December 2016 - Wine-devel - list.winehq.org

Re: [PATCH 7/7] bcrypt: Mostly implement AES decryption
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27082 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 6/7] bcrypt: Implement AES encryption
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27081 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 5/7] bcrypt: Beginnings of the AES provider
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27080 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 1/7] bcrypt: Implement BCryptGenerateSymmetricKey function
by Bruno Jesus 02 Dec '16

02 Dec '16

OOps, sorry for that. I pasted the wrong command in the terminal. This is the real deal but was only supposed to be sent if the previous 5 test patches were commited. On Fri, Dec 2, 2016 at 8:28 PM, Bruno Jesus <00cpxxx(a)gmail.com> wrote: > From: Hans Leidekker <hans(a)codeweavers.com> > Signed-off-by: Bruno Jesus <00cpxxx(a)gmail.com> > --- > dlls/bcrypt/bcrypt_main.c | 79 +++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 79 insertions(+) > > diff --git a/dlls/bcrypt/bcrypt_main.c b/dlls/bcrypt/bcrypt_main.c > index 6023c94..50c05c0 100644 > --- a/dlls/bcrypt/bcrypt_main.c > +++ b/dlls/bcrypt/bcrypt_main.c > @@ -138,6 +138,8 @@ NTSTATUS WINAPI BCryptEnumAlgorithms(ULONG dwAlgOperations, ULONG *pAlgCount, > > #define MAGIC_ALG (('A' << 24) | ('L' << 16) | ('G' << 8) | '0') > #define MAGIC_HASH (('H' << 24) | ('A' << 16) | ('S' << 8) | 'H') > +#define MAGIC_KEY (('K' << 24) | ('E' << 16) | ('Y' << 8) | '0') > + > struct object > { > ULONG magic; > @@ -172,6 +174,8 @@ struct algorithm > BOOL hmac; > }; > > +static ULONG get_block_size( enum alg_id alg ); > + > NTSTATUS WINAPI BCryptGenRandom(BCRYPT_ALG_HANDLE handle, UCHAR *buffer, ULONG count, ULONG flags) > { > const DWORD supported_flags = BCRYPT_USE_SYSTEM_PREFERRED_RNG; > @@ -541,6 +545,37 @@ static NTSTATUS hmac_finish( struct hash *hash, UCHAR *output, ULONG size ) > pgnutls_hmac_deinit( hash->u.hmac_handle, output ); > return STATUS_SUCCESS; > } > + > +struct key > +{ > + struct object hdr; > + enum alg_id alg_id; > + ULONG block_size; > + gnutls_cipher_hd_t handle; > + UCHAR *secret; > + ULONG secret_len; > +}; > + > +static NTSTATUS key_init( struct key *key, enum alg_id id, UCHAR *secret, ULONG secret_len ) > +{ > + if (!libgnutls_handle) return STATUS_INTERNAL_ERROR; > + > + switch (id) > + { > + default: > + FIXME( "algorithm %u not supported\n", id ); > + return STATUS_NOT_SUPPORTED; > + } > + > + if (!(key->block_size = get_block_size( id ))) return STATUS_INVALID_PARAMETER; > + > + key->alg_id = id; > + key->handle = 0; /* initialized on first use */ > + key->secret = secret; > + key->secret_len = secret_len; > + > + return STATUS_SUCCESS; > +} > #else > struct hash > { > @@ -584,6 +619,17 @@ static NTSTATUS hmac_finish( struct hash *hash, UCHAR *output, ULONG size ) > ERR( "support for hashes not available at build time\n" ); > return STATUS_NOT_IMPLEMENTED; > } > + > +struct key > +{ > + struct object hdr; > +}; > + > +static NTSTATUS key_init( struct key *key, enum alg_id id, const UCHAR *secret, ULONG secret_len ) > +{ > + ERR( "support for keys not available at build time\n" ); > + return STATUS_NOT_IMPLEMENTED; > +} > #endif > > #define OBJECT_LENGTH_MD5 274 > @@ -694,6 +740,13 @@ static NTSTATUS get_alg_property( enum alg_id id, const WCHAR *prop, UCHAR *buf, > return STATUS_SUCCESS; > } > > +static ULONG get_block_size( enum alg_id alg ) > +{ > + ULONG ret = 0, size = sizeof(ret); > + get_alg_property( alg, BCRYPT_BLOCK_LENGTH, (UCHAR *)&ret, sizeof(ret), &size ); > + return ret; > +} > + > static NTSTATUS get_hash_property( enum alg_id id, const WCHAR *prop, UCHAR *buf, ULONG size, ULONG *ret_size ) > { > NTSTATUS status; > @@ -854,6 +907,32 @@ NTSTATUS WINAPI BCryptHash( BCRYPT_ALG_HANDLE algorithm, UCHAR *secret, ULONG se > return BCryptDestroyHash( handle ); > } > > +NTSTATUS WINAPI BCryptGenerateSymmetricKey( BCRYPT_ALG_HANDLE algorithm, BCRYPT_KEY_HANDLE *handle, > + UCHAR *object, ULONG object_len, UCHAR *secret, ULONG secret_len, > + ULONG flags ) > +{ > + struct algorithm *alg = algorithm; > + struct key *key; > + NTSTATUS status; > + > + TRACE( "%p, %p, %p, %u, %p, %u, %08x\n", alg, handle, object, object_len, secret, secret_len, flags ); > + > + if (!alg || alg->hdr.magic != MAGIC_ALG) return STATUS_INVALID_HANDLE; > + if (object) FIXME( "ignoring object buffer\n" ); > + > + if (!(key = HeapAlloc( GetProcessHeap(), 0, sizeof(*key) ))) return STATUS_NO_MEMORY; > + key->hdr.magic = MAGIC_KEY; > + > + if ((status = key_init( key, alg->id, secret, secret_len ))) > + { > + HeapFree( GetProcessHeap(), 0, key ); > + return status; > + } > + > + *handle = key; > + return STATUS_SUCCESS; > +} > + > BOOL WINAPI DllMain( HINSTANCE hinst, DWORD reason, LPVOID reserved ) > { > switch (reason) > -- > 2.9.3 >

1 0

Re: [PATCH v3] winemapi: Directly use xdg-email if available, enabling file attachments.
by André Hentschel 02 Dec '16

02 Dec '16

Hi, comments inline Am 22.11.2016 um 20:46 schrieb Jeremy White: > -ULONG WINAPI MAPISendMail(LHANDLE session, ULONG_PTR uiparam, > +static ULONG BrowserSendMail(LHANDLE session, ULONG_PTR uiparam, > lpMapiMessage message, FLAGS flags, ULONG reserved) > { > ULONG ret = MAPI_E_FAILURE; > @@ -291,6 +370,129 @@ exit: > return ret; > } Do you need all those parameters? > +/************************************************************************** > + * XDGSendMail > + * > + * Send a message using xdg-email mail client. > + * > + */ > +ULONG XDGSendMail(LHANDLE session, ULONG_PTR uiparam, > + lpMapiMessage message, FLAGS flags, ULONG reserved) Same here Also should be a static function > + > + TRACE("(0x%08lx 0x%08lx %p 0x%08x 0x%08x)\n", session, uiparam, message, flags, reserved); I'd keep the TRACE in the WINAPI function > + > + if (!message) > + return MAPI_E_FAILURE; maybe this check, too > + max_args = 1 + (2 + message->nRecipCount + message->nFileCount) * 2; > + argv = HeapAlloc(GetProcessHeap(), 0, (max_args + 1) * sizeof(*argv)); > + if (!argv) > + return MAPI_E_INSUFFICIENT_MEMORY; > + > + ret = add_argument(argv, &argc, "xdg-email", NULL); > + if (ret != SUCCESS_SUCCESS) > + goto exit; > + > + if (message->lpOriginator) > + TRACE("From: %s (unused)\n", debugstr_a(message->lpOriginator->lpszAddress)); Why is the TRACE message different from the original function? > + > + for (i = 0; i < message->nRecipCount; i++) > + { > + if (!message->lpRecips) > + { > + WARN("Recipient %d missing\n", i); > + ret = MAPI_E_FAILURE; > + goto exit; > + } > + > + if (message->lpRecips[i].lpszAddress) > + { > + ret = add_target(argv, &argc, message->lpRecips[i].ulRecipClass, add_target is only used in one place, no need for a helper function IMO > + message->lpRecips[i].lpszAddress); > + if (ret != SUCCESS_SUCCESS) > + goto exit; > + } > + else > + FIXME("Name resolution and entry identifiers not supported\n"); > + } > + > + for (i = 0; i < message->nFileCount; i++) > + { > + TRACE("File Path: %s, name %s\n", debugstr_a(message->lpFiles[i].lpszPathName), > + debugstr_a(message->lpFiles[i].lpszFileName)); > + ret = add_file(argv, &argc, message->lpFiles[i].lpszPathName, message->lpFiles[i].lpszFileName); Same for add_file Hope that helps

2 1

Re: [PATCH 5/5] bcrypt/tests: Add BCryptEncrypt/Decrypt tests
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27070 Your paranoid android. === wvistau64 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 598 bcrypt.c:848: Test failed: got 598 === w2008s64 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 598 bcrypt.c:848: Test failed: got 598 === w7pro64 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 622 bcrypt.c:848: Test failed: got 622 === w864 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 654 bcrypt.c:848: Test failed: got 654 === w1064 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 654 bcrypt.c:848: Test failed: got 654

2 2

Re: [PATCH 4/5] bcrypt/tests: Add BCryptGenerateSymmetricKey tests
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27069 Your paranoid android. === wvistau64 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 598 bcrypt.c:848: Test failed: got 598 === w2008s64 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 598 bcrypt.c:848: Test failed: got 598 === w7pro64 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 622 bcrypt.c:848: Test failed: got 622 === w864 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 654 bcrypt.c:848: Test failed: got 654 === w1064 (64 bit bcrypt) === bcrypt.c:792: Test failed: got 654 bcrypt.c:848: Test failed: got 654

1 0

Re: [PATCH 3/5] bcrypt/tests: Add AES provider tests
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27068 Your paranoid android. === wvistau64 (64 bit bcrypt) === bcrypt.c:783: Test failed: got 598 === w2008s64 (64 bit bcrypt) === bcrypt.c:783: Test failed: got 598 === w7pro64 (64 bit bcrypt) === bcrypt.c:783: Test failed: got 622 === w864 (64 bit bcrypt) === bcrypt.c:783: Test failed: got 654 === w1064 (64 bit bcrypt) === bcrypt.c:783: Test failed: got 654

1 0

Re: [PATCH 1/3] server: Introduced iosb struct for server-side IO_STATUS_BLOCK representation and use it in irp_call.
by Alexandre Julliard 02 Dec '16

02 Dec '16

Jacek Caban <jacek(a)codeweavers.com> writes: > +/* release iosb reference */ > +void release_iosb( struct iosb *iosb ) > +{ > + if (--iosb->refcount) return; > + > + free( iosb->in_data ); > + free( iosb->out_data ); > + free( iosb ); > +} I think you should use a proper server object instead. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: server: Signal pipe server fd in create_pipe_server.
by Marvin 02 Dec '16

02 Dec '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=27047 Your paranoid android. === w2008s64 (32 bit pipe) === Unable to set the VM system time: the "nc -q0 '10.42.42.136' '4242'" command returned 1 (settime/connect). Maybe the TestAgentd process is missing the required privileges. The test VM is powered off! Did the test shut it down?

1 0