update build dependencies #3
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
sebroy
approved these changes
Sep 12, 2019
delphix-devops-bot
pushed a commit
that referenced
this pull request
Jan 1, 2020
ASAN is incredibly useful to prevent entire classes of memory errors.
For example, when ASAN is enabled, it already shows us 35 memory leaks:
```
$ ~/dev/bpftrace git:(asan) ✗ sudo ./build/src/bpftrace -e 'BEGIN { printf("hi\n"); }'
[sudo] password for dxu:
Attaching 1 probe...
hi
^C
=================================================================
==758098==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 288 byte(s) in 4 object(s) allocated from:
#0 0x7f72bbb5dcd8 in __interceptor_calloc /build/gcc/src/gcc/lib...
#1 0x7f72b96665e5 in perf_reader_new (/usr/lib/libbcc.so.0+0x3e85e5)
#2 0x7ffc5103b767 ([stack]+0x68767)
Direct leak of 80 byte(s) in 1 object(s) allocated from:
#0 0x7f72bbb5f8f8 in operator new(unsigned long) /build/gcc/src/...
#1 0x5631e1a566cd in bpftrace::Parser::parse() src/parser.yy:151
#2 0x5631e18cae7e in bpftrace::Driver::parse() /home/dxu/dev/bpft...
#3 0x5631e18dd8ac in main /home/dxu/dev/bpftrace/src/main.cpp:349
#4 0x7f72b3eb5152 in __libc_start_main (/usr/lib/libc.so.6+0x27152)
Direct leak of 80 byte(s) in 1 object(s) allocated from:
#0 0x7f72bbb5f8f8 in operator new(unsigned long) /build/gcc/src/g...
#1 0x5631e1a566cd in bpftrace::Parser::parse() src/parser.yy:151
#2 0x5631e18cae7e in bpftrace::Driver::parse() /home/dxu/dev/bpf...
#3 0x5631e18df34f in main /home/dxu/dev/bpftrace/src/main.cpp:491
#4 0x7f72b3eb5152 in __libc_start_main (/usr/lib/libc.so.6+0x27152)
[...]
```
The default is off for now b/c we have a lot of noisy leaks. Once those
are fixed, we should flip to on by default to be in preventative mode.
Note we also have to disable the rlimit protection b/c ASAN needs to
mmap a huge amount of the address space to work.
delphix-devops-bot
pushed a commit
that referenced
this pull request
Mar 21, 2020
This fixes the following error.
```
% sudo ./src/bpftrace -e 'w::0x10000000:8:: {}'
AddressSanitizer:DEADLYSIGNAL
=================================================================
==26761==ERROR: AddressSanitizer: SEGV on unknown address 0x614000010000 (pc 0x0000008dc6b6 bp 0x7ffe172080c0 sp 0x7ffe17207540 T0)
==26761==The signal is caused by a READ memory access.
#0 0x8dc6b6 in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::AttachPoint&) /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/semantic_analyser.cpp:1684:25
#1 0x87c3e3 in bpftrace::ast::AttachPoint::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/ast.cpp:323:5
#2 0x8dd1bc in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::Probe&) /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/semantic_analyser.cpp:1737:9
#3 0x87c603 in bpftrace::ast::Probe::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/ast.cpp:355:5
#4 0x8dd2eb in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::Program&) /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/semantic_analyser.cpp:1751:12
#5 0x87c6b3 in bpftrace::ast::Program::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/ast.cpp:364:5
#6 0x8dd35a in bpftrace::ast::SemanticAnalyser::analyse() /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/semantic_analyser.cpp:1760:12
#7 0x80f63f in main /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/main.cpp:577:19
#8 0x7fec4efa5b96 in __libc_start_main /build/glibc-OTsEL5/glibc-2.27/csu/../csu/libc-start.c:310
#9 0x5c1e39 in _start (/home/ubuntu/work/bpftrace/bpftrace/build_asan/src/bpftrace+0x5c1e39)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: SEGV /home/ubuntu/work/bpftrace/bpftrace/build_asan/../src/ast/semantic_analyser.cpp:1684:25 in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::AttachPoint&)
==26761==ABORTING
```
This is because `ap.mode.size()` is zero (`ap.mode` is a empty string)
and `ap.mode.size() - 1` causes underflow (`size_t` is unsigned).
https://github.com/iovisor/bpftrace/blob/96b6eb83c93668fd1e54147c8e1929fedd571dac/src/ast/semantic_analyser.cpp#L1683-L1684
With this patch,
```
% sudo ./src/bpftrace -e 'w::0x10000000:8:: {}'
stdin:1:1-18: ERROR: watchpoint mode must be combination of (r,w,x)
w::0x10000000:8:: {}
~~~~~~~~~~~~~~~~
```
delphix-devops-bot
pushed a commit
that referenced
this pull request
Apr 17, 2020
Fix the following error.
```
% sudo ./src/bpftrace -e 'k:f { exit() ? 0 : 0 }'
AddressSanitizer:DEADLYSIGNAL
=================================================================
==14010==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000 (pc 0x000000850854 bp 0x7fffffffa450 sp 0x7fffffffa180 T0)
==14010==The signal is caused by a READ memory access.
==14010==Hint: address points to the zero page.
#0 0x850853 in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Ternary&) /home/ubuntu/work/bpftrace/src/ast/codegen_llvm.cpp
#1 0x861f29 in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Probe&) /home/ubuntu/work/bpftrace/src/ast/codegen_llvm.cpp:1506:13
#2 0x865256 in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Program&) /home/ubuntu/work/bpftrace/src/ast/codegen_llvm.cpp:1597:12
#3 0x86e34b in bpftrace::ast::CodegenLLVM::compile(bpftrace::DebugLevel, std::ostream&) /home/ubuntu/work/bpftrace/src/ast/codegen_llvm.cpp:2013:10
#4 0x77623f in main /home/ubuntu/work/bpftrace/src/main.cpp:605:22
#5 0x7fffee1451e2 in __libc_start_main /build/glibc-t7JzpG/glibc-2.30/csu/../csu/libc-start.c:308:16
#6 0x4d1bcd in _start (/disk/work/bpftrace/build_san3/src/bpftrace+0x4d1bcd)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: SEGV /home/ubuntu/work/bpftrace/src/ast/codegen_llvm.cpp in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Ternary&)
==14010==ABORTING
```
delphix-devops-bot
pushed a commit
that referenced
this pull request
Apr 17, 2020
This prevents the error like the follwing:
```
% sudo ASAN_OPTIONS=detect_leaks=0 ./src/bpftrace -e 'BEGIN { @ = max(exit()); }'
AddressSanitizer:DEADLYSIGNAL
=================================================================
==25253==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000 (pc 0x00000089e3dc bp 0x7fff0bf07270 sp 0x7fff0bf07270 T0)
==25253==The signal is caused by a READ memory access.
==25253==Hint: address points to the zero page.
#0 0x89e3dc in llvm::Value::getType() const /usr/lib/llvm-10/include/llvm/IR/Value.h:246:34
#1 0x89cd6e in llvm::IRBuilder<llvm::ConstantFolder, llvm::IRBuilderDefaultInserter>::CreateIntCast(llvm::Value*, llvm::Type*, bool, llvm::Twine const&) /usr/lib/llvm-10/include/llvm/IR/IRBuilder.h:2172:12
#2 0x889ad5 in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Call&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/codegen_llvm.cpp:328:16
#3 0x880160 in bpftrace::ast::Call::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/ast.cpp:129:5
#4 0x8966c0 in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::AssignMapStatement&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/codegen_llvm.cpp:1315:20
#5 0x880fa3 in bpftrace::ast::AssignMapStatement::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/ast.cpp:284:5
#6 0x897a70 in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Probe&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/codegen_llvm.cpp:1510:13
#7 0x8815f3 in bpftrace::ast::Probe::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/ast.cpp:355:5
#8 0x89895b in bpftrace::ast::CodegenLLVM::visit(bpftrace::ast::Program&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/codegen_llvm.cpp:1601:12
#9 0x8816a3 in bpftrace::ast::Program::accept(bpftrace::ast::Visitor&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/ast.cpp:364:5
#10 0x89b397 in bpftrace::ast::CodegenLLVM::compile(bpftrace::DebugLevel, std::ostream&) /home/ubuntu/work/bpftrace/bpftrace/src/ast/codegen_llvm.cpp:2019:10
#11 0x813f3c in main /home/ubuntu/work/bpftrace/bpftrace/src/main.cpp:589:22
#12 0x7f2794c4db96 in __libc_start_main /build/glibc-OTsEL5/glibc-2.27/csu/../csu/libc-start.c:310
#13 0x5c64c9 in _start (/home/ubuntu/work/bpftrace/bpftrace/build_llvm10/src/bpftrace+0x5c64c9)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: SEGV /usr/lib/llvm-10/include/llvm/IR/Value.h:246:34 in llvm::Value::getType() const
==25253==ABORTING
```
delphix-devops-bot
pushed a commit
that referenced
this pull request
Oct 22, 2020
The expr_ of a PositionalParameter is a pointer to a buffer, and an
offset can be added to it (e.g., `str($1 + 1)`). Cast expr_ to int so
that llvm does not complain about type mismatches when operating addition.
This fixes the following error reported by AddressSanitizer.
```
% sudo ./src/bpftrace -e 'BEGIN { printf("%s", str($1 + 1)); exit(); }' hello
AddressSanitizer:DEADLYSIGNAL
=================================================================
==3979==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000f8b (pc 0x7f1947a755ba bp 0x60700000def0 sp 0x7ffc07b486b0 T0)
==3979==The signal is caused by a READ memory access.
==3979==Hint: address points to the zero page.
#0 0x7f1947a755ba in llvm::Constant::isNullValue() const (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0xa735ba)
#1 0x7f1948826177 in llvm::LazyValueInfo::getPredicateAt(unsigned int, llvm::Value*, llvm::Constant*, llvm::Instruction*) (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0x1824177)
#2 0x7f194841e4a8 (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0x141c4a8)
#3 0x7f1948420049 (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0x141e049)
#4 0x7f1947b13ed5 in llvm::FPPassManager::runOnFunction(llvm::Function&) (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0xb11ed5)
#5 0x7f194879ca90 (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0x179aa90)
#6 0x7f1947b145ff in llvm::legacy::PassManagerImpl::run(llvm::Module&) (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0xb125ff)
#7 0x851bbe in bpftrace::ast::CodegenLLVM::optimize() /home/ubuntu/work/bpftrace/bpftrace/src/ast/codegen_llvm.cpp:2720:6
#8 0x791416 in main /home/ubuntu/work/bpftrace/bpftrace/src/main.cpp:745:10
#9 0x7f1945cc8b96 in __libc_start_main /build/glibc-2ORdQG/glibc-2.27/csu/../csu/libc-start.c:310
#10 0x4d1c79 in _start (/home/ubuntu/work/bpftrace/bpftrace/build_dev/src/bpftrace+0x4d1c79)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: SEGV (/usr/lib/x86_64-linux-gnu/libLLVM-10.so.1+0xa735ba) in llvm::Constant::isNullValue() const
==3979==ABORTING
```
delphix-devops-bot
pushed a commit
that referenced
this pull request
Jan 14, 2021
I though #1572 fixes the issue of invalid tuple creation, but there is
another issue. libfuzzer found the following:
```
% sudo ./src/bpftrace -e 'BEGIN { (1, !(int1)1); }'
/disk/work/bpftrace2/src/struct.cpp:29:37: runtime error: division by zero
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /disk/work/bpftrace2/src/struct.cpp:29:37 in
AddressSanitizer:DEADLYSIGNAL
=================================================================
==46777==ERROR: AddressSanitizer: FPE on unknown address 0x00000000b6b9 (pc 0x0000009aa53c bp 0x7fffffff68d0 sp 0x7fffffff64a0 T0)
#0 0x9aa53c in bpftrace::Tuple::Create(std::vector<bpftrace::SizedType, std::allocator<bpftrace::SizedType> >) /disk/work/bpftrace2/src/struct.cpp:29:37
#1 0x9cb00a in bpftrace::CreateTuple(std::vector<bpftrace::SizedType, std::allocator<bpftrace::SizedType> > const&) /disk/work/bpftrace2/src/types.cpp:443:20
#2 0xc9dd2c in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::Tuple&) /disk/work/bpftrace2/src/ast/semantic_analyser.cpp:1994:16
#3 0xa8048b in bpftrace::ast::Tuple::accept(bpftrace::ast::Visitor&) /disk/work/bpftrace2/src/ast/ast.cpp:31:1
#4 0xc9e116 in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::ExprStatement&) /disk/work/bpftrace2/src/ast/semantic_analyser.cpp:1999:14
#5 0xa805eb in bpftrace::ast::ExprStatement::accept(bpftrace::ast::Visitor&) /disk/work/bpftrace2/src/ast/ast.cpp:32:1
#6 0xcbb643 in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::Probe&) /disk/work/bpftrace2/src/ast/semantic_analyser.cpp:2473:11
#7 0xa8124b in bpftrace::ast::Probe::accept(bpftrace::ast::Visitor&) /disk/work/bpftrace2/src/ast/ast.cpp:41:1
#8 0xcbb97d in bpftrace::ast::SemanticAnalyser::visit(bpftrace::ast::Program&) /disk/work/bpftrace2/src/ast/semantic_analyser.cpp:2481:12
#9 0xa813ab in bpftrace::ast::Program::accept(bpftrace::ast::Visitor&) /disk/work/bpftrace2/src/ast/ast.cpp:42:1
#10 0xcbbda0 in bpftrace::ast::SemanticAnalyser::analyse() /disk/work/bpftrace2/src/ast/semantic_analyser.cpp:2490:12
#11 0xa474fa in main /disk/work/bpftrace2/src/main.cpp:738:19
#12 0x7fffed57e0b2 in __libc_start_main /build/glibc-ZN95T4/glibc-2.31/csu/../csu/libc-start.c:308:16
#13 0x4fe68d in _start (/disk/work/bpftrace2/build/src/bpftrace+0x4fe68d)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: FPE /disk/work/bpftrace2/src/struct.cpp:29:37 in bpftrace::Tuple::Create(std::vector<bpftrace::SizedType, std::allocator<bpftrace::SizedType> >)
==46777==ABORTING
```
In this case, `elem->type` is `int` because of `!`, but its size is zero
and that causes the problem. To fix issue, don't create tuple if an
elemenet size is zero.
delphix-devops-bot
pushed a commit
that referenced
this pull request
Jun 29, 2021
Not entirely sure why but for some reason the destructor deallocates
less bytes than it should:
```
==14571==ERROR: AddressSanitizer: new-delete-type-mismatch on 0x60b000003740 in thread T0:
object passed to delete has wrong type:
size of the allocated type: 112 bytes;
size of the deallocated type: 104 bytes.
#0 0x7f192d4879c8 in operator delete(void*, unsigned long) (/usr/lib/x86_64-linux-gnu/libasan.so.4+0xe19c8)
#1 0x5569f26b5c4c in std::default_delete<bpftrace::AttachedProbe>::operator()(bpftrace::AttachedProbe*) const /usr/include/c++/7/bits/unique_ptr.h:78
#2 0x5569f26b1364 in std::unique_ptr<bpftrace::AttachedProbe, std::default_delete<bpftrace::AttachedProbe> >::~unique_ptr() (/home/vagrant/build/src/bpftrace+0xdd364)
#3 0x5569f26acc41 in void std::_Destroy<std::unique_ptr<bpftrace::AttachedProbe, std::default_delete<bpftrace::AttachedProbe> > >(std::unique_ptr<bpftrace::AttachedProbe, std::default_delete<bpftrace::AttachedProbe> >*) (/home/vagrant/build/src/bpftrace+0xd8c41)
```
Removing the ifdef fixes is. Having the extra field won't hurt.
Fixes #1853
prakashsurya
pushed a commit
that referenced
this pull request
Nov 8, 2022
On older kernels that don't support BTF, BPFFeature::try_load causes a segfault by passing in a NULL pointer to bpf_find_by_name_kind: Program received signal SIGSEGV, Segmentation fault. (gdb) bt #0 0x95101db8 in btf_find_by_name_kind (btf=0x0, start_id=1, [...]) #1 0x0a74695e in bpftrace::BTF::get_btf_id #2 0x0a716fe8 in bpftrace::BPFfeature::try_load #3 0x0a717546 in bpftrace::BPFfeature::detect_prog_type #4 0x0a71963c in bpftrace::BPFfeature::has_prog_kfunc #5 0x0a7181e8 in bpftrace::BPFfeature::has_kfunc [...]
delphix-devops-bot
pushed a commit
that referenced
this pull request
Mar 13, 2023
The change now fixes an invalid typecast rather than variable assignment.
When CodegenLLVM::visit calls CreateStore() with NULL expr_ as parameter,
segfault is generated, for example:
$ cat sample.bt
#include <linux/fs.h>
kprobe:vfs_open {
$file = (struct file *)
printf("Hello\n");
}
Acctually, we want to '$file = (struct file *)arg1;' here. When writing
incorrectly '$file = (struct file *)', printf() make expr_ == nullptr,
when call CreateStore(NULL, ...), segfault has occurred, and the stack
backtrace is as follows:
(gdb) bt
#0 0x00000000007098c6 in llvm::Value::getType (this=0x0) at /usr/include/llvm/IR/Value.h:255
#1 0x000000000070cc76 in llvm::IRBuilderBase::CreateAlignedStore (this=0x7fffffffbe40, Val=0x0,
Ptr=0x15e2340, Align=..., isVolatile=false) at /usr/include/llvm/IR/IRBuilder.h:1689
#2 0x000000000070cab3 in llvm::IRBuilderBase::CreateStore (this=0x7fffffffbe40, Val=0x0, Ptr=0x15e2340,
isVolatile=false) at /usr/include/llvm/IR/IRBuilder.h:1663
#3 0x00000000006f6948 in bpftrace::ast::CodegenLLVM::visit (this=0x7fffffffbe10, assignment=...)
at /home/rongtao/Git/bpftrace/src/ast/passes/codegen_llvm.cpp:2140
#4 0x0000000000751258 in bpftrace::ast::AssignVarStatement::accept (this=0x7fffe99b9a30, v=...)
at /home/rongtao/Git/bpftrace/src/ast/ast.cpp:35
#5 0x00000000006ff6ea in bpftrace::ast::CodegenLLVM::accept (this=0x7fffffffbe10, node=0x7fffe99b9a30)
at /home/rongtao/Git/bpftrace/src/ast/passes/codegen_llvm.cpp:3360
We should prompt the user for some useful information, like:
$ sudo ./sample.bt
./sample.bt:4:11-26: ERROR: Cannot cast from "none" type
$file = (struct file *)
~~~~~~~~~~~~~~~
LLVM/Clang Version: 15.0.1
Signed-off-by: Rong Tao <[email protected]>
delphix-devops-bot
pushed a commit
that referenced
this pull request
Jul 3, 2024
If it is dummy, then the visit() operation should not be executed. For
example, on aarch64, there is no syscalls:sys_{enter,exit}_open
tracepoint, however, opensnoop.bt tool will try to attch them and visit
args's field. We expect to get a warning, but instead we get a segmentation
fault. As follows:
$ sudo gdb bpftrace
...
(gdb) set args opensnoop.bt
(gdb) r
opensnoop.bt:22-24: WARNING: tracepoint not found: syscalls:sys_enter_open
opensnoop.bt:28-30: WARNING: tracepoint not found: syscalls:sys_exit_open
Thread 1 "bpftrace" received signal SIGSEGV, Segmentation fault.
0x000000000052bbd8 in __gnu_cxx::__normal_iterator<bpftrace::Field const*, std::vector<bpftrace::Field, std::allocator<bpftrace::Field> > >::__normal_iterator (this=0xffffffffb0b8, __i=<error reading variable: Cannot access memory at address 0x10>)
at /usr/include/c++/14/bits/stl_iterator.h:1068
1068 : _M_current(__i) { }
(gdb) bt
#0 0x000000000052bbd8 in __gnu_cxx::__normal_iterator<bpftrace::Field const*, std::vector<bpftrace::Field, std::allocator<bpftrace::Field> > >::__normal_iterator (this=0xffffffffb0b8,
__i=<error reading variable: Cannot access memory at address 0x10>) at /usr/include/c++/14/bits/stl_iterator.h:1068
#1 0x0000000000525bdc in std::vector<bpftrace::Field, std::allocator<bpftrace::Field> >::begin (this=0x10)
at /usr/include/c++/14/bits/stl_vector.h:884
#2 0x000000000059ce8c in bpftrace::Struct::GetField (this=0x0, name="filename")
at /home/rongtao/Git/bpftrace/bpftrace/src/struct.cpp:131
#3 0x00000000005a70a0 in bpftrace::SizedType::GetField (this=0xe9b998, name="filename")
at /home/rongtao/Git/bpftrace/bpftrace/src/types.cpp:538
#4 0x00000000007c7178 in bpftrace::ast::CodegenLLVM::visit (this=0xffffffffc650, acc=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:1973
#5 0x0000000000847db0 in bpftrace::ast::FieldAccess::accept (this=0xe9be30, v=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/ast.cpp:31
#6 0x00000000007d2404 in bpftrace::ast::CodegenLLVM::accept (this=0xffffffffc650, node=0xe9be30)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:3778
#7 0x00000000007c8bec in bpftrace::ast::CodegenLLVM::visit (this=0xffffffffc650, assignment=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:2218
#8 0x0000000000847eb4 in bpftrace::ast::AssignMapStatement::accept (this=0xffffe4429250, v=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/ast.cpp:36
#9 0x00000000007d2404 in bpftrace::ast::CodegenLLVM::accept (this=0xffffffffc650, node=0xffffe4429250)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:3778
#10 0x00000000007ca4c4 in bpftrace::ast::CodegenLLVM::generateProbe (this=0xffffffffc650, probe=..., full_func_id="dummy",
name="dummy", func_type=0xff2b50, usdt_location_index=std::optional [no contained value], dummy=true)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:2539
#11 0x00000000007cb4e4 in bpftrace::ast::CodegenLLVM::visit (this=0xffffffffc650, probe=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:2734
#12 0x00000000008480f0 in bpftrace::ast::Probe::accept (this=0xffffe4427650, v=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/ast.cpp:47
#13 0x00000000007d2404 in bpftrace::ast::CodegenLLVM::accept (this=0xffffffffc650, node=0xffffe4427650)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:3778
#14 0x00000000007cb7a4 in bpftrace::ast::CodegenLLVM::visit (this=0xffffffffc650, program=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:2752
#15 0x000000000084818c in bpftrace::ast::Program::accept (this=0xe90f50, v=...)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/ast.cpp:50
#16 0x00000000007d2404 in bpftrace::ast::CodegenLLVM::accept (this=0xffffffffc650, node=0xe90f50)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:3778
#17 0x00000000007d03fc in bpftrace::ast::CodegenLLVM::generate_ir (this=0xffffffffc650)
at /home/rongtao/Git/bpftrace/bpftrace/src/ast/passes/codegen_llvm.cpp:3450
#18 0x0000000000478138 in main (argc=2, argv=0xfffffffff4c8) at /home/rongtao/Git/bpftrace/bpftrace/src/main.cpp:898
We can simplify opensnoop.bt to:
tracepoint:syscalls:sys_enter_open_not_exist,
tracepoint:syscalls:sys_enter_openat
{
@ = args.filename;
}
This will produce the following error:
stdin:1:1-45: WARNING: tracepoint not found: syscalls:sys_enter_open_not_exist
tracepoint:syscalls:sys_enter_open_not_exist,tracepoint:syscalls:sys_enter_openat {@ = args.filename;}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Segmentation fault
We need to generate the dummy probe only for cases when none of the probe
attach points exists.
Link: bpftrace/bpftrace#3274
Signed-off-by: Viktor Malik <[email protected]>
Signed-off-by: Rong Tao <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This depends on delphix/linux-pkg#54
bpftrace is now available as a package on Ubuntu 19.04, and it has the dependencies below.
Switching to llvm7/clang7 fixes bpftrace/bpftrace#76, so we should be able to remove the custom llvm5/clang5 ppa we are using in linux-pkg to build bpftrace.
Testing
ab-pre-push: http://selfservice.jenkins.delphix.com/job/devops-gate/job/master/job/appliance-build-orchestrator-pre-push/2184/ (pass)