https://bugs.kde.org/show_bug.cgi?id=357932
Bug ID: 357932
Summary: vex amd64->IR: unhandled instruction bytes: 0xF2 0x49
0xF 0x5D and 0xF2 0x49 0xF 0x5F
Product: valgrind
Version: 3.10 SVN
Platform: Compiled Sources
OS: Linux
Status: UNCONFIRMED
Severity: normal
Priority: NOR
Component: vex
Assignee: [email protected]
Reporter: [email protected]
Created attachment 96622
--> https://bugs.kde.org/attachment.cgi?id=96622&action=edit
patch to decode both invalid instruction bytes
Valgrind terminates for application which is using the Intel IPP library.
vex amd64->IR: unhandled instruction bytes: 0xF2 0x49 0xF 0x5D 0x0 0x49 0x83
0xC0
vex amd64->IR: REX=1 REX.W=1 REX.R=0 REX.X=0 REX.B=1
vex amd64->IR: VEX=0 VEX.L=0 VEX.nVVVV=0x0 ESC=0F
vex amd64->IR: PFX.66=0 PFX.F2=1 PFX.F3=0
I tried then to patch the code from SVN (r15755). My approach was similiar to
bug #278744 and bug #307637.
Then i started Valgrind again. This time I've got another unhandled
instruction:
vex amd64->IR: unhandled instruction bytes: 0xF2 0x49 0xF 0x5F 0x0 0x49 0x83
0xC0
vex amd64->IR: REX=1 REX.W=1 REX.R=0 REX.X=0 REX.B=1
vex amd64->IR: VEX=0 VEX.L=0 VEX.nVVVV=0x0 ESC=0F
vex amd64->IR: PFX.66=0 PFX.F2=1 PFX.F3=0
Again I patched the code. Now Valgrind does not terminate anymore but it spills
out a lot of invalid reads in the IPPs log function (which uses SSE)
Invalid read of size 8
ippsLn_32f_A11
Although, I trust Valgrind a lot I doubt that our application is doing invalid
reads here because I double checked with GCCs address sanitizer and it didn't
report anything. Thus, I guess my patches are incorrect. I have attached the
patches and would be happy if someone with more competence than me would have a
look at them.
--
You are receiving this mail because:
You are watching all bug changes.
