Revision 8cd9da85d2bd87ce889043e7b1735723dd10eb89 authored by Frederic Weisbecker on 13 September 2021, 14:53:32 UTC, committed by Thomas Gleixner on 23 September 2021, 09:53:51 UTC
Resetting/stopping an itimer eventually leads to it being reprogrammed
with an actual "0" value. As a result the itimer expires on the next
tick, triggering an unexpected signal.
To fix this, make sure that
struct signal_struct::it[CPUCLOCK_PROF/VIRT]::expires is set to 0 when
setitimer() passes a 0 it_value, indicating that the timer must stop.
Fixes: 406dd42bd1ba ("posix-cpu-timers: Force next expiration recalc after itimer reset")
Reported-by: Victor Stinner <vstinner@redhat.com>
Reported-by: Chris Hixon <linux-kernel-bugs@hixontech.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210913145332.232023-1-frederic@kernel.org
1 parent e4e737b
| File | Mode | Size |
|---|---|---|
| Kconfig | -rw-r--r-- | 5.4 KB |
| Makefile | -rw-r--r-- | 918 bytes |
| dsa.c | -rw-r--r-- | 11.6 KB |
| dsa2.c | -rw-r--r-- | 34.7 KB |
| dsa_priv.h | -rw-r--r-- | 17.8 KB |
| master.c | -rw-r--r-- | 10.6 KB |
| port.c | -rw-r--r-- | 32.1 KB |
| slave.c | -rw-r--r-- | 65.7 KB |
| switch.c | -rw-r--r-- | 18.6 KB |
| tag_8021q.c | -rw-r--r-- | 17.4 KB |
| tag_ar9331.c | -rw-r--r-- | 2.3 KB |
| tag_brcm.c | -rw-r--r-- | 8.6 KB |
| tag_dsa.c | -rw-r--r-- | 11.4 KB |
| tag_gswip.c | -rw-r--r-- | 2.7 KB |
| tag_hellcreek.c | -rw-r--r-- | 1.4 KB |
| tag_ksz.c | -rw-r--r-- | 5.6 KB |
| tag_lan9303.c | -rw-r--r-- | 3.7 KB |
| tag_mtk.c | -rw-r--r-- | 2.6 KB |
| tag_ocelot.c | -rw-r--r-- | 5.2 KB |
| tag_ocelot_8021q.c | -rw-r--r-- | 1.9 KB |
| tag_qca.c | -rw-r--r-- | 2.3 KB |
| tag_rtl4_a.c | -rw-r--r-- | 3.0 KB |
| tag_sja1105.c | -rw-r--r-- | 21.1 KB |
| tag_trailer.c | -rw-r--r-- | 1.3 KB |
| tag_xrs700x.c | -rw-r--r-- | 1.3 KB |
Computing file changes ...
