Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2009-05-27 | Add cli_posix_readlink() and a torture test for it. | Jeremy Allison | 1 | -1/+183 | |
Jeremy. | |||||
2009-05-27 | Add aync POSIX hardlink and symlink and torture test for them. | Jeremy Allison | 1 | -151/+300 | |
Missing call cli_readlink() is next. Jeremy. | |||||
2009-05-26 | Introduce "struct stat_ex" as a replacement for SMB_STRUCT_STAT | Volker Lendecke | 3 | -23/+23 | |
This patch introduces struct stat_ex { dev_t st_ex_dev; ino_t st_ex_ino; mode_t st_ex_mode; nlink_t st_ex_nlink; uid_t st_ex_uid; gid_t st_ex_gid; dev_t st_ex_rdev; off_t st_ex_size; struct timespec st_ex_atime; struct timespec st_ex_mtime; struct timespec st_ex_ctime; struct timespec st_ex_btime; /* birthtime */ blksize_t st_ex_blksize; blkcnt_t st_ex_blocks; }; typedef struct stat_ex SMB_STRUCT_STAT; It is really large because due to the friendly libc headers playing macro tricks with fields like st_ino, so I renamed them to st_ex_xxx. Why this change? To support birthtime, we already have quite a few #ifdef's at places where it does not really belong. With a stat struct that we control, we can consolidate the nanosecond timestamps and the birthtime deep in the VFS stat calls. At this moment it is triggered by a request to support the birthtime field for GPFS. GPFS does not extend the system level struct stat, but instead has a separate call that gets us the additional information beyond posix. Without being able to do that within the VFS stat calls, that support would have to be scattered around the main smbd code. It will very likely break all the onefs modules, but I think the changes will be reasonably easy to do. | |||||
2009-05-24 | Add "err_on_readability" to writev_send | Volker Lendecke | 1 | -2/+2 | |
A socket where the other side has closed only becomes readable. To catch errors early when sitting in a pure writev, we need to also test for readability. | |||||
2009-05-20 | Make cli_posix_open() and cli_posix_mkdir() async. | Jeremy Allison | 1 | -56/+217 | |
Jeremy. | |||||
2009-05-18 | Fix broken pipe handling | Volker Lendecke | 1 | -2/+2 | |
Metze is right: If we have *any* error at the socket level, we just can not continue. Also, apply some defensive programming: With this async stuff someone else might already have closed the socket. | |||||
2009-05-19 | s3: tevent_req_poll() loops forever when pipe is broken | Bo Yang | 1 | -0/+8 | |
Signed-off-by: Bo Yang <boyang@samba.org> | |||||
2009-05-13 | What to do about debugging in a multi-threaded application? | Derrell Lipman | 1 | -7/+10 | |
- For now, punt. Any thread that requests debug to stderr will establish that for all threads. Derrell | |||||
2009-05-13 | Thread-safe protection: libsmbclient initialization | Derrell Lipman | 1 | -94/+133 | |
- Begin converting init functions to use SMB_THREAD_ONCE. libsmbclient module-wide initialization is now moved into a separate function and called via SMB_THREAD_ONCE. - libsmbclient counts users (contexts) so that it can release global resources when the last context is closed. That count of contexts is now protected by a mutex. Derrell | |||||
2009-05-13 | s3:libsmb: move read_smb_send/recv() static in async_smb.c | Stefan Metzmacher | 1 | -0/+87 | |
metze | |||||
2009-05-13 | s3:libsmb: let cli_smb_chain_send() also return NTSTATUS | Stefan Metzmacher | 1 | -9/+12 | |
metze | |||||
2009-05-13 | s3:libsmb: return NT_STATUS_CONNECTION_INVALID if the fd is -1 | Bo Yang | 1 | -5/+7 | |
This way we can destinguish between requests which failed because the connection broke after they were triggered and the requests which are started on an already broken connection. This also moves the check to cli_smb_req_iov_send() where it really belongs. metze | |||||
2009-05-13 | s3: return proper error code in cli_smb_req_send | Bo Yang | 5 | -38/+82 | |
Signed-off-by: Bo Yang <boyang@samba.org> | |||||
2009-05-12 | Clean up assignments to iov_base, ensure it's always cast to void *. This ↵ | Jeremy Allison | 6 | -18/+18 | |
should quieten some warnings with picky compilers on the buildfarm. Jeremy. | |||||
2009-05-10 | Provide a libsmbclient interface for programs requiring threads | Derrell Lipman | 2 | -0/+176 | |
- This adds two functions: smbc_thread_posix() which provides access to the internal threading implementation using pthread; and smbc_thread_impl() where the user provides each of the functions required by Samba, to give access to the thread implementation's native capabilities. Derrell | |||||
2009-05-09 | s3:libsmb: fix layering of cli_ntrename_internal and its callers | Stefan Metzmacher | 1 | -29/+35 | |
It's easier to have cli_ntrename_internal as a semetric async tevent_req function. cli_ntrename() and cli_nt_hardlink() should be callers on top of cli_ntrename_internal(). metze | |||||
2009-05-07 | Async API needs all parameters to be kept around until sent, | Jeremy Allison | 1 | -55/+15 | |
ensure they're attached to the state structure. Thanks to Metze for pointing this out. Jeremy. | |||||
2009-05-07 | Make cli_tcon_andx chainable | Volker Lendecke | 1 | -15/+42 | |
2009-05-07 | Make cli_session_setup_guest chainable | Volker Lendecke | 1 | -9/+31 | |
2009-05-07 | s3-credentials: protect netlogon_creds_server_step() against NULL creds. | Günther Deschner | 1 | -0/+4 | |
Found by SCHANNEL torture tests. Guenther | |||||
2009-05-06 | Make cli_setatr async. | Jeremy Allison | 3 | -2/+129 | |
Jeremy. | |||||
2009-05-06 | Make cli_setattrE async. | Jeremy Allison | 2 | -26/+98 | |
Jeremy. | |||||
2009-05-05 | Make cli_getatr() async. | Jeremy Allison | 2 | -27/+126 | |
Jeremy. | |||||
2009-05-05 | Make cli_getattrE async. | Jeremy Allison | 3 | -33/+132 | |
Jeremy. | |||||
2009-05-04 | Fix the async calls for the posix_unlink and posix_rmdir. | Jeremy Allison | 1 | -13/+49 | |
Jeremy. | |||||
2009-04-30 | Cause cli_close to return an NTSTATUS. | Jeremy Allison | 2 | -81/+4 | |
Jeremy. | |||||
2009-04-30 | Get medieval on our ass about SMB1 file descriptors being 16 bits, not an int. | Jeremy Allison | 10 | -73/+72 | |
Convert all uses of cli_open(), cli_nt_createXXX to NTSTATUS versions. This is smaller than it looks, it just fixes a lot of old code. Next up, ensure all cli_XX functions return NTSTATUS. Jeremy. | |||||
2009-04-29 | Make cli_unlink async. | Jeremy Allison | 2 | -28/+93 | |
Jeremy. | |||||
2009-04-29 | More async calls in libsmb/clifile.c | Jeremy Allison | 1 | -50/+170 | |
Jeremy. | |||||
2009-04-29 | s3: fix crash in winbindd | Bo Yang | 1 | -0/+4 | |
2009-04-28 | Doh ! Don't need "int dummy" when we have state :-). | Jeremy Allison | 1 | -1/+0 | |
Jeremy. | |||||
2009-04-28 | Convert cli_rename to async. | Jeremy Allison | 2 | -25/+108 | |
Jeremy. | |||||
2009-04-28 | Convert cli_posix_unlink() and cli_posix_rmdir() | Jeremy Allison | 1 | -63/+190 | |
to async. First trans calls I've done. Jeremy. | |||||
2009-04-28 | Fix a type-punned warning | Volker Lendecke | 1 | -1/+1 | |
2009-04-27 | [FIX Bug 6235] domain enumeration breaks if master browser has space in name | Derrell Lipman | 1 | -1/+2 | |
Jeremy: please review to ensure this doesn't appear to break anything The function name_status_find() is documented as used for finding a server's name given its IP address. It was, however, looking for the first matching name which could be a group name at times. This fix ensures that group names are skipped when scanning for a matching name. Derrell | |||||
2009-04-24 | find/replace. Change uintX types to uintX_t types to tidy up the code. | Jeremy Allison | 1 | -38/+38 | |
Jeremy. | |||||
2009-04-23 | errormap: Change the WBC_ERR_UNKNOWN_ERROR to map to NT_STATUS_UNSUCCESSFUL | Kai Blin | 1 | -1/+1 | |
This allows us to distinguish on NTSTATUS level if we got a WBC_ERR_UNKNOWN_ERROR or WBC_ERR_NSS_ERROR. | |||||
2009-04-22 | Make dskattr async. | Jeremy Allison | 1 | -38/+102 | |
Jeremy. | |||||
2009-04-22 | Make cli_chkpath async. | Jeremy Allison | 1 | -0/+121 | |
Jeremy | |||||
2009-04-22 | errormap: Add wbcErr to NTSTATUS mappings | Kai Blin | 1 | -0/+41 | |
2009-04-21 | s3:libsmb: always use the tevent_req_nomem() for checking allocation failures | Stefan Metzmacher | 1 | -6/+6 | |
This will also make sure we cleanup 'req' in case of an error. metze | |||||
2009-04-21 | Make rmdir async. | Jeremy Allison | 2 | -59/+92 | |
Jeremy. | |||||
2009-04-21 | Make cli_mkdir async. Change it to return NTSTATUS. | Jeremy Allison | 2 | -2/+111 | |
Jeremy. | |||||
2009-04-21 | s3-secdesc: use SEC_FLAG_MAXIMUM_ALLOWED instead of SEC_RIGHTS_MAXIMUM_ALLOWED. | Günther Deschner | 1 | -1/+1 | |
Guenther | |||||
2009-04-20 | Merge commit 'origin/master' into libcli-auth-merge-without-netlogond | Andrew Bartlett | 1 | -1/+1 | |
2009-04-20 | s3:ntlmssp Remove use of talloc(NULL) in NTLMSSP code | Andrew Bartlett | 2 | -12/+12 | |
2009-04-20 | s3:ntlmssp Fix segfault: msrpc_gen now uses talloc() | Andrew Bartlett | 1 | -46/+8 | |
2009-04-16 | When doing a cli_ulogoff don't invalidate the cnum, invalidate the vuid. | Jeremy Allison | 1 | -1/+1 | |
Jeremy. | |||||
2009-04-16 | Fix crash bug in NTLMSSP caused by msrpc_parse() moving to talloc | Andrew Bartlett | 1 | -4/+2 | |
2009-04-14 | More work to adapt to merged libcli/auth function prototypes | Andrew Bartlett | 4 | -16/+19 | |