d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

mca: use time_uptime instead of ticks for CMCI throttling

Browse Source 
This solves several problems.
First of all, cmc_throttle is specified in seconds and there was no
conversion between ticks and seconds when they were mixed together.
Second, we avoid potential problems with ticks wrapping around.

Resolution of time_uptime should be sufficient for the throttling
purposes.

Discussed with:	jhb
MFC after:	12 days
This commit is contained in:
avg 2017-02-14 22:46:39 +00:00
parent 2e3128b7ec
commit 3cf45460dd
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
sys/x86/x86/mca.c
View File

@ -533,7 +533,7 @@ cmci_update(enum scan_mode mode, int bank, int valid, struct mca_record *rec)
cc = &cmc_state[PCPU_GET(cpuid)][bank];
ctl = rdmsr(MSR_MC_CTL2(bank));
count = (rec->mr_status & MC_STATUS_COR_COUNT) >> 38;
delta = (u_int)(ticks - cc->last_intr);
delta = (u_int)(time_uptime - cc->last_intr);
/*
* If an interrupt was received less than cmc_throttle seconds
@ -550,7 +550,7 @@ cmci_update(enum scan_mode mode, int bank, int valid, struct mca_record *rec)
ctl |= limit;
wrmsr(MSR_MC_CTL2(bank), ctl);
}
cc->last_intr = ticks;
cc->last_intr = time_uptime;
return;
}
@ -857,7 +857,7 @@ cmci_resume(int i)
return;
cc = &cmc_state[PCPU_GET(cpuid)][i];
cc->last_intr = -ticks;
cc->last_intr = 0;
ctl = rdmsr(MSR_MC_CTL2(i));
ctl &= ~MC_CTL2_THRESHOLD;
ctl |= MC_CTL2_CMCI_EN | 1;