| Age | Commit message (Collapse) | Author |
|---|
|
in 1.126, found by nick during a test install; ok krw
|
|
|
|
again fits on one page. Suggested by, tweaked and ok deraadt@.
|
|
#if NUMBOOT > 0 section.
|
|
the -r removal. Found by drahn@ when attempting a landisk install.
ok deraadt@ drahn@
|
|
that was contingent on it being set.
|
|
|
|
the '-c' and '-d` commands.
With the removal of 'rflag' (logically making it always false), (a ||
(rflag && cflag + dflag) || b) should be (a || b), not (a || (cflag +
dflag) || b).
This fixes 'disklabel -w -d <disk>' as used in the install scripts.
Reported by Peter Kun on misc@. Thanks!
|
|
"make sense" miod@ ok deraadt@
|
|
functional change. Noticed by lint.
|
|
Problem found by ckuethe@, slightly different fix than the one
proposed by otto@ applied.
|
|
|
|
does an unsigned comparison and read() can return -1. Use '!=' instead
of '<' since read() can't return more than 'sizeof Y'. Not perfect
(that would require a separate test for -1) but a very common usage.
Do the same for a write(), and a couple of read() calls which are
probably ok but why be inconsistant?
This, as the last couple of commits, found by ian@ as a result of
poking around in fsck_msdos() when his new iPod had problems vs
OpenBSD.
Concept reviewed by miod@, beck@, otto@ and ian@.
|
|
in find_bounds() when processing a DOS MBR.
|
|
offsets and sizes of free chunks. 32 bits just won't hack new big
partitions. Simplify and clarify code while here.
|
|
returning the number of elements. Check for terminating empty entry
(i.e. NULL) instead. Code a bit clearer and shorter with fewer
variables. No functional change.
|
|
calloc() the memory.
|
|
static array that we zero before each use. heapsort(3) shouldn't
care.
"should be fine" beck@
|
|
- align usages and their descriptions (text enclosed in brackets);
- blank[], used for padding prior to r1.86, can improve readability
of usage on 80-column displays if NUMBOOT == 2
ok jsing@, krw@
|
|
ok krw@
|
|
reduce meat of editor_add() && editor_modify() to single if statements
checking for != 0 from all get_* functions. No functional change.
|
|
called during -E initialization. Thus it no longer needs the
'resolve' parameter.
|
|
new offset being in a free area.
|
|
in a semi-global. Call it whenever the count is needed rather than
trying to make sure it is called anytime the value might change.
Eliminates a bunch of now unnecessary *freep parameters and calls to
editor_countfree().
|
|
|
|
a partition can be without causing overlap. Use it to eliminate the
need to check overlap in get_size() and to make the (c)hange command
message more meaningful, i.e. mention max size allowed not the amount
of total free space.
|
|
get_offset() when the user enters an invalid value. Just abort the
partition operation and return to the main prompt. Prevents users
getting trapped because they don't know about ^D. Requested by
deraadt@.
Tweak an error message to remove ambiguity about what 'between x and
y' means.
|
|
allow editing the fsize/bsize FFS values for FS_UNUSED partitions.
ok (for at least the first bit) miod@ weingart@ beck@
|
|
editor_modify(). No functional change.
|
|
details of the free space it calculates. Gives an easy way to see all
the chunks of free sectors on the disk.
ok beck@
|
|
|
|
This removes the last need for has_overlap() to touch the free space
counter. So eliminate that parameter.
|
|
|
|
-E, zero intervening partitions as well as the specified partition so
no partitions are inadvertantly activated.
|
|
error.
1) Don't accept a partition offset in used space.
2) Make the default size the space available from the offset
to the end of the free chunk the offset is in.
3) Make (m)odify ask the questions in the same order as
the (a)dd command. i.e. filesystem type after offset/size.
4) Make it more obvious new partitions default into the largest
chunk of free space.
5) Nuke another now pointless function - next_offset().
"Looks reasonable" millert@
|
|
|
|
get_size() anyway. Allows elimination of make_contiguous() since we
no longer allow overlaps to develop.
|
|
than hand rolling another partition walking loop. No functional change.
|
|
work anyway and ignores this value. So just set it to 1 and nuke
get_cpg(), shrinking and simplifying.
ok otto@ millert@
|
|
has been removed.
ok krw
|
|
ending_sector' tests and one superfluous editor_countfree(). Rectify a
few comments.
|
|
fields.
|
|
ending_sector, so no need to check for this condition after calling
get_size().
|
|
to existing FS_BOOT and FS_UNUSED partitions via -E, tighten up error
verbiage.
Feedback from jsing@.
ok millert@.
|
|
another, valid, set while adding a partition. Lost in r1.119 when we
switched to DL_SETP* defines.
|
|
code. Makes the size prompt/helpstring/looping the same for
(a)dd/(m)odify/(c)hange.
|
|
the variable 'c' to 'partno' in editor_delete() so all editor_*
functions are even more consistant. No functional change.
|
|
counter. Call edit_countfree() instead.
Feedback from otto@.
ok millert@ "Go for it" deraadt@
|
|
there is no point in allowing the edit mode commands
(a)add/(n)ame/(m)odify/(d)elete/(c)hange to pretend to fiddle with it.
Some already checked for and avoided 'c', but do it in a consistant
manner in all cases.
ok millert@ otto@ marco@
|
|
The only visible change is allowing only one chance to specify a valid
partition letter before returning to the main edit prompt.
"This looks OK to me" millert@
|
