aer: add cond_resched to aer_isr
PATCH 1/1 aer: add cond_resched to aer_isr
From: Joern Engel <[email protected]>
Multiple nested loops. I have observed 590ms scheduler latency caused by this loop and interrupts. Interrupts were responsible for 190ms, the rest could have been avoided with a cond_resched.
Signed-off-by Joern Engel <[email protected]>
Signed-off-by Spencer Baugh <[email protected]>
--- drivers/pci/pcie/aer/aerdrv_core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/pci/pcie/aer/aerdrv_core.c +3 -1
@@ -780,8 +780,10 @@ void aer_isr(struct work_struct *work)struct aer_err_source uninitialized_var (e_src );mutex_lock (&rpc->rpc_mutex); -while (get_e_source (rpc, &e_src)) +while (get_e_source (rpc, &e_src)) {aer_isr_one_error (p_device, &e_src); +cond_resched (); + }mutex_unlock (&rpc->rpc_mutex);wake_up (&rpc->wait_release);
Signature · 1 line
2.5.0.rc3
Bjorn Helgaas
maintainerHi Spencer & Joern,
Quoted 5 lines · Spencer Baugh
From: Joern Engel <[email protected]>
Multiple nested loops. I have observed 590ms scheduler latency caused by this loop and interrupts. Interrupts were responsible for 190ms, the rest could have been avoided with a
cond_resched.
I'm not disagreeing with this patch, but it would be helpful to sketch the outline of the "multiple nested loop" problem here. This might be a hint that we could do even better by rethinking the algorithm reduce the nesting.
Quoted 25 lines
Signed-off-by Joern Engel <[email protected]>Signed-off-by Spencer Baugh <[email protected]>--- drivers/pci/pcie/aer/aerdrv_core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)drivers/pci/pcie/aer/aerdrv_core.c +3
-1@@ -780,8 +780,10 @@ void aer_isr(struct work_struct *work)struct aer_err_source uninitialized_var (e_src );mutex_lock (&rpc->rpc_mutex); -while (get_e_source (rpc, &e_src)) +while (get_e_source (rpc, &e_src)) {aer_isr_one_error (p_device, &e_src); +cond_resched (); + }mutex_unlock (&rpc->rpc_mutex);wake_up (&rpc->wait_release);Signature · 1 line
2.5.0.rc3