Bug 396001

Summary:	unhandled instruction: 0xEC51 0x0F1E; ARMv7 libcrypto 'mrrc'
Product:	[Developer tools] valgrind	Reporter:	John Reiser <jreiser>
Component:	memcheck	Assignee:	Julian Seward <jseward>
Status:	RESOLVED DUPLICATE
Severity:	normal	CC:	peter.maydell, taborkelly
Priority:	NOR
Version:	3.13.0
Target Milestone:	---
Platform:	Fedora RPMs
OS:	Linux
Latest Commit:		Version Fixed In:

Description John Reiser 2018-06-29 15:51:29 UTC

On 32-bit ARMv7 the file libcrypto.so.1.1 in package openssl-libs tries to execute 0xEC51 0x0F1E which is "mrrc	p15,0,r0,r1,c14".  valgrind does not handle this instruction.  The instruction is part of subroutine _armv7_tick() called from OPENSSL_cpuid_setup() to interrogate the hardware capabilities of the cpu.

Please handle this instruction.

Comment 1 Peter Maydell 2018-06-29 16:02:00 UTC

0xEC510F1E is actually "mrrc	p15,1,r0,r1,c14", which is an access to CNTVCT. I think this is a duplicate of #344802.

Comment 2 John Reiser 2018-06-30 01:00:25 UTC

OK, it's the same as #344802.  It would be nice to fix a bug that is 2.5 to 3 years old, depending on how you count.  Anyone who uses libcrypto.so.1.1 on ARM will stumble on this bug.

The [valgrind-users] report "Valgrind on arm system startup message" of Thu, 28 Jun 2018 20:21:05 -0400 gives some distracting information:
=====
==621== Using Valgrind-3.13.0 and LibVEX; rerun with -h for copyright info
==621== Command: ./myprogram
==621==
disInstr(thumb): unhandled instruction: 0xEC51 0x0F1E
==621== valgrind: Unrecognised instruction at address 0x4cc1767.
==621==    at 0x4CC1766: ??? (in /usr/lib/arm-linux-gnueabihf/libcrypto.so.1.1) 
=====

Notice that it says "thumb", with a not-even PC address 0x4cc1767, and that the disassembly is two 16-bit words instead of one 32-bit word.  Fooled me.

*** This bug has been marked as a duplicate of bug 344802 ***

Comment 3 Julian Seward 2018-09-03 09:40:31 UTC

*** Bug 397256 has been marked as a duplicate of this bug. ***