Post on 07-Apr-2015
Building Dynamic Instrumentation Tools
with DynamoRIO
Saman AmarasingheDerek Bruening
Qin Zhao
2DynamoRIO Tutorial at CGO 24 April 2010
Tutorial Outline
• 1:30-1:40 Welcome + DynamoRIO History• 1:40-2:40 DynamoRIO Internals• 2:40-3:00 Examples, Part 1• 3:00-3:15 Break• 3:15-4:15 DynamoRIO API• 4:15-5:15 Examples, Part 2• 5:15-5:30 Feedback
3DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO History
• Dynamo– HP Labs: PA-RISC late 1990’s– x86 Dynamo: 2000
• RIO DynamoRIO– MIT: 2001-2004
• Prior releases– 0.9.1: Jun 2002 (PLDI tutorial)– 0.9.2: Oct 2002 (ASPLOS tutorial)– 0.9.3: Mar 2003 (CGO tutorial)– 0.9.4: Feb 2005
4DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO History
– 0.9.5: Apr 2008 (CGO tutorial)– 1.0 (0.9.6): Sep 2008 (GoVirtual.org launch)
• Determina– 2003-2007– Security company
• VMware– Acquired Determina (and DynamoRIO) in 2007
• Open-source BSD license– Feb 2009: 1.3.1 release– Dec 2009: 1.5.0 release– Apr 2010: 2.0.0 release
DynamoRIO Internals1:30-1:40 Welcome + DynamoRIO History1:40-2:40 DynamoRIO Internals2:40-3:00 Examples, Part 13:00-3:15 Break3:15-4:15 DynamoRIO API4:15-5:15 Examples, Part 25:15-5:30 Feedback
6DynamoRIO Tutorial at CGO 24 April 2010
Typical Modern Application: IIS
7DynamoRIO Tutorial at CGO 24 April 2010
Runtime Interposition Layer
8DynamoRIO Tutorial at CGO 24 April 2010
Design Goals
• Efficient– Near-native performance
• Transparent– Match native behavior
• Comprehensive– Control every instruction, in any application
• Customizable– Adapt to satisfy disparate tool needs
9DynamoRIO Tutorial at CGO 24 April 2010
Challenges of Real-World Apps
• Multiple threads– Synchronization
• Application introspection– Reading of return address
• Transparency corner cases are the norm– Example: access beyond top of stack
• Scalability– Must adapt to varying code sizes, thread counts, etc.
• Dynamically generated code– Performance challenges
10DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient– Software code cache overview– Thread-shared code cache– Cache capacity limits– Data structures
• Transparent• Comprehensive• Customizable
11DynamoRIO Tutorial at CGO 24 April 2010
Direct Code Modification
Kernel32!TerminateProcess:7d4d1028 7c 05 jl 7d4d102f7d4d102a 33 c0 xor %eax,%eax7d4d102c 40 inc %eax7d4d102d eb 08 jmp 7d4d10377d4d102f 50 push %eax7d4d1030 e8 ed 7c 00 00 call 7d4d8d22
e9 37 6f 48 92 jmp <callout>
12DynamoRIO Tutorial at CGO 24 April 2010
Debugger Trap Too Expensive
Kernel32!TerminateProcess:7d4d1028 7c 05 jl 7d4d102f7d4d102a 33 c0 xor %eax,%eax7d4d102c 40 inc %eax7d4d102d eb 08 jmp 7d4d10377d4d102f 50 push %eax7d4d1030 e8 ed 7c 00 00 call 7d4d8d22
cc int3 (breakpoint)
13DynamoRIO Tutorial at CGO 24 April 2010
Variable-Length Instruction Complications
Kernel32!TerminateProcess:7d4d1028 7c 05 jl 7d4d102f7d4d102a 33 c0 xor %eax,%eax7d4d102c 40 inc %eax7d4d102d eb 08 jmp 7d4d10377d4d102f 50 push %eax7d4d1030 e8 ed 7c 00 00 call 7d4d8d22
e9 37 6f 48 92 jmp <callout>
14DynamoRIO Tutorial at CGO 24 April 2010
Entry Point Complications
Kernel32!TerminateProcess:7d4d1028 7c 05 jl 7d4d102f7d4d102a 33 c0 xor %eax,%eax7d4d102c 40 inc %eax7d4d102d eb 08 jmp 7d4d10377d4d102f 50 push %eax7d4d1030 e8 ed 7c 00 00 call 7d4d8d22
e9 37 6f 48 92 jmp <callout>
15DynamoRIO Tutorial at CGO 24 April 2010
Direct Code Modification: Too Limited
• Not transparent– Cannot write jump atomically if crosses cache line– Even if write is atomic, not safe if overwrites part of next
instruction– Jump may span code entry point
• Too limited– Not safe w/o suspending all threads and knowing all entry points– Limited to inserting callouts
• Code displaced by jump is a mini code cache– All the same consistency challenges of larger cache
• Inter-operation issues with other hooks
16DynamoRIO Tutorial at CGO 24 April 2010
We Need Indirection
• Avoid transparency and granularity limitations of directly modifying application code
• Allow arbitrary modifications at unrestricted points in code stream
• Allow systematic, fine-grained modifications to code stream• Guarantee that all code is observed
17DynamoRIO Tutorial at CGO 24 April 2010
Basic Interpreter
fetch
START
decode execute
Slowdown: ~300x
18DynamoRIO Tutorial at CGO 24 April 2010
Improvement #1: Interpreter + Basic Block Cache
basic block builderSTART
dispatch
context switch
BASIC BLOCK CACHE
non-control-flow instructions
Non-control-flow instructions executed from software code cache
Slowdown: 300x 25x
19DynamoRIO Tutorial at CGO 24 April 2010
Improvement #1: Interpreter + Basic Block Cache
basic block builderSTART
dispatch
context switch
BASIC BLOCK CACHE
A
B C
D
A B D
20DynamoRIO Tutorial at CGO 24 April 2010
Example Basic Block Fragment
add %eax, %ecxcmp $4, %eaxjle $0x40106f
add %eax, %ecxcmp $4, %eaxjle <stub0>jmp <stub1>mov %eax, eax-slotmov &dstub0, %eaxjmp context_switchmov %eax, eax-slotmov &dstub1, %eaxjmp context_switch
frag7:
stub0:
stub1:
dstub0target: 0x40106f
dstub1target: fall-thru
21DynamoRIO Tutorial at CGO 24 April 2010
Improvement #2: Linking Direct Branches
basic block builderSTART
dispatch
context switch
BASIC BLOCK CACHE
non-control-flow instructions
Direct branch to existing block can bypass dispatch
Slowdown: 300x 25x 3x
22DynamoRIO Tutorial at CGO 24 April 2010
Improvement #2: Linking Direct Branches
basic block builderSTART
dispatch
context switch
BASIC BLOCK CACHE
A
B C
D
A B D
23DynamoRIO Tutorial at CGO 24 April 2010
Direct Linking
add %eax, %ecxcmp $4, %eaxjle $0x40106f
add %eax, %ecxcmp $4, %eaxjle <frag8>jmp <stub1>mov %eax, eax-slotmov &dstub0, %eaxjmp context_switchmov %eax, eax-slotmov &dstub1, %eaxjmp context_switch
frag7:
stub0:
stub1:
dstub0target: 0x40106f
dstub1target: fall-thru
24DynamoRIO Tutorial at CGO 24 April 2010
Improvement #3: Linking Indirect Branches
basic block builderSTART
dispatch
context switch
BASIC BLOCK CACHE
non-control-flow instructions
indirect branch lookup
Application address mapped to code cache
Slowdown: 300x 25x 3x 1.2x
25DynamoRIO Tutorial at CGO 24 April 2010
Indirect Branch Transformation
retmov %ecx, ecx-slotpop %ecxjmp <ib_lookup>
frag8:
...
...
...
ib_lookup:
26DynamoRIO Tutorial at CGO 24 April 2010
Improvement #4: Trace Building
• Traces reduce branching, improve layout and locality, and facilitate optimizations across blocks– We avoid indirect branch lookup
• Next Executing Tail (NET) trace building scheme [Duesterwald 2000]
A
B
D G
E
C F
H
I
J
K
L
ABEFHD
GKJ
Basic Block Cache Trace Cache
27DynamoRIO Tutorial at CGO 24 April 2010
Incremental NET Trace Building
J
K
G
Basic Block Cache Trace Cache
G
J
K
GK
J
K
GKJ
G
G
28DynamoRIO Tutorial at CGO 24 April 2010
Improvement #4: Trace Building
basic block builder trace selectorSTART
dispatch
context switch
BASIC BLOCK CACHE
TRACE CACHE
non-control-flow instructions
non-control-flow instructions
indirect branch lookup
Slowdown: 300x 26x 3x 1.2x 1.1x
indirect branch stays on trace?
29DynamoRIO Tutorial at CGO 24 April 2010
Base Performance
SPEC CPU2000 DesktopServer
30DynamoRIO Tutorial at CGO 24 April 2010
Sources of Overhead
• Extra instructions– Indirect branch target comparisons– Indirect branch hashtable lookups
• Extra data cache pressure– Indirect branch hashtable
• Branch mispredictions– ret becomes jmp*
• Application code modification
31DynamoRIO Tutorial at CGO 24 April 2010
Time Breakdown for SPECINT
basic block builder trace selectorSTART
dispatch
context switch
BASIC BLOCK CACHE
TRACE CACHE
non-control-flow instructions
non-control-flow instructions
indirect branch lookup
indirect branch stays on trace?
< 1%
~ 0% ~ 2%~ 94%~ 4%
32DynamoRIO Tutorial at CGO 24 April 2010
Not An Ordinary Application
• An application executing in DynamoRIO’s code cache looks different from what the underlying hardware has been tuned for
• The hardware expects:– Little or no dynamic code modification
• Writes to code are expensive– call and ret instructions
• Return Stack Buffer predictor
33DynamoRIO Tutorial at CGO 24 April 2010
Performance Counter Data
34DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient– Software code cache overview– Thread-shared code cache– Cache capacity limits– Data structures
• Transparent• Comprehensive• Customizable
35DynamoRIO Tutorial at CGO 24 April 2010
Threading Model
Thread1
Running Program
Code Caching Runtime System
Operating System
Thread1
Thread1
Thread2
Thread2
Thread2
Thread3
Thread3
Thread3
ThreadN
ThreadN
ThreadN
…
…
…
36DynamoRIO Tutorial at CGO 24 April 2010
Code Space
Thread
Thread
Thread1
Thread
Thread
Thread2
Thread
Thread
Thread1
Thread
Thread
Thread2
Running Program
Thread-Shared Code Cache
Operating System
Thread-Private Code Caches
37DynamoRIO Tutorial at CGO 24 April 2010
Thread-Private versus Thread-Shared
• Thread-private– Less synchronization needed– Absolute addressing for thread-local storage– Thread-specific optimization and instrumentation
• Thread-shared– Scales to many-threaded apps
38DynamoRIO Tutorial at CGO 24 April 2010
Database and Web Server Suite
Benchmark Server Processes
ab low IIS low isolation inetinfo.exe
ab med IIS medium isolation inetinfo.exe, dllhost.exe
guest low IIS low isolation, SQL Server 2000
inetinfo.exe, sqlservr.exe
guest med IIS medium isolation, SQL Server 2000
inetinfo.exe, dllhost.exe, sqlservr.exe
39DynamoRIO Tutorial at CGO 24 April 2010
Memory Impact
ab med guest low guest med
40DynamoRIO Tutorial at CGO 24 April 2010
Performance Impact
41DynamoRIO Tutorial at CGO 24 April 2010
Scalability Limit
42DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient– Software code cache overview– Thread-shared code cache– Cache capacity limits– Data structures
• Transparent• Comprehensive• Customizable
43DynamoRIO Tutorial at CGO 24 April 2010
Added Memory Breakdown
44DynamoRIO Tutorial at CGO 24 April 2010
Code Expansion
original code66%
net jumps8%
indirect branch target handling
7%
exit stubs19%
45DynamoRIO Tutorial at CGO 24 April 2010
Cache Capacity Challenges
• How to set an upper limit on the cache size– Different applications have different working sets and different
total code sizes• Which fragments to evict when that limit is reached
– Without expensive profiling or extensive fragmentation
46DynamoRIO Tutorial at CGO 24 April 2010
Adaptive Sizing Algorithm
• Enlarge cache if warranted by percentage of new fragments that are regenerated
• Target working set of application: don’t enlarge for once-only code
• Low-overhead, incremental, and reactive
47DynamoRIO Tutorial at CGO 24 April 2010
Cache Capacity Settings
• Thread-private:– Working set size matching is on by default– Client may see blocks or traces being deleted in the absence of
any cache consistency event– Can disable capacity management via
• -no_finite_bb_cache• -no_finite_trace_cache
• Thread-shared:– Set to infinite size by default– Can enable capacity management via
• -finite_shared_bb_cache• -finite_shared_trace_cache
48DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient– Software code cache overview– Thread-shared code cache– Cache capacity limits– Data structures
• Transparent• Comprehensive• Customizable
49DynamoRIO Tutorial at CGO 24 April 2010
Two Modes of Code Cache Operation
• Fine-grained scheme– Supports individual code fragment unlink and removal– Separate data structure per code fragment and each of its exits,
memory regions spanned, and incoming links• Coarse-grained scheme
– No individual code fragment control– Permanent intra-cache links– No per-fragment data structures at all– Treat entire cache as a unit for consistency
50DynamoRIO Tutorial at CGO 24 April 2010
Data Structures
• Fine-grained scheme– Data structures are highly tuned and compact
• Coarse-grained scheme– There are no data structures– Savings on applications with large amounts of code are typically
15%-25% of committed memory and 5%-15% of working set
51DynamoRIO Tutorial at CGO 24 April 2010
Status in Current Release
• Fine-grained scheme– Current default
• Coarse-grained scheme– Select with –opt_memory runtime option– Possible performance hit on certain benchmarks– In the future will be the default option– Required for persisted and process-shared caches
52DynamoRIO Tutorial at CGO 24 April 2010
Adaptive Level of Granularity
• Start with coarse-grain caches– Plus freezing and sharing/persisting
• Switch to fine-grain for individual modules or sub-regions of modules after significant consistency events, to avoid expensive entire-module flushes– Support simultaneous fine-grain fragments within coarse-grain
regions for corner cases• Match amount of bookkeeping to amount of code change
– Majority of application code does not need fine-grain
53DynamoRIO Tutorial at CGO 24 April 2010
Many Varieties of Code Caches
• Coarse-grained versus fine-grained• Thread-shared versus thread-private• Basic blocks versus traces
54DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient• Transparent
– Rules of transparency– Cache consistency– Synchronization
• Comprehensive• Customizable
55DynamoRIO Tutorial at CGO 24 April 2010
Transparency
• Do not want to interfere with the semantics of the program• Dangerous to make any assumptions about:
– Register usage– Calling conventions– Stack layout– Memory/heap usage– I/O and other system call use
56DynamoRIO Tutorial at CGO 24 April 2010
Painful, But Necessary
• Difficult and costly to handle corner cases• Many applications will not notice…• …but some will!
– Microsoft Office: Visual Basic generated code, stack convention violations
– COM, Star Office, MMC: trampolines– Adobe Premiere: self-modifying code– VirtualDub: UPX-packed executable– etc.
57DynamoRIO Tutorial at CGO 24 April 2010
Rule 1: Avoid Resource Conflicts
• DynamoRIO system code executes at arbitrary points during application execution
• If DynamoRIO uses the same library routine as the application, it may call that routine in the middle of the same routine being called by the application
• Most library routines are not re-entrant!– Many are thread-safe, but that does not help us
58DynamoRIO Tutorial at CGO 24 April 2010
Rule 1: Avoid Resource Conflicts
Linux Windows
59DynamoRIO Tutorial at CGO 24 April 2010
Rule 2: If It’s Not Broken, Don’t Change It
• Threads• Executable on disk• Application data
– Including the stack!
60DynamoRIO Tutorial at CGO 24 April 2010
Example Transparency Violation
Error
Error
Error
Error
SPEC CPU2000 DesktopServer
Error
Error
Error
Error
Error
Error
61DynamoRIO Tutorial at CGO 24 April 2010
Rule 3: If You Change It, Emulate Original Behavior’s Visible Effects
• Application addresses• Address space• Error transparency• Code cache consistency
62DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient• Transparent
– Rules of transparency– Cache consistency– Synchronization
• Comprehensive• Customizable
63DynamoRIO Tutorial at CGO 24 April 2010
Code Change Mechanisms
I-Cache
Store BFlush BJump B
A:B:C:D:
D-CacheA:B:C:D:
RISCI-Cache
Store BJump B
A:B:C:D:
D-CacheA:B:C:D:
x86
64DynamoRIO Tutorial at CGO 24 April 2010
How Often Does Code Change?
• Not just modification of code!• Removal of code
– Shared library unloading• Replacement of code
– JIT region re-use– Trampoline on stack
65DynamoRIO Tutorial at CGO 24 April 2010
Code Change Events
Memory Unmappings
Generated Code Regions
Modified Code Regions
SPECFP 112 0 0
SPECINT 29 0 0
SPECJVM 7 3373 4591
Excel 144 21 20
Photoshop 1168 40 0
Powerpoint 367 28 33
Word 345 20 6
66DynamoRIO Tutorial at CGO 24 April 2010
Detecting Code Removal
• Example: shared library being unloaded• Requires explicit request by application to operating system• Detect by monitoring system calls (munmap,
NtUnmapViewOfSection)
67DynamoRIO Tutorial at CGO 24 April 2010
Detecting Code Modification
• On x86, no explicit app request required, as the icache is kept consistent in hardware – so any memory write could modify code!
I-Cache
Store BJump B
A:B:C:D:
D-CacheA:B:C:D:
x86
68DynamoRIO Tutorial at CGO 24 April 2010
Page Protection Plus Instrumentation
• Invariant: application code copied to code cache must be read-only– If writable, hide read-only status from application
• Some code cannot or should not be made read-only– Self-modifying code– Windows stack– Code on a page with frequently written data
• Use per-fragment instrumentation to ensure code is not stale on entry and to catch self-modification
69DynamoRIO Tutorial at CGO 24 April 2010
Adaptive Consistency Algorithm
• Use page protection by default– Most code regions are always read-only
• Subdivide written-to regions to reduce flushing cost of write-execute cycle– Large read-only regions, small written-to regions
• Switch to instrumentation if write-execute cycle repeats too often (or on same page)– Switch back to page protection if writes decrease
Bruening et al. “Maintaining Consistency and Bounding Capacity of Software Code Caches” CGO’05
70DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient• Transparent
– Rules of transparency– Cache consistency– Synchronization
• Comprehensive• Customizable
71DynamoRIO Tutorial at CGO 24 April 2010
Synchronization Transparency
• Application thread management should not interfere with the runtime system, and vice versa– Cannot allow the app to suspend a thread holding a runtime
system lock– Runtime system cannot use app locks
72DynamoRIO Tutorial at CGO 24 April 2010
Code Cache Invariant
• App thread suspension requires safe spots where no runtime system locks are held
• Time spent in the code cache can be unbounded→ Our invariant: no runtime system lock can be held while executing in the code cache
73DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient• Transparent• Comprehensive• Customizable
74DynamoRIO Tutorial at CGO 24 April 2010
Above the Operating System
75DynamoRIO Tutorial at CGO 24 April 2010
Kernel-Mediated Control Transfers
message handler
kernel modeuser mode
message pendingsave user context
no message pendingrestore context
time
majority of executed code in a typical Windows
application
76DynamoRIO Tutorial at CGO 24 April 2010
register our own signal handler
DynamoRIO handler
Intercepting Linux Signals
signal handler
kernel modeuser mode
signal pendingsave user context
no signal pendingrestore context
time
77DynamoRIO Tutorial at CGO 24 April 2010
dispatcher
Windows Messages
message handler
kernel modeuser mode
message pendingsave user context
no message pendingrestore context
time
78DynamoRIO Tutorial at CGO 24 April 2010
modify shared library memory image
dispatcher
dispatcher
Intercepting Windows Messages
message handler
kernel modeuser mode
message pendingsave user context
no message pendingrestore context
time
79DynamoRIO Tutorial at CGO 24 April 2010
Must Monitor System Calls
• To maintain control:– Calls that affect the flow of control: register signal handler,
create thread, set thread context, etc.• To maintain transparency:
– Queries of modified state app should not see• To maintain cache consistency:
– Calls that affect the address space• To support cache eviction:
– Interruptible system calls must be redirected
80DynamoRIO Tutorial at CGO 24 April 2010
Operating System Dependencies
• System calls and their numbers– Monitor application’s usage, as well as for our own resource
management– Windows changes the numbers each major rel
• Details of kernel-mediated control flow– Must emulate how kernel delivers events
• Initial injection– Once in, follow child processes
81DynamoRIO Tutorial at CGO 24 April 2010
Internals Outline
• Efficient• Transparent• Comprehensive• Customizable
82DynamoRIO Tutorial at CGO 24 April 2010
Clients
• The engine exports an API for building a client
• System details abstracted away: client focuses on manipulating the code stream
83DynamoRIO Tutorial at CGO 24 April 2010
Client Events
basic block builder trace selectorSTART
dispatch
context switch
BASIC BLOCK CACHE
TRACE CACHE
non-control-flow instructions
non-control-flow instructions
indirect branch lookup
indirect branch stays on trace?
client client
client
Examples: Part I1:30-1:40 Welcome + DynamoRIO History1:40-2:40 DynamoRIO Internals2:40-3:00 Examples, Part 13:00-3:15 Break3:15-4:15 DynamoRIO API4:15-5:15 Examples, Part 25:15-5:30 Feedback
85DynamoRIO Tutorial at CGO 24 April 2010 85
DynamoRIO Examples Part I Outline
• Common Steps of writing a DynamoRIO client
• Dynamic Instruction Counting Example
86DynamoRIO Tutorial at CGO 24 April 2010
Common Steps
• Step 1: Register Events– DR_EXPORT void dr_init(client_id_t id)
• Step2: Implementation– Initialization– Finalization– Instrumentation
• Step 3: Optimization– Optimize the instrumentation to improve the performance
86
Register Function Eventsdr_register_bb_event Basic Block Buildingdr_register_thread_init_event Thread Initializationdr_register_exit_event Process Exit
87DynamoRIO Tutorial at CGO 24 April 2010 87
DynamoRIO Examples Part I Outline
• Common Steps of writing a DynamoRIO client
• Dynamic Instruction Counting Example
88DynamoRIO Tutorial at CGO 24 April 2010 88
A Simplified View of DynamoRIO
basic block builderSTART
dispatch
context switch
BASIC BLOCK CACHE
89DynamoRIO Tutorial at CGO 24 April 2010
Step 1: Register Events
uint num_dyn_instrs;
static void event_init(void);static void event_exit(void);static dr_emit_flags_t event_basic_block(void *drcontext, void *tag, instrlist_t *ilist, bool for_trace, bool translating);
DR_EXPORT void dr_init(client_id_t id) { /* register events */ dr_register_bb_event (event_basic_block); dr_register_exit_event(event_exit); /* process initialization event */ event_init();}
89
90DynamoRIO Tutorial at CGO 24 April 2010
Step 2: Implementation (I)
static void event_init(void) { num_dyn_instrs = 0;}
static void event_exit(void) { dr_printf(“Total number of instruction executed: %u\n”, num_dyn_instrs);}
static dr_emit_flags_t event_basic_block(void *drcontext, void *tag, instrlist_t *ilist, bool for_trace, bool translating) { int num_instrs; num_instrs = ilist_num_instrs(ilist); insert_count_code(drcontext, ilist, num_instrs); return DR_EMIT_DEFAULT;}
90
91DynamoRIO Tutorial at CGO 24 April 2010
Step 2: Implementation (II)
static int ilist_num_instrs(instrlist_t *ilist) { instr_t *instr; int num_instrs = 0; /* iterate over instruction list to count number of instructions */ for (instr = instrlist_first(ilist); instr != NULL; instr = instr_get_next(instr)) num_instrs++; return num_instrs;}
static void do_ins_count(int num_instrs) { num_dyn_instrs += num_instrs; }
static void insert_count_code(void * drcontext, instrlist_t * ilist, int num_instrs) { dr_insert_clean_call(drcontext, ilist, instrlist_first(ilist), do_ins_count, false, 1, OPND_CREATE_INT32(num_instrs));}
91
92DynamoRIO Tutorial at CGO 24 April 2010
Instrumented Basic Block
# switch stack# switch aflags and errorno# save all registers# call do_ins_countpush $0x00000003call $0xb7ef73e4 (do_ins_count)# restore registers# switch aflags and errorno back# switch stack back# application codeadd $0x0000e574 %ebx %ebxtest %al $0x08jz $0xb80e8a98
92
93DynamoRIO Tutorial at CGO 24 April 2010
Step 3: Optimization (I): counter update inlining
static void insert_count_code (void * drcontext, instrlist_t * ilist, int num_instrs) { instr_t *instr, *where; opnd_t opnd1, opnd2;
where = instrlist_first(ilist); /* save aflags */ dr_save_arith_flags(drcontext, ilist, where, SPILL_SLOT_1); /* num_dyn_instrs += num_instrs */ opnd1 = OPND_CREATE_ABSMEM(&num_dyn_instrs, OPSZ_PTR); opnd2 = OPND_CREATE_INT32(num_instrs); instr = INSTR_CREATE_add(drcontext, opnd1, opnd2); instrlist_meta_preinsert(ilist, where, instr); /* restore aflags */ dr_restore_arith_flags(drcontext, ilist, where, SPILL_SLOT_1); }
93
94DynamoRIO Tutorial at CGO 24 April 2010
Instrumented Basic Block
mov %eax %fs:0x0clahf %ahseto %aladd $0x00000003, 0xb7d25030add $0x7f %al %alsahf %ahmov %fs:0x0c %eax# application codeadd $0x0000e574 %ebx %ebxtest %al $0x08jz $0xb7f14a98
94
95DynamoRIO Tutorial at CGO 24 April 2010
Step 3: Optimization (II): aflags stealing
static void insert_count_code (void * drcontext, instrlist_t * ilist, int num_instrs) { … save_aflags = aflags_analysis(ilist); /* save aflags */ if (save_aflags) dr_save_arith_flags(drcontext, ilist, where, SPILL_SLOT_1); /* num_dyn_instrs += num_instrs */ opnd1 = OPND_CREATE_ABSMEM(&num_dyn_instrs, OPSZ_PTR); opnd2 = OPND_CREATE_INT32(num_instrs); instr = INSTR_CREATE_add(drcontext, opnd1, opnd2); instrlist_meta_preinsert(ilist, where, instr); /* restore aflags */ if (save_aflags) dr_restore_arith_flags(drcontext, ilist, where, SPILL_SLOT_1); }
95
96DynamoRIO Tutorial at CGO 24 April 2010
Instrumented Basic Block
add $0x00000003, 0xb7d25030# application codeadd $0x0000e574 %ebx %ebxtest %al $0x08jz $0xb7f14a98
96
97DynamoRIO Tutorial at CGO 24 April 2010
Step 3: Optimization (III): more optimizations
• Using lea (load effective address) instead of addlea [%reg, num_instr] %reg
• Register liveness analysis– Using dead register to avoid register save/restore for lea
• Global aflags/registers analysis– Analyze aflags/registers liveness over CFG
• Trace Optimization– Trace: single-entry multi-exit– Update counters only at trace exits
97
98DynamoRIO Tutorial at CGO 24 April 2010
Other Issues
• Data race on counter update in multithreaded programs– Global lock for every update– Atomic update (lock prefixed add)
• LOCK(instr);– Thread private counter
• Thread-private code cache: different variable at different address• Thread-shared code cache: thread local storage
• 32-bit counter overflow– 64-bit counter:
• Two instructions on 32-bit architecture: add, adc– One 32-bit local counter and one 64-bit global counter
• Instrument to update 32-bit local counter• Update 64-bit global counter using time interrupt
98
DynamoRIO API1:30-1:40 Welcome + DynamoRIO History1:40-2:40 DynamoRIO Internals2:40-3:00 Examples, Part 13:00-3:15 Break3:15-4:15 DynamoRIO API4:15-5:15 Examples, Part 25:15-5:30 Feedback
100DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
101DynamoRIO Tutorial at CGO 24 April 2010
Clients
• The engine exports an API for building a client
• System details abstracted away: client focuses on manipulating the code stream
102DynamoRIO Tutorial at CGO 24 April 2010
Cross-Platform Clients
• DynamoRIO API presents a consistent interface that works across platforms– Windows versus Linux– 32-bit versus 64-bit– Thread-private versus thread-shared
• Same client source code generally works on all combinations of platforms
• Some exceptions, noted in the documentation
103DynamoRIO Tutorial at CGO 24 April 2010
Building a Client
• Include DR API header file– #include “dr_api.h”
• Set platform defines– WINDOWS or LINUX – X86_32 or X86_64
• Export a dr_init function– DR_EXPORT void dr_init (client_id_t client_id)
• Build a shared library
104DynamoRIO Tutorial at CGO 24 April 2010
Auto-Configure Using CMake
add_library(myclient SHARED myclient.c) find_package(DynamoRIO)if (NOT DynamoRIO_FOUND) message(FATAL_ERROR "DynamoRIO package
required to build")endif(NOT DynamoRIO_FOUND)configure_DynamoRIO_client(myclient)
105DynamoRIO Tutorial at CGO 24 April 2010
CMake
• Build system converted to CMake when open-sourced– Switch from frozen toolchain to supporting range of tools
• CMake generates build files for native compiler of choice– Makefiles for UNIX, nmake, etc.– Visual Studio project files
• http://www.cmake.org/
106DynamoRIO Tutorial at CGO 24 April 2010
Library Usage and Transparency
• For best transparency: completely self-contained client– Imports only from DynamoRIO API– -nodefaultlibs or /nodefaultlib
• On Windows:– String and utility routines provided by forwards to ntdll– Cl.exe /MT static copy of C/C++ libraries– Custom loader loads private copy of client dependences
• On Linux:– Use ld –wrap to redirect malloc calls to DR’s heap– Older distributions shipped suitable static C/C++ lib– Newer distros: need to build yourself– Coming soon: custom loader for private copy of libs
107DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO Extensions
• DynamoRIO API is extended via libraries called Extensions• Both static and shared supported• Built and packaged with DynamoRIO• Easy for a client to use
– use_DynamoRIO_extension(myclient drsyms)• Current Extensions:
– drsyms: symbol lookup (currently Windows-only)– drcontainers: hashtable
• Coming soon:– Umbra: shadow memory framework– Your utility library or framework contribution!
108DynamoRIO Tutorial at CGO 24 April 2010
Application Configuration
• File-based scheme• Per-user local files
– $HOME/.dynamorio/ on Linux– $USERPROFILE/dynamorio/ on Windows
• Global files– /etc/dynamorio/ on Linux– Registry-specified directory on Windows
• Files are lists of var=value
109DynamoRIO Tutorial at CGO 24 April 2010
Deploying Clients
• One-step configure-and-run usage model:– drrun <client> <options> <app cmdline>– Uses an invisible temporary one-time configuration file– Overrides any regular config file
• Two-step usage model giving control over children:– drconfig –reg <appname> <client> <options>– drinject <app cmdline>
• Systemwide injection:– drconfig –syswide_on –reg <appname> <client> <options>– <run app normally>
110DynamoRIO Tutorial at CGO 24 April 2010
Deploying Clients On Linux
• drrun and drinject scripts: LD_PRELOAD-based– Take over after statically-dependent shared libs but before exe
• Suid apps ignore LD_PRELOAD– Place libdrpreload.so's full path in /etc/ld.so.preload– Copy libdynamorio.so to /usr/lib
• In the future:– Attach– Early injection
111DynamoRIO Tutorial at CGO 24 April 2010
Deploying Clients On Windows
• drinject and drrun injection– Currently after all shared libs are initialized
• From-parent injection– Early: before any shared libs are loaded
• Systemwide injection via –syswide_on– Requires administrative privileges– Launch app normally: no need to run via drinject/drrun– Moderately early: during user32.dll initialization
• In the future:– Earliest injection for drrun/drinject and from-parent
112DynamoRIO Tutorial at CGO 24 April 2010
Non-Standard Deployment
• Standalone API– Use DynamoRIO as a library of IA-32/AMD64 manipulation
routines• Start/Stop API
– Can instrument source code with where DynamoRIO should control the application
113DynamoRIO Tutorial at CGO 24 April 2010
Runtime Options
• Pass options to drconfig/drrun• A large number of options; the most relevant are:
– -code_api– -client <client lib> <client ops> <client id>– -thread_private– -tracedump_text and –tracedump_binary– -prof_pcs
114DynamoRIO Tutorial at CGO 24 April 2010
Runtime Options For Debugging
• Notifications:– -stderr_mask 0xN– -msgbox_mask 0xN
• Windows:– -no_hide
• Debug-build-only:– -loglevel N– -ignore_assert_list ‘*’
115DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
116DynamoRIO Tutorial at CGO 24 April 2010
Client Events
basic block builder trace selectorSTART
dispatch
context switch
BASIC BLOCK CACHE
TRACE CACHE
non-control-flow instructions
non-control-flow instructions
indirect branch lookup
indirect branch stays on trace?
client client
client
117DynamoRIO Tutorial at CGO 24 April 2010
Client Events: Code Stream
• Client has opportunity to inspect and potentially modify every single application instruction, immediately before it executes
• Entire application code stream– Basic block creation event: can modify the block– For comprehensive instrumentation tools
• Or, focus on hot code only– Trace creation event: can modify the trace– Custom trace creation: can determine trace end condition– For optimization and profiling tools
118DynamoRIO Tutorial at CGO 24 April 2010
Simplifying Client View
• Several optimizations disabled– Elision of unconditional branches– Indirect call to direct call conversion– Shared cache sizing– Process-shared and persistent code caches
• Future release will give client control over optimizations
119DynamoRIO Tutorial at CGO 24 April 2010
Basic Block Event
static dr_emit_flags_tevent_basic_block(void *drcontext, void *tag, instrlist_t *bb, bool for_trace, bool translating) { instr_t *inst; for (inst = instrlist_first(bb); inst != NULL; inst = instr_get_next(inst)) { /* … */ }
return DR_EMIT_DEFAULT;}DR_EXPORT void dr_init(client_id_t id) { dr_register_bb_event(event_basic_block);}
120DynamoRIO Tutorial at CGO 24 April 2010
Trace Event
static dr_emit_flags_tevent_trace(void *drcontext, void *tag, instrlist_t *trace, bool translating) { instr_t *inst; for (inst = instrlist_first(trace); inst != NULL; inst = instr_get_next(inst)) { /* … */ }
return DR_EMIT_DEFAULT;}DR_EXPORT void dr_init(client_id_t id) { dr_register_trace_event(event_trace);}
121DynamoRIO Tutorial at CGO 24 April 2010
Client Events: Application Actions
• Application thread creation and deletion• Application library load and unload• Application exception (Windows)
– Client chooses whether to deliver or suppress• Application signal (Linux)
– Client chooses whether to deliver, suppress, bypass the app handler, or redirect control
122DynamoRIO Tutorial at CGO 24 April 2010
Client Events: Application System Calls
• Application pre- and post- system call– Platform-independent system call parameter access– Client can modify:
• Return value in post-, or set value and skip syscall in pre-• Call number• Params
– Client can invoke an additional system call as the app
123DynamoRIO Tutorial at CGO 24 April 2010
Client Events: Bookkeeping
• Initialization and Exit– Entire process– Each thread– Child of fork (Linux-only)
• Basic block and trace deletion during cache management• Nudge received
– Used for communication into client• Itimer fired (Linux-only)
124DynamoRIO Tutorial at CGO 24 April 2010
Multiple Clients
• It is each client's responsibility to ensure compatibility with other clients– Instruction stream modifications made by one client are visible to
other clients• At client registration each client is given a priority
– dr_init() called in priority order (priority 0 called first and thus registers its callbacks first)
• Event callbacks called in reverse order of registration– Gives precedence to first registered callback, which is given the
final opportunity to modify the instruction stream or influence DynamoRIO's operation
125DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
126DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API: General Utilities
• Transparency support– Separate memory allocation and I/O– Alternate stack
• Thread support– Thread-local memory– Simple mutexes– Thread-private code caches, if requested
127DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API: General Utilities, Cont’d
• Communication– Nudges: ping from external process– File creation, reading, and writing
• Sideline support– Create new client-only thread– Thread-private itimer (Linux-only)
128DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API: General Utilities, Cont’d
• Application inspection– Address space querying– Module iterator– Processor feature identification– Symbol lookup (currently Windows-only)
• Third-party library support– If transparency is maintained!– -wrap support on Linux– ntdll.dll link support on Windows– Custom loader for private library copy on Windows
129DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO Heap
• Three flavors:– Thread-private: no synchronization; thread lifetime– Global: synchronized, process lifetime– “Non-heap”: for generated code, etc.– No header on allocated memory: low overhead but must pass
size on free• Leak checking
– Debug build complains at exit if memory was not deallocated
130DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
131DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API: Instruction Representation
• Full IA-32/AMD64 instruction representation• Instruction creation with auto-implicit-operands• Operand iteration• Instruction lists with iteration, insertion, removal• Decoding at various levels of detail• Encoding
132DynamoRIO Tutorial at CGO 24 April 2010
Instruction Representation
cmp
shl
movzx
sub
mov
lea
jnl
8b 46 0c
8d 34 01
2b 46 1c
0f b7 4e 08
c1 e1 07
3b c1
0f 8d a2 0a 00 00 $0x77f52269
%eax %ecx
$0x07 %ecx -> %ecx
0x8(%esi) -> %ecx
0x1c(%esi) %eax -> %eax
0xc(%esi) -> %eax
(%ecx,%eax,1) -> %esi
WCPAZSO
WCPAZSO
-
WCPAZSO
-
-
RSOopcode operandsraw bytes eflags
133DynamoRIO Tutorial at CGO 24 April 2010
Instruction Representation
cmp
shl
movzx
sub
mov
lea
jnl
c1 e1 07
3b c1
0f 8d a2 0a 00 00 $0x77f52269
%eax %ecx
$0x07 %ecx -> %ecx
0x8(%edi) -> %ecx
0x1c(%edi) %eax -> %eax
0xc(%edi) -> %eax
(%ecx,%eax,1) -> %edi
WCPAZSO
WCPAZSO
-
WCPAZSO
-
-
RSOopcode operandsraw bytes eflags
134DynamoRIO Tutorial at CGO 24 April 2010
Instruction Creation
• Method 1: use the INSTR_CREATE_opcode macros that fill in implicit operands automatically:instr_t *instr = INSTR_CREATE_dec(dcontext,
opnd_create_reg(REG_EDX)); • Method 2: specify opcode + all operands (including implicit
operands):instr_t *instr = instr_create(dcontext);instr_set_opcode(instr, OP_dec);instr_set_num_opnds(dcontext, instr, 1, 1);instr_set_dst(instr, 0, opnd_create_reg(REG_EDX));instr_set_src(instr, 0, opnd_create_reg(REG_EDX));
135DynamoRIO Tutorial at CGO 24 April 2010
Linear Control Flow
• Both basic blocks and traces are linear
• Instruction sequences are all single-entrance, multiple-exit
• Greatly simplifies analysis algorithms
136DynamoRIO Tutorial at CGO 24 April 2010
64-Bit Versus 32-Bit
• 32-bit build of DynamoRIO only handles 32-bit code• 64-bit build of DynamoRIO decodes/encodes both 32-bit and
64-bit code– Current release does not support executing applications that mix
the two• IR is universal: covers both 32-bit and 64-bit
– Abstracts away underlying mode
137DynamoRIO Tutorial at CGO 24 April 2010
64-Bit Thread and Instruction Modes
• When going to or from the IR, the thread mode and instruction mode determine how instrs are interpreted
• When decoding, current thread’s mode is used– Default is 64-bit for 64-bit DynamoRIO– Can be changed with set_x86_mode()
• When encoding, that instruction’s mode is used– When created, set to mode of current thread– Can be changed with instr_set_x86_mode()
138DynamoRIO Tutorial at CGO 24 April 2010
64-Bit Clients
• Define X86_64 before including header files when building a 64-bit client
• Convenience macros for printf formats, etc. are provided– E.g.:
• printf(“Pointer is ”PFX“\n”, p);• Use “X” macros for cross-platform registers
– REG_XAX is REG_EAX when compiled 32-bit, and REG_RAX when compiled 64-bit
139DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API: Code Manipulation
• Processor information• State preservation
– Eflags, arith flags, floating-point state, MMX/SSE state– Spill slots, TLS
• Clean calls to C code• Dynamic instrumentation
– Replace code in the code cache• Branch instrumentation
– Convenience routines
140DynamoRIO Tutorial at CGO 24 April 2010
Processor Information
• Processor type– proc_get_vendor(), proc_get_family(), proc_get_type(),
proc_get_model(), proc_get_stepping(), proc_get_brand_string()• Processor features
– proc_has_feature(), proc_get_all_feature_bits()• Cache information
– proc_get_cache_line_size(), proc_is_cache_aligned(), proc_bump_to_end_of_cache_line(), proc_get_containing_page()
– proc_get_L1_icache_size(), proc_get_L1_dcache_size(), proc_get_L2_cache_size(), proc_get_cache_size_str()
141DynamoRIO Tutorial at CGO 24 April 2010
State Preservation
• Spill slots for registers– 3 fast slots, 6/14 slower slots– dr_save_reg(), dr_restore_reg(), and dr_reg_spill_slot_opnd() – from C code: dr_read_saved_reg(), dr_write_saved_reg()
• Dedicated TLS field for thread-local data– dr_insert_read_tls_field(), dr_insert_write_tls_field() – from C code: dr_get_tls_field(), dr_set_tls_field()
• Arithmetic flag preservation– dr_save_arith_flags(), dr_restore_arith_flags()
• Floating-point/MMX/SSE state– dr_insert_save_fpstate(), dr_insert_restore_fpstate()
142DynamoRIO Tutorial at CGO 24 April 2010
Thread-Local Storage (TLS)
• Absolute addressing– Thread-private only
• Application stack– Not reliable or transparent
• Stolen register– Performance hit
• Segment– Best solution for thread-shared
143DynamoRIO Tutorial at CGO 24 April 2010
Clean Calls
if (instr_is_mbr(instr)) { app_pc address = instr_get_app_pc(instr); uint opcode = instr_get_opcode(instr); instr_t *nxt = instr_get_next(instr); dr_insert_clean_call(drcontext, ilist, nxt, (void *) at_mbr, false/*don't need to save fp state*/, 2 /* 2 parameters */, /* opcode is 1st parameter */ OPND_CREATE_INT32(opcode), /* address is 2nd parameter */ OPND_CREATE_INTPTR(address));}
• Saved interrupted application state can be accessed using dr_get_mcontext() and modified using dr_set_mcontext()
144DynamoRIO Tutorial at CGO 24 April 2010
Dynamic Instrumentation
• Thread-shared: flush all code corresponding to application address and then re-instrument when re-executed– Can flush from clean call, and use dr_redirect_execution() since
cannot return to potentially flushed cache fragment• Thread-private: can also replace particular fragment (does not
affect other potential copies of the source app code)– dr_replace_fragment()
145DynamoRIO Tutorial at CGO 24 April 2010
Flushing the Cache
• Immediately deleting or replacing individual code cache fragments is available for thread-private caches– Only removes from that thread’s cache
• Two basic types of thread-shared flush:– Non-precise: remove all entry points but let target cache code be
invalidated and freed lazily– Precise/synchronous:
• Suspend the world• Relocate threads inside the target cache code• Invalidate and free the target code immediately
146DynamoRIO Tutorial at CGO 24 April 2010
Flushing the Cache
• Thread-shared flush API routines:– dr_unlink_flush_region(): non-precise flush– dr_flush_region(): synchronous flush– dr_delay_flush_region():
• No action until a thread exits code cache on its own• If provide a completion callback, synchronous once triggered• Without a callback, non-precise
147DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
148DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API: Translation
• Translation refers to the mapping of a code cache machine state (program counter, registers, and memory) to its corresponding application state– The program counter always needs to be translated– Registers and memory may also need to be translated
depending on the transformations applied when copying into the code cache
149DynamoRIO Tutorial at CGO 24 April 2010
Translation Case 1: Fault
• Exception and signal handlers are passed machine context of the faulting instruction.
• For transparency, that context must be translated from the code cache to the original code location
• Translated location should be where the application would have had the fault or where execution should be resumed
user context
faulting instr.
user context
faulting instr.
150DynamoRIO Tutorial at CGO 24 April 2010
Translation Case 2: Relocation
• If one application thread suspends another, or DynamoRIO suspends all threads for a synchronous cache flush:– Need suspended target thread in a safe spot– Not always practical to wait for it to arrive at a safe spot (if in a
system call, e.g.)• DynamoRIO forcibly relocates the thread
– Must translate its state to the proper application state at which to resume execution
151DynamoRIO Tutorial at CGO 24 April 2010
Translation Approaches
• Two approaches to program counter translation:– Store mappings generated during fragment building
• High memory overhead (> 20% for some applications, because it prevents internal storage optimizations) even with highly optimized difference-based encoding. Costly for something rarely used.
– Re-create mapping on-demand from original application code• Cache consistency guarantees mean the corresponding application
code is unchanged• Requires idempotent code transformations
• DynamoRIO supports both approaches– The engine mostly uses the on-demand approach, but stored
mappings are occasionally needed
152DynamoRIO Tutorial at CGO 24 April 2010
Instruction Translation Field
• Each instruction contains a translation field• Holds the application address that the instruction corresponds
to• Set via instr_set_translation()
153DynamoRIO Tutorial at CGO 24 April 2010
Context Translation Via Re-Creation
C1: mov %ebx, %ecxC2: add %eax, (%ecx)C3: cmp $4, (%eax)C4: jle <stub0>C5: jmp <stub1>
A1: mov %ebx, %ecxA2: add %eax, (%ecx)A3: cmp $4, (%eax)A4: jle 710349fb
D1: (A1) mov %ebx, %ecxD2: (A2) add %eax, (%ecx)D3: (A3) cmp $4, (%eax)D4: (A4) jle <stub0>D5: (A4) jmp <stub1>
154DynamoRIO Tutorial at CGO 24 April 2010
Meta vs. Non-Meta Instructions
• Non-Meta instructions are treated as application instructions– They must have translations– Control flow changing instructions are modified to retain
DynamoRIO control and result in cache populating• Meta instructions are added instrumentation code
– Not treated as part of the application (e.g., calls run natively)– Cannot fault, so translations not needed
• Meta-may-fault instructions– Can fault, but should not be “interpreted”: won’t modify app code– Fault typically deliberate and handled by client
• Xref instr_set_ok_to_mangle() and instr_set_meta_may_fault()
155DynamoRIO Tutorial at CGO 24 April 2010
Client Translation Support
• Instruction lists passed to clients are annotated with translation information– Read via instr_get_translation()– Clients are free to delete instructions, change instructions and
their translations, and add new meta and non-meta instructions (see dr_register_bb_event() for restrictions)
– An idempotent client that restricts itself to deleting app instructions and adding non-faulting meta instructions can ignore translation concerns
– DynamoRIO takes care of instructions added by API routines (insert_clean_call(), etc.)
• Clients can choose between storing or regenerating translations on a fragment by fragment basis.
156DynamoRIO Tutorial at CGO 24 April 2010
Client Regenerated Translations
• Client returns DR_EMIT_DEFAULT from its bb or trace event callback
• Client bb & trace event callbacks are re-called when translations are needed with translating==true
• Client must exactly duplicate transformations performed when the block was generated
• Client must set translation field for all added non-meta instructions and all meta-may-fault instructions– This is true even if translating==false since DynamoRIO may
decide it needs to store translations anyway
157DynamoRIO Tutorial at CGO 24 April 2010
Client Stored Translations
• Client returns DR_EMIT_STORE_TRANSLATIONS from its bb or trace event callback
• Client must set translation field for all added non-meta instructions and all meta-may-fault instructions
• Client bb or trace hook will not be re-called with translating==true
158DynamoRIO Tutorial at CGO 24 April 2010
Register State Translation
• Translation may be needed at a point where some registers are spilled to memory– During indirect branch or RIP-relative mangling, e.g.
• DynamoRIO walks fragment up to translation point, tracking register spills and restores– Special handling for stack pointer around indirect calls and
returns• DynamoRIO tracks client spills and restores implicitly added
by API routines– Clean calls, etc.– Explicit spill/restore (e.g., dr_save_reg()) client’s responsibility
159DynamoRIO Tutorial at CGO 24 April 2010
Client Register State Translation
• If a client adds its own register spilling/restoring code or changes register mappings it must register for the restore state event to correct the context
• The same event can also be used to fix up the application’s view of memory
• DynamoRIO does not internally store this kind of translation information ahead of time when the fragment is built– The client must maintain its own data structures
160DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
161DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO versus Pin
• Basic interface is fundamentally different• Pin = insert callout/trampoline only
– Not so different from tools that modify the original code: Dyninst, Vulcan, Detours
– Uses code cache only for transparency• DynamoRIO = arbitrary code stream modifications
– Only feasible with a code cache– Takes full advantage of power of code cache– General IA-32/AMD64 decode/encode/IR support
162DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO versus Pin
• Pin = insert callout/trampoline only– Pin tries to inline and optimize– Client has little control or guarantee over final performance
• DynamoRIO = arbitrary code stream modifications– Client has full control over all inserted instrumentation– Result can be significant performance difference
• PiPA Memory Profiler + Cache Simulator: 3.27x speedup w/ DynamoRIO vs 2.6x w/ Pin
163DynamoRIO Tutorial at CGO 24 April 2010
Base Performance Comparison (No Tool)
171%
121%
164DynamoRIO Tutorial at CGO 24 April 2010
Base Memory Comparison
15MB
2.6MB
44MB
3.5MB
165DynamoRIO Tutorial at CGO 24 April 2010
BBCount Pin Tool
static int bbcount;VOID PIN_FAST_ANALYSIS_CALL docount() { bbcount++; }VOID Trace(TRACE trace, VOID *v) { for (BBL bbl = TRACE_BblHead(trace); BBL_Valid(bbl); bbl = BBL_Next(bbl)) { BBL_InsertCall(bbl, IPOINT_ANYWHERE, AFUNPTR(docount), IARG_FAST_ANALYSIS_CALL, IARG_END); }}int main(int argc, CHAR *argv[]) { PIN_InitSymbols(); PIN_Init(argc, argv); TRACE_AddInstrumentFunction(Trace, 0); PIN_StartProgram(); return 0;}
166DynamoRIO Tutorial at CGO 24 April 2010
BBCount DynamoRIO Tool
static int global_count;static dr_emit_flags_tevent_basic_block(void *drcontext, void *tag, instrlist_t *bb, bool for_trace, bool translating) { instr_t *instr, *first = instrlist_first(bb); uint flags; /* Our inc can go anywhere, so find a spot where flags are dead. * Technically this can be unsafe if app reads flags on fault => * stop at instr that can fault, or supply runtime op */ for (instr = first; instr != NULL; instr = instr_get_next(instr)) { flags = instr_get_arith_flags(instr); /* OP_inc doesn't write CF but not worth distinguishing */ if (TESTALL(EFLAGS_WRITE_6, flags) && !TESTANY(EFLAGS_READ_6, flags)) break; } if (instr == NULL) dr_save_arith_flags(drcontext, bb, first, SPILL_SLOT_1); instrlist_meta_preinsert(bb, (instr == NULL) ? first : instr, INSTR_CREATE_inc(drcontext, OPND_CREATE_ABSMEM((byte *)&global_count, OPSZ_4))); if (instr == NULL) dr_restore_arith_flags(drcontext, bb, first, SPILL_SLOT_1); return DR_EMIT_DEFAULT;}DR_EXPORT void dr_init(client_id_t id) { dr_register_bb_event(event_basic_block);}
167DynamoRIO Tutorial at CGO 24 April 2010
BBCount: Pin Inlining Importance
569%
231%
168DynamoRIO Tutorial at CGO 24 April 2010
BBCount Performance Comparison
226%185%
233%
169DynamoRIO Tutorial at CGO 24 April 2010
DynamoRIO API Outline
• Building and Deploying• Events• Utilities• Instruction Manipulation• State Translation• Comparison with Pin• Troubleshooting
170DynamoRIO Tutorial at CGO 24 April 2010
Obtaining Help
• Read the documentation– http://dynamorio.org/docs/
• Look at the sample clients– In the documentation– In the release package: samples/
• Ask on the DynamoRIO Users discussion forum/mailing list– http://groups.google.com/group/dynamorio-users
171DynamoRIO Tutorial at CGO 24 April 2010
Debugging Clients
• Use the DynamoRIO debug build for asserts– Often point out the problem
• Use logging– -loglevel N– stored in logs/ subdir of DR install dir
• Attach a debugger– gdb or windbg– -msgbox_mask 0xN– -no_hide– windbg: .reload myclient.dll=0xN
• More tips:– http://code.google.com/p/dynamorio/wiki/Debugging
172DynamoRIO Tutorial at CGO 24 April 2010
Reporting Bugs
• Search the Issue Tracker off http://dynamorio.org first– http://code.google.com/p/dynamorio/issues/list
• File a new Issue if not found• Follow conventions on wiki
– http://code.google.com/p/dynamorio/wiki/BugReporting – CRASH, APP CRASH, HANG, ASSERT
• Example titles:– CRASH (1.3.1 calc.exe)
vm_area_add_fragment:vmareas.c(4466) – ASSERT (1.3.0 suite/tests/common/segfault)
study_hashtable:fragment.c:1745 ASSERT_NOT_REACHED
173DynamoRIO Tutorial at CGO 24 April 2010
Changes From Prior Releases
• Backward compatible with 1.0 (0.9.6) and above– Except configuration and deployment scheme and tools:
switched to file-based scheme to support unprivileged and parallel execution on Windows
• Not backward compatible with 0.9.1-0.9.5
Examples: Part 21:30-1:40 Welcome + DynamoRIO History1:40-2:40 DynamoRIO Internals2:40-3:00 Examples, Part 13:00-3:15 Break3:15-4:15 DynamoRIO API4:15-5:15 Examples, Part 25:15-5:30 Feedback
175DynamoRIO Tutorial at CGO 24 April 2010 175
More Examples
• Dynamic Optimization– Strength Reduction– Software Prefetching
• Profiling– Memory Reference Trace– PiPA
• Shadow Memory– Umbra
• Dr. Memory
176DynamoRIO Tutorial at CGO 24 April 2010 176
Dynamic Optimization Opportunities
• Traditional compiler optimizations– Compiler has limited view: application assembled at runtime– Some shipped products are built without optimizations
• Microarchitecture-specific optimizations– Feature set and relative performance of instructions varies– Combinatorial blowup if done statically
• Adaptive optimizations– Need runtime information: prior profiling runs not always
representative– Execution phase changes during execution
177DynamoRIO Tutorial at CGO 24 April 2010 177
Dynamic Optimization in DynamoRIO
• Traces are natural unit for optimization– Focus only on hot code– Cross procedure, file and module boundaries
• Linear control flow– Single-entry, multi-exit simplifies analysis
• Support for adaptive optimization– Can replace traces dynamically
178DynamoRIO Tutorial at CGO 24 April 2010 178
DynamoRIO with Trace Optimization
basic block builder trace selectorSTART
dispatch
context switch
BASIC BLOCK CACHE
TRACE CACHE
indirect branch lookup
indirect branch stays on trace?
179DynamoRIO Tutorial at CGO 24 April 2010 179
Strength Reduction: inc to add
• Pentium 4– inc is slower add 1 – dec is slower than sub 1
• Pentium 3– inc is faster add 1 – dec is faster than sub 1
• Microarchitecture-specific optimization best performed dynamically
180DynamoRIO Tutorial at CGO 24 April 2010 180
Look for inc / dec
EXPORT void dr_init() { if (proc_get_family() == FAMILY_PENTIUM_IV) dr_register_trace_event(event_trace);}static void event_trace(void *drcontext, app_pc tag, instrlist_t *trace, bool xl8) { instr_t *instr, *next_instr; int opcode; for (instr = instrlist_first(bb); instr != NULL; instr = next_instr) { next_instr = instr_get_next(instr); opcode = instr_get_opcode(instr); if (opcode == OP_inc || opcode == OP_dec) replace_inc_with_add(drcontext, instr, trace); } }}static bool replace_inc_with_add(void *drcontext, instr_t *instr, instrlist_t *trace) { instr_t *in; uint eflags; int opcode = instr_get_opcode(instr); bool ok_to_replace = false;
for (in = instr; in != NULL; in = instr_get_next(in)) { eflags = instr_get_arith_flags(in); if ((eflags & EFLAGS_READ_CF) != 0) return false; if ((eflags & EFLAGS_WRITE_CF) != 0) { ok_to_replace = true; break; } if (instr_is_exit_cti(in)) return false; } if (!ok_to_replace) return false; if (opcode == OP_inc) in = INSTR_CREATE_add(drcontext, instr_get_dst(instr, 0), OPND_CREATE_INT8(1)); else in = INSTR_CREATE_sub(drcontext, instr_get_dst(instr, 0), OPND_CREATE_INT8(1)); instr_set_prefixes(in, instr_get_prefixes(instr)); instrlist_replace(trace, instr, in); instr_destroy(drcontext, instr); return true;}
Pentium 4?
Replace with add / sub
Ensure eflags change ok
181DynamoRIO Tutorial at CGO 24 April 2010 181
Strength Reduction Results
0.00.10.20.30.40.50.60.70.80.91.01.11.2
amm
p
appl
u
apsi art
equa
ke
mes
a
mgr
id
sixt
rack
swim
wup
wis
e
har.
mea
n
B e nchmark
Nor
mal
ized
Exe
cutio
n Ti
me
base inc2add
2% mean speedup
182DynamoRIO Tutorial at CGO 24 April 2010
Software Prefetching
• Ubiquitous Memory Introspection (cgo 2007)– Sampling to select hot traces– Instrument to collect memory references– Analyze to discover reference patterns
• Stride– Insert software prefetching instruction
182
183DynamoRIO Tutorial at CGO 24 April 2010 183
Software Prefetching Results
184DynamoRIO Tutorial at CGO 24 April 2010 184
More Examples
• Dynamic Optimization– Strength Reduction– Software Prefetching
• Profiling– Memory Reference Trace– PiPA
• Shadow Memory– Umbra
• Dr. Memory
185DynamoRIO Tutorial at CGO 24 April 2010
Memory Reference Trace
• Memtrace– Profile format
• <r/w, addr>– Steps
• Thread initialization– Allocate buffer per thread
• Instrumentation– Fill the buffer– Dump to file if buffer is full
• Thread exit– Delete the buffer
– Optimization• Inline buffer filling code• Out-line the clean call invocation code
185
186DynamoRIO Tutorial at CGO 24 April 2010
Register Events
DR_EXPORT voiddr_init(client_id_t id){
…mutex = dr_mutex_create();dr_register_exit_event(event_exit);dr_register_thread_init_event(event_thread_init);dr_register_thread_exit_event(event_thread_exit);dr_register_bb_event(event_basic_block);/* out-lined clean call invocation code */code_cache_init();…
}
186
187DynamoRIO Tutorial at CGO 24 April 2010
Own Code Cache Initialization
static void code_cache_init(void) { drcontext = dr_get_current_drcontext();
code_cache = dr_nonheap_alloc(PAGE_SIZE, DR_MEMPROT_READ | DR_MEMPROT_WRITE |
DR_MEMPROT_EXEC); ilist = instrlist_create(drcontext);
where = INSTR_CREATE_jmp_ind(drcontext, opnd_create_reg(REG_XCX)); instrlist_meta_append(ilist, where);
dr_insert_clean_call(drcontext, ilist, where, (void *)clean_call, false, 0);end = instrlist_encode(drcontext, ilist, code_cache, false);DR_ASSERT((end - code_cache) < PAGE_SIZE);instrlist_clear_and_destroy(drcontext, ilist);dr_memory_protect(code_cache, PAGE_SIZE, DR_MEMPROT_READ |
DR_MEMPROT_EXEC);}
187
188DynamoRIO Tutorial at CGO 24 April 2010
Clean Call
static void clean_call() {…drcontext = dr_get_current_drcontext();
data = dr_get_tls_field(drcontext); mem_ref = (mem_ref_t *)databuf_base; num_refs = (int)((mem_ref_t *)databuf_ptr - mem_ref);
for (i = 0; i < num_refs; i++) {dr_fprintf(datalog, "%c:"PFX"\n", mem_refwrite ? 'w' : 'r', mem_refaddr);
++mem_ref;}datanum_refs += num_refs;…
}
188
189DynamoRIO Tutorial at CGO 24 April 2010
Thread initialization & exit
void event_thread_init(void *drcontext) {…data = dr_thread_alloc(drcontext, sizeof(per_thread_t));dr_set_tls_field(drcontext, data);databuf_base = dr_thread_alloc(drcontext, MAX_MEM_BUF_SIZE);…datanum_refs = 0;
}void event_thread_exit(void *drcontext) {
data = dr_get_tls_field(drcontext);dr_mutex_lock(mutex);num_refs += datanum_refs;dr_mutex_unlock(mutex);dr_thread_free(drcontext, databuf_base, MAX_MEM_BUF_SIZE);dr_thread_free(drcontext, data, sizeof(per_thread_t));
}
189
190DynamoRIO Tutorial at CGO 24 April 2010
Basic Block Instrumentation
static dr_emit_flags_t event_basic_block(void *drcontext, void *tag, instrlist_t *bb,
bool for_trace, bool translating) { … for (instr = instrlist_first(bb); instr != NULL; instr = instr_get_next(instr)) { if (instr_get_app_pc(instr) == NULL) continue; if (instr_reads_memory(instr)) for (i = 0; i < instr_num_srcs(instr); i++) if (opnd_is_memory_reference(instr_get_src(instr, i))) instrument_mem(drcontext, bb, instr, i, false); if (instr_writes_memory(instr)) for (i = 0; i < instr_num_dsts(instr); i++) if (opnd_is_memory_reference(instr_get_dst(instr, i))) instrument_mem(drcontext, bb, instr, i, true);}
190
191DynamoRIO Tutorial at CGO 24 April 2010
instrument_mem (I)
/* get memory reference address into reg1 */opnd1 = opnd_create_reg(reg1);
if (opnd_is_base_disp(ref)) { /* lea [ref] reg */opnd2 = ref;
opnd_set_size(&opnd2, OPSZ_lea); instr = INSTR_CREATE_lea(drcontext, opnd1, opnd2);
} else if(IF_X64(opnd_is_rel_addr(ref) ||) opnd_is_abs_addr(ref)) { /* mov addr reg */
opnd2 = OPND_CREATE_INTPTR(opnd_get_addr(ref)); instr = INSTR_CREATE_mov_imm(drcontext, opnd1, opnd2);
} else {instr = NULL;DR_ASSERT_MSG(false, "Unhandled instructions");
} instrlist_meta_preinsert(ilist, where, instr);
191
192DynamoRIO Tutorial at CGO 24 April 2010
instrument_mem (II)
/* Move write/read to write field */ opnd1 = OPND_CREATE_MEM32(reg2, offsetof(mem_ref_t, write)); opnd2 = OPND_CREATE_INT32(write); instr = INSTR_CREATE_mov_imm(drcontext, opnd1, opnd2); instrlist_meta_preinsert(ilist, where, instr); /* Store address in memory ref */ opnd1 = OPND_CREATE_MEMPTR(reg2, offsetof(mem_ref_t, addr)); opnd2 = opnd_create_reg(reg1); instr = INSTR_CREATE_mov_st(drcontext, opnd1, opnd2); instrlist_meta_preinsert(ilist, where, instr);
/* Increment reg value by pointer size using lea instr */ opnd1 = opnd_create_reg(reg2); opnd2 = opnd_create_base_disp(reg2, REG_NULL, 0, sizeof(mem_ref_t), OPSZ_lea); instr = INSTR_CREATE_lea(drcontext, opnd1, opnd2); instrlist_meta_preinsert(ilist, where, instr);
192
193DynamoRIO Tutorial at CGO 24 April 2010
instrument_mem (III)
/* jecxz call */call = INSTR_CREATE_label(drcontext);instrlist_meta_preinsert(ilist, where, INSTR_CREATE_jecxz(drcontext, opnd_create_instr(call)));/* jump restore to skip clean call */restore = INSTR_CREATE_label(drcontext);instrlist_meta_preinsert(ilist, where,
INSTR_CREATE_jmp(drcontext, opnd_create_instr(restore)));instrlist_meta_preinsert(ilist, where, call);/* mov restore REG_XCX */instr = INSTR_CREATE_mov_st(drcontext,opnd_create_reg(reg2),opnd_create_instr(restore));instrlist_meta_preinsert(ilist, where, instr);/* jmp code_cache */opnd1 = opnd_create_pc(code_cache);instrlist_meta_preinsert(ilist, where, INSTR_CREATE_jmp(drcontext, opnd1));/* restore %reg */instrlist_meta_preinsert(ilist, where, restore);
193
194DynamoRIO Tutorial at CGO 24 April 2010
PiPA
• Pipelined Profiling and Analysis (cgo 2008)– Stages (thread/process)
• Profiling• Reconstruction/extraction• Analysis
– Profiling• Runtime Execution Profile (REP)
– Communication• Double buffer
– Analysis• Parallel cache simulation
194
195DynamoRIO Tutorial at CGO 24 April 2010 195
More Examples
• Dynamic Optimization– Strength Reduction
• Profiling– Memory Reference Trace
• Shadow Memory– Umbra
• Dr. Memory
196DynamoRIO Tutorial at CGO 24 April 2010 196
Shadow Memory
• Application– Store meta-data to track properties of application memory
• Millions of software watchpoints• Dynamic information flow tracking (taint propagation)• Race detection• Memory usage debugging tool (MemCheck/Dr. Memory)
• Issues– Performance– Multi-thread applications– Flexibility– Platform dependent– Development challenges
197DynamoRIO Tutorial at CGO 24 April 2010
Umbra (CGO 2010)
• Design• Implementation• Optimization• Download
– http://people.csail.mit.edu/qin_zhao/umbra/
197
198DynamoRIO Tutorial at CGO 24 April 2010
Design
• Address Space– A collection of fixed size units
• 4G (64-bit)• Application, Shadow, Unused
• Translation Table– Translation from application memory unit to
corresponding shadow memory unit
198
App Mem Shd Mem Offset[0x00000000, 0x10000000)
[0x20000000, 0x30000000)
0x20000000
[0x60000000, 0x70000000)
[0x40000000, 0x50000000)
-0x20000000
[0x80000000, 0x90000000)
[0x50000000, 0x60000000)
-0x20000000
App Mem 1
Shd Mem 1
Unused
Unused
App Mem 3
App Mem 2
Shd Mem 2
Shd Mem 3offsetscaleaddraddr appshd +×=
199DynamoRIO Tutorial at CGO 24 April 2010
Implementation
• Memory Manager– Monitor and control application memory allocation
• brk, mmap, munmap, mremap• dr_register_pre_syscall_event• dr_register_post_syscall_event
– Allocate shadow memory– Maintain translation table
• Instrumenter– Instrument every memory reference
• Context save• Address calculation• Address translation• Shadow memory update• Context restore
199
200DynamoRIO Tutorial at CGO 24 April 2010
Instrument Code Example
Context Save mov %ecx [ECX_SLOT]mov %edx [EDX_SLOT] mov %eax [EAX_SLOT]lahf %ahseto %al
Address Calculation lea [%ebx, 16] %ecx
Address Translation mov 0 %edx… # table lookup codeadd %ecx, table[%edx].offset %ecx
Shadow Memory Update mov 1 [%ecx]
Context Restore add %al 0x7fsahf mov [ECX_SLOT] %ecxmov [EDX_SLOT] %edx mov [EAX_SLOT] %eax
Application memory reference mov 0 [%ebx, 16]
200
201DynamoRIO Tutorial at CGO 24 April 2010
Optimization
• Translation Optimization– Thread Local Translation Table– Memoization Check– Reference Check
• Instrumentation Optimization– Context Switch Reduction– Reference Grouping– 3-stage Code Layout
201
202DynamoRIO Tutorial at CGO 24 April 2010
Translation Optimization
• Caching
202
Global translation table
App 1
Shd 1
Shd 2
App 2
203DynamoRIO Tutorial at CGO 24 April 2010
Translation Optimization
• Thread Local Translation Optimization– Local translation table per thread– Synchronize with global translation table when necessary– Avoid lock contention
203
Global translation table
Thread Local translation table
Thread 1
Thread 2
204DynamoRIO Tutorial at CGO 24 April 2010
Translation Optimization
• Memoization Cache– Software cache per thread– Stores frequently used translation entries
• Stack• Units found in last table lookup
204
Global translation table
Thread Local translation table
Memoization Cache
Thread 1
Thread 2
205DynamoRIO Tutorial at CGO 24 April 2010
Translation Optimization
• Reference Cache– Software cache per static application memory reference
• Last reference unit tag• Last translation offset
205
Reference cache
Thread 1
Thread 2
Global translation table
Thread Local translation table
Memoization Cache
206DynamoRIO Tutorial at CGO 24 April 2010
Instrumentation Optimization
• Context Switch Reduction– Registers liveness analysis
• Reference Grouping– One translation lookup for multiple references
• Stack local variables• Different members of the same object
• 3-stage Code Layout– Inline stub
• Quick inline check code with minimal context switch– Lean procedure
• Simple assembly procedure with partial context switch– Callout
• C function with complete context switch
206
207DynamoRIO Tutorial at CGO 24 April 2010
3-stage Code Layout
• Inline stub– Reference cache check– Jump to lean procedure if miss
• Lean procedure– Memoization cache check – Local table lookup– Clean call to call out
• Callout– Global table synchronization– Local table lookup
207
208DynamoRIO Tutorial at CGO 24 April 2010
Instrumentation Optimization
208
# reference cache check lea [ref] %r1 %r1 & 0xffffffff00000000 %r1 cmp %r1, ref.tag je .update_shadow_memory# jmp-and-link to lean procedure mov %r1 ref.tag mov .update_ref_cache [ret_pc] jmp lean_procedure.update_ref_cache mov %r1 ref.offset# shadow memory update.update_shadow_memory lea [ref] %r1 add %r1 + ref.offset %r1 mov 1 [%r1]
# memorization check cmp %r1, cache1.tag jne .cache1_miss mov cache1.offset %r1 jmp [ret_pc].cache1_miss cmp %r1, cache2.tag jne .cache2_miss mov cache1.offset %r1 jmp [ret_pc].cache2_miss# table lookup mov %r1 cache2.tag mov %r2 [R2_SLOT] … mov [R2_SLOT] %r2 mov %r1 cache2.offset jmp [ret_pc]
Inline Stub Lean Procedure
209DynamoRIO Tutorial at CGO 24 April 2010
Performance Evaluation
209
210DynamoRIO Tutorial at CGO 24 April 2010
Umbra Client: Shared Memory Detection
static void instrument_update(void *drcontext, umbra_info_t *umbra_info, mem_ref_t *ref, instrlist_t *ilist, instr_t *where) { … /* test [%reg].tid_map, tid_map*/ opnd1 = OPND_CREATE_MEM32(umbra_inforeg, 0, OPSZ_4); opnd2 = OPND_CREATE_INT32(client_tls_datatid_map); instrlist_meta_preinsert(ilist, where, INSTR_CREATE_test(drcontext, opnd1, opnd2)); /* jnz where */ opnd1 = opnd_create_instr(where); instrlist_meta_preinsert(ilist, where, INSTR_CREATE_jcc(drcontext, OP_jnz, opnd1)); /* or */ opnd1 = OPND_CREATE_MEM32(umbra_inforeg, 0, OPSZ_4); opnd2 = OPND_CREATE_INT32(client_tls_datatid_map | 1); instr = INSTR_CREATE_or(drcontext, opnd1, opnd2); LOCK(instr); instrlist_meta_preinsert(ilist, label, instr);}
210
211DynamoRIO Tutorial at CGO 24 April 2010 211
More Examples
• Dynamic Optimization– Strength Reduction
• Profiling– Memory Reference Trace
• Shadow Memory– Umbra
• Dr. Memory
212DynamoRIO Tutorial at CGO 24 April 2010
Dr. Memory
• Detects reads of uninitialized memory• Detects heap errors
– Out-of-bounds accesses (underflow, overflow)– Access to freed memory– Invalid frees– Memory leaks
• Detects other accesses to invalid memory– Stack tracking– Thread-local storage slot tracking
• Operates at runtime on unmodified Windows & Linux binaries
213DynamoRIO Tutorial at CGO 24 April 2010
Dr. Memory Instrumentation
• Monitor all memory accesses, stack adjustments, and heap allocations
• Shadow each byte of app memory• Each byte’s shadow stores one of 4 values:
– Unaddressable– Uninitialized– Defined at byte level– Defined at bit level escape to extra per-bit shadow values
214DynamoRIO Tutorial at CGO 24 April 2010
Dr. Memory
defined
invalid
undefineddefined
Shadow StackStack
Shadow HeapHeap
redzone
malloc
freed
redzone
invalid
invalid
invalid
defined
undefineddefined
215DynamoRIO Tutorial at CGO 24 April 2010
Partial-Word Defines But Whole-Word Transfers
• Sub-dword variables are moved around as whole dwords• Cannot raise error when a move reads uninitialized bits• Must propagate on moves and thus must shadow registers
– Propagate shadow values by mirroring app data flow• Check system call reads and propagate system call writes
– Else, false negatives (reads) or positives (writes)• Raise errors instead of propagating at certain points
– Report errors only on “significant” reads
216DynamoRIO Tutorial at CGO 24 April 2010
Shadowing Registers
• Use multiple TLS slots– dr_raw_tls_calloc()– Alternative: steal register
• Can read and write w/o spilling• Bring into spilled register to combine w/ other args
– Defined=0, uninitialized=1– Combine via bitwise or
217DynamoRIO Tutorial at CGO 24 April 2010
Monitoring Stack Changes
• As stack is extended and contracts again, must update stack shadow as unaddressable vs uninitialized
• Push, pop, or any write to stack pointer• Try to distinguish large alloc/dealloc from stack swap
218DynamoRIO Tutorial at CGO 24 April 2010
Kernel-Mediated Stack Changes
• Kernel places data on the stack and removes it again– Windows: APC, callback, and exception– Linux: signals
• Linux signals as an example:– intercept sigaltstack changes– intercept handler registration to instrument handler code– use DR's signal event to record app xsp at interruption point– when see event followed by handler, check which stack and
mark from either interrupted xsp or altstack base to cur xsp as defined (ignoring padding)
– record cur xsp in handler, and use to undo on sigreturn
219DynamoRIO Tutorial at CGO 24 April 2010
Types Of Instrumentation
• Clean call– Simplest, but expensive in both time and space
• Shared clean call– Saves space
• Lean procedure– Shared routine with smaller context switch than full clean call
• Inlined– Smallest context switch, but should limit to small sequences of
instrumentation
220DynamoRIO Tutorial at CGO 24 April 2010
Non-Code-Cache Code
• Use dr_nonheap_alloc() to allocate space to store code• Generate code using DR’s IR and emit to target space• Mark read-only once emitted via dr_memory_protect()
221DynamoRIO Tutorial at CGO 24 April 2010
Jump-and-Link
• Rather than using call+return, avoid stack swap cost by using jump-and-link– Store return address in a register or TLS slot– Direct jump to target– Indirect jump back to source
PRE(bb, inst, INSTR_CREATE_mov_st(drcontext, spill_slot_opnd(drcontext, SPILL_SLOT_2), opnd_create_instr(appinst)));PRE(bb, inst, INSTR_CREATE_jmp(drcontext, opnd_create_pc(shared_slowpath_region)));...PRE(ilist, NULL, INSTR_CREATE_jmp_ind(drcontext, spill_slot_opnd(SPILL_SLOT_2)));
222DynamoRIO Tutorial at CGO 24 April 2010
Inter-Instruction Storage
• Spill slots provided by DR are only guaranteed to be live during a single app instr– In practice, live until next selfmod instr
• Allocate own TLS for spill slots– dr_raw_tls_calloc()
• Steal registers across whole bb– Restore before each app read– Update spill slot after each app write– Restore on fault
223DynamoRIO Tutorial at CGO 24 April 2010
Using Faults For Faster Common Case Code
• Instead of explicitly checking for rare cases, use faults to handle them and keep common case code path fast
• Signal and exception event and restore state extended event all provide pre- and post-translation contexts and containing fragment information
• Client can return failure for extended restore state event– When can support re-execution of faulting cache instr, but not
re-start translation for relocation
224DynamoRIO Tutorial at CGO 24 April 2010
Address Space Iteration
• Repeated calls to dr_query_memory_ex()• Check dr_memory_is_in_client() and
dr_memory_is_dr_internal()• Heap walk
– API on Windows• Initial structures on Windows
– TEB, TLS, etc.– PEB, ProcessParameters, etc.
225DynamoRIO Tutorial at CGO 24 April 2010
Intercepting Library Routines
• Common task• Dr. Memory monitors malloc, calloc, realloc, free,
malloc_usable_size, etc.– Alternative is to replace w/ own copies
• Locating entry point– Module API
• Pre-hooks are easy• Post-hooks are hard
– Three techniques, each with its own limitations
226DynamoRIO Tutorial at CGO 24 April 2010
Intercepting Library Routines: Technique 1
• CFG analysis at init time– Statically analyze code from entry point and find return
instruction(s)– Post-hook placed at each return instruction
• Complications:– Not always easy or even possible to statically analyze
• Hot/cold and other layout optimizations• Switches or other indirection• Mixed code/data
– Tailcall from hooked routine A to hooked routine B will skip a return in A
– Longjmp or SEH unwind can skip any post-hook
227DynamoRIO Tutorial at CGO 24 April 2010
Intercepting Library Routines: Technique 2
• At call site identify target– Direct calls/jmp: easy– Indirect through PLT/IAT: easy– Indirect through register/unknown memory: not always easy– Post-hook is placed at post-call-site– Flush post-call-site if it exists
• Complications:– Indirect targets that are not “statically” analyzable– Same call targets multiple hooked routines– Tailcall from hooked routine A to hooked routine B will skip post-
call of site in A– Longjmp or SEH unwind can skip any post-hook
228DynamoRIO Tutorial at CGO 24 April 2010
Intercepting Library Routines: Technique 3
• Inside callee, obtain return address– Flush that address if it already exists– Post-hooks are placed at return address
• Complications:– Same call targets multiple hooked routines
• Store which inside callee– Tailcall from hooked routine A to hooked routine B will skip
return address of B• Identify tailcall and store target B; process B and then A at post-A
– Longjmp or SEH unwind can skip any post-hook• Try to intercept
229DynamoRIO Tutorial at CGO 24 April 2010
Modifying Library Routine Parameters
• Use cases for Dr. Memory:– Add redzone to heap allocations– Delay frees
• Simply clobber the actual parameter– Need to know calling convention– Use dr_safe_write() for robustness
230DynamoRIO Tutorial at CGO 24 April 2010
Replacing Library Routines
• Dr. Memory replaces libc routines containing optimized code that raises false positives– memcpy, strlen, strchr, etc.
• Simplification: arrange for routines to always be entered in a new bb– Do not request elision or indcall2direct from DR
• Want to interpret replaced routines– DR treats native execution differently: aborts on fault, etc.
• Replace entire bb with jump to replacement routine
231DynamoRIO Tutorial at CGO 24 April 2010
State Across Windows Callbacks
• Per-thread state varies in whether should be shared or private across callbacks– Pre-to-post syscall data must be callback-private
• Callback entry:– Ntdll!KiUserCallbackDispatcher
• Callback exit:– NtCallbackReturn system call– Int 0x2b
• Make these DynamoRIO Events? (Issue 241)
232DynamoRIO Tutorial at CGO 24 April 2010
Delayed Fragment Deletion
• Due to non-precise flushing we can have a flushed bb made inaccessible but not actually freed for some time
• When keeping state per bb, if a duplicate bb is seen, replace the state and increment a counter ignore_next_delete
• On a deletion event, decrement and ignore unless below 0• Can't tell apart from duplication due to thread-private copies:
but this mechanism handles that if saved info is deterministic and identical for each copy
233DynamoRIO Tutorial at CGO 24 April 2010
Callstack Walking
• Use case: error reporting• Technique:
– Start with xbp as frame pointr (fp)– Look for <fp,retaddr> pairs where retaddr = inside a module
• Interesting issues:– When scanning for frame pointer (in frameless func, or at bottom
of stack), querying whether in a module dominates performance– msvcr80!malloc pushes ebx and then ebp, requiring special
handling– When displaying, use retaddr-1 for symbol lookup
234DynamoRIO Tutorial at CGO 24 April 2010
Client Files Closed By Application
• Some applications close all file descriptors• Solution:
– Keep table of file descriptors owned by client– Intercept close system call– Turn close into nop when called on client descriptors
235DynamoRIO Tutorial at CGO 24 April 2010
Suspending The World
• Use case: Dr. Memory leak check– GC-like memory scan
• Use dr_suspend_all_other_threads() and dr_resume_all_other_threads()
• Cannot hold locks while suspending
236DynamoRIO Tutorial at CGO 24 April 2010
Using Nudges
• Daemon apps do not exit• Request results mid-run• Cross-platform
– Signal on Linux– Remote thread on Windows
237DynamoRIO Tutorial at CGO 24 April 2010
Tool Packaging
• DynamoRIO is redistributable, so can include a copy with your tool
• Front end to configure and launch app– On Linux use a script that execs drrun– On Windows use drinjectlib.dll
Feedback1:30-1:40 Welcome + DynamoRIO History1:40-2:40 DynamoRIO Internals2:40-3:00 Examples, Part 13:00-3:15 Break3:15-4:15 DynamoRIO API4:15-5:15 Examples, Part 25:15-5:30 Feedback
239DynamoRIO Tutorial at CGO 24 April 2010
Contributors
• Looking for contributors to DynamoRIO– Work on major features
• Windows 7 support• Auto-inline callouts• Attach to running process• MacOS port
– Add new Extensions– Run nightly test suites– Maintain particular platforms– Etc.
240DynamoRIO Tutorial at CGO 24 April 2010
Future Releases
• Better Linux library support (STL, etc.)– Custom loader
• Auto-inline callouts • Persistent and process-shared caches• Attach to a process• Symbol table lookup support on Linux• 64-bit client controlling 32-bit app• Your suggestion here
241DynamoRIO Tutorial at CGO 24 April 2010
Feedback
• Questions for you– Feedback on what you want to see in API
• Thank you!