You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
suricata/src/detect.c

1704 lines
64 KiB
C

/* Copyright (C) 2007-2017 Open Information Security Foundation
*
* You can copy, redistribute or modify this Program under the terms of
* the GNU General Public License version 2 as published by the Free
* Software Foundation.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* version 2 along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
* 02110-1301, USA.
*/
/**
* \file
*
* \author Victor Julien <victor@inliniac.net>
*
* Basic detection engine
*/
#include "suricata-common.h"
#include "suricata.h"
#include "conf.h"
#include "decode.h"
#include "flow.h"
#include "stream-tcp.h"
#include "app-layer.h"
#include "app-layer-parser.h"
#include "detect.h"
#include "detect-engine.h"
#include "detect-engine-profile.h"
#include "detect-engine-alert.h"
#include "detect-engine-siggroup.h"
#include "detect-engine-address.h"
#include "detect-engine-proto.h"
#include "detect-engine-port.h"
#include "detect-engine-mpm.h"
#include "detect-engine-iponly.h"
15 years ago
#include "detect-engine-threshold.h"
#include "detect-engine-prefilter.h"
#include "detect-engine-state.h"
#include "detect-engine-analyzer.h"
detect: rewrite of the detect engine 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.
8 years ago
#include "detect-engine-payload.h"
#include "detect-engine-event.h"
#include "detect-filestore.h"
#include "detect-flowvar.h"
#include "detect-replace.h"
#include "util-validate.h"
#include "util-detect.h"
detect: rewrite of the detect engine 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.
8 years ago
typedef struct DetectRunScratchpad {
const AppProto alproto;
const uint8_t flow_flags; /* flow/state flags: STREAM_* */
const bool app_decoder_events;
const SigGroupHead *sgh;
SignatureMask pkt_mask;
} DetectRunScratchpad;
/* prototypes */
static DetectRunScratchpad DetectRunSetup(const DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx, Packet * const p, Flow * const pflow);
static void DetectRunInspectIPOnly(ThreadVars *tv, const DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx, Flow * const pflow, Packet * const p);
static inline void DetectRunGetRuleGroup(const DetectEngineCtx *de_ctx,
Packet * const p, Flow * const pflow, DetectRunScratchpad *scratch);
static inline void DetectRunPrefilterPkt(ThreadVars *tv,
DetectEngineCtx *de_ctx, DetectEngineThreadCtx *det_ctx, Packet *p,
DetectRunScratchpad *scratch);
static inline void DetectRulePacketRules(ThreadVars * const tv,
DetectEngineCtx * const de_ctx, DetectEngineThreadCtx * const det_ctx,
Packet * const p, Flow * const pflow, const DetectRunScratchpad *scratch);
static void DetectRunTx(ThreadVars *tv, DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx, Packet *p,
Flow *f, DetectRunScratchpad *scratch);
static inline void DetectRunPostRules(ThreadVars *tv, DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx, Packet * const p, Flow * const pflow,
DetectRunScratchpad *scratch);
static void DetectRunCleanup(DetectEngineThreadCtx *det_ctx,
Packet *p, Flow * const pflow);
detect: rewrite of the detect engine 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.
8 years ago
/** \internal
*/
static void DetectRun(ThreadVars *th_v,
DetectEngineCtx *de_ctx, DetectEngineThreadCtx *det_ctx,
Packet *p)
{
SCEnter();
SCLogDebug("pcap_cnt %"PRIu64, p->pcap_cnt);
/* bail early if packet should not be inspected */
if (p->flags & PKT_NOPACKET_INSPECTION) {
/* nothing to do */
SCReturn;
}
/* Load the Packet's flow early, even though it might not be needed.
* Mark as a constant pointer, although the flow itself can change. */
Flow * const pflow = p->flow;
DetectRunScratchpad scratch = DetectRunSetup(de_ctx, det_ctx, p, pflow);
/* run the IPonly engine */
DetectRunInspectIPOnly(th_v, de_ctx, det_ctx, pflow, p);
/* get our rule group */
DetectRunGetRuleGroup(de_ctx, p, pflow, &scratch);
/* if we didn't get a sig group head, we
* have nothing to do.... */
if (scratch.sgh == NULL) {
SCLogDebug("no sgh for this packet, nothing to match against");
goto end;
}
/* run the prefilters for packets */
DetectRunPrefilterPkt(th_v, de_ctx, det_ctx, p, &scratch);
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_RULES);
/* inspect the rules against the packet */
DetectRulePacketRules(th_v, de_ctx, det_ctx, p, pflow, &scratch);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_RULES);
/* run tx/state inspection. Don't call for ICMP error msgs. */
if (pflow && pflow->alstate && likely(pflow->proto == p->proto)) {
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_TX);
detect: rewrite of the detect engine 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.
8 years ago
DetectRunTx(th_v, de_ctx, det_ctx, p, pflow, &scratch);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_TX);
detect: rewrite of the detect engine 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.
8 years ago
}
end:
DetectRunPostRules(th_v, de_ctx, det_ctx, p, pflow, &scratch);
DetectRunCleanup(det_ctx, p, pflow);
SCReturn;
}
static void DetectRunPostMatch(ThreadVars *tv,
DetectEngineThreadCtx *det_ctx, Packet *p,
const Signature *s)
{
/* run the packet match functions */
detect: rewrite of the detect engine 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.
8 years ago
const SigMatchData *smd = s->sm_arrays[DETECT_SM_LIST_POSTMATCH];
if (smd != NULL) {
KEYWORD_PROFILING_SET_LIST(det_ctx, DETECT_SM_LIST_POSTMATCH);
SCLogDebug("running match functions, sm %p", smd);
while (1) {
KEYWORD_PROFILING_START;
(void)sigmatch_table[smd->type].Match(det_ctx, p, s, smd->ctx);
KEYWORD_PROFILING_END(det_ctx, smd->type, 1);
if (smd->is_last)
break;
smd++;
}
}
}
/**
* \brief Get the SigGroupHead for a packet.
*
* \param de_ctx detection engine context
* \param p packet
*
* \retval sgh the SigGroupHead or NULL if non applies to the packet
*/
const SigGroupHead *SigMatchSignaturesGetSgh(const DetectEngineCtx *de_ctx,
const Packet *p)
{
SCEnter();
int f;
SigGroupHead *sgh = NULL;
/* if the packet proto is 0 (not set), we're inspecting it against
* the decoder events sgh we have. */
if (p->proto == 0 && p->events.cnt > 0) {
SCReturnPtr(de_ctx->decoder_event_sgh, "SigGroupHead");
} else if (p->proto == 0) {
if (!(PKT_IS_IPV4(p) || PKT_IS_IPV6(p))) {
/* not IP, so nothing to do */
SCReturnPtr(NULL, "SigGroupHead");
}
}
/* select the flow_gh */
if (p->flowflags & FLOW_PKT_TOCLIENT)
f = 0;
else
f = 1;
int proto = IP_GET_IPPROTO(p);
if (proto == IPPROTO_TCP) {
DetectPort *list = de_ctx->flow_gh[f].tcp;
SCLogDebug("tcp toserver %p, tcp toclient %p: going to use %p",
de_ctx->flow_gh[1].tcp, de_ctx->flow_gh[0].tcp, de_ctx->flow_gh[f].tcp);
uint16_t port = f ? p->dp : p->sp;
SCLogDebug("tcp port %u -> %u:%u", port, p->sp, p->dp);
DetectPort *sghport = DetectPortLookupGroup(list, port);
if (sghport != NULL)
sgh = sghport->sh;
SCLogDebug("TCP list %p, port %u, direction %s, sghport %p, sgh %p",
list, port, f ? "toserver" : "toclient", sghport, sgh);
} else if (proto == IPPROTO_UDP) {
DetectPort *list = de_ctx->flow_gh[f].udp;
uint16_t port = f ? p->dp : p->sp;
DetectPort *sghport = DetectPortLookupGroup(list, port);
if (sghport != NULL)
sgh = sghport->sh;
SCLogDebug("UDP list %p, port %u, direction %s, sghport %p, sgh %p",
list, port, f ? "toserver" : "toclient", sghport, sgh);
} else {
sgh = de_ctx->flow_gh[f].sgh[proto];
}
16 years ago
SCReturnPtr(sgh, "SigGroupHead");
}
static inline void DetectPrefilterMergeSort(DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx)
{
SigIntId mpm, nonmpm;
SigIntId *mpm_ptr = det_ctx->pmq.rule_id_array;
SigIntId *nonmpm_ptr = det_ctx->non_pf_id_array;
uint32_t m_cnt = det_ctx->pmq.rule_id_array_cnt;
uint32_t n_cnt = det_ctx->non_pf_id_cnt;
SigIntId *final_ptr;
uint32_t final_cnt;
SigIntId id;
SigIntId previous_id = (SigIntId)-1;
Signature **sig_array = de_ctx->sig_array;
Signature **match_array = det_ctx->match_array;
Signature *s;
SCLogDebug("PMQ rule id array count %d", det_ctx->pmq.rule_id_array_cnt);
/* Load first values. */
if (likely(m_cnt)) {
mpm = *mpm_ptr;
} else {
/* mpm list is empty */
final_ptr = nonmpm_ptr;
final_cnt = n_cnt;
goto final;
}
if (likely(n_cnt)) {
nonmpm = *nonmpm_ptr;
} else {
/* non-mpm list is empty. */
final_ptr = mpm_ptr;
final_cnt = m_cnt;
goto final;
}
while (1) {
if (mpm < nonmpm) {
/* Take from mpm list */
id = mpm;
s = sig_array[id];
/* As the mpm list can contain duplicates, check for that here. */
if (likely(id != previous_id)) {
*match_array++ = s;
previous_id = id;
}
if (unlikely(--m_cnt == 0)) {
/* mpm list is now empty */
final_ptr = nonmpm_ptr;
final_cnt = n_cnt;
goto final;
}
mpm_ptr++;
mpm = *mpm_ptr;
} else if (mpm > nonmpm) {
id = nonmpm;
s = sig_array[id];
/* As the mpm list can contain duplicates, check for that here. */
if (likely(id != previous_id)) {
*match_array++ = s;
previous_id = id;
}
if (unlikely(--n_cnt == 0)) {
final_ptr = mpm_ptr;
final_cnt = m_cnt;
goto final;
}
nonmpm_ptr++;
nonmpm = *nonmpm_ptr;
} else { /* implied mpm == nonmpm */
/* special case: if on both lists, it's a negated mpm pattern */
/* mpm list may have dups, so skip past them here */
while (--m_cnt != 0) {
mpm_ptr++;
mpm = *mpm_ptr;
if (mpm != nonmpm)
break;
}
/* if mpm is done, update nonmpm_ptrs and jump to final */
if (unlikely(m_cnt == 0)) {
n_cnt--;
/* mpm list is now empty */
final_ptr = ++nonmpm_ptr;
final_cnt = n_cnt;
goto final;
}
/* otherwise, if nonmpm is done jump to final for mpm
* mpm ptrs alrady updated */
if (unlikely(--n_cnt == 0)) {
final_ptr = mpm_ptr;
final_cnt = m_cnt;
goto final;
}
/* not at end of the lists, update nonmpm. Mpm already
* updated in while loop above. */
nonmpm_ptr++;
nonmpm = *nonmpm_ptr;
}
}
final: /* Only one list remaining. Just walk that list. */
while (final_cnt-- > 0) {
id = *final_ptr++;
s = sig_array[id];
/* As the mpm list can contain duplicates, check for that here. */
if (likely(id != previous_id)) {
*match_array++ = s;
previous_id = id;
}
}
det_ctx->match_array_cnt = match_array - det_ctx->match_array;
DEBUG_VALIDATE_BUG_ON((det_ctx->pmq.rule_id_array_cnt + det_ctx->non_pf_id_cnt) < det_ctx->match_array_cnt);
}
/** \internal
* \brief build non-prefilter list based on the rule group list we've set.
*/
static inline void
DetectPrefilterBuildNonPrefilterList(DetectEngineThreadCtx *det_ctx,
const SignatureMask mask, const uint8_t alproto)
{
for (uint32_t x = 0; x < det_ctx->non_pf_store_cnt; x++) {
/* only if the mask matches this rule can possibly match,
* so build the non_mpm array only for match candidates */
detect: rewrite of the detect engine 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.
8 years ago
const SignatureMask rule_mask = det_ctx->non_pf_store_ptr[x].mask;
const uint8_t rule_alproto = det_ctx->non_pf_store_ptr[x].alproto;
if ((rule_mask & mask) == rule_mask && (rule_alproto == 0 || rule_alproto == alproto)) {
det_ctx->non_pf_id_array[det_ctx->non_pf_id_cnt++] = det_ctx->non_pf_store_ptr[x].id;
}
}
}
/** \internal
* \brief select non-mpm list
detect: rewrite of the detect engine 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.
8 years ago
* Based on the packet properties, select the non-mpm list to use
* \todo move non_pf_store* into scratchpad */
static inline void
detect: rewrite of the detect engine 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.
8 years ago
DetectPrefilterSetNonPrefilterList(const Packet *p, DetectEngineThreadCtx *det_ctx, DetectRunScratchpad *scratch)
{
if ((p->proto == IPPROTO_TCP) && (p->tcph != NULL) && (p->tcph->th_flags & TH_SYN)) {
detect: rewrite of the detect engine 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.
8 years ago
det_ctx->non_pf_store_ptr = scratch->sgh->non_pf_syn_store_array;
det_ctx->non_pf_store_cnt = scratch->sgh->non_pf_syn_store_cnt;
} else {
detect: rewrite of the detect engine 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.
8 years ago
det_ctx->non_pf_store_ptr = scratch->sgh->non_pf_other_store_array;
det_ctx->non_pf_store_cnt = scratch->sgh->non_pf_other_store_cnt;
}
SCLogDebug("sgh non_pf ptr %p cnt %u (syn %p/%u, other %p/%u)",
det_ctx->non_pf_store_ptr, det_ctx->non_pf_store_cnt,
detect: rewrite of the detect engine 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.
8 years ago
scratch->sgh->non_pf_syn_store_array, scratch->sgh->non_pf_syn_store_cnt,
scratch->sgh->non_pf_other_store_array, scratch->sgh->non_pf_other_store_cnt);
}
/** \internal
* \brief update flow's file tracking flags based on the detection engine
* A set of flags is prepared that is sent to the File API. The
File API may reject one or more based on the global force settings.
*/
static inline void
DetectPostInspectFileFlagsUpdate(Flow *f, const SigGroupHead *sgh, uint8_t direction)
{
uint16_t flow_file_flags = FLOWFILE_INIT;
if (sgh == NULL) {
SCLogDebug("requesting disabling all file features for flow");
flow_file_flags = FLOWFILE_NONE;
} else {
if (sgh->filestore_cnt == 0) {
SCLogDebug("requesting disabling filestore for flow");
flow_file_flags |= (FLOWFILE_NO_STORE_TS|FLOWFILE_NO_STORE_TC);
}
#ifdef HAVE_MAGIC
if (!(sgh->flags & SIG_GROUP_HEAD_HAVEFILEMAGIC)) {
SCLogDebug("requesting disabling magic for flow");
flow_file_flags |= (FLOWFILE_NO_MAGIC_TS|FLOWFILE_NO_MAGIC_TC);
}
#endif
#ifdef HAVE_NSS
if (!(sgh->flags & SIG_GROUP_HEAD_HAVEFILEMD5)) {
SCLogDebug("requesting disabling md5 for flow");
flow_file_flags |= (FLOWFILE_NO_MD5_TS|FLOWFILE_NO_MD5_TC);
}
if (!(sgh->flags & SIG_GROUP_HEAD_HAVEFILESHA1)) {
SCLogDebug("requesting disabling sha1 for flow");
flow_file_flags |= (FLOWFILE_NO_SHA1_TS|FLOWFILE_NO_SHA1_TC);
}
if (!(sgh->flags & SIG_GROUP_HEAD_HAVEFILESHA256)) {
SCLogDebug("requesting disabling sha256 for flow");
flow_file_flags |= (FLOWFILE_NO_SHA256_TS|FLOWFILE_NO_SHA256_TC);
}
#endif
if (!(sgh->flags & SIG_GROUP_HEAD_HAVEFILESIZE)) {
SCLogDebug("requesting disabling filesize for flow");
flow_file_flags |= (FLOWFILE_NO_SIZE_TS|FLOWFILE_NO_SIZE_TC);
}
}
if (flow_file_flags != 0) {
FileUpdateFlowFileFlags(f, flow_file_flags, direction);
}
}
static inline void
detect: rewrite of the detect engine 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.
8 years ago
DetectRunPostGetFirstRuleGroup(const Packet *p, Flow *pflow, const SigGroupHead *sgh)
{
if ((p->flowflags & FLOW_PKT_TOSERVER) && !(pflow->flags & FLOW_SGH_TOSERVER)) {
/* first time we see this toserver sgh, store it */
pflow->sgh_toserver = sgh;
pflow->flags |= FLOW_SGH_TOSERVER;
if (p->proto == IPPROTO_TCP && (sgh == NULL || !(sgh->flags & SIG_GROUP_HEAD_HAVERAWSTREAM))) {
if (pflow->protoctx != NULL) {
TcpSession *ssn = pflow->protoctx;
SCLogDebug("STREAMTCP_STREAM_FLAG_DISABLE_RAW ssn.client");
ssn->client.flags |= STREAMTCP_STREAM_FLAG_DISABLE_RAW;
}
}
DetectPostInspectFileFlagsUpdate(pflow,
pflow->sgh_toserver, STREAM_TOSERVER);
} else if ((p->flowflags & FLOW_PKT_TOCLIENT) && !(pflow->flags & FLOW_SGH_TOCLIENT)) {
pflow->sgh_toclient = sgh;
pflow->flags |= FLOW_SGH_TOCLIENT;
if (p->proto == IPPROTO_TCP && (sgh == NULL || !(sgh->flags & SIG_GROUP_HEAD_HAVERAWSTREAM))) {
if (pflow->protoctx != NULL) {
TcpSession *ssn = pflow->protoctx;
SCLogDebug("STREAMTCP_STREAM_FLAG_DISABLE_RAW ssn.server");
ssn->server.flags |= STREAMTCP_STREAM_FLAG_DISABLE_RAW;
}
}
DetectPostInspectFileFlagsUpdate(pflow,
pflow->sgh_toclient, STREAM_TOCLIENT);
}
}
detect: rewrite of the detect engine 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.
8 years ago
static inline void DetectRunGetRuleGroup(
const DetectEngineCtx *de_ctx,
Packet * const p, Flow * const pflow,
DetectRunScratchpad *scratch)
{
const SigGroupHead *sgh = NULL;
if (pflow) {
bool use_flow_sgh = false;
/* Get the stored sgh from the flow (if any). Make sure we're not using
* the sgh for icmp error packets part of the same stream. */
if (IP_GET_IPPROTO(p) == pflow->proto) { /* filter out icmp */
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_GETSGH);
if ((p->flowflags & FLOW_PKT_TOSERVER) && (pflow->flags & FLOW_SGH_TOSERVER)) {
sgh = pflow->sgh_toserver;
SCLogDebug("sgh = pflow->sgh_toserver; => %p", sgh);
use_flow_sgh = true;
} else if ((p->flowflags & FLOW_PKT_TOCLIENT) && (pflow->flags & FLOW_SGH_TOCLIENT)) {
sgh = pflow->sgh_toclient;
SCLogDebug("sgh = pflow->sgh_toclient; => %p", sgh);
use_flow_sgh = true;
}
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_GETSGH);
}
if (!(use_flow_sgh)) {
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_GETSGH);
sgh = SigMatchSignaturesGetSgh(de_ctx, p);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_GETSGH);
/* HACK: prevent the wrong sgh (or NULL) from being stored in the
* flow's sgh pointers */
if (PKT_IS_ICMPV4(p) && ICMPV4_DEST_UNREACH_IS_VALID(p)) {
; /* no-op */
} else {
/* store the found sgh (or NULL) in the flow to save us
* from looking it up again for the next packet.
* Also run other tasks */
DetectRunPostGetFirstRuleGroup(p, pflow, sgh);
}
}
} else { /* p->flags & PKT_HAS_FLOW */
/* no flow */
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_GETSGH);
sgh = SigMatchSignaturesGetSgh(de_ctx, p);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_GETSGH);
}
scratch->sgh = sgh;
}
static void DetectRunInspectIPOnly(ThreadVars *tv, const DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx,
Flow * const pflow, Packet * const p)
{
if (pflow) {
/* set the iponly stuff */
if (pflow->flags & FLOW_TOCLIENT_IPONLY_SET)
p->flowflags |= FLOW_PKT_TOCLIENT_IPONLY_SET;
if (pflow->flags & FLOW_TOSERVER_IPONLY_SET)
p->flowflags |= FLOW_PKT_TOSERVER_IPONLY_SET;
if (((p->flowflags & FLOW_PKT_TOSERVER) && !(p->flowflags & FLOW_PKT_TOSERVER_IPONLY_SET)) ||
((p->flowflags & FLOW_PKT_TOCLIENT) && !(p->flowflags & FLOW_PKT_TOCLIENT_IPONLY_SET)))
{
SCLogDebug("testing against \"ip-only\" signatures");
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_IPONLY);
IPOnlyMatchPacket(tv, de_ctx, det_ctx, &de_ctx->io_ctx, &det_ctx->io_ctx, p);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_IPONLY);
/* save in the flow that we scanned this direction... */
FlowSetIPOnlyFlag(pflow, p->flowflags & FLOW_PKT_TOSERVER ? 1 : 0);
} else if (((p->flowflags & FLOW_PKT_TOSERVER) &&
(pflow->flags & FLOW_TOSERVER_IPONLY_SET)) ||
((p->flowflags & FLOW_PKT_TOCLIENT) &&
(pflow->flags & FLOW_TOCLIENT_IPONLY_SET)))
{
/* If we have a drop from IP only module,
* we will drop the rest of the flow packets
* This will apply only to inline/IPS */
if (pflow->flags & FLOW_ACTION_DROP) {
PACKET_DROP(p);
}
}
} else { /* p->flags & PKT_HAS_FLOW */
/* no flow */
/* Even without flow we should match the packet src/dst */
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_IPONLY);
IPOnlyMatchPacket(tv, de_ctx, det_ctx, &de_ctx->io_ctx,
&det_ctx->io_ctx, p);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_IPONLY);
}
}
/* returns 0 if no match, 1 if match */
static inline int DetectRunInspectRuleHeader(
const Packet *p,
const Flow *f,
const Signature *s,
const uint32_t sflags,
const uint8_t s_proto_flags)
{
/* check if this signature has a requirement for flowvars of some type
* and if so, if we actually have any in the flow. If not, the sig
* can't match and we skip it. */
if ((p->flags & PKT_HAS_FLOW) && (sflags & SIG_FLAG_REQUIRE_FLOWVAR)) {
DEBUG_VALIDATE_BUG_ON(f == NULL);
int m = f->flowvar ? 1 : 0;
/* no flowvars? skip this sig */
if (m == 0) {
SCLogDebug("skipping sig as the flow has no flowvars and sig "
"has SIG_FLAG_REQUIRE_FLOWVAR flag set.");
return 0;
}
}
if ((s_proto_flags & DETECT_PROTO_IPV4) && !PKT_IS_IPV4(p)) {
SCLogDebug("ip version didn't match");
return 0;
}
if ((s_proto_flags & DETECT_PROTO_IPV6) && !PKT_IS_IPV6(p)) {
SCLogDebug("ip version didn't match");
return 0;
}
if (DetectProtoContainsProto(&s->proto, IP_GET_IPPROTO(p)) == 0) {
SCLogDebug("proto didn't match");
return 0;
}
/* check the source & dst port in the sig */
if (p->proto == IPPROTO_TCP || p->proto == IPPROTO_UDP || p->proto == IPPROTO_SCTP) {
if (!(sflags & SIG_FLAG_DP_ANY)) {
if (p->flags & PKT_IS_FRAGMENT)
return 0;
DetectPort *dport = DetectPortLookupGroup(s->dp,p->dp);
if (dport == NULL) {
SCLogDebug("dport didn't match.");
return 0;
}
}
if (!(sflags & SIG_FLAG_SP_ANY)) {
if (p->flags & PKT_IS_FRAGMENT)
return 0;
DetectPort *sport = DetectPortLookupGroup(s->sp,p->sp);
if (sport == NULL) {
SCLogDebug("sport didn't match.");
return 0;
}
}
} else if ((sflags & (SIG_FLAG_DP_ANY|SIG_FLAG_SP_ANY)) != (SIG_FLAG_DP_ANY|SIG_FLAG_SP_ANY)) {
SCLogDebug("port-less protocol and sig needs ports");
return 0;
}
/* check the destination address */
if (!(sflags & SIG_FLAG_DST_ANY)) {
if (PKT_IS_IPV4(p)) {
if (DetectAddressMatchIPv4(s->addr_dst_match4, s->addr_dst_match4_cnt, &p->dst) == 0)
return 0;
} else if (PKT_IS_IPV6(p)) {
if (DetectAddressMatchIPv6(s->addr_dst_match6, s->addr_dst_match6_cnt, &p->dst) == 0)
return 0;
}
}
/* check the source address */
if (!(sflags & SIG_FLAG_SRC_ANY)) {
if (PKT_IS_IPV4(p)) {
if (DetectAddressMatchIPv4(s->addr_src_match4, s->addr_src_match4_cnt, &p->src) == 0)
return 0;
} else if (PKT_IS_IPV6(p)) {
if (DetectAddressMatchIPv6(s->addr_src_match6, s->addr_src_match6_cnt, &p->src) == 0)
return 0;
}
}
return 1;
}
detect: rewrite of the detect engine 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.
8 years ago
/** \internal
* \brief run packet/stream prefilter engines
*/
detect: rewrite of the detect engine 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.
8 years ago
static inline void DetectRunPrefilterPkt(
ThreadVars *tv,
DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx,
Packet *p,
DetectRunScratchpad *scratch
)
{
detect: rewrite of the detect engine 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.
8 years ago
DetectPrefilterSetNonPrefilterList(p, det_ctx, scratch);
/* create our prefilter mask */
detect: rewrite of the detect engine 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.
8 years ago
PacketCreateMask(p, &scratch->pkt_mask, scratch->alproto, scratch->app_decoder_events);
/* build and prefilter non_pf list against the mask of the packet */
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_NONMPMLIST);
det_ctx->non_pf_id_cnt = 0;
if (likely(det_ctx->non_pf_store_cnt > 0)) {
detect: rewrite of the detect engine 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.
8 years ago
DetectPrefilterBuildNonPrefilterList(det_ctx, scratch->pkt_mask, scratch->alproto);
}
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_NONMPMLIST);
/* run the prefilter engines */
detect: rewrite of the detect engine 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.
8 years ago
Prefilter(det_ctx, scratch->sgh, p, scratch->flow_flags);
/* create match list if we have non-pf and/or pf */
if (det_ctx->non_pf_store_cnt || det_ctx->pmq.rule_id_array_cnt) {
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_PF_SORT2);
DetectPrefilterMergeSort(de_ctx, det_ctx);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_PF_SORT2);
}
#ifdef PROFILING
detect: rewrite of the detect engine 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.
8 years ago
if (tv) {
StatsAddUI64(tv, det_ctx->counter_mpm_list,
(uint64_t)det_ctx->pmq.rule_id_array_cnt);
detect: rewrite of the detect engine 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.
8 years ago
StatsAddUI64(tv, det_ctx->counter_nonmpm_list,
(uint64_t)det_ctx->non_pf_store_cnt);
/* non mpm sigs after mask prefilter */
detect: rewrite of the detect engine 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.
8 years ago
StatsAddUI64(tv, det_ctx->counter_fnonmpm_list,
(uint64_t)det_ctx->non_pf_id_cnt);
}
#endif
detect: rewrite of the detect engine 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.
8 years ago
}
static inline void DetectRulePacketRules(
ThreadVars * const tv,
DetectEngineCtx * const de_ctx,
DetectEngineThreadCtx * const det_ctx,
Packet * const p,
Flow * const pflow,
const DetectRunScratchpad *scratch
)
{
const Signature *s = NULL;
const Signature *next_s = NULL;
/* inspect the sigs against the packet */
/* Prefetch the next signature. */
SigIntId match_cnt = det_ctx->match_array_cnt;
#ifdef PROFILING
detect: rewrite of the detect engine 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.
8 years ago
if (tv) {
StatsAddUI64(tv, det_ctx->counter_match_list,
(uint64_t)match_cnt);
}
#endif
Signature **match_array = det_ctx->match_array;
detect: rewrite of the detect engine 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.
8 years ago
SGH_PROFILING_RECORD(det_ctx, scratch->sgh);
#ifdef PROFILING
if (match_cnt >= de_ctx->profile_match_logging_threshold)
detect: rewrite of the detect engine 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.
8 years ago
RulesDumpMatchArray(det_ctx, scratch->sgh, p);
#endif
uint32_t sflags, next_sflags = 0;
if (match_cnt) {
next_s = *match_array++;
next_sflags = next_s->flags;
}
while (match_cnt--) {
RULE_PROFILING_START(p);
detect: rewrite of the detect engine 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.
8 years ago
uint8_t alert_flags = 0;
bool state_alert = false;
#ifdef PROFILING
bool smatch = false; /* signature match */
#endif
s = next_s;
sflags = next_sflags;
if (match_cnt) {
next_s = *match_array++;
next_sflags = next_s->flags;
}
const uint8_t s_proto_flags = s->proto.flags;
SCLogDebug("inspecting signature id %"PRIu32"", s->id);
if (s->app_inspect != NULL) {
goto next; // handle sig in DetectRunTx
detect: rewrite of the detect engine 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.
8 years ago
}
detect: rewrite of the detect engine 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.
8 years ago
/* don't run mask check for stateful rules.
* There we depend on prefilter */
if ((s->mask & scratch->pkt_mask) != s->mask) {
SCLogDebug("mask mismatch %x & %x != %x", s->mask, scratch->pkt_mask, s->mask);
goto next;
}
if (unlikely(sflags & SIG_FLAG_DSIZE)) {
if (likely(p->payload_len < s->dsize_low || p->payload_len > s->dsize_high)) {
SCLogDebug("kicked out as p->payload_len %u, dsize low %u, hi %u",
p->payload_len, s->dsize_low, s->dsize_high);
goto next;
}
}
/* if the sig has alproto and the session as well they should match */
if (likely(sflags & SIG_FLAG_APPLAYER)) {
detect: rewrite of the detect engine 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.
8 years ago
if (s->alproto != ALPROTO_UNKNOWN && s->alproto != scratch->alproto) {
if (s->alproto == ALPROTO_DCERPC) {
if (scratch->alproto != ALPROTO_SMB) {
SCLogDebug("DCERPC sig, alproto not SMB");
goto next;
}
} else {
SCLogDebug("alproto mismatch");
goto next;
}
}
}
if (DetectRunInspectRuleHeader(p, pflow, s, sflags, s_proto_flags) == 0) {
goto next;
}
if (DetectEnginePktInspectionRun(tv, det_ctx, s, pflow, p, &alert_flags) == false) {
goto next;
}
#ifdef PROFILING
smatch = true;
#endif
detect: rewrite of the detect engine 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.
8 years ago
DetectRunPostMatch(tv, det_ctx, p, s);
if (!(sflags & SIG_FLAG_NOALERT)) {
/* stateful sigs call PacketAlertAppend from DeStateDetectStartDetection */
if (!state_alert)
PacketAlertAppend(det_ctx, s, p, 0, alert_flags);
} else {
/* apply actions even if not alerting */
DetectSignatureApplyActions(p, s, alert_flags);
}
next:
DetectVarProcessList(det_ctx, pflow, p);
DetectReplaceFree(det_ctx);
RULE_PROFILING_END(det_ctx, s, smatch, p);
13 years ago
det_ctx->flags = 0;
continue;
}
detect: rewrite of the detect engine 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.
8 years ago
}
detect: rewrite of the detect engine 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.
8 years ago
static DetectRunScratchpad DetectRunSetup(
const DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx,
Packet * const p, Flow * const pflow)
{
AppProto alproto = ALPROTO_UNKNOWN;
uint8_t flow_flags = 0; /* flow/state flags */
bool app_decoder_events = false;
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_SETUP);
detect: rewrite of the detect engine 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.
8 years ago
#ifdef UNITTESTS
p->alerts.cnt = 0;
#endif
det_ctx->ticker++;
det_ctx->filestore_cnt = 0;
det_ctx->base64_decoded_len = 0;
det_ctx->raw_stream_progress = 0;
det_ctx->match_array_cnt = 0;
detect: rewrite of the detect engine 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.
8 years ago
#ifdef DEBUG
if (p->flags & PKT_STREAM_ADD) {
det_ctx->pkt_stream_add_cnt++;
}
#endif
/* grab the protocol state we will detect on */
if (p->flags & PKT_HAS_FLOW) {
DEBUG_VALIDATE_BUG_ON(pflow == NULL);
detect: rewrite of the detect engine 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.
8 years ago
if (p->flowflags & FLOW_PKT_TOSERVER) {
flow_flags = STREAM_TOSERVER;
SCLogDebug("flag STREAM_TOSERVER set");
} else if (p->flowflags & FLOW_PKT_TOCLIENT) {
flow_flags = STREAM_TOCLIENT;
SCLogDebug("flag STREAM_TOCLIENT set");
}
SCLogDebug("p->flowflags 0x%02x", p->flowflags);
if (p->flags & PKT_STREAM_EOF) {
flow_flags |= STREAM_EOF;
SCLogDebug("STREAM_EOF set");
}
/* store tenant_id in the flow so that we can use it
* for creating pseudo packets */
if (p->tenant_id > 0 && pflow->tenant_id == 0) {
pflow->tenant_id = p->tenant_id;
}
/* live ruleswap check for flow updates */
if (pflow->de_ctx_version == 0) {
/* first time this flow is inspected, set id */
pflow->de_ctx_version = de_ctx->version;
} else if (pflow->de_ctx_version != de_ctx->version) {
/* first time we inspect flow with this de_ctx, reset */
pflow->flags &= ~FLOW_SGH_TOSERVER;
pflow->flags &= ~FLOW_SGH_TOCLIENT;
pflow->sgh_toserver = NULL;
pflow->sgh_toclient = NULL;
pflow->de_ctx_version = de_ctx->version;
GenericVarFree(pflow->flowvar);
pflow->flowvar = NULL;
DetectEngineStateResetTxs(pflow);
}
/* Retrieve the app layer state and protocol and the tcp reassembled
* stream chunks. */
if ((p->proto == IPPROTO_TCP && (p->flags & PKT_STREAM_EST)) ||
(p->proto == IPPROTO_UDP) ||
(p->proto == IPPROTO_SCTP && (p->flowflags & FLOW_PKT_ESTABLISHED)))
{
/* update flow flags with knowledge on disruptions */
flow_flags = FlowGetDisruptionFlags(pflow, flow_flags);
alproto = FlowGetAppProtocol(pflow);
if (p->proto == IPPROTO_TCP && pflow->protoctx &&
StreamReassembleRawHasDataReady(pflow->protoctx, p)) {
p->flags |= PKT_DETECT_HAS_STREAMDATA;
detect/http: flush bodies when inspecting stream The HTTP bodies (http_client_body and http_server_body/file_data) use settings to control how much data we have before doing first inspection: request-body-minimal-inspect-size response-body-minimal-inspect-size These settings default to 32k as quite some existing rules need this. At the same time, the 'raw stream' inspection uses its own limits. By default it inspects the data in blocks of about 2.5k. This could lead to a situation where rules would not match. For example, with 2 rules like this: content:"abc"; content:"data="; http_client_body; depth:5; sid:1; content:"xyz"; sid:2; Sid 1 would only be inspected when the POST body reached the 32k limit or when it was complete. Observed case shows the POST body to be 18k. Sid 2 is inspected as soon as the 2.5k limit is reached, and then again for each 2.5k increment. This moves the raw stream tracker forward. So by the time sid 1 is inspected, some 18/19k into the stream, the raw stream tracker is actually already moved forward for approximately 17.5k, this leads to the stream match of sid 1 possibly not matching. Since the body match is at the start of the buffer, it makes sense that the body and stream are inspected together. The body inspection uses a tracker 'body_inspected', that keeps track of how far into the body both MPM and per signature inspection has moved. This patch updates the logic in 2 ways: 1. it triggers earlier HTTP body inspection, which is matched to the stream inspection. When the detection engine finds it has stream data available for inspection, it passes the new 'STREAM_FLUSH' flag to the HTTP body inspection code. Which will then do an early inspection, even if still before the min inspect size. 2. to still somewhat adhere to the min inspect size, the body tracker is not updated until the min inspect size is reached. This will lead to some re-evaluation of the same body data. If raw stream reassembly is disabled, this 'STREAM_FLUSH' flag is never set, and the old behavior is used. Bug #2522.
7 years ago
flow_flags |= STREAM_FLUSH;
detect: rewrite of the detect engine 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.
8 years ago
}
SCLogDebug("alproto %u", alproto);
} else {
SCLogDebug("packet doesn't have established flag set (proto %d)", p->proto);
}
app_decoder_events = AppLayerParserHasDecoderEvents(pflow->alparser);
}
DetectRunScratchpad pad = { alproto, flow_flags, app_decoder_events, NULL, 0 };
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_SETUP);
detect: rewrite of the detect engine 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.
8 years ago
return pad;
}
static inline void DetectRunPostRules(
ThreadVars *tv,
DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx,
Packet * const p,
Flow * const pflow,
DetectRunScratchpad *scratch)
{
/* see if we need to increment the inspect_id and reset the de_state */
if (pflow && pflow->alstate) {
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_TX_UPDATE);
detect: rewrite of the detect engine 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.
8 years ago
DeStateUpdateInspectTransactionId(pflow, scratch->flow_flags, (scratch->sgh == NULL));
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_TX_UPDATE);
}
/* so now let's iterate the alerts and remove the ones after a pass rule
* matched (if any). This is done inside PacketAlertFinalize() */
/* PR: installed "tag" keywords are handled after the threshold inspection */
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_ALERT);
PacketAlertFinalize(de_ctx, det_ctx, p);
if (p->alerts.cnt > 0) {
detect: rewrite of the detect engine 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.
8 years ago
StatsAddUI64(tv, det_ctx->counter_alerts, (uint64_t)p->alerts.cnt);
}
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_ALERT);
}
detect: rewrite of the detect engine 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.
8 years ago
static void DetectRunCleanup(DetectEngineThreadCtx *det_ctx,
Packet *p, Flow * const pflow)
{
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_CLEANUP);
/* cleanup pkt specific part of the patternmatcher */
PacketPatternCleanup(det_ctx);
InspectionBufferClean(det_ctx);
if (pflow != NULL) {
/* update inspected tracker for raw reassembly */
if (p->proto == IPPROTO_TCP && pflow->protoctx != NULL &&
(p->flags & PKT_STREAM_EST))
{
StreamReassembleRawUpdateProgress(pflow->protoctx, p,
det_ctx->raw_stream_progress);
}
}
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_CLEANUP);
SCReturn;
}
detect: rewrite of the detect engine 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.
8 years ago
void RuleMatchCandidateTxArrayInit(DetectEngineThreadCtx *det_ctx, uint32_t size)
{
DEBUG_VALIDATE_BUG_ON(det_ctx->tx_candidates);
det_ctx->tx_candidates = SCCalloc(size, sizeof(RuleMatchCandidateTx));
if (det_ctx->tx_candidates == NULL) {
FatalError(SC_ERR_MEM_ALLOC, "failed to allocate %"PRIu64" bytes",
(uint64_t)(size * sizeof(RuleMatchCandidateTx)));
}
det_ctx->tx_candidates_size = size;
SCLogDebug("array initialized to %u elements (%"PRIu64" bytes)",
size, (uint64_t)(size * sizeof(RuleMatchCandidateTx)));
}
void RuleMatchCandidateTxArrayFree(DetectEngineThreadCtx *det_ctx)
{
SCFree(det_ctx->tx_candidates);
det_ctx->tx_candidates_size = 0;
}
/* if size >= cur_space */
static inline bool RuleMatchCandidateTxArrayHasSpace(const DetectEngineThreadCtx *det_ctx,
const uint32_t need)
{
if (det_ctx->tx_candidates_size >= need)
return 1;
return 0;
}
/* realloc */
static int RuleMatchCandidateTxArrayExpand(DetectEngineThreadCtx *det_ctx, const uint32_t needed)
{
const uint32_t old_size = det_ctx->tx_candidates_size;
uint32_t new_size = needed;
void *ptmp = SCRealloc(det_ctx->tx_candidates, (new_size * sizeof(RuleMatchCandidateTx)));
if (ptmp == NULL) {
FatalError(SC_ERR_MEM_ALLOC, "failed to expand to %"PRIu64" bytes",
(uint64_t)(new_size * sizeof(RuleMatchCandidateTx)));
// TODO can this be handled more gracefully?
}
det_ctx->tx_candidates = ptmp;
det_ctx->tx_candidates_size = new_size;
SCLogDebug("array expanded from %u to %u elements (%"PRIu64" bytes -> %"PRIu64" bytes)",
old_size, new_size, (uint64_t)(old_size * sizeof(RuleMatchCandidateTx)),
(uint64_t)(new_size * sizeof(RuleMatchCandidateTx))); (void)old_size;
return 1;
}
/* TODO maybe let one with flags win if equal? */
static int
DetectRunTxSortHelper(const void *a, const void *b)
{
const RuleMatchCandidateTx *s0 = a;
const RuleMatchCandidateTx *s1 = b;
if (s1->id == s0->id)
return 0;
else
return s0->id > s1->id ? -1 : 1;
}
#if 0
#define TRACE_SID_TXS(sid,txs,...) \
do { \
char _trace_buf[2048]; \
snprintf(_trace_buf, sizeof(_trace_buf), __VA_ARGS__); \
SCLogNotice("%p/%"PRIu64"/%u: %s", txs->tx_ptr, txs->tx_id, sid, _trace_buf); \
} while(0)
#else
#define TRACE_SID_TXS(sid,txs,...)
#endif
/** \internal
* \brief inspect a rule against a transaction
*
* Inspect a rule. New detection or continued stateful
* detection.
*
* \param stored_flags pointer to stored flags or NULL.
* If stored_flags is set it means we're continueing
* inspection from an earlier run.
*
* \retval bool true sig matched, false didn't match
*/
static bool DetectRunTxInspectRule(ThreadVars *tv,
DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx,
Packet *p,
Flow *f,
const uint8_t in_flow_flags, // direction, EOF, etc
detect: rewrite of the detect engine 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.
8 years ago
void *alstate,
DetectTransaction *tx,
const Signature *s,
uint32_t *stored_flags,
RuleMatchCandidateTx *can,
DetectRunScratchpad *scratch)
{
uint8_t flow_flags = in_flow_flags;
detect: rewrite of the detect engine 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.
8 years ago
const int direction = (flow_flags & STREAM_TOSERVER) ? 0 : 1;
uint32_t inspect_flags = stored_flags ? *stored_flags : 0;
int total_matches = 0;
int file_no_match = 0;
bool retval = false;
bool mpm_before_progress = false; // is mpm engine before progress?
bool mpm_in_progress = false; // is mpm engine in a buffer we will revisit?
/* see if we want to pass on the FLUSH flag */
if ((s->flags & SIG_FLAG_FLUSH) == 0)
flow_flags &=~ STREAM_FLUSH;
detect: rewrite of the detect engine 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.
8 years ago
TRACE_SID_TXS(s->id, tx, "starting %s", direction ? "toclient" : "toserver");
TRACE_SID_TXS(s->id, tx, "FLUSH? %s", (flow_flags & STREAM_FLUSH)?"true":"false");
detect: rewrite of the detect engine 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.
8 years ago
/* for a new inspection we inspect pkt header and packet matches */
if (likely(stored_flags == NULL)) {
TRACE_SID_TXS(s->id, tx, "first inspect, run packet matches");
if (DetectRunInspectRuleHeader(p, f, s, s->flags, s->proto.flags) == 0) {
TRACE_SID_TXS(s->id, tx, "DetectRunInspectRuleHeader() no match");
return false;
}
if (DetectEnginePktInspectionRun(tv, det_ctx, s, f, p, NULL) == false) {
TRACE_SID_TXS(s->id, tx, "DetectEnginePktInspectionRun no match");
detect: rewrite of the detect engine 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.
8 years ago
return false;
}
/* stream mpm and negated mpm sigs can end up here with wrong proto */
if (!(f->alproto == s->alproto || s->alproto == ALPROTO_UNKNOWN)) {
TRACE_SID_TXS(s->id, tx, "alproto mismatch");
return false;
}
}
const DetectEngineAppInspectionEngine *engine = s->app_inspect;
while (engine != NULL) { // TODO could be do {} while as s->app_inspect cannot be null
TRACE_SID_TXS(s->id, tx, "engine %p inspect_flags %x", engine, inspect_flags);
if (!(inspect_flags & BIT_U32(engine->id)) &&
direction == engine->dir)
{
const bool skip_engine = (engine->alproto != 0 && engine->alproto != f->alproto);
/* special case: file_data on 'alert tcp' will have engines
* in the list that are not for us. */
if (unlikely(skip_engine)) {
engine = engine->next;
continue;
}
detect: rewrite of the detect engine 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.
8 years ago
/* engines are sorted per progress, except that the one with
* mpm/prefilter enabled is first */
if (tx->tx_progress < engine->progress) {
SCLogDebug("tx progress %d < engine progress %d",
tx->tx_progress, engine->progress);
break;
}
if (engine->mpm) {
if (tx->tx_progress > engine->progress) {
mpm_before_progress = true;
} else if (tx->tx_progress == engine->progress) {
mpm_in_progress = true;
}
}
/* run callback: but bypass stream callback if we can */
int match;
if (unlikely(engine->stream && can->stream_stored)) {
match = can->stream_result;
TRACE_SID_TXS(s->id, tx, "stream skipped, stored result %d used instead", match);
} else {
KEYWORD_PROFILING_SET_LIST(det_ctx, engine->sm_list);
if (engine->Callback) {
match = engine->Callback(tv, de_ctx, det_ctx,
s, engine->smd, f, flow_flags, alstate, tx->tx_ptr, tx->tx_id);
} else {
BUG_ON(engine->v2.Callback == NULL);
match = engine->v2.Callback(de_ctx, det_ctx, engine,
s, f, flow_flags, alstate, tx->tx_ptr, tx->tx_id);
}
detect: rewrite of the detect engine 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.
8 years ago
TRACE_SID_TXS(s->id, tx, "engine %p match %d", engine, match);
if (engine->stream) {
can->stream_stored = true;
can->stream_result = match;
TRACE_SID_TXS(s->id, tx, "stream ran, store result %d for next tx (if any)", match);
}
}
if (match == DETECT_ENGINE_INSPECT_SIG_MATCH) {
inspect_flags |= BIT_U32(engine->id);
engine = engine->next;
total_matches++;
continue;
} else if (match == DETECT_ENGINE_INSPECT_SIG_MATCH_MORE_FILES) {
/* if the file engine matched, but indicated more
* files are still in progress, we don't set inspect
* flags as these would end inspection for this tx */
engine = engine->next;
total_matches++;
continue;
} else if (match == DETECT_ENGINE_INSPECT_SIG_CANT_MATCH) {
inspect_flags |= DE_STATE_FLAG_SIG_CANT_MATCH;
inspect_flags |= BIT_U32(engine->id);
} else if (match == DETECT_ENGINE_INSPECT_SIG_CANT_MATCH_FILES) {
detect: rewrite of the detect engine 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.
8 years ago
inspect_flags |= DE_STATE_FLAG_SIG_CANT_MATCH;
inspect_flags |= BIT_U32(engine->id);
file_no_match = 1;
}
/* implied DETECT_ENGINE_INSPECT_SIG_NO_MATCH */
if (engine->mpm && mpm_before_progress) {
inspect_flags |= DE_STATE_FLAG_SIG_CANT_MATCH;
inspect_flags |= BIT_U32(engine->id);
}
break;
}
engine = engine->next;
}
TRACE_SID_TXS(s->id, tx, "inspect_flags %x, total_matches %u, engine %p",
inspect_flags, total_matches, engine);
if (engine == NULL && total_matches) {
inspect_flags |= DE_STATE_FLAG_FULL_INSPECT;
TRACE_SID_TXS(s->id, tx, "MATCH");
retval = true;
}
if (stored_flags) {
*stored_flags = inspect_flags;
TRACE_SID_TXS(s->id, tx, "continue inspect flags %08x", inspect_flags);
} else {
// store... or? If tx is done we might not want to come back to this tx
// also... if mpmid tracking is enabled, we won't do a sig again for this tx...
TRACE_SID_TXS(s->id, tx, "start inspect flags %08x", inspect_flags);
if (inspect_flags & DE_STATE_FLAG_SIG_CANT_MATCH) {
if (file_no_match) {
/* if we have a mismatch on a file sig, we need to keep state.
* We may get another file on the same tx (for http and smtp
* at least), so for a new file we need to re-eval the sig.
* Thoughts / TODO:
* - not for some protos that have 1 file per tx (e.g. nfs)
* - maybe we only need this for file sigs that mix with
* other matches? E.g. 'POST + filename', is different than
* just 'filename'.
*/
DetectRunStoreStateTx(scratch->sgh, f, tx->tx_ptr, tx->tx_id, s,
inspect_flags, flow_flags, file_no_match);
detect: rewrite of the detect engine 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.
8 years ago
}
} else if ((inspect_flags & DE_STATE_FLAG_FULL_INSPECT) && mpm_before_progress) {
TRACE_SID_TXS(s->id, tx, "no need to store match sig, "
"mpm won't trigger for it anymore");
if (inspect_flags & DE_STATE_FLAG_FILE_INSPECT) {
TRACE_SID_TXS(s->id, tx, "except that for new files, "
"we may have to revisit anyway");
DetectRunStoreStateTx(scratch->sgh, f, tx->tx_ptr, tx->tx_id, s,
inspect_flags, flow_flags, file_no_match);
}
detect: rewrite of the detect engine 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.
8 years ago
} else if ((inspect_flags & DE_STATE_FLAG_FULL_INSPECT) == 0 && mpm_in_progress) {
TRACE_SID_TXS(s->id, tx, "no need to store no-match sig, "
"mpm will revisit it");
} else {
TRACE_SID_TXS(s->id, tx, "storing state: flags %08x", inspect_flags);
DetectRunStoreStateTx(scratch->sgh, f, tx->tx_ptr, tx->tx_id, s,
inspect_flags, flow_flags, file_no_match);
}
}
return retval;
}
/** \internal
* \brief get a DetectTransaction object
* \retval struct filled with relevant info or all nulls/0s
*/
static DetectTransaction GetDetectTx(const uint8_t ipproto, const AppProto alproto,
void *alstate, const uint64_t tx_id, void *tx_ptr, const int tx_end_state,
detect: rewrite of the detect engine 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.
8 years ago
const uint8_t flow_flags)
{
uint64_t detect_flags;
AppLayerTxData *txd = AppLayerParserGetTxData(ipproto, alproto, tx_ptr);
if (likely(txd != NULL)) {
detect_flags = (flow_flags & STREAM_TOSERVER) ? txd->detect_flags_ts : txd->detect_flags_tc;
} else {
detect_flags = 0;
}
detect: rewrite of the detect engine 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.
8 years ago
if (detect_flags & APP_LAYER_TX_INSPECTED_FLAG) {
SCLogDebug("%"PRIu64" tx already fully inspected for %s. Flags %016"PRIx64,
tx_id, flow_flags & STREAM_TOSERVER ? "toserver" : "toclient",
detect_flags);
DetectTransaction no_tx = { NULL, 0, NULL, NULL, 0, 0, 0, 0, 0, };
detect: rewrite of the detect engine 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.
8 years ago
return no_tx;
}
const int tx_progress = AppLayerParserGetStateProgress(ipproto, alproto, tx_ptr, flow_flags);
const int dir_int = (flow_flags & STREAM_TOSERVER) ? 0 : 1;
DetectEngineState *tx_de_state = AppLayerParserGetTxDetectState(ipproto, alproto, tx_ptr);
DetectEngineStateDirection *tx_dir_state = tx_de_state ? &tx_de_state->dir_state[dir_int] : NULL;
uint64_t prefilter_flags = detect_flags & APP_LAYER_TX_PREFILTER_MASK;
DetectTransaction tx = {
.tx_ptr = tx_ptr,
.tx_id = tx_id,
.tx_data_ptr = (struct AppLayerTxData *)txd,
detect: rewrite of the detect engine 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.
8 years ago
.de_state = tx_dir_state,
.detect_flags = detect_flags,
.prefilter_flags = prefilter_flags,
.prefilter_flags_orig = prefilter_flags,
.tx_progress = tx_progress,
.tx_end_state = tx_end_state,
};
return tx;
}
static inline void StoreDetectFlags(DetectTransaction *tx, const uint8_t flow_flags,
const uint8_t ipproto, const AppProto alproto, const uint64_t detect_flags)
{
AppLayerTxData *txd = (AppLayerTxData *)tx->tx_data_ptr;
if (likely(txd != NULL)) {
if (flow_flags & STREAM_TOSERVER) {
txd->detect_flags_ts = detect_flags;
} else {
txd->detect_flags_tc = detect_flags;
}
}
}
detect: rewrite of the detect engine 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.
8 years ago
static void DetectRunTx(ThreadVars *tv,
DetectEngineCtx *de_ctx,
DetectEngineThreadCtx *det_ctx,
Packet *p,
Flow *f,
DetectRunScratchpad *scratch)
{
const uint8_t flow_flags = scratch->flow_flags;
const SigGroupHead * const sgh = scratch->sgh;
void * const alstate = f->alstate;
const uint8_t ipproto = f->proto;
const AppProto alproto = f->alproto;
const uint64_t total_txs = AppLayerParserGetTxCnt(f, alstate);
uint64_t tx_id_min = AppLayerParserGetTransactionInspectId(f->alparser, flow_flags);
detect: rewrite of the detect engine 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.
8 years ago
const int tx_end_state = AppLayerParserGetStateProgressCompletionStatus(alproto, flow_flags);
AppLayerGetTxIteratorFunc IterFunc = AppLayerGetTxIterator(ipproto, alproto);
AppLayerGetTxIterState state;
memset(&state, 0, sizeof(state));
while (1) {
AppLayerGetTxIterTuple ires = IterFunc(ipproto, alproto, alstate, tx_id_min, total_txs, &state);
if (ires.tx_ptr == NULL)
break;
DetectTransaction tx = GetDetectTx(ipproto, alproto,
alstate, ires.tx_id, ires.tx_ptr, tx_end_state, flow_flags);
detect: rewrite of the detect engine 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.
8 years ago
if (tx.tx_ptr == NULL) {
SCLogDebug("%p/%"PRIu64" no transaction to inspect",
tx.tx_ptr, tx_id_min);
tx_id_min++; // next (if any) run look for +1
goto next;
detect: rewrite of the detect engine 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.
8 years ago
}
tx_id_min = tx.tx_id + 1; // next look for cur + 1
detect: rewrite of the detect engine 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.
8 years ago
uint32_t array_idx = 0;
uint32_t total_rules = det_ctx->match_array_cnt;
total_rules += (tx.de_state ? tx.de_state->cnt : 0);
/* run prefilter engines and merge results into a candidates array */
if (sgh->tx_engines) {
PACKET_PROFILING_DETECT_START(p, PROF_DETECT_PF_TX);
detect: rewrite of the detect engine 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.
8 years ago
DetectRunPrefilterTx(det_ctx, sgh, p, ipproto, flow_flags, alproto,
alstate, &tx);
PACKET_PROFILING_DETECT_END(p, PROF_DETECT_PF_TX);
detect: rewrite of the detect engine 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.
8 years ago
SCLogDebug("%p/%"PRIu64" rules added from prefilter: %u candidates",
tx.tx_ptr, tx.tx_id, det_ctx->pmq.rule_id_array_cnt);
detect: rewrite of the detect engine 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.
8 years ago
total_rules += det_ctx->pmq.rule_id_array_cnt;
if (!(RuleMatchCandidateTxArrayHasSpace(det_ctx, total_rules))) {
RuleMatchCandidateTxArrayExpand(det_ctx, total_rules);
}
for (uint32_t i = 0; i < det_ctx->pmq.rule_id_array_cnt; i++) {
const Signature *s = de_ctx->sig_array[det_ctx->pmq.rule_id_array[i]];
const SigIntId id = s->num;
det_ctx->tx_candidates[array_idx].s = s;
det_ctx->tx_candidates[array_idx].id = id;
det_ctx->tx_candidates[array_idx].flags = NULL;
det_ctx->tx_candidates[array_idx].stream_reset = 0;
array_idx++;
}
} else {
if (!(RuleMatchCandidateTxArrayHasSpace(det_ctx, total_rules))) {
RuleMatchCandidateTxArrayExpand(det_ctx, total_rules);
}
}
/* merge 'state' rules from the regular prefilter */
uint32_t x = array_idx;
for (uint32_t i = 0; i < det_ctx->match_array_cnt; i++) {
const Signature *s = det_ctx->match_array[i];
if (s->app_inspect != NULL) {
detect: rewrite of the detect engine 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.
8 years ago
const SigIntId id = s->num;
det_ctx->tx_candidates[array_idx].s = s;
det_ctx->tx_candidates[array_idx].id = id;
det_ctx->tx_candidates[array_idx].flags = NULL;
det_ctx->tx_candidates[array_idx].stream_reset = 0;
array_idx++;
SCLogDebug("%p/%"PRIu64" rule %u (%u) added from 'match' list",
tx.tx_ptr, tx.tx_id, s->id, id);
detect: rewrite of the detect engine 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.
8 years ago
}
}
SCLogDebug("%p/%"PRIu64" rules added from 'match' list: %u",
tx.tx_ptr, tx.tx_id, array_idx - x); (void)x;
detect: rewrite of the detect engine 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.
8 years ago
/* merge stored state into results */
if (tx.de_state != NULL) {
const uint32_t old = array_idx;
/* if tx.de_state->flags has 'new file' set and sig below has
* 'file inspected' flag, reset the file part of the state */
const bool have_new_file = (tx.de_state->flags & DETECT_ENGINE_STATE_FLAG_FILE_NEW);
if (have_new_file) {
SCLogDebug("%p/%"PRIu64" destate: need to consider new file",
tx.tx_ptr, tx.tx_id);
tx.de_state->flags &= ~DETECT_ENGINE_STATE_FLAG_FILE_NEW;
}
detect: rewrite of the detect engine 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.
8 years ago
SigIntId state_cnt = 0;
DeStateStore *tx_store = tx.de_state->head;
for (; tx_store != NULL; tx_store = tx_store->next) {
SCLogDebug("tx_store %p", tx_store);
SigIntId store_cnt = 0;
for (store_cnt = 0;
store_cnt < DE_STATE_CHUNK_SIZE && state_cnt < tx.de_state->cnt;
store_cnt++, state_cnt++)
{
DeStateStoreItem *item = &tx_store->store[store_cnt];
SCLogDebug("rule id %u, inspect_flags %u", item->sid, item->flags);
if (have_new_file && (item->flags & DE_STATE_FLAG_FILE_INSPECT)) {
/* remove part of the state. File inspect engine will now
* be able to run again */
item->flags &= ~(DE_STATE_FLAG_SIG_CANT_MATCH|DE_STATE_FLAG_FULL_INSPECT|DE_STATE_FLAG_FILE_INSPECT);
SCLogDebug("rule id %u, post file reset inspect_flags %u", item->sid, item->flags);
}
detect: rewrite of the detect engine 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.
8 years ago
det_ctx->tx_candidates[array_idx].s = de_ctx->sig_array[item->sid];
det_ctx->tx_candidates[array_idx].id = item->sid;
det_ctx->tx_candidates[array_idx].flags = &item->flags;
det_ctx->tx_candidates[array_idx].stream_reset = 0;
array_idx++;
}
}
if (old && old != array_idx) {
qsort(det_ctx->tx_candidates, array_idx, sizeof(RuleMatchCandidateTx),
DetectRunTxSortHelper);
SCLogDebug("%p/%"PRIu64" rules added from 'continue' list: %u",
tx.tx_ptr, tx.tx_id, array_idx - old);
detect: rewrite of the detect engine 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.
8 years ago
}
}
det_ctx->tx_id = tx.tx_id;
detect: rewrite of the detect engine 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.
8 years ago
det_ctx->tx_id_set = 1;
det_ctx->p = p;
/* run rules: inspect the match candidates */
for (uint32_t i = 0; i < array_idx; i++) {
RuleMatchCandidateTx *can = &det_ctx->tx_candidates[i];
const Signature *s = det_ctx->tx_candidates[i].s;
uint32_t *inspect_flags = det_ctx->tx_candidates[i].flags;
/* deduplicate: rules_array is sorted, but not deduplicated:
* both mpm and stored state could give us the same sid.
* As they are back to back in that case we can check for it
* here. We select the stored state one. */
if ((i + 1) < array_idx) {
if (det_ctx->tx_candidates[i].s == det_ctx->tx_candidates[i+1].s) {
if (det_ctx->tx_candidates[i].flags != NULL) {
i++;
SCLogDebug("%p/%"PRIu64" inspecting SKIP NEXT: sid %u (%u), flags %08x",
tx.tx_ptr, tx.tx_id, s->id, s->num, inspect_flags ? *inspect_flags : 0);
detect: rewrite of the detect engine 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.
8 years ago
} else if (det_ctx->tx_candidates[i+1].flags != NULL) {
SCLogDebug("%p/%"PRIu64" inspecting SKIP CURRENT: sid %u (%u), flags %08x",
tx.tx_ptr, tx.tx_id, s->id, s->num, inspect_flags ? *inspect_flags : 0);
detect: rewrite of the detect engine 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.
8 years ago
continue;
} else {
// if it's all the same, inspect the current one and skip next.
i++;
SCLogDebug("%p/%"PRIu64" inspecting SKIP NEXT: sid %u (%u), flags %08x",
tx.tx_ptr, tx.tx_id, s->id, s->num, inspect_flags ? *inspect_flags : 0);
detect: rewrite of the detect engine 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.
8 years ago
}
}
}
SCLogDebug("%p/%"PRIu64" inspecting: sid %u (%u), flags %08x",
tx.tx_ptr, tx.tx_id, s->id, s->num, inspect_flags ? *inspect_flags : 0);
detect: rewrite of the detect engine 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.
8 years ago
if (inspect_flags) {
if (*inspect_flags & (DE_STATE_FLAG_FULL_INSPECT|DE_STATE_FLAG_SIG_CANT_MATCH)) {
SCLogDebug("%p/%"PRIu64" inspecting: sid %u (%u), flags %08x ALREADY COMPLETE",
tx.tx_ptr, tx.tx_id, s->id, s->num, *inspect_flags);
detect: rewrite of the detect engine 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.
8 years ago
continue;
}
}
if (inspect_flags) {
/* continue previous inspection */
SCLogDebug("%p/%"PRIu64" Continueing sid %u", tx.tx_ptr, tx.tx_id, s->id);
detect: rewrite of the detect engine 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.
8 years ago
} else {
/* start new inspection */
SCLogDebug("%p/%"PRIu64" Start sid %u", tx.tx_ptr, tx.tx_id, s->id);
detect: rewrite of the detect engine 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.
8 years ago
}
/* call individual rule inspection */
RULE_PROFILING_START(p);
detect: rewrite of the detect engine 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.
8 years ago
const int r = DetectRunTxInspectRule(tv, de_ctx, det_ctx, p, f, flow_flags,
alstate, &tx, s, inspect_flags, can, scratch);
if (r == 1) {
/* match */
DetectRunPostMatch(tv, det_ctx, p, s);
uint8_t alert_flags = (PACKET_ALERT_FLAG_STATE_MATCH|PACKET_ALERT_FLAG_TX);
if (s->action & ACTION_DROP)
alert_flags |= PACKET_ALERT_FLAG_DROP_FLOW;
SCLogDebug("%p/%"PRIu64" sig %u (%u) matched", tx.tx_ptr, tx.tx_id, s->id, s->num);
detect: rewrite of the detect engine 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.
8 years ago
if (!(s->flags & SIG_FLAG_NOALERT)) {
PacketAlertAppend(det_ctx, s, p, tx.tx_id, alert_flags);
detect: rewrite of the detect engine 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.
8 years ago
} else {
DetectSignatureApplyActions(p, s, alert_flags);
}
}
DetectVarProcessList(det_ctx, p->flow, p);
RULE_PROFILING_END(det_ctx, s, r, p);
detect: rewrite of the detect engine 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.
8 years ago
}
det_ctx->tx_id = 0;
det_ctx->tx_id_set = 0;
det_ctx->p = NULL;
/* see if we have any updated state to store in the tx */
uint64_t new_detect_flags = 0;
/* this side of the tx is done */
if (tx.tx_progress >= tx.tx_end_state) {
new_detect_flags |= APP_LAYER_TX_INSPECTED_FLAG;
SCLogDebug("%p/%"PRIu64" tx is done for direction %s. Flag %016"PRIx64,
tx.tx_ptr, tx.tx_id,
detect: rewrite of the detect engine 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.
8 years ago
flow_flags & STREAM_TOSERVER ? "toserver" : "toclient",
new_detect_flags);
}
if (tx.prefilter_flags != tx.prefilter_flags_orig) {
new_detect_flags |= tx.prefilter_flags;
SCLogDebug("%p/%"PRIu64" updated prefilter flags %016"PRIx64" "
"(was: %016"PRIx64") for direction %s. Flag %016"PRIx64,
tx.tx_ptr, tx.tx_id, tx.prefilter_flags, tx.prefilter_flags_orig,
detect: rewrite of the detect engine 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.
8 years ago
flow_flags & STREAM_TOSERVER ? "toserver" : "toclient",
new_detect_flags);
}
if (new_detect_flags != 0 &&
(new_detect_flags | tx.detect_flags) != tx.detect_flags)
{
new_detect_flags |= tx.detect_flags;
SCLogDebug("%p/%"PRIu64" Storing new flags %016"PRIx64" (was %016"PRIx64")",
tx.tx_ptr, tx.tx_id, new_detect_flags, tx.detect_flags);
StoreDetectFlags(&tx, flow_flags, ipproto, alproto, new_detect_flags);
detect: rewrite of the detect engine 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.
8 years ago
}
next:
InspectionBufferClean(det_ctx);
if (!ires.has_next)
break;
detect: rewrite of the detect engine 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.
8 years ago
}
}
/** \brief Apply action(s) and Set 'drop' sig info,
* if applicable */
void DetectSignatureApplyActions(Packet *p,
const Signature *s, const uint8_t alert_flags)
{
PACKET_UPDATE_ACTION(p, s->action);
if (s->action & ACTION_DROP) {
if (p->alerts.drop.action == 0) {
p->alerts.drop.num = s->num;
p->alerts.drop.action = s->action;
p->alerts.drop.s = (Signature *)s;
}
} else if (s->action & ACTION_PASS) {
/* if an stream/app-layer match we enforce the pass for the flow */
if ((p->flow != NULL) &&
(alert_flags & (PACKET_ALERT_FLAG_STATE_MATCH|PACKET_ALERT_FLAG_STREAM_MATCH)))
{
FlowSetNoPacketInspectionFlag(p->flow);
}
}
}
static DetectEngineThreadCtx *GetTenantById(HashTable *h, uint32_t id)
{
/* technically we need to pass a DetectEngineThreadCtx struct with the
* tentant_id member. But as that member is the first in the struct, we
* can use the id directly. */
return HashTableLookup(h, &id, 0);
}
static void DetectFlow(ThreadVars *tv,
DetectEngineCtx *de_ctx, DetectEngineThreadCtx *det_ctx,
Packet *p)
{
if (p->flags & PKT_NOPACKET_INSPECTION) {
/* hack: if we are in pass the entire flow mode, we need to still
* update the inspect_id forward. So test for the condition here,
* and call the update code if necessary. */
const int pass = ((p->flow->flags & FLOW_NOPACKET_INSPECTION));
if (pass) {
uint8_t flags;
if (p->flowflags & FLOW_PKT_TOSERVER) {
flags = STREAM_TOSERVER;
} else {
flags = STREAM_TOCLIENT;
}
flags = FlowGetDisruptionFlags(p->flow, flags);
detect: rewrite of the detect engine 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.
8 years ago
DeStateUpdateInspectTransactionId(p->flow, flags, true);
}
SCLogDebug("p->pcap %"PRIu64": no detection on packet, "
"PKT_NOPACKET_INSPECTION is set", p->pcap_cnt);
return;
}
/* see if the packet matches one or more of the sigs */
detect: rewrite of the detect engine 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.
8 years ago
(void)DetectRun(tv, de_ctx, det_ctx, p);
}
static void DetectNoFlow(ThreadVars *tv,
DetectEngineCtx *de_ctx, DetectEngineThreadCtx *det_ctx,
Packet *p)
{
/* No need to perform any detection on this packet, if the the given flag is set.*/
if ((p->flags & PKT_NOPACKET_INSPECTION) ||
(PACKET_TEST_ACTION(p, ACTION_DROP)))
{
return;
}
/* see if the packet matches one or more of the sigs */
detect: rewrite of the detect engine 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.
8 years ago
DetectRun(tv, de_ctx, det_ctx, p);
return;
}
/** \brief Detection engine thread wrapper.
* \param tv thread vars
* \param p packet to inspect
* \param data thread specific data
* \param pq packet queue
* \retval TM_ECODE_FAILED error
* \retval TM_ECODE_OK ok
*/
TmEcode Detect(ThreadVars *tv, Packet *p, void *data)
{
DEBUG_VALIDATE_PACKET(p);
DetectEngineCtx *de_ctx = NULL;
DetectEngineThreadCtx *det_ctx = (DetectEngineThreadCtx *)data;
if (det_ctx == NULL) {
printf("ERROR: Detect has no thread ctx\n");
goto error;
}
if (unlikely(SC_ATOMIC_GET(det_ctx->so_far_used_by_detect) == 0)) {
(void)SC_ATOMIC_SET(det_ctx->so_far_used_by_detect, 1);
SCLogDebug("Detect Engine using new det_ctx - %p",
det_ctx);
}
/* if in MT mode _and_ we have tenants registered, use
* MT logic. */
if (det_ctx->mt_det_ctxs_cnt > 0 && det_ctx->TenantGetId != NULL)
{
uint32_t tenant_id = p->tenant_id;
if (tenant_id == 0)
tenant_id = det_ctx->TenantGetId(det_ctx, p);
if (tenant_id > 0 && tenant_id < det_ctx->mt_det_ctxs_cnt) {
p->tenant_id = tenant_id;
det_ctx = GetTenantById(det_ctx->mt_det_ctxs_hash, tenant_id);
if (det_ctx == NULL)
return TM_ECODE_OK;
de_ctx = det_ctx->de_ctx;
if (de_ctx == NULL)
return TM_ECODE_OK;
if (unlikely(SC_ATOMIC_GET(det_ctx->so_far_used_by_detect) == 0)) {
(void)SC_ATOMIC_SET(det_ctx->so_far_used_by_detect, 1);
SCLogDebug("MT de_ctx %p det_ctx %p (tenant %u)", de_ctx, det_ctx, tenant_id);
}
} else {
/* use default if no tenants are registered for this packet */
de_ctx = det_ctx->de_ctx;
}
} else {
de_ctx = det_ctx->de_ctx;
}
if (p->flow) {
DetectFlow(tv, de_ctx, det_ctx, p);
} else {
DetectNoFlow(tv, de_ctx, det_ctx, p);
}
return TM_ECODE_OK;
error:
return TM_ECODE_FAILED;
}
/** \brief disable file features we don't need
* Called if we have no detection engine.
*/
void DisableDetectFlowFileFlags(Flow *f)
{
DetectPostInspectFileFlagsUpdate(f, NULL /* no sgh */, STREAM_TOSERVER);
DetectPostInspectFileFlagsUpdate(f, NULL /* no sgh */, STREAM_TOCLIENT);
}
detect: rewrite of the detect engine 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.
8 years ago
#ifdef UNITTESTS
/**
* \brief wrapper for old tests
*/
void SigMatchSignatures(ThreadVars *th_v,
DetectEngineCtx *de_ctx, DetectEngineThreadCtx *det_ctx,
Packet *p)
{
DetectRun(th_v, de_ctx, det_ctx, p);
}
#endif
/*
* TESTS
*/
#ifdef UNITTESTS
#include "tests/detect.c"
#endif