GNUBoot/coreboot-kgpe-d16
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
coreboot-kgpe-d16/src/soc/amd/stoneyridge/monotonic_timer.c

26 lines
447 B
C
Raw Normal View History

soc/amd: replace remaining license headers with SPDX ones Change-Id: Ib45e93faebc2d24389f8739911419dfec437bd59 Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/40505 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
2020-04-18 22:26:39 +02:00
/* SPDX-License-Identifier: GPL-2.0-only */
soc/amd/stoneyridge: provide alternate monotonic timer The TSC has been observed to be ticking at a non-constant rate in early boot. The root cause is still not known, but this misbehavior necessitates an alternative monotonic timer source. Use the perf TSC which ticks at 100 MHz. This also means the timestamp table is not accurate as well. Root cause of TSC rate instability needs to be resolved in order to fix that. BUG=b:72170796 Change-Id: Ie052169868a9d9f25f8cc0ce8dd8251b560e671f Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23397 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-01-23 18:53:05 +01:00
#include <cpu/x86/msr.h>
#include <timer.h>
soc/amd/stoneyridge: remove dependence on TSC The TSC rate is empirically swinging during early boot. That leaves timestamps and udelay()s to not be correct. To rectify this stop using TSC for all of these time sources. Instead use the performance TSC which is at a fixed 100MHz clock. That provides stable time sources and legit timestamps. BUG=b:72378235,b:72170796 Change-Id: Ia2693c415c557aac687bcb48ee69358ea1c53d67 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23424 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2018-01-25 01:42:51 +01:00
#include <timestamp.h>
soc/amd/stoneyridge: provide alternate monotonic timer The TSC has been observed to be ticking at a non-constant rate in early boot. The root cause is still not known, but this misbehavior necessitates an alternative monotonic timer source. Use the perf TSC which ticks at 100 MHz. This also means the timestamp table is not accurate as well. Root cause of TSC rate instability needs to be resolved in order to fix that. BUG=b:72170796 Change-Id: Ie052169868a9d9f25f8cc0ce8dd8251b560e671f Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23397 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-01-23 18:53:05 +01:00
#define CU_PTSC_MSR 0xc0010280
#define PTSC_FREQ_MHZ 100
void timer_monotonic_get(struct mono_time *mt)
soc/amd/stoneyridge: remove dependence on TSC The TSC rate is empirically swinging during early boot. That leaves timestamps and udelay()s to not be correct. To rectify this stop using TSC for all of these time sources. Instead use the performance TSC which is at a fixed 100MHz clock. That provides stable time sources and legit timestamps. BUG=b:72378235,b:72170796 Change-Id: Ia2693c415c557aac687bcb48ee69358ea1c53d67 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23424 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2018-01-25 01:42:51 +01:00
{
mono_time_set_usecs(mt, timestamp_get());
}
uint64_t timestamp_get(void)
soc/amd/stoneyridge: provide alternate monotonic timer The TSC has been observed to be ticking at a non-constant rate in early boot. The root cause is still not known, but this misbehavior necessitates an alternative monotonic timer source. Use the perf TSC which ticks at 100 MHz. This also means the timestamp table is not accurate as well. Root cause of TSC rate instability needs to be resolved in order to fix that. BUG=b:72170796 Change-Id: Ie052169868a9d9f25f8cc0ce8dd8251b560e671f Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23397 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-01-23 18:53:05 +01:00
{
unsigned long long val;
msr_t msr;
msr = rdmsr(CU_PTSC_MSR);
val = ((unsigned long long)msr.hi << 32) | msr.lo;
soc/amd/stoneyridge: remove dependence on TSC The TSC rate is empirically swinging during early boot. That leaves timestamps and udelay()s to not be correct. To rectify this stop using TSC for all of these time sources. Instead use the performance TSC which is at a fixed 100MHz clock. That provides stable time sources and legit timestamps. BUG=b:72378235,b:72170796 Change-Id: Ia2693c415c557aac687bcb48ee69358ea1c53d67 Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23424 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2018-01-25 01:42:51 +01:00
return val / PTSC_FREQ_MHZ;
soc/amd/stoneyridge: provide alternate monotonic timer The TSC has been observed to be ticking at a non-constant rate in early boot. The root cause is still not known, but this misbehavior necessitates an alternative monotonic timer source. Use the perf TSC which ticks at 100 MHz. This also means the timestamp table is not accurate as well. Root cause of TSC rate instability needs to be resolved in order to fix that. BUG=b:72170796 Change-Id: Ie052169868a9d9f25f8cc0ce8dd8251b560e671f Signed-off-by: Aaron Durbin <adurbin@chromium.org> Reviewed-on: https://review.coreboot.org/23397 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Justin TerAvest <teravest@chromium.org>
2018-01-23 18:53:05 +01:00
}
Reference in a new issue Copy permalink