diff options
author | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2015-01-25 14:28:28 -0500 |
---|---|---|
committer | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2015-02-26 14:57:29 -0500 |
commit | c25197841efe53258abb22cfd894a729a272edf9 (patch) | |
tree | 2aee40a0447d3380378b9d0e77148c3a015c352f /Documentation/timers | |
parent | 89bf5d82ed451f02329bbbb06ac365e96b18804d (diff) |
documentation: Update NO_HZ_FULL interaction with POSIX timers
POSIX timers are no longer starved on adaptive-ticks CPUs. Instead, they
prevent affected CPUs from entering adaptive-ticks mode. This commit
therefore updates the NO_HZ.txt documentation.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Diffstat (limited to 'Documentation/timers')
-rw-r--r-- | Documentation/timers/NO_HZ.txt | 10 |
1 files changed, 3 insertions, 7 deletions
diff --git a/Documentation/timers/NO_HZ.txt b/Documentation/timers/NO_HZ.txt index cca122f25120..6eaf576294f3 100644 --- a/Documentation/timers/NO_HZ.txt +++ b/Documentation/timers/NO_HZ.txt | |||
@@ -158,13 +158,9 @@ not come for free: | |||
158 | to the need to inform kernel subsystems (such as RCU) about | 158 | to the need to inform kernel subsystems (such as RCU) about |
159 | the change in mode. | 159 | the change in mode. |
160 | 160 | ||
161 | 3. POSIX CPU timers on adaptive-tick CPUs may miss their deadlines | 161 | 3. POSIX CPU timers prevent CPUs from entering adaptive-tick mode. |
162 | (perhaps indefinitely) because they currently rely on | 162 | Real-time applications needing to take actions based on CPU time |
163 | scheduling-tick interrupts. This will likely be fixed in | 163 | consumption need to use other means of doing so. |
164 | one of two ways: (1) Prevent CPUs with POSIX CPU timers from | ||
165 | entering adaptive-tick mode, or (2) Use hrtimers or other | ||
166 | adaptive-ticks-immune mechanism to cause the POSIX CPU timer to | ||
167 | fire properly. | ||
168 | 164 | ||
169 | 4. If there are more perf events pending than the hardware can | 165 | 4. If there are more perf events pending than the hardware can |
170 | accommodate, they are normally round-robined so as to collect | 166 | accommodate, they are normally round-robined so as to collect |