Some rules need to inspect both raw stream data and higher level
buffers together. When this higher level buffer is a streaming
buffer itself, the risk of mismatch exists.
This patch allows an app-layer parser to set a 'min inspect depth'.
The value is used by the stream engine to keep at least this
depth worth of data, so that the detection engine can request
all of it for inspection.
For rules that have the SIG_FLAG_FLUSH flag set, data is inspected
not from offset raw_progress, but from raw_progress minus
min_inspect_depth.
At this time this is only used for sigs that have their fast_pattern
in a HTTP body and have raw stream match as well.
Use per tx detect_flags to track prefilter. Detect flags are used for 2
things:
1. marking tx as fully inspected
2. tracking already run prefilter (incl mpm) engines
This supercedes the MpmIDs API for directionless tracking
of the prefilter engines.
When we have no SGH we have to flag the txs that are 'complete'
as inspected as well.
Special handling for the stream engine:
If a rule mixes TX inspection and STREAM inspection, we can encounter
the case where the rule is evaluated against multiple transactions
during a single inspection run. As the stream data is exactly the same
for each of those runs, it's wasteful to rerun inspection of the stream
portion of the rule.
This patch enables caching of the stream 'inspect engine' result in
the local 'RuleMatchCandidateTx' array. This is valid only during the
live of a single inspection run.
Remove stateful inspection from 'mask' (SignatureMask). The mask wasn't
used in most cases for those rules anyway, as there we rely on the
prefilter. Add a alproto check to catch the remaining cases.
When building the active non-mpm/non-prefilter list check not just
the mask, but also the alproto. This especially helps stateful rules
with negated mpm.
Simplify AppLayerParserHasDecoderEvents usage in detection to only
return true if protocol detection events are set. Other detection is done
in inspect engines.
Move rule group lookup and handling into it's own function. Handle
'post lookup' tasks immediately, instead of after the first detect
run. The tasks were independent of the initial detection.
Many cleanups and much refactoring.
If 'raw' content patterns were used in a dns_query rule, the raw
patterns would only be evaluated for TCP, but not for UDP.
This patch adds the inspection for UDP as well.
Bug #2263.
Certain rules can apply to both TCP and UDP. For example 'alert dns'
rules are inspected against both TCP and UDP. This lead to the
stream inspect engine being called on a UDP packet.
This patch fixes the issue by exiting early from the stream inspect
engine if a) proto is not TCP or b) ssn is not available
Bug #2158.
Set flags by default:
-Wmissing-prototypes
-Wmissing-declarations
-Wstrict-prototypes
-Wwrite-strings
-Wcast-align
-Wbad-function-cast
-Wformat-security
-Wno-format-nonliteral
-Wmissing-format-attribute
-funsigned-char
Fix minor compiler warnings for these new flags on gcc and clang.
Now that MPM runs when the TX progress is right, stateful detection
operates differently.
Changes:
1. raw stream inspection is now also an inspect engine
Since this engine doesn't take the transactions into account, it
could potentially run multiple times on the same data. To avoid
this, basic result caching is in place.
2. the engines are sorted by progress, but the 'MPM' engine is first
even if the progress is higher
If MPM flags a rule to be inspected, the inspect engine for that
buffer runs first. If this step fails, the rule is no longer
evaluated. No state is stored.
Implement the inline mode for raw content inspection. Packets
are leading, and when a packet's payload has been added to the
stream, the packet is inspected in the context of the stream.
Reassembly will return a buffer with the packet data with older
data in front of it and after it, if available.
Remove the 'StreamMsg' approach from the engine. In this approach the
stream engine would create a list of chunks for inspection by the
detection engine. There were several issues:
1. the messages had a fixed size, so blocks of data bigger than ~4k
would be cut into multiple messages
2. it lead to lots of data copying and unnecessary memory use
3. the StreamMsgs used a central pool
The Stream engine switched over to the streaming buffer API, which
means that the reassembled data is always available. This made the
StreamMsg approach even clunkier.
The new approach exposes the streaming buffer data to the detection
engine. It has to pay attention to an important issue though: packet
loss. The data may have gaps. The streaming buffer API tracks the
blocks of continuous data.
To access the data for inspection a callback approach is used. The
'StreamReassembleRaw' function is called with a callback and data.
This way it runs the MPM and individual rule inspection code. At
the end of each detection run the stream engine is notified that it
can move forward it's 'progress'.
The new Hyperscan 4.4 API provides a function to check for SSSE3
presence at runtime. This allows us to fall back to non-Hyperscan
matchers on systems without SSSE3 even when the suricata executable
is built with Hyperscan support. Addresses Redmine issue #2010.
Signed-off-by: Sascha Steinbiss <sascha@steinbiss.name>
Tested-by: Arturo Borrero Gonzalez <arturo@debian.org>
SGH's for tcp and udp are now always only per proto and per direction.
This means we can simply reuse the packet and stream mpm pointers.
The SGH's for the other protocols already used a directionless catch
all mpm pointer.
For all mpm wrapper functions, check minlen vs the input buffer to see
if we can bypass the mpm search.
Next to this, make all the function inline. Also constify the input and
do other minor cleanups.
FreeBSD 10 32-bit with clang 3.3:
log-tlslog.c:172:14: error: format specifies type 'long' but the argument has type 'time_t' (aka 'int') [-Werror,-Wformat]
p->ts.tv_sec,
^~~~~~~~~~~~
1 error generated.
detect-engine-payload.c:508:27: warning: format specifies type 'long' but the argument has type 'time_t' (aka 'int') [-Wformat]
printf("%ld.%06ld\n", tv_diff.tv_sec, (long int)tv_diff.tv_usec);
~~~ ^~~~~~~~~~~~~~
%d
1 warning generated.
The old behaviour of returning a failure if we found a pattern while
matching on negated content is now changed to continuing searching
for other combinations where we don't find the pattern for the
negated content.
Thanks to Will Metcalf for reporting this.
Aho-Corasick mpm optimized for Tilera Tile-Gx architecture. Based on the
util-mpm-ac.c code base. The primary optimizations are:
1) Matching function used Tilera specific instructions.
2) Alphabet compression to reduce delta table size to increase cache
utilization and performance.
The basic observation is that not all 256 ASCII characters are used by
the set of multiple patterns in a group for which a DFA is
created. The first reason is that Suricata's pattern matching is
case-insensitive, so all uppercase characters are converted to
lowercase, leaving a hole of 26 characters in the
alphabet. Previously, this hole was simply left in the middle of the
alphabet and thus in the generated Next State (delta) tables.
A new, smaller, alphabet is created using a translation table of 256
bytes per mpm group. Previously, there was one global translation
table for converting upper case to lowercase.
Additional, unused characters are found by creating a histogram of all
the characters in all the patterns. Then all the characters with zero
counts are mapped to one character (0) in the new alphabet. Since
These characters appear in no pattern, they can all be mapped to a
single character and still result in the same matches being
found. Zero was chosen for the value in the new alphabet since this
"character" is more likely to appear in the input. The unused
character always results in the next state being state zero, but that
fact is not currently used by the code, since special casing takes
additional instructions.
The characters that do appear in some pattern are mapped to
consecutive characters in the new alphabet, starting at 1. This
results in a dense packing of next state values in the delta tables
and additionally can allow for a smaller number of columns in that
table, thus using less memory and better packing into the cache. The
size of the new alphabet is the number of used characters plus 1 for
the unused catch-all character.
The alphabet size is rounded up to the next larger power-of-2 so that
multiplication by the alphabet size can be done with a shift. It
might be possible to use a multiply instruction, so that the exact
alphabet size could be used, which would further reduce the size of
the delta tables, increase cache density and not require the
specialized search functions. The multiply would likely add 1 cycle to
the inner search loop.
Since the multiply by alphabet-size is cleverly merged with a mask
instruction (in the SINDEX macro), specialized versions of the
SCACSearch function are generated for alphabet sizes 256, 128, 64, 32
and 16. This is done by including the file util-mpm-ac-small.c
multiple times with a redefined SINDEX macro. A function pointer is
then stored in the mpm context for the search function. For alpha bit
sizes of 8 or smaller, the number of states usually small, so the DFA
is already very small, so there is little difference using the 16
state search function.
The SCACSearch function is also specialized by the size of the value
stored in the next state (delta) tables, either 16-bits or 32-bits.
This removes a conditional inside the Search function. That
conditional is only called once, but doesn't hurt to remove
it. 16-bits are used for up to 32K states, with the sign bit set for
states with matches.
Future optimization:
The state-has-match values is only needed per state, not per next
state, so checking the next-state sign bit could be replaced with
reading a different value, at the cost of an additional load, but
increasing the 16-bit next state span to 64K.
Since the order of the characters in the new alphabet doesn't matter,
the new alphabet could be sorted by the frequency of the characters in
the expected input stream for that multi-pattern matcher. This would
group more frequent characters into the same cache lines, thus
increasing the probability of reusing a cache-line.
All the next state values for each state live in their own set of
cache-lines. With power-of-two sizes alphabets, these don't overlap.
So either 32 or 16 character's next states are loaded in each cache
line load. If the alphabet size is not an exact power-of-2, then the
last cache-line is not completely full and up to 31*2 bytes of that
line could be wasted per state.
The next state table could be transposed, so that all the next states
for a specific character are stored sequentially, this could be better
if some characters, for example the unused character, are much more
frequent.
Now depth is kept in mind when we inspect chunks in client/server body.
This takes care of FPs originating from inspecting subsequent chunks that
match with depth, but shouldn't.
Victor Julien
Philippe Antoine
Philippe Antoine
Sam Muhammed
Victor Julien
Sascha Steinbiss
Jason Ish
Ken Steele
Anoop Saldanha