long kernel pause on boot sometimes a panic

kernel: Linux version 3.19.3-1.gf10e7fc-default (geeko@buildhost) (gcc version 4.8.3 20141208 [gcc-4_8-branch revision 218481] (SUSE Linux) )
DMI: /DX79SI, BIOS SIX7910J.86A.0590.2013.0717.2343 07/17/2013

I have had his issue for a very long time over several kernels (3.x), the kernel will boot to the point of bringing up the cpu’s and then stop for up to 3 minutes or so. Sometimes it then proceeds ok and sometimes it panics
My last dmesg shows this. Why do the printk timings jump to 107 then back to 0.33?:

0.082413] smpboot: CPU0: Intel(R) Core(TM) i7-3960X CPU @ 3.30GHz (fam: 06, model: 2d, stepping: 07)
0.082762] TSC deadline timer enabled
0.082781] Performance Events: PEBS fmt1+, 16-deep LBR, SandyBridge events, full-width counters, Intel PMU driver.
0.083230] ... version:                3
0.083358] ... bit width:              48
0.083487] ... generic registers:      4
0.083615] ... value mask:             0000ffffffffffff
0.083747] ... max period:             0000ffffffffffff
0.083880] ... fixed-purpose events:   3
0.084008] ... event mask:             000000070000000f
0.084670] x86: Booting SMP configuration:
0.084800] .... node  #0, CPUs:        #1
0.098730] TSC synchronization [CPU#0 -> CPU#1]:
0.098982] Measured 521353464 cycles TSC warp between CPUs, turning off TSC clock.
0.099199] tsc: Marking TSC unstable due to check_tsc_sync_source failed
0.099469]   #2  #3  #4  #5  #6  #7  #8  #9 #10 #11

107.700418] x86: Booted up 1 node, 12 CPUs
107.700665] smpboot: Total of 12 processors activated (79206.33 BogoMIPS)
107.713419] NMI watchdog: enabled on all CPUs, permanently consumes one hw-PMU counter.
0.332037] devtmpfs: initialized
0.337187] PM: Registering ACPI NVS region [mem 0xd84a6000-0xd8589fff] (933888 bytes)

The pause point is when it prints out the cpu #, so it might go almost instantly to say cpu #4 then sit for 1 to 2 minutes and then proceed to boot ok, or it may panic instead of proceeding.
On other boots it rips right past this wtih no apparent problem. Is that TSC sync a problem? I’ve never seen that succeed on any of my pc’s, so maybe not?
Any thoughts? Would anyone know where is this in the kernel 3.19 source?