* [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
@ 2020-02-13 18:23 Mike Christie
2020-02-13 20:08 ` Darrick J. Wong
2020-02-14 12:54 ` Michal Hocko
0 siblings, 2 replies; 8+ messages in thread
From: Mike Christie @ 2020-02-13 18:23 UTC (permalink / raw)
To: linux-api, david, mhocko, masato.suzuki, damien.lemoal,
darrick.wong, bvanassche, mtk.manpages, linux-man
Cc: Mike Christie
This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
prctl commands added to the linux kernel for 5.6 in commit:
commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
Author: Mike Christie <mchristi@redhat.com>
Date: Mon Nov 11 18:19:00 2019 -0600
prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
---
V3:
- Replace emulation device example.
V2:
- My initial patch for this was very bad. This version is almost 100%
taken word for word from Dave Chinner's review comments.
Signed-off-by: Mike Christie <mchristi@redhat.com>
---
man2/prctl.2 | 25 +++++++++++++++++++++++++
1 file changed, 25 insertions(+)
diff --git a/man2/prctl.2 b/man2/prctl.2
index 720ec04e4..58d77bf2e 100644
--- a/man2/prctl.2
+++ b/man2/prctl.2
@@ -1381,6 +1381,30 @@ system call on Tru64).
for information on versions and architectures.)
Return unaligned access control bits, in the location pointed to by
.IR "(unsigned int\ *) arg2" .
+.TP
+.B PR_SET_IO_FLUSHER (Since Linux 5.6)
+An IO_FLUSHER is a user process that the kernel uses to issue IO
+that cleans dirty page cache data and/or filesystem metadata. The
+kernel may need to clean this memory when under memory pressure in
+order to free it. This means there is potential for a memory reclaim
+recursion deadlock if the user process attempts to allocate memory
+and the kernel then blocks waiting for it to clean memory before it
+can make reclaim progress.
+
+The kernel avoids these recursion problems internally via a special
+process state that prevents recursive reclaim from issuing new IO.
+If \fIarg2\fP is 1, the \fPPR_SET_IO_FLUSHER\fP control allows a userspace
+process to set up this same process state and hence avoid the memory
+reclaim recursion deadlocks in the same manner the kernel avoids them.
+If \fIarg2\fP is 0, the process will clear the IO_FLUSHER state, and the
+default behavior will be used.
+
+Examples of IO_FLUSHER applications are FUSE daemons, SCSI device
+emulation daemons, etc."
+.TP
+.B PR_GET_IO_FLUSHER (Since Linux 5.6)
+Return as the function result 1 if the caller is in the IO_FLUSHER state and
+0 if not.
.SH RETURN VALUE
On success,
.BR PR_GET_DUMPABLE ,
@@ -1395,6 +1419,7 @@ On success,
.BR PR_GET_SPECULATION_CTRL ,
.BR PR_MCE_KILL_GET ,
.BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
+.BR PR_GET_IO_FLUSHER ,
and (if it returns)
.BR PR_GET_SECCOMP
return the nonnegative values described above.
--
2.21.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
2020-02-13 18:23 [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER Mike Christie
@ 2020-02-13 20:08 ` Darrick J. Wong
2020-02-13 20:14 ` Mike Christie
2020-02-14 12:54 ` Michal Hocko
1 sibling, 1 reply; 8+ messages in thread
From: Darrick J. Wong @ 2020-02-13 20:08 UTC (permalink / raw)
To: Mike Christie
Cc: linux-api, david, mhocko, masato.suzuki, damien.lemoal,
bvanassche, mtk.manpages, linux-man
On Thu, Feb 13, 2020 at 12:23:36PM -0600, Mike Christie wrote:
> This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
> prctl commands added to the linux kernel for 5.6 in commit:
>
> commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
> Author: Mike Christie <mchristi@redhat.com>
> Date: Mon Nov 11 18:19:00 2019 -0600
>
> prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
> ---
>
> V3:
> - Replace emulation device example.
>
> V2:
> - My initial patch for this was very bad. This version is almost 100%
> taken word for word from Dave Chinner's review comments.
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> ---
> man2/prctl.2 | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
>
> diff --git a/man2/prctl.2 b/man2/prctl.2
> index 720ec04e4..58d77bf2e 100644
> --- a/man2/prctl.2
> +++ b/man2/prctl.2
> @@ -1381,6 +1381,30 @@ system call on Tru64).
> for information on versions and architectures.)
> Return unaligned access control bits, in the location pointed to by
> .IR "(unsigned int\ *) arg2" .
> +.TP
> +.B PR_SET_IO_FLUSHER (Since Linux 5.6)
> +An IO_FLUSHER is a user process that the kernel uses to issue IO
> +that cleans dirty page cache data and/or filesystem metadata. The
> +kernel may need to clean this memory when under memory pressure in
> +order to free it. This means there is potential for a memory reclaim
> +recursion deadlock if the user process attempts to allocate memory
> +and the kernel then blocks waiting for it to clean memory before it
> +can make reclaim progress.
> +
> +The kernel avoids these recursion problems internally via a special
> +process state that prevents recursive reclaim from issuing new IO.
> +If \fIarg2\fP is 1, the \fPPR_SET_IO_FLUSHER\fP control allows a userspace
> +process to set up this same process state and hence avoid the memory
> +reclaim recursion deadlocks in the same manner the kernel avoids them.
> +If \fIarg2\fP is 0, the process will clear the IO_FLUSHER state, and the
> +default behavior will be used.
I forget, does a program have to have special capabilities (e.g.
CAP_SYS_ADMIN) to be able to PR_SET_IO_FLUSHER?
--D
> +Examples of IO_FLUSHER applications are FUSE daemons, SCSI device
> +emulation daemons, etc."
> +.TP
> +.B PR_GET_IO_FLUSHER (Since Linux 5.6)
> +Return as the function result 1 if the caller is in the IO_FLUSHER state and
> +0 if not.
> .SH RETURN VALUE
> On success,
> .BR PR_GET_DUMPABLE ,
> @@ -1395,6 +1419,7 @@ On success,
> .BR PR_GET_SPECULATION_CTRL ,
> .BR PR_MCE_KILL_GET ,
> .BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
> +.BR PR_GET_IO_FLUSHER ,
> and (if it returns)
> .BR PR_GET_SECCOMP
> return the nonnegative values described above.
> --
> 2.21.0
>
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
2020-02-13 20:08 ` Darrick J. Wong
@ 2020-02-13 20:14 ` Mike Christie
0 siblings, 0 replies; 8+ messages in thread
From: Mike Christie @ 2020-02-13 20:14 UTC (permalink / raw)
To: Darrick J. Wong
Cc: linux-api, david, mhocko, masato.suzuki, damien.lemoal,
bvanassche, mtk.manpages, linux-man
On 02/13/2020 02:08 PM, Darrick J. Wong wrote:
> On Thu, Feb 13, 2020 at 12:23:36PM -0600, Mike Christie wrote:
>> This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
>> prctl commands added to the linux kernel for 5.6 in commit:
>>
>> commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
>> Author: Mike Christie <mchristi@redhat.com>
>> Date: Mon Nov 11 18:19:00 2019 -0600
>>
>> prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
>>
>> Signed-off-by: Mike Christie <mchristi@redhat.com>
>> Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
>> ---
>>
>> V3:
>> - Replace emulation device example.
>>
>> V2:
>> - My initial patch for this was very bad. This version is almost 100%
>> taken word for word from Dave Chinner's review comments.
>>
>> Signed-off-by: Mike Christie <mchristi@redhat.com>
>> ---
>> man2/prctl.2 | 25 +++++++++++++++++++++++++
>> 1 file changed, 25 insertions(+)
>>
>> diff --git a/man2/prctl.2 b/man2/prctl.2
>> index 720ec04e4..58d77bf2e 100644
>> --- a/man2/prctl.2
>> +++ b/man2/prctl.2
>> @@ -1381,6 +1381,30 @@ system call on Tru64).
>> for information on versions and architectures.)
>> Return unaligned access control bits, in the location pointed to by
>> .IR "(unsigned int\ *) arg2" .
>> +.TP
>> +.B PR_SET_IO_FLUSHER (Since Linux 5.6)
>> +An IO_FLUSHER is a user process that the kernel uses to issue IO
>> +that cleans dirty page cache data and/or filesystem metadata. The
>> +kernel may need to clean this memory when under memory pressure in
>> +order to free it. This means there is potential for a memory reclaim
>> +recursion deadlock if the user process attempts to allocate memory
>> +and the kernel then blocks waiting for it to clean memory before it
>> +can make reclaim progress.
>> +
>> +The kernel avoids these recursion problems internally via a special
>> +process state that prevents recursive reclaim from issuing new IO.
>> +If \fIarg2\fP is 1, the \fPPR_SET_IO_FLUSHER\fP control allows a userspace
>> +process to set up this same process state and hence avoid the memory
>> +reclaim recursion deadlocks in the same manner the kernel avoids them.
>> +If \fIarg2\fP is 0, the process will clear the IO_FLUSHER state, and the
>> +default behavior will be used.
>
> I forget, does a program have to have special capabilities (e.g.
> CAP_SYS_ADMIN) to be able to PR_SET_IO_FLUSHER?
Yes, CAP_SYS_RESOURCE. I will add that info.
>
> --D
>
>> +Examples of IO_FLUSHER applications are FUSE daemons, SCSI device
>> +emulation daemons, etc."
>> +.TP
>> +.B PR_GET_IO_FLUSHER (Since Linux 5.6)
>> +Return as the function result 1 if the caller is in the IO_FLUSHER state and
>> +0 if not.
>> .SH RETURN VALUE
>> On success,
>> .BR PR_GET_DUMPABLE ,
>> @@ -1395,6 +1419,7 @@ On success,
>> .BR PR_GET_SPECULATION_CTRL ,
>> .BR PR_MCE_KILL_GET ,
>> .BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
>> +.BR PR_GET_IO_FLUSHER ,
>> and (if it returns)
>> .BR PR_GET_SECCOMP
>> return the nonnegative values described above.
>> --
>> 2.21.0
>>
>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
2020-02-13 18:23 [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER Mike Christie
2020-02-13 20:08 ` Darrick J. Wong
@ 2020-02-14 12:54 ` Michal Hocko
2020-04-01 9:47 ` Michael Kerrisk (man-pages)
1 sibling, 1 reply; 8+ messages in thread
From: Michal Hocko @ 2020-02-14 12:54 UTC (permalink / raw)
To: Mike Christie
Cc: linux-api, david, masato.suzuki, damien.lemoal, darrick.wong,
bvanassche, mtk.manpages, linux-man
On Thu 13-02-20 12:23:36, Mike Christie wrote:
> This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
> prctl commands added to the linux kernel for 5.6 in commit:
>
> commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
> Author: Mike Christie <mchristi@redhat.com>
> Date: Mon Nov 11 18:19:00 2019 -0600
>
> prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
> ---
>
> V3:
> - Replace emulation device example.
>
> V2:
> - My initial patch for this was very bad. This version is almost 100%
> taken word for word from Dave Chinner's review comments.
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> ---
> man2/prctl.2 | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
>
> diff --git a/man2/prctl.2 b/man2/prctl.2
> index 720ec04e4..58d77bf2e 100644
> --- a/man2/prctl.2
> +++ b/man2/prctl.2
> @@ -1381,6 +1381,30 @@ system call on Tru64).
> for information on versions and architectures.)
> Return unaligned access control bits, in the location pointed to by
> .IR "(unsigned int\ *) arg2" .
> +.TP
> +.B PR_SET_IO_FLUSHER (Since Linux 5.6)
> +An IO_FLUSHER is a user process that the kernel uses to issue IO
> +that cleans dirty page cache data and/or filesystem metadata. The
> +kernel may need to clean this memory when under memory pressure in
> +order to free it. This means there is potential for a memory reclaim
> +recursion deadlock if the user process attempts to allocate memory
> +and the kernel then blocks waiting for it to clean memory before it
> +can make reclaim progress.
> +
> +The kernel avoids these recursion problems internally via a special
> +process state that prevents recursive reclaim from issuing new IO.
I would refrain from describing the internal implementation. The
important part is that this flag tells the kernel that IO_FLUSHER
process gets a special treatment to workaround the deadlock.
So anytime a process is involved in the IO path and the kernel cannot
make a forward progress without it then the flag should be set.
--
Michal Hocko
SUSE Labs
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
2020-02-14 12:54 ` Michal Hocko
@ 2020-04-01 9:47 ` Michael Kerrisk (man-pages)
0 siblings, 0 replies; 8+ messages in thread
From: Michael Kerrisk (man-pages) @ 2020-04-01 9:47 UTC (permalink / raw)
To: Mike Christie
Cc: Linux API, Michal Hocko, Dave Chinner, masato.suzuki,
damien.lemoal, Darrick J. Wong, bvanassche, linux-man
Mike,
Do you have a refresh of this patch, in the light of the comments?
Thanks,
Michael
On Fri, 14 Feb 2020 at 13:54, Michal Hocko <mhocko@kernel.org> wrote:
>
> On Thu 13-02-20 12:23:36, Mike Christie wrote:
> > This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
> > prctl commands added to the linux kernel for 5.6 in commit:
> >
> > commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
> > Author: Mike Christie <mchristi@redhat.com>
> > Date: Mon Nov 11 18:19:00 2019 -0600
> >
> > prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
> >
> > Signed-off-by: Mike Christie <mchristi@redhat.com>
> > Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
> > ---
> >
> > V3:
> > - Replace emulation device example.
> >
> > V2:
> > - My initial patch for this was very bad. This version is almost 100%
> > taken word for word from Dave Chinner's review comments.
> >
> > Signed-off-by: Mike Christie <mchristi@redhat.com>
> > ---
> > man2/prctl.2 | 25 +++++++++++++++++++++++++
> > 1 file changed, 25 insertions(+)
> >
> > diff --git a/man2/prctl.2 b/man2/prctl.2
> > index 720ec04e4..58d77bf2e 100644
> > --- a/man2/prctl.2
> > +++ b/man2/prctl.2
> > @@ -1381,6 +1381,30 @@ system call on Tru64).
> > for information on versions and architectures.)
> > Return unaligned access control bits, in the location pointed to by
> > .IR "(unsigned int\ *) arg2" .
> > +.TP
> > +.B PR_SET_IO_FLUSHER (Since Linux 5.6)
> > +An IO_FLUSHER is a user process that the kernel uses to issue IO
> > +that cleans dirty page cache data and/or filesystem metadata. The
> > +kernel may need to clean this memory when under memory pressure in
> > +order to free it. This means there is potential for a memory reclaim
> > +recursion deadlock if the user process attempts to allocate memory
> > +and the kernel then blocks waiting for it to clean memory before it
> > +can make reclaim progress.
> > +
> > +The kernel avoids these recursion problems internally via a special
> > +process state that prevents recursive reclaim from issuing new IO.
>
> I would refrain from describing the internal implementation. The
> important part is that this flag tells the kernel that IO_FLUSHER
> process gets a special treatment to workaround the deadlock.
>
> So anytime a process is involved in the IO path and the kernel cannot
> make a forward progress without it then the flag should be set.
>
> --
> Michal Hocko
> SUSE Labs
--
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/
^ permalink raw reply [flat|nested] 8+ messages in thread
* [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
@ 2020-04-02 2:07 Mike Christie
2020-04-02 2:08 ` Mike Christie
0 siblings, 1 reply; 8+ messages in thread
From: Mike Christie @ 2020-04-02 2:07 UTC (permalink / raw)
To: linux-api, david, mhocko, masato.suzuki, damien.lemoal,
darrick.wong, bvanassche, mtk.manpages, linux-man
Cc: Mike Christie
This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
prctl commands added to the linux kernel for 5.6 in commit:
commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
Author: Mike Christie <mchristi@redhat.com>
Date: Mon Nov 11 18:19:00 2019 -0600
prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
Signed-off-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
---
V3:
- Replace emulation device example.
V2:
- My initial patch for this was very bad. This version is almost 100%
taken word for word from Dave Chinner's review comments.
Signed-off-by: Mike Christie <mchristi@redhat.com>
---
man2/prctl.2 | 25 +++++++++++++++++++++++++
1 file changed, 25 insertions(+)
diff --git a/man2/prctl.2 b/man2/prctl.2
index 720ec04e4..58d77bf2e 100644
--- a/man2/prctl.2
+++ b/man2/prctl.2
@@ -1381,6 +1381,30 @@ system call on Tru64).
for information on versions and architectures.)
Return unaligned access control bits, in the location pointed to by
.IR "(unsigned int\ *) arg2" .
+.TP
+.B PR_SET_IO_FLUSHER (Since Linux 5.6)
+An IO_FLUSHER is a user process that the kernel uses to issue IO
+that cleans dirty page cache data and/or filesystem metadata. The
+kernel may need to clean this memory when under memory pressure in
+order to free it. This means there is potential for a memory reclaim
+recursion deadlock if the user process attempts to allocate memory
+and the kernel then blocks waiting for it to clean memory before it
+can make reclaim progress.
+
+The kernel avoids these recursion problems internally via a special
+process state that prevents recursive reclaim from issuing new IO.
+If \fIarg2\fP is 1, the \fPPR_SET_IO_FLUSHER\fP control allows a userspace
+process to set up this same process state and hence avoid the memory
+reclaim recursion deadlocks in the same manner the kernel avoids them.
+If \fIarg2\fP is 0, the process will clear the IO_FLUSHER state, and the
+default behavior will be used.
+
+Examples of IO_FLUSHER applications are FUSE daemons, SCSI device
+emulation daemons, etc."
+.TP
+.B PR_GET_IO_FLUSHER (Since Linux 5.6)
+Return as the function result 1 if the caller is in the IO_FLUSHER state and
+0 if not.
.SH RETURN VALUE
On success,
.BR PR_GET_DUMPABLE ,
@@ -1395,6 +1419,7 @@ On success,
.BR PR_GET_SPECULATION_CTRL ,
.BR PR_MCE_KILL_GET ,
.BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
+.BR PR_GET_IO_FLUSHER ,
and (if it returns)
.BR PR_GET_SECCOMP
return the nonnegative values described above.
--
2.21.0
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
2020-04-02 2:07 Mike Christie
@ 2020-04-02 2:08 ` Mike Christie
0 siblings, 0 replies; 8+ messages in thread
From: Mike Christie @ 2020-04-02 2:08 UTC (permalink / raw)
To: linux-api, david, mhocko, masato.suzuki, damien.lemoal,
darrick.wong, bvanassche, mtk.manpages, linux-man
Sorry. Ignore this patch. It was an old version.
On 04/01/2020 09:07 PM, Mike Christie wrote:
> This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
> prctl commands added to the linux kernel for 5.6 in commit:
>
> commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
> Author: Mike Christie <mchristi@redhat.com>
> Date: Mon Nov 11 18:19:00 2019 -0600
>
> prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
> ---
>
> V3:
> - Replace emulation device example.
>
> V2:
> - My initial patch for this was very bad. This version is almost 100%
> taken word for word from Dave Chinner's review comments.
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> ---
> man2/prctl.2 | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
>
> diff --git a/man2/prctl.2 b/man2/prctl.2
> index 720ec04e4..58d77bf2e 100644
> --- a/man2/prctl.2
> +++ b/man2/prctl.2
> @@ -1381,6 +1381,30 @@ system call on Tru64).
> for information on versions and architectures.)
> Return unaligned access control bits, in the location pointed to by
> .IR "(unsigned int\ *) arg2" .
> +.TP
> +.B PR_SET_IO_FLUSHER (Since Linux 5.6)
> +An IO_FLUSHER is a user process that the kernel uses to issue IO
> +that cleans dirty page cache data and/or filesystem metadata. The
> +kernel may need to clean this memory when under memory pressure in
> +order to free it. This means there is potential for a memory reclaim
> +recursion deadlock if the user process attempts to allocate memory
> +and the kernel then blocks waiting for it to clean memory before it
> +can make reclaim progress.
> +
> +The kernel avoids these recursion problems internally via a special
> +process state that prevents recursive reclaim from issuing new IO.
> +If \fIarg2\fP is 1, the \fPPR_SET_IO_FLUSHER\fP control allows a userspace
> +process to set up this same process state and hence avoid the memory
> +reclaim recursion deadlocks in the same manner the kernel avoids them.
> +If \fIarg2\fP is 0, the process will clear the IO_FLUSHER state, and the
> +default behavior will be used.
> +
> +Examples of IO_FLUSHER applications are FUSE daemons, SCSI device
> +emulation daemons, etc."
> +.TP
> +.B PR_GET_IO_FLUSHER (Since Linux 5.6)
> +Return as the function result 1 if the caller is in the IO_FLUSHER state and
> +0 if not.
> .SH RETURN VALUE
> On success,
> .BR PR_GET_DUMPABLE ,
> @@ -1395,6 +1419,7 @@ On success,
> .BR PR_GET_SPECULATION_CTRL ,
> .BR PR_MCE_KILL_GET ,
> .BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
> +.BR PR_GET_IO_FLUSHER ,
> and (if it returns)
> .BR PR_GET_SECCOMP
> return the nonnegative values described above.
>
^ permalink raw reply [flat|nested] 8+ messages in thread
[parent not found: <20200210221557.8021-1-mchristi@redhat.com>]
* Re: [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER
[not found] <20200210221557.8021-1-mchristi@redhat.com>
@ 2020-02-11 14:17 ` Christian Brauner
0 siblings, 0 replies; 8+ messages in thread
From: Christian Brauner @ 2020-02-11 14:17 UTC (permalink / raw)
To: Mike Christie, mtk.manpages
Cc: linux-api, david, mhocko, masato.suzuki, damien.lemoal,
darrick.wong, bvanassche, linux-man
I think you've missed:
mtk.manpages@gmail.com
linux-man@vger.kernel.org
:)
Christian
On Mon, Feb 10, 2020 at 04:15:57PM -0600, Mike Christie wrote:
> This patch documents the PR_SET_IO_FLUSHER and PR_GET_IO_FLUSHER
> prctl commands added to the linux kernel for 5.6 in commit:
>
> commit 8d19f1c8e1937baf74e1962aae9f90fa3aeab463
> Author: Mike Christie <mchristi@redhat.com>
> Date: Mon Nov 11 18:19:00 2019 -0600
>
> prctl: PR_{G,S}ET_IO_FLUSHER to support controlling memory reclaim
>
> Signed-off-by: Mike Christie <mchristi@redhat.com>
> ---
>
> V2:
> - My initial patch for this was very bad. This version is almost 100%
> taken word for word from Dave Chinner's review comments.
>
>
> man2/prctl.2 | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
>
> diff --git a/man2/prctl.2 b/man2/prctl.2
> index 720ec04e4..b481d186b 100644
> --- a/man2/prctl.2
> +++ b/man2/prctl.2
> @@ -1381,6 +1381,30 @@ system call on Tru64).
> for information on versions and architectures.)
> Return unaligned access control bits, in the location pointed to by
> .IR "(unsigned int\ *) arg2" .
> +.TP
> +.B PR_SET_IO_FLUSHER (Since Linux 5.6)
> +An IO_FLUSHER is a user process that the kernel uses to issue IO
> +that cleans dirty page cache data and/or filesystem metadata. The
> +kernel may need to clean this memory when under memory pressure in
> +order to free it. This means there is potential for a memory reclaim
> +recursion deadlock if the user process attempts to allocate memory
> +and the kernel then blocks waiting for it to clean memory before it
> +can make reclaim progress.
> +
> +The kernel avoids these recursion problems internally via a special
> +process state that prevents recursive reclaim from issuing new IO.
> +If \fIarg2\fP is 1, the \fPPR_SET_IO_FLUSHER\fP control allows a userspace
> +process to set up this same process state and hence avoid the memory
> +reclaim recursion deadlocks in the same manner the kernel avoids them.
> +If \fIarg2\fP is 0, the process will clear the IO_FLUSHER state, and the
> +default behavior will be used.
> +
> +Examples of IO_FLUSHER applications are FUSE daemons, zoned disk
> +emulation daemons, etc."
> +.TP
> +.B PR_GET_IO_FLUSHER (Since Linux 5.6)
> +Return as the function result 1 if the caller is in the IO_FLUSHER state and
> +0 if not.
> .SH RETURN VALUE
> On success,
> .BR PR_GET_DUMPABLE ,
> @@ -1395,6 +1419,7 @@ On success,
> .BR PR_GET_SPECULATION_CTRL ,
> .BR PR_MCE_KILL_GET ,
> .BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
> +.BR PR_GET_IO_FLUSHER ,
> and (if it returns)
> .BR PR_GET_SECCOMP
> return the nonnegative values described above.
> --
> 2.21.0
>
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2020-04-02 2:08 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-13 18:23 [PATCH 1/1] prctl.2: doc PR_SET/GET_IO_FLUSHER Mike Christie
2020-02-13 20:08 ` Darrick J. Wong
2020-02-13 20:14 ` Mike Christie
2020-02-14 12:54 ` Michal Hocko
2020-04-01 9:47 ` Michael Kerrisk (man-pages)
-- strict thread matches above, loose matches on Subject: below --
2020-04-02 2:07 Mike Christie
2020-04-02 2:08 ` Mike Christie
[not found] <20200210221557.8021-1-mchristi@redhat.com>
2020-02-11 14:17 ` Christian Brauner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).