Sorry I didn't get a chance to look at this before I went away on holiday, however I've just posted a patchset at https://lists.gnu.org/archive/html/qemu-devel/2019-09/msg05577.html which should resolve the issue for you.
With the above patchset applied I now see the following results with your test program: LE host: $ ../qemu-ppc64le -L /usr/powerpc64le-linux-gnu -cpu power9 test-denbcdqle 0x00000000000000000000000000000000 0x0000000000000000000000000000000c 0x22080000000000000000000000000000 $ ../qemu-ppc64le -L /usr/powerpc64le-linux-gnu -cpu power9 test-denbcdqle 1 0x00000000000000000000000000000001 0x0000000000000000000000000000001c 0x22080000000000000000000000000001 $ ../qemu-ppc64le -L /usr/powerpc64le-linux-gnu -cpu power9 test-denbcdqle $(seq 0 99) 0x00000000000000000000000000000064 0x0000000000000000000000000000100c 0x22080000000000000000000000000080 BE host: $ ../qemu-ppc64 -L /usr/powerpc64-linux-gnu -cpu power9 test-denbcdq 0x00000000000000000000000000000000 0x000000000000000c0000000000000000 0x00000000000000002208000000000000 $ ../qemu-ppc64 -L /usr/powerpc64-linux-gnu -cpu power9 test-denbcdq 1 0x00000000000000010000000000000000 0x000000000000001c0000000000000000 0x00000000000000012208000000000000 $ ../qemu-ppc64 -L /usr/powerpc64-linux-gnu -cpu power9 test-denbcdq $(seq 0 99) 0x00000000000000640000000000000000 0x000000000000100c0000000000000000 0x00000000000000802208000000000000 If you could confirm that the BE host results above match those on real hardware then that would be great as I've switched over to use macros that should do the right thing regardless of host endian. Finally if you have access to a more comprehensive test suite then that would be helpful to test more of the 64-bit DFP number paths and some of more esoteric DFP instructions. -- You received this bug notification because you are a member of qemu- devel-ml, which is subscribed to QEMU. https://bugs.launchpad.net/bugs/1841990 Title: instruction 'denbcdq' misbehaving Status in QEMU: New Bug description: Instruction 'denbcdq' appears to have no effect. Test case attached. On ppc64le native: -- gcc -g -O -mcpu=power9 bcdcfsq.c test-denbcdq.c -o test-denbcdq $ ./test-denbcdq 0x00000000000000000000000000000000 0x0000000000000000000000000000000c 0x22080000000000000000000000000000 $ ./test-denbcdq 1 0x00000000000000000000000000000001 0x0000000000000000000000000000001c 0x22080000000000000000000000000001 $ ./test-denbcdq $(seq 0 99) 0x00000000000000000000000000000064 0x0000000000000000000000000000100c 0x22080000000000000000000000000080 -- With "qemu-ppc64le -cpu power9" -- $ qemu-ppc64le -cpu power9 -L [...] ./test-denbcdq 0x00000000000000000000000000000000 0x0000000000000000000000000000000c 0x0000000000000000000000000000000c $ qemu-ppc64le -cpu power9 -L [...] ./test-denbcdq 1 0x00000000000000000000000000000001 0x0000000000000000000000000000001c 0x0000000000000000000000000000001c $ qemu-ppc64le -cpu power9 -L [...] ./test-denbcdq $(seq 100) 0x00000000000000000000000000000064 0x0000000000000000000000000000100c 0x0000000000000000000000000000100c -- I started looking at the code, but I got confused rather quickly. Could be related to endianness? I think denbcdq arrived on the scene before little-endian was a big deal. Maybe something to do with utilizing implicit floating-point register pairs... I don't think the right data is getting to helper_denbcdq, which would point back to the gen_fprp_ptr uses in dfp-impl.inc.c (GEN_DFP_T_FPR_I32_Rc). (Maybe?) To manage notifications about this bug go to: https://bugs.launchpad.net/qemu/+bug/1841990/+subscriptions