April 2012 Archives by thread
Starting: Sun Apr 1 00:26:27 CEST 2012
Ending: Mon Apr 30 14:16:58 CEST 2012
Messages: 90
- [stgt] [PATCHes] Updated patches for thin-provisioning support
ronnie sahlberg
- [stgt] [PATCH] spc.c: fix compiler warnings
Arne Redlich
- [stgt] [PATCH] Fix PR OUT with REGISTER AND MOVE service action
Arne Redlich
- [stgt] [PATCH] Rework param_set_val and friends
Arne Redlich
- [stgt] [PATCH] bs_ssc.c: fix format string
FUJITA Tomonori
- [stgt] Fix memset calls in tgtimg.c and iscsi/md5.c
FUJITA Tomonori
- [stgt] isns.c: fix compiler warnings
FUJITA Tomonori
- [stgt] [PATCH] Makefile: use 'rm -f' instead of rmdir
FUJITA Tomonori
- [stgt] Infinite Loop on 1.0.26?
Brad.Goodman at emc.com
- [stgt] [PATCH] fix check of sent len in mtask_recv_send_handler(), report exceptions in concat_write()
nezhinsky at gmail.com
- [stgt] [PATCH] Update LBPPBE automatically
Ronnie Sahlberg
- [stgt] [PATCH 0/3] Implement WRITE_SAME10/16
Ronnie Sahlberg
- [stgt] [PATCH 0/1] VPD BLOCK LIMITS. Update with UNMAP and CAW limits
Ronnie Sahlberg
- [stgt] [PATCH 0/5] WRITE_SAME10/16 and thin-provisioning patches
Ronnie Sahlberg
- [stgt] [PATCH 0/8] few patches prompted by compiler warnings
nezhinsky at gmail.com
- [stgt] [PATCH 1/8] fix error handling in get_bsg_major() return error code if the device file in /sys/class/bsg failed to open; use fscanf to read an integer major number directly (instead of reading and parsing the string), return error and print appropriate messages when scanf fails.
nezhinsky at gmail.com
- [stgt] [PATCH 2/8] If param check fails during iscsi login text processing, answer that key with Reject. According to iscsi spec (5.2) the target can choose between responding bad values with key=Reject, sending an admissible value and closing the connection by returning a login error. Here the Reject option is implemented. Any objections?
nezhinsky at gmail.com
- [stgt] [PATCH 3/8] Store iscsi initiator alias in connection, copy it to iscsi session, show it in session info
nezhinsky at gmail.com
- [stgt] [PATCH 4/8] In sbc.c implementation of synchronize cache, set resid to 0 unconditionally and remove unused len=0, as const zero is used in set resid
nezhinsky at gmail.com
- [stgt] [PATCH 5/8] Improve str_to_int() There is a potential problem when a user-supplied value overflows the accepting variable; to cope with this the string is converted to a 64-bit value first then assigned to the variable, then the macro checks if they represent the same value. Thus the natural type boundaries are automatically guarded. This means that there is no need to use type's range limits (e.g. 0 and USHRT_MAX for unsigned short). Thus the range limit parameters are removed from str_to_int() -- it's enough for the (most frequent) cases when any value fitting a certain integer type is admissible. When the min/max values are relevant they can be passed to one of the new macros str_to_int_gt, str_to_int_ge, str_to_int_lt, str_to_int_range, requiring that the int value should be greater than, greater or equal, less than, or within a range, correspondingly.
nezhinsky at gmail.com
- [stgt] [PATCH 6/8] Fix "set but not used" warning in iser.c: comment out irdsl related lines; irdsl value is not used meanwhile, as we use a shortcut to cover up an initiator glitch
nezhinsky at gmail.com
- [stgt] [PATCH 7/8] If adding Unit Attention sense fails upon target destroy, just print error message and still return TGTADM_SUCCESS
nezhinsky at gmail.com
- [stgt] [PATCH 8/8] when using eventfd, each time the fd becomes readable, the number of io completions can be read; this number is provided as a 64-bit integer, thus enough space (at least 8 bytes) should be given to read(). This patch adds a comment which explicitely states the above requirement. The actual number of completions in bs_aio is limited by a very small value, thus when reading, a 64-bit variable should be used, then it can be casted to unsigned int and processed as such.
nezhinsky at gmail.com
- [stgt] [PATCH 0/8] few patches prompted by compiler warnings
FUJITA Tomonori
- [stgt] Improving iSCSI passthrough tape performance
Phillip Frost
- [stgt] 1.0.27 released
FUJITA Tomonori
Last message date:
Mon Apr 30 14:16:58 CEST 2012
Archived on: Fri May 15 07:03:29 CEST 2015
This archive was generated by
Pipermail 0.09 (Mailman edition).