http://defect.opensolaris.org/bz/show_bug.cgi?id=6265
Summary: mismerge in c-state wakeup could lead to poking the
wrong CPU
Classification: Development
Product: power-mgmt
Version: unspecified
Platform: Other
OS/Version: Solaris
Status: ACCEPTED
Severity: minor
Priority: P4
Component: c-state
AssignedTo: eric.saxe at sun.com
ReportedBy: eric.saxe at sun.com
CC: tesla-dev at opensolaris.org
While fixing a lint issue in cstate_wakeup() I found a bug:
/*
* c-state wakeup function.
* Similar to cpu_wakeup and cpu_wakeup_mwait except this function deals
* with CPUs asleep in MWAIT, HLT, or ACPI Deep C-State.
*/
void
cstate_wakeup(cpu_t *cp, int bound)
{
/*
* See if there's any other halted CPUs. If there are, then
* select one, and awaken it.
* It's possible that after we find a CPU, somebody else
* will awaken it before we get the chance.
* In that case, look again.
*/
do {
cpu_found = bitset_find(&cpu_part->cp_haltset); <------ 1
if (cpu_found == (uint_t) -1)
return;
} while (bitset_atomic_test_and_del(&cpu_part->cp_haltset, cpu_found) <
0);
/*
* Must use correct wakeup mechanism to avoid lost wakeup of
* alternate cpu.
*/
if (cpu_found != CPU->cpu_seqid) {
mcpu_mwait = cpu[cpu_found]->cpu_m.mcpu_mwait;
if ((mcpu_mwait != NULL) && (*mcpu_mwait == MWAIT_HALTED))
MWAIT_WAKEUP(cpu[cpu_found]);
else
poke_cpu(cpu_found); <------ 2
}
}
1: cpu_found now holds the sequential id returned by bitset_find().
2: The sequential id is passed to poke_cpu(). This is a problem because
poke_cpu() needs cp->cpu_id (and not cp->cpu_seqid).
The poke_cpu() should be: poke_cpu(cpu_seq[cpu_found]->cpu_id);
--
Configure bugmail: http://defect.opensolaris.org/bz/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
