size - KASAN_GRANULE_SIZE, GFP_KERNEL); > > + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr); > > + OPTIMIZER_HIDE_VAR(ptr); > > + > > + /* > > + * We test copy_to_kernel_nofault() to detect corrupted memory that is > > + * being written into the kernel. In contrast, copy_from_kernel_nofault() > > + * is primarily used in kernel helper functions where the source address > > + * might be random or uninitialized. Applying KASAN instrumentation to > > + * copy_from_kernel_nofault() could lead to false positives. > > + * By focusing KASAN checks only on copy_to_kernel_nofault(), > > + * we ensure that only valid memory is written to the kernel, > > + * minimizing the risk of kernel corruption while avoiding > > + * false positives in the reverse case. > > + */ > > + KUNIT_EXPECT_KASAN_FAIL(test, > > + copy_to_kernel_nofault(&buf[0], ptr, size)); > > + KUNIT_EXPECT_KASAN_FAIL(test, > > + copy_to_kernel_nofault(ptr, &buf[0], size)); > > + > > + kfree(ptr); > > +} > > + > > static struct kunit_case kasan_kunit_test_cases[] = { > > KUNIT_CASE(kmalloc_oob_right), > > KUNIT_CASE(kmalloc_oob_left), > > @@ -2027,6 +2060,7 @@ static struct kunit_case kasan_kunit_test_cases[] = { > > KUNIT_CASE(match_all_not_assigned), > > KUNIT_CASE(match_all_ptr_tag), > > KUNIT_CASE(match_all_mem_tag), > > + KUNIT_CASE(copy_to_kernel_nofault_oob), > > KUNIT_CASE(rust_uaf), > > {} > > }; > > diff --git a/mm/kmsan/kmsan_test.c b/mm/kmsan/kmsan_test.c > > index 13236d579eba..9733a22c46c1 100644 > > --- a/mm/kmsan/kmsan_test.c > > +++ b/mm/kmsan/kmsan_test.c > > @@ -640,6 +640,22 @@ static void test_unpoison_memory(struct kunit *test) > > KUNIT_EXPECT_TRUE(test, report_matches(&expect)); > > } > > > > +static void test_copy_from_kernel_nofault(struct kunit *test) > > +{ > > + long ret; > > + char buf[4], src[4]; > > + size_t size = sizeof(buf); > > + > > + EXPECTATION_UNINIT_VALUE_FN(expect, "copy_from_kernel_nofault"); > > + kunit_info( > > + test, > > + "testing copy_from_kernel_nofault with uninitialized memory\n"); > > + > > + ret = copy_from_kernel_nofault((char *)&buf[0], (char *)&src[0], size); > > + USE(ret); > > + KUNIT_EXPECT_TRUE(test, report_matches(&expect)); > > +} > > + > > static struct kunit_case kmsan_test_cases[] = { > > KUNIT_CASE(test_uninit_kmalloc), > > KUNIT_CASE(test_init_kmalloc), > > @@ -664,6 +680,7 @@ static struct kunit_case kmsan_test_cases[] = { > > KUNIT_CASE(test_long_origin_chain), > > KUNIT_CASE(test_stackdepot_roundtrip), > > KUNIT_CASE(test_unpoison_memory), > > + KUNIT_CASE(test_copy_from_kernel_nofault), > > {}, > > }; > > > > diff --git a/mm/maccess.c b/mm/maccess.c > > index 518a25667323..3ca55ec63a6a 100644 > > --- a/mm/maccess.c > > +++ b/mm/maccess.c > > @@ -13,9 +13,14 @@ bool __weak copy_from_kernel_nofault_allowed(const void *unsafe_src, > > return true; > > } > > > > +/* > > + * The below only uses kmsan_check_memory() to ensure uninitialized kernel > > + * memory isn't leaked. > > + */ > > #define copy_from_kernel_nofault_loop(dst, src, len, type, err_label) \ > > while (len >= sizeof(type)) { \ > > - __get_kernel_nofault(dst, src, type, err_label); \ > > + __get_kernel_nofault(dst, src, type, err_label); \ > > + kmsan_check_memory(src, sizeof(type)); \ > > dst += sizeof(type); \ > > src += sizeof(type); \ > > len -= sizeof(type); \ > > @@ -49,7 +54,8 @@ EXPORT_SYMBOL_GPL(copy_from_kernel_nofault); > > > > #define copy_to_kernel_nofault_loop(dst, src, len, type, err_label) \ > > while (len >= sizeof(type)) { \ > > - __put_kernel_nofault(dst, src, type, err_label); \ > > + __put_kernel_nofault(dst, src, type, err_label); \ > > + instrument_write(dst, sizeof(type)); \ > > dst += sizeof(type); \ > > src += sizeof(type); \ > > len -= sizeof(type); \ > > -- > > 2.34.1 > > > > Reviewed-by: Andrey Konovalov <andreyknvl@xxxxxxxxx> > Tested-by: Andrey Konovalov <andreyknvl@xxxxxxxxx> > > For KASAN parts. Andrew, Please let me know if the last v6 is ready for -mm tree. Previous version was removed here: https://lore.kernel.org/mm-commits/20241010214955.DBEB7C4CEC5@xxxxxxxxxxxxxxx/ Hopefully, they won't conflict in mm/kasan/kasan_test_c.c per another patch: https://lore.kernel.org/linux-mm/20241014025701.3096253-3-snovitoll@xxxxxxxxx/ Thanks > > Thank you!
