crypto: fix OOB read in pefile_digest_pe_contents
Subsystems (3)
PATCH 1/1 crypto: fix OOB read in pefile_digest_pe_contents
pefile_digest_pe_contents() computes the trailing-data hash length as pelen - (hashed_bytes + certs_size). A crafted PE can make the addition exceed pelen, causing the unsigned subtraction to underflow to ~4 GiB. This is passed to crypto_shash_update() which reads out of bounds and panics on unmapped vmalloc guard pages.
BUG: unable to handle page fault for address: ffffc900038d8000 Oops: Oops: 0000 [#1] SMP KASAN NOPTI RIP: 0010:sha256_blocks_generic (lib/crypto/sha256.c:152) Call Trace: <TASK> __sha256_update (lib/crypto/sha256.c:208) crypto_sha256_update (crypto/sha256.c:142) verify_pefile_signature (crypto/asymmetric_keys/verify_pefile.c:436) kexec_kernel_verify_pe_sig (kernel/kexec_file.c:151) __do_sys_kexec_file_load (kernel/kexec_file.c:406) do_syscall_64 (arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) </TASK> Kernel panic - not syncing: Fatal exception
Validate that the addition does not overflow and the result does not exceed pelen before the subtraction. Return -ELIBBAD on failure.
--- crypto/asymmetric_keys/verify_pefile.c | 2 ++ 1 file changed, 2 insertions(+)
crypto/asymmetric_keys/verify_pefile.c +2 -0
@@ -305,6 +305,8 @@ static int pefile_digest_pe_contents(const void *pebuf, unsigned int pelen,if (pelen > hashed_bytes) { tmp = hashed_bytes + ctx->certs_size; +if (tmp <= hashed_bytes || pelen < tmp) +return -ELIBBAD ; ret =crypto_shash_update (desc, pebuf + hashed_bytes, pelen - tmp);
Signature · 1 line
2.43.0
Herbert Xu
maintainercrypto/asymmetric_keys/verify_pefile.c static int pefile_digest_pe_contents(const void *pebuf, unsigned int pelen, · line 309
Quoted 11 lines
crypto/asymmetric_keys/verify_pefile.c +2
-0@@ -305,6 +305,8 @@ static int pefile_digest_pe_contents(const void *pebuf, unsigned int pelen,if (pelen > hashed_bytes) { tmp = hashed_bytes + ctx->certs_size; +if (tmp <= hashed_bytes || pelen < tmp) +return -ELIBBAD ;
I know nothing about this but why should pelen == tmp fail?
Cheers,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 7 lines · Herbert Xu
Quoted 7 lines
crypto/asymmetric_keys/verify_pefile.c +0
-0@@ -305,6 +305,8 @@ static int pefile_digest_pe_contents(const void *pebuf, unsigned int pelen,
Quoted 12 lines
crypto/asymmetric_keys/verify_pefile.c static int pefile_digest_pe_contents(const void *pebuf, unsigned int pelen, · line 309
Quoted 4 lines
crypto/asymmetric_keys/verify_pefile.c +2
-0if (pelen > hashed_bytes) { tmp = hashed_bytes + ctx->certs_size; +if (tmp <= hashed_bytes || pelen < tmp) +return -ELIBBAD ;I know nothing about this but why should pelen == tmp fail?
Cheers,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Hi Herbert,
Do you mean this should be pelen <= tmp ?
pelen == tmp means the cert table sits right at EOF with no trailing data in between - that's a legitimate layout.
Weiming Shi
Herbert Xu
maintainerQuoted 5 lines · Weiming Shi
Do you mean this should be
pelen <= tmp?pelen == tmp means the cert table sits right at EOF with no trailing data in between - that's a legitimate layout.
Nevermind, I misread the patch.
Thanks,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt