January 2013 - Wine-devel - list.winehq.org

[PATCH v2 1/8] locks: make flock_lock_file take is_conflict callback parm
by Pavel Shilovsky 07 Feb '13

07 Feb '13

This parm demetermines how to check if locks have conflicts. This let us use flock_lock_file funtions further to add O_DENY* flags support through flocks. Signed-off-by: Pavel Shilovsky <piastry(a)etersoft.ru> --- fs/locks.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/fs/locks.c b/fs/locks.c index a94e331..9edfec4 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -697,14 +697,20 @@ static int posix_locks_deadlock(struct file_lock *caller_fl, return 0; } -/* Try to create a FLOCK lock on filp. We always insert new FLOCK locks +/* + * Try to create a FLOCK lock on filp. We always insert new FLOCK locks * after any leases, but before any posix locks. * * Note that if called with an FL_EXISTS argument, the caller may determine * whether or not a lock was successfully freed by testing the return * value for -ENOENT. + * + * Take @is_conflict callback that determines how to check if locks have + * conflicts or not. */ -static int flock_lock_file(struct file *filp, struct file_lock *request) +static int +flock_lock_file(struct file *filp, struct file_lock *request, + int (*is_conflict)(struct file_lock *, struct file_lock *)) { struct file_lock *new_fl = NULL; struct file_lock **before; @@ -760,7 +766,7 @@ find_conflict: break; if (IS_LEASE(fl)) continue; - if (!flock_locks_conflict(request, fl)) + if (!is_conflict(request, fl)) continue; error = -EAGAIN; if (!(request->fl_flags & FL_SLEEP)) @@ -1589,7 +1595,7 @@ int flock_lock_file_wait(struct file *filp, struct file_lock *fl) int error; might_sleep(); for (;;) { - error = flock_lock_file(filp, fl); + error = flock_lock_file(filp, fl, flock_locks_conflict); if (error != FILE_LOCK_DEFERRED) break; error = wait_event_interruptible(fl->fl_wait, !fl->fl_next); -- 1.8.1.1

2 20

[PATCH v2 0/8] Add O_DENY* support for VFS and CIFS/NFS
by Pavel Shilovsky 07 Feb '13

07 Feb '13

This patchset adds support of O_DENY* flags for Linux fs layer. These flags can be used by any application that needs share reservations to organize a file access. VFS already has some sort of this capability - now it's done through flock/LOCK_MAND mechanis, but that approach is non-atomic. This patchset build new capabilities on top of the existing one but doesn't bring any changes into the flock call semantic. These flags can be used by NFS (built-in-kernel) and CIFS (Samba) servers and Wine applications through VFS (for local filesystems) or CIFS/NFS modules. This will help when e.g. Samba and NFS server share the same directory for Windows and Linux users or Wine applications use Samba/NFS share to access the same data from different clients. According to the previous discussions the most problematic question is how to prevent situations like DoS attacks where e.g /lib/liba.so file can be open with DENYREAD, or smth like this. That's why one extra flag O_DENYMAND is added. It indicates to underlying layer that an application want to use O_DENY* flags semantic. It allows us not affect native Linux applications (that don't use O_DENYMAND flag) - so, these flags (and the semantic of open syscall that they bring) are used only for those applications that really want it proccessed that way. So, we have four new flags: O_DENYREAD - to prevent other opens with read access, O_DENYWRITE - to prevent other opens with write access, O_DENYDELETE - to prevent delete operations (this flag is not implemented in VFS and NFS part and only suitable for CIFS module), O_DENYMAND - to switch on/off three flags above. The #1 patch bring infrastructure to let us add this capability easier. #2 patch adds flags to fcntl while #3 patch implement VFS part. Patches #4, #5, #6 are related to CIFS-specific changes. #7 and #8 describe NFS and NFSD parts. Also, I created the preliminary patch for Samba that replaces the existing use of flock/LOCK_MAND mechanism with O_DENY* flags: http://git.etersoft.ru/people/piastry/packages/?p=samba.git;a=commitdiff;h=… Pavel Shilovsky (8): locks: make flock_lock_file take is_conflict callback parm fcntl: Introduce new O_DENY* open flags vfs: Add O_DENYREAD/WRITE flags support for open syscall CIFS: Add O_DENY* open flags support CIFS: Use NT_CREATE_ANDX command for forcemand mounts CIFS: Translate SHARING_VIOLATION to -ETXTBSY error code for SMB2 NFSv4: Add O_DENY* open flags support NFSD: Pass share reservations flags to VFS fs/cifs/cifsacl.c | 10 ++-- fs/cifs/cifsglob.h | 12 +++- fs/cifs/cifsproto.h | 9 +-- fs/cifs/cifssmb.c | 47 ++++++++------- fs/cifs/dir.c | 14 +++-- fs/cifs/file.c | 18 ++++-- fs/cifs/inode.c | 11 ++-- fs/cifs/link.c | 10 ++-- fs/cifs/readdir.c | 2 +- fs/cifs/smb1ops.c | 15 ++--- fs/cifs/smb2file.c | 10 ++-- fs/cifs/smb2inode.c | 4 +- fs/cifs/smb2maperror.c | 2 +- fs/cifs/smb2ops.c | 10 ++-- fs/cifs/smb2pdu.c | 6 +- fs/cifs/smb2proto.h | 14 +++-- fs/fcntl.c | 5 +- fs/locks.c | 121 ++++++++++++++++++++++++++++++++++++--- fs/namei.c | 10 +++- fs/nfs/nfs4xdr.c | 24 ++++++-- fs/nfsd/nfs4state.c | 46 ++++++++++++++- include/linux/fs.h | 6 ++ include/uapi/asm-generic/fcntl.h | 14 +++++ 23 files changed, 324 insertions(+), 96 deletions(-) -- 1.8.1.1

3 6

Support for game DRM which overwrite the GS segment selector
by Alessandro Pignotti 03 Feb '13

03 Feb '13

Hi everyone, I'm trying to get a specific game which employs a seemingly custom protection scheme to work. The DRM does various bad things as usual, but a very bad one is manipulating to GS segment selector and setting it to a NULL segment. The GS segment is used by libc though in various ways (stack protection and syscall support, and probably others). I managed to get the activation procedure to go further and further by enclosing each offending syscall using the following 2 macros. #define SAFE_GS_START \ do { \ wine_set_gs(ntdll_get_thread_data()->gs); \ do #define SAFE_GS_END \ while(0); \ } while(0) Still, this method is very cumbersome since system calls happens in many places even outside of ntdll. Fixing the GS is also needed to support sigsetjmp which is used by wine's exception handling. I'd like to ask for feedback about what would be a sane way of supporting this application. A possible solution would be to modify wine's segfault handler to check if the instruction has a GS prefix (0x65 IIRC) and try to execute the instruction again after fixing the GS. Please keep me in CC since I'm not subscribed to the ML. Regards, Alessandro Pignotti

2 3

Re: kernel32/tests : added tests for copyfileex try6
by Marvin 31 Jan '13

31 Jan '13

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 http://testbot.winehq.org/JobDetails.pl?Key=24237 Your paranoid android. === WNT4WSSP6 (32 bit path) === Timeout

1 0

Re: [website] Added Simplified Chinese translation
by Jeremy Newman 31 Jan '13

31 Jan '13

Nice work. I did end up moving a few things around, and had to patch my language detection function, but it worked when I switched my browser over to zn-ch. I fixed the image as well. Also, please set your text editor to output 4 spaces instead of tabs. Thanks! -N On 01/30/2013 09:34 PM, Jactry Zeng wrote: > Hi folks, > this is Simplified Chinese translation for Wine's website, it's my first > patch, thanks Qian Hong, Aron Xu, Jeremy and Hin-Tak for review. > I appreciate for any comment. :-)

2 1

[website] Added simplified chinese
by Jactry 31 Jan '13

31 Jan '13

5 7

Re: kernel32/tests : added tests for copyfileex try5
by Marvin 31 Jan '13

31 Jan '13

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 http://testbot.winehq.org/JobDetails.pl?Key=24228 Your paranoid android. === WNT4WSSP6 (32 bit path) === Timeout === W2KPROSP4 (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === WXPPROSP3 (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === W2K3R2SESP2 (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === WVISTAADM (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === W2K8SE (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === W7PRO (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === W7PROX64 (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === TEST64_W7SP1 (32 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === W7PROX64 (64 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1 === TEST64_W7SP1 (64 bit path) === path.c:1814: Test failed: Expected GetFileAttributesEx to return zero, got 1

1 0

Re: [3/3] msi: Improve parsing of the supported platforms string.
by Marvin 31 Jan '13

31 Jan '13

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 http://testbot.winehq.org/JobDetails.pl?Key=24227 Your paranoid android. === WNT4WSSP6 (32 bit install) === install.c:6051: Test failed: Per-user icon file isn't where it's expected (C:\WINNT\Profiles\Default User\Application Data\Microsoft\Installer\{7DF88A49-996F-4EC8-A022-BF956F9B2CBB}\testicon)

1 0

Re: Makedll.rules.in: Explicitly specify DLL name when building importlib
by Alexandre Julliard 31 Jan '13

31 Jan '13

Jacek Caban <jacek(a)codeweavers.com> writes: > --- > dlls/Makedll.rules.in | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) Why do you need this? -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: fusion: Allow null to be the value of the public key
by Hans Leidekker 31 Jan '13

31 Jan '13

On Thu, 2013-01-31 at 19:35 +1100, Alistair Leslie-Hughes wrote: > diff --git a/dlls/fusion/asmname.c b/dlls/fusion/asmname.c > index 162a76b..2c4b379 100644 > --- a/dlls/fusion/asmname.c > +++ b/dlls/fusion/asmname.c > @@ -561,6 +561,10 @@ static HRESULT parse_pubkey(IAssemblyNameImpl *name, LPCWSTR pubkey) > { > int i; > BYTE val; > + static WCHAR nullstr[] = {'n','u','l','l',0}; This should be const. > + if(lstrcmpiW(pubkey, nullstr) == 0) > + return S_OK; Please add a test to show that it's case insensitive. > if (lstrlenW(pubkey) < CHARS_PER_PUBKEY) > return FUSION_E_INVALID_NAME; > diff --git a/dlls/fusion/tests/asmcache.c b/dlls/fusion/tests/asmcache.c > index e97cfbe..79f8709 100644 > --- a/dlls/fusion/tests/asmcache.c > +++ b/dlls/fusion/tests/asmcache.c > @@ -1016,6 +1016,10 @@ static void test_QueryAssemblyInfo(void) > 'p','u','b','l','i','c','K','e','y','T','o','k','e','n','=', > '2','d','0','3','6','1','7','b','1','c','3','1','e','2','f','5',',', > 'c','u','l','t','u','r','e','=','n','e','u','t','r','a','l',0}; > + static const WCHAR nullpublickey[] = { > + 'm','s','c','o','r','l','i','n',',','v','e','r','s','i','o','n','=','0', > + 'p','u','b','l','i','c','K','e','y','T','o','k','e','n','=','n','u','l', > + 'c','u','l','t','u','r','e','=','n','e','u','t','r','a','l',0}; > > size = MAX_PATH; > hr = pGetCachePath(ASM_CACHE_GAC, asmpath, &size); > @@ -1386,6 +1390,17 @@ static void test_QueryAssemblyInfo(void) > "Assembly path was changed\n"); > ok(info.cchBuf == MAX_PATH, "Expected MAX_PATH, got %d\n", info.cchBuf); > > + /* display name is "mscorlib.dll,version=0.0.0.0,publicKeyToken=null,culture=neutral" */ This comment doesn't agree with the string initialized above. > + INIT_ASM_INFO(); > + lstrcpyW(name, nullpublickey); > + hr = IAssemblyCache_QueryAssemblyInfo(cache, 0, name, &info); > + ok(info.cbAssemblyInfo == sizeof(ASSEMBLY_INFO), > + "Expected sizeof(ASSEMBLY_INFO), got %d\n", info.cbAssemblyInfo); You should check the return value of QueryAssemblyInfo but CreateAssemblyNameObject is actually a better way to test this.

1 0