summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2004-07-28Nuke option JEFF_DEBUG for good.Miod Vallat
(This code relies upon undefined C behaviour. Every day in ddb is your lucky day...)
2004-07-28Remove EH_DEBUG code.Miod Vallat
2004-07-28Keep exception specific defines in the exception handling code, instead ofMiod Vallat
polluting the kernel namespace.
2004-07-28Only put a space in the client buffer if the request has a non-NULLJean-Francois Brousseau
argument. This is only for prettying and makes us closer to the GNU behaviour
2004-07-28more s/illegal/invalid/Markus Friedl
2004-07-28add -ttttt (timestamp difference since the first packet); ok thierryMarkus Friedl
2004-07-28call setsid() _before_ re-execMarkus Friedl
2004-07-28Use less stack space for temporary storage and C code invocation, afterMiod Vallat
the ABI change.
2004-07-28Update stack frame processing after ABI change.Miod Vallat
2004-07-28In the never-ending saga of getting gcc to work reliably on m88k, today'sMiod Vallat
changes address incorrect stack usage, when optimization needs more nameless temporary values than available registers, and has to save them on stack. In some (rare) circumstances, it will compute a stack address _outside_ the current function local storage space, overwriting the caller's stack. Most of the time, this only affects the "outgoing argument area", which is harmless if it has not been populated; this explains why it has not been noticed earlier. Since I see no easy way to fix this, I decided to go the simpler way of removing this ougoing argument area. This not only reduces stack usage, but also makes varargs/stdarg code smaller and faster; also functions which get their first few arguments in registers, then some on the stack, then some in registers again, will not allocate stack space for the second set of arguments passed through registers. This is an ABI change, we are no longer 88Open compliant (have we ever been?).
2004-07-28fix path to sendmail's README;Jason McIntyre
from Jonathan Gray (PR #3870); ok millert@
2004-07-28remove dead Xr to uucp(1); from misc@;Jason McIntyre
ok millert@
2004-07-28Xetc causes the number of sets to change, better reflect it in the text...Miod Vallat
2004-07-28Remove a debugging messageJean-Francois Brousseau
2004-07-28Fix $OpenBSD$ tagJean-Francois Brousseau
2004-07-28Pass the appropriate argument to cvs_client_connect()Jean-Francois Brousseau
2004-07-28Add a small cache for parsed CVSROOT strings and return a pointer to aJean-Francois Brousseau
previously cached entry when we encounter the same string more than once.
2004-07-28Pass the appropriate arguments to cvs_client_connect() andJean-Francois Brousseau
cvs_client_disconnect()
2004-07-28* prepare for an overhaul of the connection code by adding a cvsrootJean-Francois Brousseau
parameter to the cvs_client_connect() and cvs_client_disconnect() functions, so we rely less and less on a global variable
2004-07-28Add a newline at the end of the strings put in CVS/Root andJean-Francois Brousseau
CVS/Repository, like GNU CVS does
2004-07-28Fix the timestamp parsing for Mod-time. Months are now being detectedJean-Francois Brousseau
correctly and there is no assumption about Daylight Savings Time
2004-07-27art.html -> art1.html, PR3872 From: ak@ebi.ac.ukHenning Brauer
2004-07-27add support for {} expansion for prefix in the filter rules, claudio okHenning Brauer
2004-07-27Don't override PREFERRED_DEBUGGING_TYPE. This makes DWARF2 the defaultMark Kettenis
again; stabs doesn't work for 64-bit code. ok miod@, espie@
2004-07-27the `.#*' pattern actually still matches, include it in standardJean-Francois Brousseau
ignore patterns
2004-07-27Fix directory loading with the new file structure allocation codeJean-Francois Brousseau
2004-07-27Be less retarded when parsing the contents of a CVS/Root fileJean-Francois Brousseau
2004-07-27Comment out unused code and some minor cleanup. Add the prototype andJean-Francois Brousseau
skeleton for cvs_printf(), which will be used to output strings that do not need all of the handling that cvs_log() performs. This will be used among other things to prepend the `M' and `E' family of responses to output made by the `server' command.
2004-07-27* use a mix of mktime() and ctime_r() instead of asctime_r() to calculateJean-Francois Brousseau
the Mod-time timestamp. This way, mktime() fills in information about the week day and day of year which was not calculated with the previous method * calculate the GMT offset This fixes a bug where each CVS entry in the Entries files would have `Sun' (for Sunday) as the day of the week.
2004-07-27do not define the prefix, prefixlen and community structs used forHenning Brauer
the filters from within struct filter_match but explicit; no functional change
2004-07-27* parse the timestamp given to Mod-time correctly and store it so theJean-Francois Brousseau
next operation on a file can use it (actually, we need to fix the problem with weekdays) * when getting a sticky command and the directory doesn't exist, create it and populate it with CVS admin files
2004-07-27macro fixesJared Yanovich
punctuation use 'Bl -diag' for diagnostic lists ok jmc
2004-07-27suport macro expansion for peer spec in filter rulesHenning Brauer
things like deny from { $peer1 $peer2 } prefix 192.168.0.0/16 are now possible.
2004-07-27* when opening an Entries file for writing, open the file in append modeJean-Francois Brousseau
* when parsing the entries, break if we encounter the last line 'D' * when adding the line to the file, generate it from the fields instead of taking the line
2004-07-27* don't always return 0 in cvs_client_sendreq()Jean-Francois Brousseau
2004-07-27Add per-device evcount interrupt counters; also use LIST_xxx macros in isr.cMiod Vallat
ok aaron@, tested aoyama@
2004-07-27Zero out the cvs root structure before assigning the fields, otherwiseJean-Francois Brousseau
we can end up with pretty strange results
2004-07-27Factorize exception handling code. The only different part between luna88kMiod Vallat
and mvme88k is the retrieval of the CMMU fault registers. Tested on mvme88k by myself and luna88k by aoyama@
2004-07-27* create cvs_file_alloc() to keep one copy of the file allocation codeJean-Francois Brousseau
and reuse it * add cvs_file_create() to explicitly create a file or directory * add cvs_mkadmin() to fill the administrative files for a CVS directory
2004-07-26* make checkout work in remote mode (almost)Jean-Francois Brousseau
2004-07-26Fix headerJean-Francois Brousseau
2004-07-26* fix two offset bugs in MT stack handlingJean-Francois Brousseau
* create the directory if it does not exist when getting a `Clear-sticky' or `Set-sticky' * fix the cvs_client_getln() that was missing in cvs_resp_statdir() * when receiving a `Created', read the proper lines, then generate a valid pathname to pass to cvs_recvfile() * don't try to compare checksums if the checksum pointer is NULL * in cvs_recvfile(), open the file with the proper flags
2004-07-26Fix pointer passed to fclose() in cvs_ent_close()Jean-Francois Brousseau
2004-07-26Typos in comments.Miod Vallat
2004-07-26document behaviour of mg +-number fileMichael Coulter
ok vincent@ better wording from jmc@
2004-07-26Remove the code that would avoid sending a `Directory' request if theJean-Francois Brousseau
previous `Directory' request had sent the same dir, our code isn't as stupid as it used to be
2004-07-26* start working on the MT supportJean-Francois Brousseau
* add dummy handlers for `Set-sticky', `Clear-sticky' and `Module-expansion'
2004-07-26* when loading a file, cvs_file_get() can now tell whether the fileJean-Francois Brousseau
is up-to-date or modified * add cvs_file_examine() to run a command on a file and its subfiles
2004-07-26Add the handler for the checkout commandJean-Francois Brousseau
2004-07-26Start explaining what the commands do and tweak the mdoc to have aJean-Francois Brousseau
prettier description layout
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-23 19:52:26 +0000