linux-kernel - Re: [RESEND] Fast TSC calibration fails with v3.14-rc1 and later

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 10 Mar 2014 11:39:44 +0100 (CET)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Julian Wollrath <jwollrath@....de>
cc:	x86@...nel.org, linux-kernel@...r.kernel.org
Subject: Re: [RESEND] Fast TSC calibration fails with v3.14-rc1 and later

On Mon, 10 Mar 2014, Julian Wollrath wrote:
> on a Thinkpad x121e (AMD E-450 APU) with version 3.13.5 and earlier of
> the kernel I get:
> [...]
> [    0.000000] tsc: Fast TSC calibration using PIT
> [    0.001000] tsc: Detected 1646.619 MHz processor
> [    0.000004] Calibrating delay loop (skipped), value calculated using
> timer frequency.. 3293.23 BogoMIPS (lpj=1646619)
> [...]
> 
> 
> but starting with v3.14-rc1 the fast TSC calibration fails:
> [...]
> [    0.000000] tsc: Fast TSC calibration failed
> [    0.000000] tsc: PIT calibration matches HPET. 1 loops
> [    0.000000] tsc: Detected 1646.490 MHz processor
> [...]

The only change in the calibration code post 3.13 is that we added the
MSR based calibration for ATOM:

@@ -419,6 +651,13 @@ unsigned long native_calibrate_tsc(void)
 	unsigned long flags, latch, ms, fast_calibrate;
 	int hpet = is_hpet_enabled(), i, loopmin;
 
+	/* Calibrate TSC using MSR for Intel Atom SoCs */
+	local_irq_save(flags);
+	fast_calibrate = try_msr_calibrate_tsc();
+	local_irq_restore(flags);
+	if (fast_calibrate)
+		return fast_calibrate;
+
 	local_irq_save(flags);
 	fast_calibrate = quick_pit_calibrate();
 	local_irq_restore(flags);

And that definitely does not affect the quick calibration. No idea,
except bisecting.

Thanks,

	tglx


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/