| Age | Commit message (Collapse) | Author |
|---|
|
Linux has more states than we were giving it credit for, which led to
our numbers not adding up. Since the exact details seem to change
between versions, and since having code specific to each kernel version
is unattractive, go with the heuristic that there are relatively fewer
"stopped" states (and they change less frequently), so all unknowns are
"sleeping".
|
|
|
|
|
|
This also makes everything more readable by separating out the -b no
formatting case.
The whitespace trimming for -b isn't strictly necessary, but it looks
weird if we output unnecessary spaces at the end of the line in -b mode,
and we've fixed it before, so let's not regress here. It also seems more
logical to do this in the one place where it actually happens.
|
|
|
|
|
|
|
|
Both refresh the display in the traditional implementation.
|
|
Even phones have enough RAM these days that KiB is not a reasonable
unit. Traditional top always uses MiB instead of always using KiB, but
we may as well just let human_readable pick a unit (that way if KiB is
reasonable on your box, that's what you'll get).
Before:
Tasks: 967 total, 1 running, 581 sleeping, 0 stopped, 0 zombie
Mem: 196734820k total,183891564k used, 12843256k free, 5805008k buffers
Swap:199888892k total, 719104k used,199169788k free,130367280k cached
After:
Tasks: 965 total, 2 running, 577 sleeping, 0 stopped, 0 zombie
Mem: 188G total, 175G used, 13G free, 5.5G buffers
Swap: 191G total, 702M used, 190G free, 124G cached
|
|
Matches traditional top.
|
|
|
|
|
|
|
|
Plus new tests.
|
|
Very few places actually check for errors from emit, and I actually see
the same endless loop from "sed (GNU sed) 4.4" on current Debian, so I'm
not sure this isn't Broken As Designed, but an endless loop spewing
"short write" (or saying nothing, in the case of GNU sed) really doesn't
feel like useful behavior in face of EPIPE, which really isn't going to
fix itself. Certainly not being able to run the sed tests to completion
is pretty annoying --- which is why, unless we remove this test as
invalid, we should probably also add a SKIP_HOST=1 to the "b loop" test.
Note that even with this fix you'll see the error twice:
sed: short write: Broken pipe
sed: short write: Broken pipe
Once from the first = command to fail, and then another from the !FLAG(n)
flush of the pattern space.
|
|
The test for \N where N was larger than the number of capturing groups
in the regular expression was incorrect, and firing for cases such as
matching __(ARM_)?NR_([a-z]*) against __NR_read, where the first group is
empty (because it failed to match) but the second group did match "read".
Use regex_t's re_nsub for the error check, and treat rm_so == -1 as a
signal to just copy nothing into the result.
(Found trying to build minijail in AOSP.)
|
|
Add support for more input formats, primarily the ISO formats used by
the AOSP build.
I've improved/added to the tests a bit to cover these changes, and to
explain the reasons for the various remaining test failures (none of
which are regressions caused by this patch).
|
|
|
|
|
|
remove more unnecessary typecasts.
|
|
reversed so the mask is 1<<(x&7) instead of 1<<(7-(x&7)). Can't _quite_
make printString() use unescape() out of lib because \q is a thing?
|
|
|
|
|
|
|
|
|
|
I don't know why NDK llvm is complaining about adjtime(), toys.h is #including
<sys/time.h> which http://man7.org/linux/man-pages/man3/adjtime.3.html says
is the right header...?
|
|
|
|
|
|
|
|
|
|
Bug: http://b/123902291
|
|
|
|
Also add a TODO for the (undocumented, half-finished) -a option.
|
|
record-commands: Delete old log, only delete $WRAPDIR at end if path wasn't
externally supplied, don't add the rm at the end to the log.
logwrapper.c: don't skip filename when measuring space for command line malloc,
use argv[0] instead of /proc/self/exe (which is realpath -f).
|
|
|
|
|
|
|
|
Set SO_REUSEADDR when listening so that we can immediately reuse ports
that are no longer being listened upon, instead of having to wait 60
seconds for the socket to be shutdown after being closed (even on
localhost!).
|
|
|
|
|
|
into display function, use unsigned length so output lines can be up to 4g each.
|
|
|
|
I accidentally added a tab in xargs.c, so as penance I'll clean up all
the tabs.
|
|
Broke the bionic build:
external/toybox/toys/net/netcat.c:188:37: error: incompatible pointer types assigning to 'sigjmp_buf *' (aka 'long (*)[33]') from 'jmp_buf *' (aka 'long (*)[32]') [-Werror,-Wincompatible-pointer-types]
if (toys.optflags&FLAG_L) NOEXIT(child = XVFORK());
^~~~~~~~~~~~~~~~~~~~~~~~
external/toybox/lib/lib.h:375:19: note: expanded from macro 'NOEXIT'
#define NOEXIT(x) WOULD_EXIT(_noexit_res, x)
^~~~~~~~~~~~~~~~~~~~~~~~~~
external/toybox/lib/lib.h:367:16: note: expanded from macro 'WOULD_EXIT'
toys.rebound = &_noexit; \
^ ~~~~~~~~
1 error generated.
|
|
Not yet setting, no server/multicast, should validate source address...
|
|
If table grows >64 entries XX needs to change, but it can't be 128 yet
because the slot value is a signed char. Currently 59 entries in table...
|
|
Only -d and -f should cause a DNS lookup. The rest should just act
directly on the result of gethostname(3). Encountered with the AOSP
buildbots' use of nsjail, but tested with both the Debian hostname and
toybox hostname thus:
```
unshare -Uunr sh
hostname android-build
hostname
hostname -s
hostname -d
hostname -f
```
(Not sure how to add that to the tests.)
Also fix a SEGV with -s if the hostname doesn't contain a '.'.
Also switch to the FLAG() macro.
Also add the missing -s to the synopsis.
Bug: http://b/123123255
|
|
Implement -p, -t, and -r.
Add some missing tests.
Move -L and -x back to TODO since they're not implemented and I haven't
yet even understood what they're supposed to do.
|
|
|
|
|
