Round up fractional percentages, as per POSIX.
From nabijaczleweli, OK deraadt@

When system calls indicate an error they return -1, not some arbitrary
value < 0. errno is only updated in this case. Change all (most?)
callers of syscalls to follow this better, and let's see if this strictness
helps us in the future.

Another %qd -> %lld + (long long) for off_t's.

replace __progname with getprogname(3)

from Michal Mazurek, ok tb@

simplify print formatting logic, replace exit() in main() with return

from Michal Mazurek, ok tb@

tidy up function declarations and definitions

from Michal Mazurek, ok tb@

remove needless headers

ok tb@, from Michal Mazurek

Change all tame callers to namechange to pledge(2).

df is a tame "stdio rpath" program, the rpath due to getfsstat and statfs.
those two system calls were put into the "rpath" catagory because they
expose pathname information.

Replace <sys/param.h> with <limits.h> and other less dirty headers where
possible. Annotate <sys/param.h> lines with their current reasons. Switch
to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change
MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where
sensible to avoid pulling in the pollution. These are the files confirmed
through binary verification.
ok guenther, millert, doug (helped with the verification protocol)

As suggested by guenther@ and millert@, replace seek+[read|write] with
p[read|write]. Makes the code much clearer by eliminating extra error
checking and verbiage.

No intentional functional change.

Tweaks by and ok guenther@

rcsid[] and sccsid[] and copyright[] are essentially unmaintained (and
unmaintainable). these days, people use source. these id's do not provide
any benefit, and do hurt the small install media
(the 33,000 line diff is essentially mechanical)
ok with the idea millert, ok dms

actually print the now available large numbers

fix df posix mode bug

from otto@

prompted by otto@ who is aftk for now

use calloc() to avoid malloc(n * m) overflows; checked by djm canacar jsg

like -i, -h is incompatible with -P; suggested by jmc@; ok ray@

Remove LFS-related code, no binary change

handle malloc failure; jan.niemann@tu-bs.de

Better handling of large filesystems, okay henning@ and millert@, with
input from martin@ and miod@.

widen the avail/used columns by one space to accomodate for >100g filesystems. the total is already wide enough due to the header length; dhartmei@ ok

Sync usage with man page; Paul de Weerd

Don't print header if we didn't print any usage lines. Noticed and
tested by robert@

fd leak (and who said lint never finds real bugs)

remove unused variable

spacing

missing protos

Remove the advertising clause in the UCB license which Berkeley
rescinded 22 July 1999. Proofed by myself and Theo.

Use prt_scaled from libutil. ok pjanzen@, millert@

LFS is dead. saves some space on install floppies too.

ok deraadt@

ansi

Part one of userland __P removal. Done with a simple regexp with some minor hand editing to make comments line up correctly. Another pass is forthcoming that handles the cases that could not be done automatically.

o) __progname aria;

millert@ ok.

Compute %used in -P mode as a double, just like we do in normal printing
mode. Fixes an int wraparound problem noted by mike@erdelynet.com

Avoid potential uninitialized variable

sort out printf formats according to types, 64bit arithmetics in ffs_df; millert@ ok

Make f_bsize signed. Having it unsigned causes surprising (and unwanted)
type coercions. With this fixed I can back out the hack in df.

In prtstat() use u_int32_t and int32_t, not long and cast
sfsp->f_bsize to be signed when using the fsbtoblk() macro so
we get signed divide, not unsigned divide. Fixes a bug in df
when the filesystem has < 0 blocks available to the user.
Guess I should have just made f_bsize signed when struct statfs
was changed.

New struct statfs with mount options. NOTE: this replaces statfs(2),
fstatfs(2), and getfsstat(2) so you will need to build a new kernel
before doing a "make build" or you will get "unimplemented syscall" errors.

The new struct statfs has the following featuires:
o Has a u_int32_t flags field--now softdep can have a real flag.

o Uses u_int32_t instead of longs (nicer on the alpha). Note: the man
page used to lie about setting invalid/unused fields to -1. SunOS does
that but our code never has.

o Gets rid of f_type completely. It hasn't been used since NetBSD 0.9
and having it there but always 0 is confusing. It is conceivable
that this may cause some old code to not compile but that is better
than silently breaking.

o Adds a mount_info union that contains the FSTYPE_args struct. This
means that "mount" can now tell you all the options a filesystem was
mounted with. This is especially nice for NFS.

Other changes:
o The linux statfs emulation didn't convert between BSD fs names
and linux f_type numbers. Now it does, since the BSD f_type
number is useless to linux apps (and has been removed anyway)

o FreeBSD's struct statfs is different from our (both old and new)
and thus needs conversion. Previously, the OpenBSD syscalls
were used without any real translation.

o mount(8) will now show extra info when invoked with no arguments.
However, to see *everything* you need to use the -v (verbose) flag.

When adjusting units for -h mode, use the absolute value to check for
kb, mega, giga, etc. Based on a change from kstailey@openbsd.org but
fixed differently.

people who fail to update man pages and usage lines will henceforth be shot on sight

fix df -P on 0-size filesystems

Fix units conversion wrt -h flag to be correct (and consistent with
the GNU df's -h flag).

i am bored enough to fix terminal space/tab uglies

Add -P option, the POSIX 1003.2 output format (strictly specified for parsing).

tabify

Better support for unmounted filesystems (i.e. df /dev/rsd0a):
- interpret the superblock of lfs, ext2fs, and ffs filesystems.
- never mount() an offline filesystem, always read its superblock.
- use the same algorithm as statfs() for ffs, get the same answers.

Even so, now I'm really unsure this code should remain in df.
Similar functionality should be added to dumpfs or fsck instead.

From denny@c200234-a.frmt1.sfba.home.com, check superblock magic number.

I'm unsure if this code should even remain in df.

make -h columns more compact

switch from <blank> to "B" for byte suffix; columns look too ugly otherwise

fix Avail column in -h mode (again)

fix Avail column in -h mode

new "-h" (human-readable) output flag now works, please test. thx

start of -h (human readable) flag. more to come

when mktemp() is hard to fix, use 10 X

indicate this mktemp is vulnerable to a DOS attack, however the impact is very low

-Wall'ing.

handle -t & -l in more cases; netbsd pr#2869, hubert.feyrer@rz.uni-regensburg.de

zap getopt() case of -?, come on, it is the default!

update rcsid

complete ufs -> ffs change (From John Kohl; PR #1403)

branches: 1.1.1;
Initial revision

When system calls indicate an error they return -1, not some arbitrary
value < 0. errno is only updated in this case. Change all (most?)
callers of syscalls to follow this better, and let's see if this strictness
helps us in the future.

Another %qd -> %lld + (long long) for off_t's.

replace __progname with getprogname(3)

from Michal Mazurek, ok tb@

simplify print formatting logic, replace exit() in main() with return

from Michal Mazurek, ok tb@

tidy up function declarations and definitions

from Michal Mazurek, ok tb@

remove needless headers

ok tb@, from Michal Mazurek

Change all tame callers to namechange to pledge(2).

df is a tame "stdio rpath" program, the rpath due to getfsstat and statfs.
those two system calls were put into the "rpath" catagory because they
expose pathname information.

Replace <sys/param.h> with <limits.h> and other less dirty headers where
possible. Annotate <sys/param.h> lines with their current reasons. Switch
to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change
MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where
sensible to avoid pulling in the pollution. These are the files confirmed
through binary verification.
ok guenther, millert, doug (helped with the verification protocol)

As suggested by guenther@ and millert@, replace seek+[read|write] with
p[read|write]. Makes the code much clearer by eliminating extra error
checking and verbiage.

No intentional functional change.

Tweaks by and ok guenther@

rcsid[] and sccsid[] and copyright[] are essentially unmaintained (and
unmaintainable). these days, people use source. these id's do not provide
any benefit, and do hurt the small install media
(the 33,000 line diff is essentially mechanical)
ok with the idea millert, ok dms

actually print the now available large numbers

fix df posix mode bug

from otto@

prompted by otto@ who is aftk for now

use calloc() to avoid malloc(n * m) overflows; checked by djm canacar jsg

like -i, -h is incompatible with -P; suggested by jmc@; ok ray@

Remove LFS-related code, no binary change

handle malloc failure; jan.niemann@tu-bs.de

Better handling of large filesystems, okay henning@ and millert@, with
input from martin@ and miod@.

widen the avail/used columns by one space to accomodate for >100g filesystems. the total is already wide enough due to the header length; dhartmei@ ok

Sync usage with man page; Paul de Weerd

Don't print header if we didn't print any usage lines. Noticed and
tested by robert@

fd leak (and who said lint never finds real bugs)

remove unused variable

spacing

missing protos

Remove the advertising clause in the UCB license which Berkeley
rescinded 22 July 1999. Proofed by myself and Theo.

Use prt_scaled from libutil. ok pjanzen@, millert@

LFS is dead. saves some space on install floppies too.

ok deraadt@

ansi

Part one of userland __P removal. Done with a simple regexp with some minor hand editing to make comments line up correctly. Another pass is forthcoming that handles the cases that could not be done automatically.

o) __progname aria;

millert@ ok.

Compute %used in -P mode as a double, just like we do in normal printing
mode. Fixes an int wraparound problem noted by mike@erdelynet.com

Avoid potential uninitialized variable

sort out printf formats according to types, 64bit arithmetics in ffs_df; millert@ ok

Make f_bsize signed. Having it unsigned causes surprising (and unwanted)
type coercions. With this fixed I can back out the hack in df.

In prtstat() use u_int32_t and int32_t, not long and cast
sfsp->f_bsize to be signed when using the fsbtoblk() macro so
we get signed divide, not unsigned divide. Fixes a bug in df
when the filesystem has < 0 blocks available to the user.
Guess I should have just made f_bsize signed when struct statfs
was changed.

New struct statfs with mount options. NOTE: this replaces statfs(2),
fstatfs(2), and getfsstat(2) so you will need to build a new kernel
before doing a "make build" or you will get "unimplemented syscall" errors.

The new struct statfs has the following featuires:
o Has a u_int32_t flags field--now softdep can have a real flag.

o Uses u_int32_t instead of longs (nicer on the alpha). Note: the man
page used to lie about setting invalid/unused fields to -1. SunOS does
that but our code never has.

o Gets rid of f_type completely. It hasn't been used since NetBSD 0.9
and having it there but always 0 is confusing. It is conceivable
that this may cause some old code to not compile but that is better
than silently breaking.

o Adds a mount_info union that contains the FSTYPE_args struct. This
means that "mount" can now tell you all the options a filesystem was
mounted with. This is especially nice for NFS.

Other changes:
o The linux statfs emulation didn't convert between BSD fs names
and linux f_type numbers. Now it does, since the BSD f_type
number is useless to linux apps (and has been removed anyway)

o FreeBSD's struct statfs is different from our (both old and new)
and thus needs conversion. Previously, the OpenBSD syscalls
were used without any real translation.

o mount(8) will now show extra info when invoked with no arguments.
However, to see *everything* you need to use the -v (verbose) flag.

When adjusting units for -h mode, use the absolute value to check for
kb, mega, giga, etc. Based on a change from kstailey@openbsd.org but
fixed differently.

people who fail to update man pages and usage lines will henceforth be shot on sight

fix df -P on 0-size filesystems

Fix units conversion wrt -h flag to be correct (and consistent with
the GNU df's -h flag).

i am bored enough to fix terminal space/tab uglies

Add -P option, the POSIX 1003.2 output format (strictly specified for parsing).

tabify

Better support for unmounted filesystems (i.e. df /dev/rsd0a):
- interpret the superblock of lfs, ext2fs, and ffs filesystems.
- never mount() an offline filesystem, always read its superblock.
- use the same algorithm as statfs() for ffs, get the same answers.

Even so, now I'm really unsure this code should remain in df.
Similar functionality should be added to dumpfs or fsck instead.

From denny@c200234-a.frmt1.sfba.home.com, check superblock magic number.

I'm unsure if this code should even remain in df.

make -h columns more compact

switch from <blank> to "B" for byte suffix; columns look too ugly otherwise

fix Avail column in -h mode (again)

fix Avail column in -h mode

new "-h" (human-readable) output flag now works, please test. thx

start of -h (human readable) flag. more to come

when mktemp() is hard to fix, use 10 X

indicate this mktemp is vulnerable to a DOS attack, however the impact is very low

-Wall'ing.

handle -t & -l in more cases; netbsd pr#2869, hubert.feyrer@rz.uni-regensburg.de

zap getopt() case of -?, come on, it is the default!

update rcsid

complete ufs -> ffs change (From John Kohl; PR #1403)

branches: 1.1.1;
Initial revision

Another %qd -> %lld + (long long) for off_t's.

replace __progname with getprogname(3)

from Michal Mazurek, ok tb@

simplify print formatting logic, replace exit() in main() with return

from Michal Mazurek, ok tb@

tidy up function declarations and definitions

from Michal Mazurek, ok tb@

remove needless headers

ok tb@, from Michal Mazurek

Change all tame callers to namechange to pledge(2).

df is a tame "stdio rpath" program, the rpath due to getfsstat and statfs.
those two system calls were put into the "rpath" catagory because they
expose pathname information.

Replace <sys/param.h> with <limits.h> and other less dirty headers where
possible. Annotate <sys/param.h> lines with their current reasons. Switch
to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change
MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where
sensible to avoid pulling in the pollution. These are the files confirmed
through binary verification.
ok guenther, millert, doug (helped with the verification protocol)

As suggested by guenther@ and millert@, replace seek+[read|write] with
p[read|write]. Makes the code much clearer by eliminating extra error
checking and verbiage.

No intentional functional change.

Tweaks by and ok guenther@

rcsid[] and sccsid[] and copyright[] are essentially unmaintained (and
unmaintainable). these days, people use source. these id's do not provide
any benefit, and do hurt the small install media
(the 33,000 line diff is essentially mechanical)
ok with the idea millert, ok dms

actually print the now available large numbers

fix df posix mode bug

from otto@

prompted by otto@ who is aftk for now

use calloc() to avoid malloc(n * m) overflows; checked by djm canacar jsg

like -i, -h is incompatible with -P; suggested by jmc@; ok ray@

Remove LFS-related code, no binary change

handle malloc failure; jan.niemann@tu-bs.de

Better handling of large filesystems, okay henning@ and millert@, with
input from martin@ and miod@.

widen the avail/used columns by one space to accomodate for >100g filesystems. the total is already wide enough due to the header length; dhartmei@ ok

Sync usage with man page; Paul de Weerd

Don't print header if we didn't print any usage lines. Noticed and
tested by robert@

fd leak (and who said lint never finds real bugs)

remove unused variable

spacing

missing protos

Remove the advertising clause in the UCB license which Berkeley
rescinded 22 July 1999. Proofed by myself and Theo.

Use prt_scaled from libutil. ok pjanzen@, millert@

LFS is dead. saves some space on install floppies too.

ok deraadt@

ansi

Part one of userland __P removal. Done with a simple regexp with some minor hand editing to make comments line up correctly. Another pass is forthcoming that handles the cases that could not be done automatically.

o) __progname aria;

millert@ ok.

Compute %used in -P mode as a double, just like we do in normal printing
mode. Fixes an int wraparound problem noted by mike@erdelynet.com

Avoid potential uninitialized variable

sort out printf formats according to types, 64bit arithmetics in ffs_df; millert@ ok

Make f_bsize signed. Having it unsigned causes surprising (and unwanted)
type coercions. With this fixed I can back out the hack in df.

In prtstat() use u_int32_t and int32_t, not long and cast
sfsp->f_bsize to be signed when using the fsbtoblk() macro so
we get signed divide, not unsigned divide. Fixes a bug in df
when the filesystem has < 0 blocks available to the user.
Guess I should have just made f_bsize signed when struct statfs
was changed.

New struct statfs with mount options. NOTE: this replaces statfs(2),
fstatfs(2), and getfsstat(2) so you will need to build a new kernel
before doing a "make build" or you will get "unimplemented syscall" errors.

The new struct statfs has the following featuires:
o Has a u_int32_t flags field--now softdep can have a real flag.

o Uses u_int32_t instead of longs (nicer on the alpha). Note: the man
page used to lie about setting invalid/unused fields to -1. SunOS does
that but our code never has.

o Gets rid of f_type completely. It hasn't been used since NetBSD 0.9
and having it there but always 0 is confusing. It is conceivable
that this may cause some old code to not compile but that is better
than silently breaking.

o Adds a mount_info union that contains the FSTYPE_args struct. This
means that "mount" can now tell you all the options a filesystem was
mounted with. This is especially nice for NFS.

Other changes:
o The linux statfs emulation didn't convert between BSD fs names
and linux f_type numbers. Now it does, since the BSD f_type
number is useless to linux apps (and has been removed anyway)

o FreeBSD's struct statfs is different from our (both old and new)
and thus needs conversion. Previously, the OpenBSD syscalls
were used without any real translation.

o mount(8) will now show extra info when invoked with no arguments.
However, to see *everything* you need to use the -v (verbose) flag.

When adjusting units for -h mode, use the absolute value to check for
kb, mega, giga, etc. Based on a change from kstailey@openbsd.org but
fixed differently.

people who fail to update man pages and usage lines will henceforth be shot on sight

fix df -P on 0-size filesystems

Fix units conversion wrt -h flag to be correct (and consistent with
the GNU df's -h flag).

i am bored enough to fix terminal space/tab uglies

Add -P option, the POSIX 1003.2 output format (strictly specified for parsing).

tabify

Better support for unmounted filesystems (i.e. df /dev/rsd0a):
- interpret the superblock of lfs, ext2fs, and ffs filesystems.
- never mount() an offline filesystem, always read its superblock.
- use the same algorithm as statfs() for ffs, get the same answers.

Even so, now I'm really unsure this code should remain in df.
Similar functionality should be added to dumpfs or fsck instead.

From denny@c200234-a.frmt1.sfba.home.com, check superblock magic number.

I'm unsure if this code should even remain in df.

make -h columns more compact

switch from <blank> to "B" for byte suffix; columns look too ugly otherwise

fix Avail column in -h mode (again)

fix Avail column in -h mode

new "-h" (human-readable) output flag now works, please test. thx

start of -h (human readable) flag. more to come

when mktemp() is hard to fix, use 10 X

indicate this mktemp is vulnerable to a DOS attack, however the impact is very low

-Wall'ing.

handle -t & -l in more cases; netbsd pr#2869, hubert.feyrer@rz.uni-regensburg.de

zap getopt() case of -?, come on, it is the default!

update rcsid

complete ufs -> ffs change (From John Kohl; PR #1403)

branches: 1.1.1;
Initial revision