Re: [PATCH] rcu/nocb: Fix dump_tree hierarchy print always active
On 07/10/2019 03:26, Paul E. McKenney wrote:
>
> On Fri, Oct 04, 2019 at 03:24:02PM -0700, Paul E. McKenney wrote:
>> On Fri, Oct 04, 2019 at 07:49:10PM +, Stefan Reiter wrote:
>>> Commit 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if
>>> dump_tree") added print statements to rcu_organize_nocb_kthreads for
>>> debugging, but incorrectly guarded them, causing the function to always
>>> spew out its message.
>>>
>>> This patch fixes it by guarding both pr_alert statements with dump_tree,
>>> while also changing the second pr_alert to a pr_cont, to print the
>>> hierarchy in a single line (assuming that's how it was supposed to
>>> work).
>>>
>>> Fixes: 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if dump_tree")
>>> Signed-off-by: Stefan Reiter
>>
>> Queued for testing and review, thank you!
>
> And here is an updated version to make the special case of a nocb GP
> kthread having no other nocb CB kthreads look less strange. Does this
> work for you?
Tested just now, works for me. Thanks for the additional fix!
>
> Thanx, Paul
>
>
>
> commit e6223b0705369750990c32ddc80251942e61be30
> Author: Stefan Reiter
> Date: Fri Oct 4 19:49:10 2019 +
>
> rcu/nocb: Fix dump_tree hierarchy print always active
>
> Commit 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if
> dump_tree") added print statements to rcu_organize_nocb_kthreads for
> debugging, but incorrectly guarded them, causing the function to always
> spew out its message.
>
> This patch fixes it by guarding both pr_alert statements with dump_tree,
> while also changing the second pr_alert to a pr_cont, to print the
> hierarchy in a single line (assuming that's how it was supposed to
> work).
>
> Fixes: 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if
> dump_tree")
> Signed-off-by: Stefan Reiter
> [ paulmck: Make single-nocbs-CPU GP kthreads look less erroneous. ]
> Signed-off-by: Paul E. McKenney
>
> diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
> index d5334e4..d43f4e0 100644
> --- a/kernel/rcu/tree_plugin.h
> +++ b/kernel/rcu/tree_plugin.h
> @@ -2295,6 +2295,8 @@ static void __init rcu_organize_nocb_kthreads(void)
> {
> int cpu;
> bool firsttime = true;
> + bool gotnocbs = false;
> + bool gotnocbscbs = true;
> int ls = rcu_nocb_gp_stride;
> int nl = 0; /* Next GP kthread. */
> struct rcu_data *rdp;
> @@ -2317,21 +2319,31 @@ static void __init rcu_organize_nocb_kthreads(void)
> rdp = per_cpu_ptr(_data, cpu);
> if (rdp->cpu >= nl) {
> /* New GP kthread, set up for CBs & next GP. */
> + gotnocbs = true;
> nl = DIV_ROUND_UP(rdp->cpu + 1, ls) * ls;
> rdp->nocb_gp_rdp = rdp;
> rdp_gp = rdp;
> - if (!firsttime && dump_tree)
> - pr_cont("\n");
> - firsttime = false;
> - pr_alert("%s: No-CB GP kthread CPU %d:", __func__, cpu);
> + if (dump_tree) {
> + if (!firsttime)
> + pr_cont("%s\n", gotnocbscbs
> + ? "" : " (self only)");
> + gotnocbscbs = false;
> + firsttime = false;
> + pr_alert("%s: No-CB GP kthread CPU %d:",
> + __func__, cpu);
> + }
> } else {
> /* Another CB kthread, link to previous GP kthread. */
> + gotnocbscbs = true;
> rdp->nocb_gp_rdp = rdp_gp;
> rdp_prev->nocb_next_cb_rdp = rdp;
> - pr_alert(" %d", cpu);
> + if (dump_tree)
> + pr_cont(" %d", cpu);
> }
> rdp_prev = rdp;
> }
> + if (gotnocbs && dump_tree)
> + pr_cont("%s\n", gotnocbscbs ? "" : " (self only)");
> }
>
> /*
>
Re: [PATCH] rcu/nocb: Fix dump_tree hierarchy print always active
On Fri, Oct 04, 2019 at 03:24:02PM -0700, Paul E. McKenney wrote:
> On Fri, Oct 04, 2019 at 07:49:10PM +, Stefan Reiter wrote:
> > Commit 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if
> > dump_tree") added print statements to rcu_organize_nocb_kthreads for
> > debugging, but incorrectly guarded them, causing the function to always
> > spew out its message.
> >
> > This patch fixes it by guarding both pr_alert statements with dump_tree,
> > while also changing the second pr_alert to a pr_cont, to print the
> > hierarchy in a single line (assuming that's how it was supposed to
> > work).
> >
> > Fixes: 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if dump_tree")
> > Signed-off-by: Stefan Reiter
>
> Queued for testing and review, thank you!
And here is an updated version to make the special case of a nocb GP
kthread having no other nocb CB kthreads look less strange. Does this
work for you?
Thanx, Paul
commit e6223b0705369750990c32ddc80251942e61be30
Author: Stefan Reiter
Date: Fri Oct 4 19:49:10 2019 +
rcu/nocb: Fix dump_tree hierarchy print always active
Commit 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if
dump_tree") added print statements to rcu_organize_nocb_kthreads for
debugging, but incorrectly guarded them, causing the function to always
spew out its message.
This patch fixes it by guarding both pr_alert statements with dump_tree,
while also changing the second pr_alert to a pr_cont, to print the
hierarchy in a single line (assuming that's how it was supposed to
work).
Fixes: 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if dump_tree")
Signed-off-by: Stefan Reiter
[ paulmck: Make single-nocbs-CPU GP kthreads look less erroneous. ]
Signed-off-by: Paul E. McKenney
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index d5334e4..d43f4e0 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -2295,6 +2295,8 @@ static void __init rcu_organize_nocb_kthreads(void)
{
int cpu;
bool firsttime = true;
+ bool gotnocbs = false;
+ bool gotnocbscbs = true;
int ls = rcu_nocb_gp_stride;
int nl = 0; /* Next GP kthread. */
struct rcu_data *rdp;
@@ -2317,21 +2319,31 @@ static void __init rcu_organize_nocb_kthreads(void)
rdp = per_cpu_ptr(_data, cpu);
if (rdp->cpu >= nl) {
/* New GP kthread, set up for CBs & next GP. */
+ gotnocbs = true;
nl = DIV_ROUND_UP(rdp->cpu + 1, ls) * ls;
rdp->nocb_gp_rdp = rdp;
rdp_gp = rdp;
- if (!firsttime && dump_tree)
- pr_cont("\n");
- firsttime = false;
- pr_alert("%s: No-CB GP kthread CPU %d:", __func__, cpu);
+ if (dump_tree) {
+ if (!firsttime)
+ pr_cont("%s\n", gotnocbscbs
+ ? "" : " (self only)");
+ gotnocbscbs = false;
+ firsttime = false;
+ pr_alert("%s: No-CB GP kthread CPU %d:",
+__func__, cpu);
+ }
} else {
/* Another CB kthread, link to previous GP kthread. */
+ gotnocbscbs = true;
rdp->nocb_gp_rdp = rdp_gp;
rdp_prev->nocb_next_cb_rdp = rdp;
- pr_alert(" %d", cpu);
+ if (dump_tree)
+ pr_cont(" %d", cpu);
}
rdp_prev = rdp;
}
+ if (gotnocbs && dump_tree)
+ pr_cont("%s\n", gotnocbscbs ? "" : " (self only)");
}
/*
Re: [PATCH] rcu/nocb: Fix dump_tree hierarchy print always active
On Fri, Oct 04, 2019 at 07:49:10PM +, Stefan Reiter wrote:
> Commit 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if
> dump_tree") added print statements to rcu_organize_nocb_kthreads for
> debugging, but incorrectly guarded them, causing the function to always
> spew out its message.
>
> This patch fixes it by guarding both pr_alert statements with dump_tree,
> while also changing the second pr_alert to a pr_cont, to print the
> hierarchy in a single line (assuming that's how it was supposed to
> work).
>
> Fixes: 18cd8c93e69e ("rcu/nocb: Print gp/cb kthread hierarchy if dump_tree")
> Signed-off-by: Stefan Reiter
Queued for testing and review, thank you!
> ---
>
> First time contributing to the kernel, hope I'm doing this right :)
Looks good so far. Then again, I just kicked off the tests. ;-)
Thanx, Paul
> kernel/rcu/tree_plugin.h | 14 +-
> 1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
> index 2defc7fe74c3..7cbf4a0f3eff 100644
> --- a/kernel/rcu/tree_plugin.h
> +++ b/kernel/rcu/tree_plugin.h
> @@ -2346,15 +2346,19 @@ static void __init rcu_organize_nocb_kthreads(void)
> nl = DIV_ROUND_UP(rdp->cpu + 1, ls) * ls;
> rdp->nocb_gp_rdp = rdp;
> rdp_gp = rdp;
> - if (!firsttime && dump_tree)
> - pr_cont("\n");
> - firsttime = false;
> - pr_alert("%s: No-CB GP kthread CPU %d:", __func__, cpu);
> + if (dump_tree) {
> + if (!firsttime)
> + pr_cont("\n");
> + firsttime = false;
> + pr_alert("%s: No-CB GP kthread CPU %d:",
> + __func__, cpu);
> + }
> } else {
> /* Another CB kthread, link to previous GP kthread. */
> rdp->nocb_gp_rdp = rdp_gp;
> rdp_prev->nocb_next_cb_rdp = rdp;
> - pr_alert(" %d", cpu);
> + if (dump_tree)
> + pr_cont(" %d", cpu);
> }
> rdp_prev = rdp;
> }
> --
> 2.23.0
>
>
