http://bugs.winehq.org/show_bug.cgi?id=21443
Summary: CryptSignHash() fails when CryptCreateHash() ALG_ID is set to CALG_SSL3_SHAMD5 Product: Wine Version: 1.1.36 Platform: x86-64 URL: http://www.secureneterm.com OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: support@securenetterm.com
Created an attachment (id=25823) --> (http://bugs.winehq.org/attachment.cgi?id=25823) Problem discussion on how to duplicate.
CryptSignHash() fails when CryptCreateHash() ALG_ID is set to CALG_SSL3_SHAMD5
This can be duplicated using the latest version of OpenSSL as outlined in the attachment. Under Wine, the test outlined in the attachment is:
CONNECTED(00000040) depth=0 CN = InterSoft International Inc. verify error:num=18:self signed certificate verify return:1 depth=0 CN = InterSoft International Inc. verify return:1 43:error:8907006F:lib(137):CAPI_RSA_SIGN:error signing hash:.\engines\e_capi.c:844:Error code= 0x80090008 43:error:14099004:SSL routines:SSL3_SEND_CLIENT_VERIFY:RSA lib:.\ssl\s3_clnt.c:2707:
Note that this error does not happen under Microsoft Windows.
http://bugs.winehq.org/show_bug.cgi?id=21443
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |advapi32
http://bugs.winehq.org/show_bug.cgi?id=21443
Juan Lang juan_lang@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|advapi32 |rsaenh
--- Comment #1 from Juan Lang juan_lang@yahoo.com 2010-01-21 13:18:40 --- This bug is likely to be in rsaenh, as advapi32 just forwards the call to the provider.
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #2 from Kenneth Robinette support@securenetterm.com 2010-01-30 09:32:29 --- (In reply to comment #1)
This bug is likely to be in rsaenh, as advapi32 just forwards the call to the provider.
Yes, the bug is in rsaenh.c in the RSAENH_CPHashData() function at line 3959:
if (!get_algid_info(hProv, pCryptHash->aiAlgid) || pCryptHash->aiAlgid == CALG_SSL3_SHAMD5) { SetLastError(NTE_BAD_ALGID); return FALSE; }
This check forces an error, but why? There is no comment or documentation on why the Wine implementation does not support CALG_SSL3_SHAMD5.
Again, it can be confirmed that the Microsoft CryptSighHash function supports this.
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #3 from Juan Lang juan_lang@yahoo.com 2010-03-11 15:21:26 --- That check was present in the initial commit of rsaenh: http://source.winehq.org/git/wine.git/?a=blob;f=dlls/rsaenh/rsaenh.c;hb=64dc...
As you say, no comment or test to justify it.
What happens if you remove the check from rsaenh.c?
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #4 from Juan Lang juan_lang@yahoo.com 2010-03-11 15:50:26 --- Actually, I'm not sure that the line you highlighted is in error. It needs tests to determine the correct behavior. As Microsoft states in MSDN at "Creating a CALG_SSL3_SHAMD5 Hash" [1],
3. Get a handle to a hash object by calling CryptCreateHash with CALG_SSL3_SHAMD5 passed in the Algid parameter. 4. Set the hash value with a call to CryptSetHashParam. The concatenated hash values are passed as a BYTE* in the pbData parameter, and the HP_HASHVAL value must be passed in the dwParam parameter. Calling CryptHashData using the handle returned by CryptCreateHash in step 3 will fail.
This is indeed what the OpenSSL source does: it calls CryptCreateHash, CryptSetHashParam, and CryptSignHash, but not CryptHashData. Internally, Wine's CPSignHash doesn't call CPHashData, either.
I think the more likely source of the problem is in build_hash_signature: http://source.winehq.org/git/wine.git/?a=blob;f=dlls/rsaenh/rsaenh.c;h=665bc...
aOIDDescriptor has implementations for CALG_MD2, CALG_MD4, CALG_MD5, and CALG_SHA, but not for CALG_SSL3_SHAMD5. Likewise, encrypt_block_impl doesn't have an implementation for CALG_SSL3_SHAMD5.
[1] Assuming MSDN doesn't go breaking its links again, http://msdn.microsoft.com/en-us/library/aa379865(VS.85).aspx
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #5 from Kenneth Robinette support@securenetterm.com 2010-03-11 19:47:32 --- (In reply to comment #4)
Actually, I'm not sure that the line you highlighted is in error. It needs tests to determine the correct behavior. As Microsoft states in MSDN at "Creating a CALG_SSL3_SHAMD5 Hash" [1],
- Get a handle to a hash object by calling CryptCreateHash with
CALG_SSL3_SHAMD5 passed in the Algid parameter. 4. Set the hash value with a call to CryptSetHashParam. The concatenated hash values are passed as a BYTE* in the pbData parameter, and the HP_HASHVAL value must be passed in the dwParam parameter. Calling CryptHashData using the handle returned by CryptCreateHash in step 3 will fail.
This is indeed what the OpenSSL source does: it calls CryptCreateHash, CryptSetHashParam, and CryptSignHash, but not CryptHashData. Internally, Wine's CPSignHash doesn't call CPHashData, either.
I think the more likely source of the problem is in build_hash_signature: http://source.winehq.org/git/wine.git/?a=blob;f=dlls/rsaenh/rsaenh.c;h=665bc...
aOIDDescriptor has implementations for CALG_MD2, CALG_MD4, CALG_MD5, and CALG_SHA, but not for CALG_SSL3_SHAMD5. Likewise, encrypt_block_impl doesn't have an implementation for CALG_SSL3_SHAMD5.
[1] Assuming MSDN doesn't go breaking its links again, http://msdn.microsoft.com/en-us/library/aa379865(VS.85).aspx
Yea, I went down the wrong path trying to trace the source of the problem. The code I have that is failing does not call CryptHashData(). However, you can duplicate the problem using the OpenSSL test case I documented in my attachment.
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #6 from Kenneth Robinette support@securenetterm.com 2010-05-20 11:31:18 --- (In reply to comment #5)
(In reply to comment #4)
Actually, I'm not sure that the line you highlighted is in error. It needs tests to determine the correct behavior. As Microsoft states in MSDN at "Creating a CALG_SSL3_SHAMD5 Hash" [1],
- Get a handle to a hash object by calling CryptCreateHash with
CALG_SSL3_SHAMD5 passed in the Algid parameter. 4. Set the hash value with a call to CryptSetHashParam. The concatenated hash values are passed as a BYTE* in the pbData parameter, and the HP_HASHVAL value must be passed in the dwParam parameter. Calling CryptHashData using the handle returned by CryptCreateHash in step 3 will fail.
This is indeed what the OpenSSL source does: it calls CryptCreateHash, CryptSetHashParam, and CryptSignHash, but not CryptHashData. Internally, Wine's CPSignHash doesn't call CPHashData, either.
I think the more likely source of the problem is in build_hash_signature: http://source.winehq.org/git/wine.git/?a=blob;f=dlls/rsaenh/rsaenh.c;h=665bc...
aOIDDescriptor has implementations for CALG_MD2, CALG_MD4, CALG_MD5, and CALG_SHA, but not for CALG_SSL3_SHAMD5. Likewise, encrypt_block_impl doesn't have an implementation for CALG_SSL3_SHAMD5.
[1] Assuming MSDN doesn't go breaking its links again, http://msdn.microsoft.com/en-us/library/aa379865(VS.85).aspx
Yea, I went down the wrong path trying to trace the source of the problem. The code I have that is failing does not call CryptHashData(). However, you can duplicate the problem using the OpenSSL test case I documented in my attachment.
Futher testing confirms that the problem does happen in the build_hash_signature() called from RSAENH_CPSignHash(). The following is a partial trace showing trace data displayed from within the RSAENH_CPSighHash function.
trace:crypt:RSAENH_CPSignHash (hProv=00000002, hHash=00000004, dwKeySpec=00000001, sDescription=(null), dwFlags=00000000, pbSignature=0x12ec478, pdwSigLen=0x33e844) trace:crypt:RSAENH_CPGetUserKey (hProv=00000002, dwKeySpec=00000001, phUserKey=0x33e764) trace:crypt:RSAENH_CPGetHashParam (hProv=00000002, hHash=00000004, dwParam=00000001, pbData=0x33e758, pdwDataLen=0x33e75c, dwFlags=00000000) trace:crypt:RSAENH_CPGetHashParam (hProv=00000002, hHash=00000004, dwParam=00000002, pbData=0x33e6f0, pdwDataLen=0x33e75c, dwFlags=00000000) trace:crypt:RSAENH_CPDestroyKey (hProv=00000002, hKey=00000006)
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #7 from Juan Lang juan_lang@yahoo.com 2010-05-20 11:57:48 --- Created an attachment (id=28127) --> (http://bugs.winehq.org/attachment.cgi?id=28127) Patch: dummy implementation
Out of curiosity, how does this work for you? I haven't tested it at all, it's just a quick check.
http://bugs.winehq.org/show_bug.cgi?id=21443
--- Comment #8 from Kenneth Robinette support@securenetterm.com 2010-05-20 13:33:40 --- (In reply to comment #7)
Created an attachment (id=28127)
--> (http://bugs.winehq.org/attachment.cgi?id=28127) [details]
Patch: dummy implementation Out of curiosity, how does this work for you? I haven't tested it at all, it's just a quick check.
That fixed the problem!
http://bugs.winehq.org/show_bug.cgi?id=21443
Juan Lang juan_lang@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |patch
--- Comment #9 from Juan Lang juan_lang@yahoo.com 2010-05-20 14:20:59 --- (In reply to comment #8)
That fixed the problem!
Thanks! I'll try to write some tests so this can go in.
http://bugs.winehq.org/show_bug.cgi?id=21443
Juan Lang juan_lang@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |FIXED
--- Comment #10 from Juan Lang juan_lang@yahoo.com 2010-05-21 12:18:37 --- Fixed by commit 62d806601bfa992f2e36ff4d208ea9781218bcd6.
http://bugs.winehq.org/show_bug.cgi?id=21443
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #11 from Alexandre Julliard julliard@winehq.org 2010-05-21 14:39:37 --- Closing bugs fixed in 1.2-rc1.