m2-internship/report
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Update source code example

Browse source
This commit is contained in:
Théophile Bastian 2018-08-03 13:59:11 +02:00
parent 7a4ad4454e
commit a06b1a915a
7 changed files with 79 additions and 37 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
report/TODO
View file

@ -1,3 +1,4 @@
- Add a sample of geenrated code?
- Explicitly tell that we only support unwinding, and this only for a few
registers
- Make consistant -ise / -ize (eg. optimize)

71
report/report.tex
View file

@ -43,12 +43,6 @@ Under supervision of Francesco Zappa-Nardelli\\
%% Fiche de synthèse %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\input{fiche_synthese}
%% Abstract %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{abstract}
\todo{Is there a need for an abstract, given the presence above of the
``fiche de synthèse''?}
\end{abstract}
%% Table of contents %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\tableofcontents
@ -185,34 +179,52 @@ valid from its start IP to the start IP of the next row, or the end IP of the
FDE if it is the last row.
\begin{minipage}{0.45\textwidth}
\lstinputlisting[language=C, firstline=3, lastline=12]
\lstinputlisting[language=C, firstline=3, lastline=12,
caption={Original C},label={lst:ex1_c}]
{src/fib7/fib7.c}
\end{minipage} \hfill \begin{minipage}{0.45\textwidth}
\lstinputlisting[language=C]{src/fib7/fib7.fde}
\lstinputlisting[language=C,caption={Processed DWARF},label={lst:ex1_dw}]
{src/fib7/fib7.fde}
\lstinputlisting[language=C,caption={Raw DWARF},label={lst:ex1_dwraw}]
{src/fib7/fib7.raw_fde}
\end{minipage}
For instance, the C source code above, when compiled with \lstbash{gcc -O0
-fomit-frame-pointer}, gives the table at its right. During the function
prelude, \ie{} for $\mhex{675} \leq \reg{rip} < \mhex{679}$, the stack frame
only contains the return address, thus the CFA is 8 bytes above \reg{rsp}
(which was the value of \reg{rsp} before the call), and the return address is
precisely at \reg{rsp}. Then, 9 integers of 8 bytes each (8 for \lstc{fibo},
one for \lstc{pos}) are allocated on the stack, which puts the CFA 80 bytes
above \reg{rsp}, and the return address still 8 bytes below the CFA\@. Then, by
the end of the function, the local variables are discarded and \reg{rsp} is
reset to its value from the first row.
\begin{minipage}{0.45\textwidth}
\lstinputlisting[language={[x86masm]Assembler},lastline=11,
caption={Generated assembly},label={lst:ex1_asm}]
{src/fib7/fib7.s}
\end{minipage} \hfill \begin{minipage}{0.45\textwidth}
\lstinputlisting[language={[x86masm]Assembler},firstline=12,
firstnumber=last]
{src/fib7/fib7.s}
\end{minipage}
However, DWARF data isn't actually stored as a table in the binary files. The
first row has the location of the first IP in the FDE, and must define at least
its CFA\@. Then, when all relevant registers are defined, it is possible to
define a new row by providing a location offset (\eg{} here $4$), and the new
row is defined as a clone of the previous one, which can then be altered (\eg{}
here by setting \lstc{CFA} to $\reg{rsp} + 80$). This means that every line is
defined \wrt{} the previous one, and that the IPs of the successive rows cannot
be determined before evaluating every row before. Thus, unwinding a frame from
an IP close to the end of the frame will require evaluating pretty much every
DWARF row in the table before reaching the relevant information, slowing down
drastically the unwinding process.
For instance, the C source code in Listing~\ref{lst:ex1_c} above, when compiled
with \lstbash{gcc -O1 -fomit-frame-pointer -fno-stack-protector}, yields the
assembly code in Listing~\ref{lst:ex1_asm}. When interpreting the generated
\ehframe{} with \lstbash{readelf -wF}, we obtain the (slightly edited)
Listing~\ref{lst:ex1_dw}. During the function prelude, \ie{} for $\mhex{615}
\leq \reg{rip} < \mhex{619}$, the stack frame only contains the return address,
thus the CFA is 8 bytes above \reg{rsp} (which was the value of \reg{rsp}
before the call), and the return address is precisely at \reg{rsp}. Then, 9
integers of 8 bytes each (8 for \lstc{fibo}, one for \lstc{pos}) are allocated
on the stack, which puts the CFA 80 bytes above \reg{rsp}, and the return
address still 8 bytes below the CFA\@. Then, by the end of the function, the
local variables are discarded and \reg{rsp} is reset to its value from the
first row.
However, DWARF data isn't actually stored as a table in the binary files, but
is instead stored as in Listing~\ref{lst:ex1_dwraw}. The first row has the
location of the first IP in the FDE, and must define at least its CFA\@. Then,
when all relevant registers are defined, it is possible to define a new row by
providing a location offset (\eg{} here $4$), and the new row is defined as a
clone of the previous one, which can then be altered (\eg{} here by setting
\lstc{CFA} to $\reg{rsp} + 80$). This means that every line is defined \wrt{}
the previous one, and that the IPs of the successive rows cannot be determined
before evaluating every row before. Thus, unwinding a frame from an IP close to
the end of the frame will require evaluating pretty much every DWARF row in the
table before reaching the relevant information, slowing down drastically the
unwinding process.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{How big are FDEs?}
@ -565,6 +577,7 @@ The major drawback of this approach, without any particular care taken, is the
space waste.
\begin{table}[h]
\centering
\begin{tabular}{r r r r r r}
\toprule
\thead{Shared object} & \thead{Original \\ program size}

2
report/src/fib7/Makefile
View file

@ -1,4 +1,4 @@
all: fib7.bin
fib7.bin: fib7.c
gcc -O1 $< -o $@
gcc -fomit-frame-pointer -fno-stack-protector -O1 $< -o $@

6
report/src/fib7/fib7.c
View file

@ -1,6 +1,6 @@
#include <stdio.h>
int fib7() {
void fib7() {
int fibo[8];
fibo[0] = 1;
fibo[1] = 1;
@ -8,10 +8,10 @@ int fib7() {
fibo[pos] =
fibo[pos - 1]
+ fibo[pos - 2];
return fibo[7];
printf("%d\n", fibo[7]);
}
int main(void) {
printf("%d\n", fib7());
fib7();
return 0;
}

8
report/src/fib7/fib7.fde
View file

@ -1,5 +1,5 @@
[...] FDE [...] pc=675..6f3
[...] FDE [...] pc=615..65a
LOC CFA ra
0000000000000675 rsp+8 c-8
0000000000000679 rsp+80 c-8
00000000000006f2 rsp+8 c-8
0000000000000615 rsp+8 c-8
0000000000000619 rsp+48 c-8
0000000000000659 rsp+8 c-8

7
report/src/fib7/fib7.raw_fde Normal file
View file

@ -0,0 +1,7 @@
[...] FDE [...] pc=615..65a
DW_CFA_def_cfa: r7 (rsp) ofs 8
DW_CFA_offset: r16 (rip) at cfa-8
DW_CFA_advance_loc: 4 to 0619
DW_CFA_def_cfa_offset: 48
DW_CFA_advance_loc1: 64 to 0659
DW_CFA_def_cfa_offset: 8

21
report/src/fib7/fib7.s Normal file
View file

@ -0,0 +1,21 @@
0000000000000615 <fib7>:
615: sub $0x28,%rsp ; Alloc stack
619: movl $0x1,(%rsp) ; fibo[0]
620: movl $0x1,0x4(%rsp) ; fibo[1]
627:
; Begin for loop
628: mov %rsp,%rax
62b: lea 0x18(%rax),%rcx
62f: mov (%rax),%edx
631: add 0x4(%rax),%edx
634: mov %edx,0x8(%rax)
637: add $0x4,%rax
63b: cmp %rcx,%rax
63e: jne 62f <fib7+0x1a>
; End for loop
640: mov 0x1c(%rsp),%esi
644: lea 0xb9(%rip),%rdi
64b: mov $0x0,%eax
650: callq 520 <printf@plt>
655: add $0x28,%rsp ; Restore rsp
659: retq