Commit graph

25 commits

Author SHA1 Message Date
Mark Salyzyn
2ad0bd0a9b logd: Allow (some) headers to be individually importable
LogReader.h needs to be individually importable.

Fix a few others, drop includes of local includes, let them be
included in source instead and allow headers to be included
alphabetically. Was not a complete audit since goal was to
separate LogReader.h out from the pack.

Bug: 27242723
Change-Id: Ic7759ef90995e5bd285810706af33550c73cf5b5
2016-02-24 10:15:22 -08:00
Sami Tolvanen
d122ee65b6 logd: enforce policy integrity
If a SELinux policy change or a switch to permissive mode is detected
on a user build, restart the device into safe mode, and keep it there
until an OTA is applied or user data is wiped.

This change deprecates the ro.logd.auditd property.

Needs matching changes from
  I781c3059ea8d4fb2f0c923e4488b1932d69678d3
  Ica825cf2af74f5624cf4091544bd24bb5482dbe7
  Id3ca7889ede30b54b7af73dd50653ca1a20d59aa

Bug: 26902605
Change-Id: Idcdc5bff133f13c1267f0ec0a75cc8cf1ddbda0d
2016-02-16 12:54:54 -08:00
Mark Salyzyn
b6bee33182 liblog: logd: support logd.timestamp = monotonic
if ro.logd.timestamp or persist.logd.timestamp are set to the value
monotonic then liblog writer, liblog printing and logd all switch to
recording/printing monotonic time rather than realtime. If reinit
detects a change for presist.logd.timestamp, correct the older entry
timestamps in place.

ToDo: A corner case condition where new log entries in monotonic time
      occur before logd reinit detects persist.logd.timestamp, there
      will be a few out-of-order entries, but with accurate
      timestamps. This problem does not happen for ro.logd.timestamp
      as it is set before logd starts.

NB: This offers a nano second time accuracy on all log entries
    that may be more suitable for merging with other system
    activities, such as systrace, that also use monotonic time. This
    feature is for debugging.

Bug: 23668800
Change-Id: Iee6dab7140061b1a6627254921411f61b01aa5c2
2015-11-03 15:15:51 -08:00
Mark Salyzyn
758058ffd8 logd: object layer format statistics
Simplify table generation by placing the line and header formatting
into each type's (UID, PID, TID) object. Switch to const return
values for the ownership passing functions (*ToName() functions
and methods). Use longer variable names to reduce confusion.

Switch from LINES To NUM for pruned column as that more accurately
reflects what is dropped since one entry can contain several lines.

Bug: 22855208
Change-Id: Ib110dce98a68cf5f844eb30f8a192a1f691eeba2
2015-10-05 09:27:57 -07:00
Mark Salyzyn
ddda212faa logd: optimize code hotspots
Discovered that we had a few libc hotspots. Adjust code to generally
reduce or nullify the number of calls to malloc, free, strlen,
strcmp, strncmp, memcmp & strncasecmp. Total gain looks to be about
3% of logd's processing time. malloc still contributes to 3%, but all
others are now total 0.5%.

Bug: 23685592
Change-Id: Ife721121667969260cdb8b055524ae90f5911278
2015-10-02 16:45:22 -07:00
Mark Salyzyn
151beac76d logd: klogd deal with nuls in dmesg
Switch to using string and length in all transactions, treating
trailing nuls the same as spaces.

ToDo: change dumpstate (bugreport) to use logcat -b printable _regardless_

Bug: 23517551
Change-Id: I42162365e6bf8ed79d356e7b689a673902116fdb
2015-09-29 15:51:38 -07:00
Mark Salyzyn
ed777e9eec logd: serialize accesses to stats helpers
Quick low-risk to resolve possible hash table corruption.
Resolved an unlikely path memory leak.

ToDo: replace lock with nested lock so no lock
      helpers are required.

Bug: 22068332
Change-Id: I303ab06608502c7d61d42f111a9c43366f184d0c
2015-06-25 07:39:24 -07:00
Mark Salyzyn
7718778793 logd: Cleanup
- Android Coding Standard for Constructors
- Side effects NONE

Change-Id: I2cda9dd73f3ac3ab58f394015cb810820093d47b
2015-05-12 15:51:46 -07:00
Mark Salyzyn
ae4d928d81 logd: Add klogd
- Add a klogd to collect the kernel logs and place them into a
  new kernel log buffer
- Parse priority, tag and message from the kernel log messages.
- Turn off pruning for worst UID for the kernel log buffer
- Sniff for 'PM: suspend exit', 'PM: suspend enter' and
  'Suspended for' messages and correct the internal definition
  time correction against monotonic dynamically.
- Discern if we have monotonic or real time (delineation 1980) in
  audit messages.
- perform appropriate math to correct the timestamp to be real time
- filter out any external sources of kernel logging

Change-Id: I8d4c7c5ac19f1f3218079ee3a05a50e2ca55f60d
2015-05-12 15:51:46 -07:00
Mark Salyzyn
202e153f94 logd: propagate ::log status
Add a return value for the ::log() methods, this allows
us to optimize the wakeup for the readers to only occur
when the log message is actually placed.

This is for a future where we may dedupe identical log
messages, filter out log messages, and certainly if we
filter the messages out with an internal logd check of
__android_log_is_loggable().

Change-Id: I763b2a7c29502ab7fa0a5d5022c7b60244fcfde4
2015-04-15 19:43:28 -07:00
Mark Salyzyn
56c6575da7 Merge "logd: syscall optimization" 2015-04-15 14:45:24 +00:00
Nick Kralevich
58ba58a97c logd: Don't embed a flexible array member within another struct
C (but not C++) has a concept of a flexible array member, which
is documented at https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html .
Using a flexible array member indicates that the structure is
really a header for a variable length object.

In logd's case, the variable length structure android_event_string_t
was embedded within another structure called
android_log_event_string_t. This makes gcc's __builtin_object_size()
function really confused. When compiling with C++,
__builtin_object_size(android_log_event_string_t.payload.data, 1)
would return 0, whereas if you compiled the code with C, the same
call would (properly) return -1.

Code which does automatic bounds checking, such as the proposed
patch at https://android-review.googlesource.com/145411 , will
cause problems for logd if this syntax is used.

Don't try to embed a variable length structure within another
structure. This doesn't appear to be valid C nor C++, and
while it's worked, it seems problematic.

Instead, inline the structure so it's one big happy structure.

Change-Id: I8ac02b7142a4f6560f5f80df2effcf720f9896fc
2015-04-07 10:12:20 -07:00
Mark Salyzyn
e3aeeeeccc logd: syscall optimization
- prset(PR_SET_NAME) call once
- No need to call getuid(), should be AID_LOGD

Change-Id: I4dde0b178bc84e711b355cd7677b0dbf905a0634
2015-04-01 19:41:59 +00:00
Mark Salyzyn
ccbadc6be0 logd: report reinit
Bug: 19681572
Change-Id: I343b9d108f064f87df79512a0fdf1b35513c3136
2015-03-12 20:38:53 +00:00
Mark Salyzyn
29eb57066c logd: use <endian.h>
Change-Id: Iba843c054ea4fbe1a26c7821b5613fdb5e8001a1
2015-03-04 13:21:41 -08:00
Mark Salyzyn
eb06de716b logd: auditd remove logDmesg method
- logDmesg method consumes considerable memory
  resources (typically 128KB depending on kernel)
- In the future (eg: klogd, syslogd) there may be need to
  feed multiple logs or threads with the retrieved data.
- By moving the actions of logDmesg into the mainline that
  instantiates the thread objects, we can leverage a single
  allocation of the the kernel log allocation.
- logDmesg (private) is replaced with log (public) which
  has a more useful and descriptive purpose for the class.

Change-Id: Ie2dd0370661493c1e596a7e486904a0e8caab9ff
2014-12-05 20:56:44 +00:00
Nick Kralevich
c234a1b879 logd: throttle SELinux denials to 20/sec
Impose a limit of 20 selinux denials per second. Denials beyond
that point don't add any value, and have the potential to cause
crashes or denial of service attacks.

Do some other misc cleanup while I'm here.

Bug: 18341932
Change-Id: I6125d629ae4d6ae131d2e53bfa41e1f50277d402
2014-11-19 13:35:36 -08:00
Mark Salyzyn
7ee2aef8e0 logd: auditd: report facility LOG_AUTH
Change-Id: Ie325e1b58f52b6c728d5cfd6f6b87287fcf32e10
2014-09-28 14:42:05 -07:00
Mark Salyzyn
6bdeee0ce6 logd: auditd: kmsg priority
Change-Id: I2016fe140e2daf6c69efbd10aef205fffb931aa1
2014-09-19 11:59:42 -07:00
Mark Salyzyn
e4369d68a2 logd: logcat: debuggerd: audit logs to events and main
- auditd spawn log copy to events and main
- logcat delete events as one of the default logs
- debuggerd do not collect events.
- squish multiple spaces
- switch from strcpy to memmove for overlapping buffers

BUG: 14626551
Change-Id: I89b30273ce931ed2b25a53ea9be48e77f4c1bbf4
2014-05-27 14:29:56 -07:00
Mark Salyzyn
989980c55d logd: logcat: debuggerd: auditd logs to events
- auditd switch to recording logs to events log id
- logcat add events as one of the default logs
- debuggerd collect events log as well.

ToDo: debuggerd & bugreport collect intermixed logs.

BUG: 14626551
Change-Id: I958f0e729b7596748be57488a38824db5645be7b
2014-05-15 08:50:50 -07:00
Mark Salyzyn
e0fa291e89 logd: add logd.auditd property
- permit us a mechanism to disable auditd
- standardize property boolean

Bug: 14275676
Change-Id: I76f245c6aee511ed44274159e0ea55915b484dda
2014-04-29 07:25:27 -07:00
Mark Salyzyn
8daa9af02d logd: add thread setname
- permits easier determination of logd thread at
  fault in a stack trace from debuggerd.

Bug: 14275676
Change-Id: Iac2c523147e2bcce34ab7ddcecd02582c5fa7cc0
2014-04-29 00:37:37 +00:00
Mark Salyzyn
e9bebd0eb1 logd: auditd: add logd.auditd.dmesg property
Change-Id: If4a579c2221eec99cf3f6acf59ead8c2d5230517
2014-04-07 10:51:00 -07:00
William Roberts
29d238d2a8 logd: selinux auditd initial commit
Initial commit for an audit daemon that writes kernel audit
messages to the Android logger. The daemon searches dmesg
for all lines that contain "audit" and writes them. Then
receiving the messages from the netlink socket.

It also formats the messages so they are compatable with
ausearch (type=<t> <m> format)

Modified: Mark Salyzyn <salyzyn@google.com>

- do not start auditd
- merge into logd, stripping unnecessary file logging.
- Convert headers and code to support C++
- Fix bugs in libaudit
- squash timestamp (replace with 0.0) due to duplication
- squash pid due to duplication
- squash comm due to duplication

Change-Id: I421bcf33e7e670d596628b1b5c7c25536ce2d3fe
2014-04-07 10:51:00 -07:00