In 'prec1' commit, modified imm encoding was changed to use constant amount of time for any immediate.
Sadly, new encoding became not canonical i.e. rotate value may be not minimal as possible, which is not recommended, but allowed by ARM. Start of -XX:+PrintInterpreter looks like
  0x72df8320: mov r3, sp
  0x72df8324: sub sp, sp, #88 ; 0x58
  0x72df8328: str lr, [sp]
  0x72df832c: mov r2, r6
  0x72df8330: mov r1, sl
  0x72df8334: str r5, [fp, #-32] ; 0xffffffe0
  0x72df8338: ldr r9, [fp, #-12]
  0x72df833c: cmp r9, #0, 8
  0x72df8340: beq 0x72df8374
  0x72df8344: push {r0, r1, r2, r3, r4, r5, r6, r7, r8, r9, sl, fp, ip, lr}
  0x72df8348: add ip, sp, #224, 2 ; 0x38
  0x72df834c: push {ip} ; (str ip, [sp, #-4]!)
  0x72df8350: movw r0, #33092 ; 0x8144
  0x72df8354: movt r0, #30085 ; 0x7585
  0x72df8358: mov r1, pc
  0x72df835c: sub r1, r1, #128, 4

Note instructions like
  0x72df833c: cmp r9, #0, 8 // zero encoded as 0 rotated at 8.

The patch restores recommended encoding (still constant time encoding), after applying listing looks like
  0x72dae320: mov r3, sp
  0x72dae324: sub sp, sp, #88 ; 0x58
  0x72dae328: str lr, [sp]
  0x72dae32c: mov r2, r6
  0x72dae330: mov r1, sl
  0x72dae334: str r5, [fp, #-32] ; 0xffffffe0
  0x72dae338: ldr r9, [fp, #-12]
  0x72dae33c: cmp r9, #0
  0x72dae340: beq 0x72dae374
  0x72dae344: push {r0, r1, r2, r3, r4, r5, r6, r7, r8, r9, sl, fp, ip, lr}
  0x72dae348: add ip, sp, #56 ; 0x38
  0x72dae34c: push {ip} ; (str ip, [sp, #-4]!)
  0x72dae350: movw r0, #45548 ; 0xb1ec
  0x72dae354: movt r0, #30080 ; 0x7580
  0x72dae358: mov r1, pc
  0x72dae35c: sub r1, r1, #8

Sorry for temporary inconvenience in debugging. Happy hacking again!