http://sourceware.org/bugzilla/show_bug.cgi?id=12608
--- Comment #4 from Witold Baryluk <baryluk at smp dot if.uj.edu.pl> 2011-03-31
00:32:44 UTC ---
Additionally, .o files are only differing in this places
$ cmp --verbose source-ev{4,5}.o
871 35 175
872 42 247
875 175 35
876 247 42
8409 44 50
8433 50 44
$
This actually do not exactly much my understaning relating object code and this
numers (they do not much when doing, hex<->dec, but I will blame cmp reporting
in strange way). Here is the same file piped thru hd, and then compared.
baryluk@a09:~/tls-relocation-bug$ diff -Nu source-ev4.o.hd source-ev5.o.hd
--- source-ev4.o.hd 2011-03-31 01:37:45.000000000 +0200
+++ source-ev5.o.hd 2011-03-31 01:37:49.000000000 +0200
@@ -52,7 +52,7 @@
00000330 2d 00 20 e4 00 00 3d a4 0d 00 5f 20 22 37 44 48 |-. ...=..._ "7DH|
00000340 01 04 20 40 19 00 42 20 00 00 61 a4 23 36 7e 48 |.. @..B ..a.#6~H|
00000350 a2 05 62 40 24 00 40 e4 08 00 21 a4 a1 75 23 40 |..b@$.@...!..u#@|
-00000360 21 00 20 e4 00 00 1d 22 00 00 7d a7 00 40 5b 6b |!. ...."..}..@[k|
+00000360 21 00 20 e4 00 00 7d a7 00 00 1d 22 00 40 5b 6b |!. ...}....".@[k|
00000370 00 00 ba 27 00 00 00 24 11 00 3f 20 00 00 40 a4 |...'...$..? ..@.|
00000380 00 00 bd 23 21 17 24 48 22 36 5e 48 10 00 21 20 |...#!.$H"6^H..! |
00000390 a1 05 41 40 14 00 20 e4 00 00 00 20 08 00 20 a4 |..A@.. .... .. .|
@@ -521,9 +521,9 @@
000020a0 10 00 00 00 00 00 00 00 bc 02 00 00 00 00 00 00 |................|
000020b0 25 00 00 00 18 00 00 00 00 00 00 00 00 00 00 00 |%...............|
000020c0 f4 02 00 00 00 00 00 00 25 00 00 00 1a 00 00 00 |........%.......|
-000020d0 00 00 00 00 00 00 00 00 24 03 00 00 00 00 00 00 |........$.......|
+000020d0 00 00 00 00 00 00 00 00 28 03 00 00 00 00 00 00 |........(.......|
000020e0 1e 00 00 00 07 00 00 00 00 00 00 00 00 00 00 00 |................|
-000020f0 28 03 00 00 00 00 00 00 04 00 00 00 13 00 00 00 |(...............|
+000020f0 24 03 00 00 00 00 00 00 04 00 00 00 13 00 00 00 |$...............|
00002100 00 00 00 00 00 00 00 00 2c 03 00 00 00 00 00 00 |........,.......|
00002110 05 00 00 00 01 00 00 00 05 00 00 00 00 00 00 00 |................|
00002120 2c 03 00 00 00 00 00 00 08 00 00 00 13 00 00 00 |,...............|
PS. Ah, sorry, cmp --verbose, output in base-8. Now it matches.
This is what I got in resulting .so file in my case (remember binutils 2.18),
$ hd tst-tlsmod7-z-ev4.so | grep '75 23' -A 2
00000ec0 a2 05 62 40 24 00 40 e4 08 00 21 a4 a1 75 23 40 |..b@$.@...!..u#@|
00000ed0 21 00 20 e4 58 80 1d a6 00 00 fe 2f 9e 00 00 00 |!. .X....../....|
00000ee0 00 04 00 42 00 00 00 24 11 00 3f 20 58 00 40 a4 |...B...$..? X.@.|
After locating it in objdump -d, I have this
ebc: 23 36 7e 48 zapnot t2,0xf1,t2
ec0: a2 05 62 40 cmpeq t2,t1,t1
ec4: 24 00 40 e4 beq t1,f58 <check1+0x178>
ec8: 08 00 21 a4 ldq t0,8(t0)
ecc: a1 75 23 40 cmpeq t0,0x1b,t0
ed0: 21 00 20 e4 beq t0,f58 <check1+0x178>
ed4: 58 80 1d a6 ldq a0,-32680(gp)
ed8: 00 00 fe 2f unop
edc: 9e 00 00 00 rduniq
ee0: 00 04 00 42 addq a0,v0,v0
ee4: 00 00 00 24 ldah v0,0(v0)
for both ev4 and ev5 .so. This matches a result of Aurelien's ev4.so
I will try bissecting problem.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
_______________________________________________
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
