test/interrupt: account for race with callback
PATCH 1/1 test/interrupt: account for race with callback
Because the eal interrupt framework can race when invoking the callback and a separate unregister call, the test needs to accommodate the chance that the two collide. Do this by checking the return value of unregister against the race-condition flag (EAGAIN).
---
NOTE: it's difficult to reproduce this race. I tried a bit, but have
only seen it sporadically. In Travis environment, the CPU resource can be very limited and this test is quite racy.
app/test/test_interrupts.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
app/test/test_interrupts.c +7 -3
@@ -370,9 +370,13 @@ test_interrupt_full_path_check(enum test_interrupt_handle_type intr_type)rte_delay_ms (TEST_INTERRUPT_CHECK_INTERVAL );rte_delay_ms (TEST_INTERRUPT_CHECK_INTERVAL ); -if (rte_intr_callback_unregister (&test_intr_handle, - test_interrupt_callback, &test_intr_handle) <0 ) -return -1 ; +while ((count = +rte_intr_callback_unregister (&test_intr_handle, + test_interrupt_callback, + &test_intr_handle)) <0 ) { +if (count != -EAGAIN ) +return -1 ; + }if (flag ==0 ) {printf ("callback has not been called\n" );
Signature · 1 line
2.21.0
Quoted 7 lines · Aaron Conole
Because the eal interrupt framework can race when invoking the callback and a separate unregister call, the test needs to accommodate the chance that the two collide. Do this by checking the return value of unregister against the race-condition flag (EAGAIN).
Fixes f1a6c22424ce ("app/test: update interrupts test")
Not too sure about this tag, but anyway, this is old enough to apply to every stable releases we have.
Cc: [email protected] ?
Quoted 5 lines
Signed-off-by Aaron Conole <[email protected]>---
NOTE: it's difficult to reproduce this race. I tried a bit, but have
only seen it sporadically. In Travis environment, the CPU resource can be very limited and this test is quite racy.
Managed to reproduce it:
# time (log=/tmp/$$.log; while true; do echo interrupt_autotest |taskset -c 0-1 ./build-gcc-static/app/test/dpdk-test -l 0-1 >$log 2>&1; grep -q 'Test OK' $log || break; done; cat $log; rm -f $log) EAL: Detected 8 lcore(s) EAL: Detected 1 NUMA nodes EAL: Multi-process socket /var/run/dpdk/rte/mp_socket EAL: Selected IOVA mode 'PA' EAL: No available hugepages reported in hugepages-1048576kB EAL: Probing VFIO support... EAL: PCI device 0000:00:1f.6 on NUMA socket -1 EAL: Invalid NUMA socket, default to 0 EAL: probe driver: 8086:15d7 net_e1000_em APP: HPET is not enabled, using TSC as default timer RTE>>interrupt_autotest Check unknown valid interrupt full path Check valid UIO interrupt full path Check valid device event interrupt full path count=-11 Resource temporarily unavailable failure occurred during checking valid device event interrupt full path Clearing for interrupt tests Test Failed RTE>> real 0m38.081s user 0m35.836s sys 0m2.171s
Quoted 26 lines
app/test/test_interrupts.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)app/test/test_interrupts.c +7
-4@@ -370,9 +370,13 @@ test_interrupt_full_path_check(enum test_interrupt_handle_type intr_type)rte_delay_ms (TEST_INTERRUPT_CHECK_INTERVAL );rte_delay_ms (TEST_INTERRUPT_CHECK_INTERVAL ); -if (rte_intr_callback_unregister (&test_intr_handle, - test_interrupt_callback, &test_intr_handle) <0 ) -return -1 ; +while ((count = +rte_intr_callback_unregister (&test_intr_handle, + test_interrupt_callback, + &test_intr_handle)) <0 ) { +if (count != -EAGAIN ) +return -1 ; + }if (flag ==0 ) {printf ("callback has not been called\n" ); --
Quoted 2 lines
2.21.0
With this patch, my loop has been running for more than 10 minutes now.
--
David Marchand
Thomas Monjalon
Applied09/08/2019 11:18, David Marchand:
Quoted 47 lines · Aaron Conole
Quoted 7 lines
Because the eal interrupt framework can race when invoking the callback and a separate unregister call, the test needs to accommodate the chance that the two collide. Do this by checking the return value of unregister against the race-condition flag (EAGAIN).
Fixes f1a6c22424ce ("app/test: update interrupts test")Not too sure about this tag, but anyway, this is old enough to apply to every stable releases we have.
Cc: [email protected] ?Quoted 5 lines
Signed-off-by Aaron Conole <[email protected]>---
NOTE: it's difficult to reproduce this race. I tried a bit, but have
only seen it sporadically. In Travis environment, the CPU resource can be very limited and this test is quite racy.Managed to reproduce it:
# time (log=/tmp/$$.log; while true; do echo interrupt_autotest |taskset -c 0-1 ./build-gcc-static/app/test/dpdk-test -l 0-1 >$log 2>&1; grep -q 'Test OK' $log || break; done; cat $log; rm -f $log) EAL: Detected 8 lcore(s) EAL: Detected 1 NUMA nodes EAL: Multi-process socket /var/run/dpdk/rte/mp_socket EAL: Selected IOVA mode 'PA' EAL: No available hugepages reported in hugepages-1048576kB EAL: Probing VFIO support... EAL: PCI device 0000:00:1f.6 on NUMA socket -1 EAL: Invalid NUMA socket, default to 0 EAL: probe driver: 8086:15d7 net_e1000_em APP: HPET is not enabled, using TSC as default timer RTE>>interrupt_autotest Check unknown valid interrupt full path Check valid UIO interrupt full path Check valid device event interrupt full path count=-11 Resource temporarily unavailable failure occurred during checking valid device event interrupt full path Clearing for interrupt tests Test Failed RTE>> real 0m38.081s user 0m35.836s sys 0m2.171s
Quoted 2 lines
With this patch, my loop has been running for more than 10 minutes now.
Reviewed-by David Marchand <[email protected]>
Applied, thanks
Quoted 8 lines · Aaron Conole
Because the eal interrupt framework can race when invoking the callback and a separate unregister call, the test needs to accommodate the chance that the two collide. Do this by checking the return value of unregister against the race-condition flag (EAGAIN).
Fixes f1a6c22424ce ("app/test: update interrupts test")Signed-off-by Aaron Conole <[email protected]>---
Ping. Still see these failures.