Thanks for the report Paul (and also the investigation work Philippe). So yes it seems the DFP code is another fallout from the conversion of the floating point registers over to host-endian/VSR format. I've had a quick look at this and it seems that the simple fix to compensate for the FP registers not being contiguous anymore still won't work on ppc64le.
In order to fix this properly I think the best solution is to use an approach similar to that used in my last set of VSX patches, i.e. using macros to avoid having separate code paths for big and little endian hosts. I can certainly come up with some patches for this, however I don't have any ppc64le hardware to test it myself. If I were to do a trial conversion of denbcdq would you be able to test it for me? -- 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
