* [PATCH] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none
@ 2020-04-11 6:19 Catherine Ho
2020-04-11 8:35 ` [PATCH v2] " Catherine Ho
0 siblings, 1 reply; 5+ messages in thread
From: Catherine Ho @ 2020-04-11 6:19 UTC (permalink / raw)
To: Dr. David Alan Gilbert, Stefan Hajnoczi, qemu-devel; +Cc: Catherine Ho
cache=none means to bypass host cache. So we can't remove O_DIRECT flag in
unconditionally in update_open_flags();
Signed-off-by: Catherine Ho <catherine.hecx@gmail.com>
---
tools/virtiofsd/passthrough_ll.c | 16 ++++++++++------
1 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/tools/virtiofsd/passthrough_ll.c b/tools/virtiofsd/passthrough_ll.c
index 4c35c95..889e144 100644
--- a/tools/virtiofsd/passthrough_ll.c
+++ b/tools/virtiofsd/passthrough_ll.c
@@ -1677,7 +1677,8 @@ static void lo_releasedir(fuse_req_t req, fuse_ino_t ino,
fuse_reply_err(req, 0);
}
-static void update_open_flags(int writeback, struct fuse_file_info *fi)
+static void update_open_flags(int writeback, int cache_mode,
+ struct fuse_file_info *fi)
{
/*
* With writeback cache, kernel may send read requests even
@@ -1702,10 +1703,13 @@ static void update_open_flags(int writeback, struct fuse_file_info *fi)
/*
* O_DIRECT in guest should not necessarily mean bypassing page
- * cache on host as well. If somebody needs that behavior, it
- * probably should be a configuration knob in daemon.
+ * cache on host as well. If cache=none, set the flag to O_DIRECT
*/
- fi->flags &= ~O_DIRECT;
+ if (cache_mode == CACHE_NONE) {
+ fi->flags |= O_DIRECT;
+ } else {
+ fi->flags &= ~O_DIRECT;
+ }
}
static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name,
@@ -1737,7 +1741,7 @@ static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name,
goto out;
}
- update_open_flags(lo->writeback, fi);
+ update_open_flags(lo->writeback, lo->cache, fi);
fd = openat(parent_inode->fd, name, (fi->flags | O_CREAT) & ~O_NOFOLLOW,
mode);
@@ -1947,7 +1951,7 @@ static void lo_open(fuse_req_t req, fuse_ino_t ino, struct fuse_file_info *fi)
fuse_log(FUSE_LOG_DEBUG, "lo_open(ino=%" PRIu64 ", flags=%d)\n", ino,
fi->flags);
- update_open_flags(lo->writeback, fi);
+ update_open_flags(lo->writeback, lo->cache, fi);
sprintf(buf, "%i", lo_fd(req, ino));
fd = openat(lo->proc_self_fd, buf, fi->flags & ~O_NOFOLLOW);
--
1.7.1
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH v2] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none
2020-04-11 6:19 [PATCH] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none Catherine Ho
@ 2020-04-11 8:35 ` Catherine Ho
2020-04-21 1:55 ` Catherine Ho
2020-04-30 8:35 ` [Virtio-fs] " Stefan Hajnoczi
0 siblings, 2 replies; 5+ messages in thread
From: Catherine Ho @ 2020-04-11 8:35 UTC (permalink / raw)
To: Dr. David Alan Gilbert, Stefan Hajnoczi, qemu-devel; +Cc: Catherine Ho
cache=none means to bypass host cache. So we can't remove O_DIRECT flag in
unconditionally in update_open_flags();
Signed-off-by: Catherine Ho <catherine.hecx@gmail.com>
---
v2: Fix to keep flags unchanged if cache=none, otherwise changed the file
without O_DIRECT incorrectly.
tools/virtiofsd/passthrough_ll.c | 14 ++++++++------
1 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/tools/virtiofsd/passthrough_ll.c b/tools/virtiofsd/passthrough_ll.c
index 4c35c95..59e64dd 100644
--- a/tools/virtiofsd/passthrough_ll.c
+++ b/tools/virtiofsd/passthrough_ll.c
@@ -1677,7 +1677,8 @@ static void lo_releasedir(fuse_req_t req, fuse_ino_t ino,
fuse_reply_err(req, 0);
}
-static void update_open_flags(int writeback, struct fuse_file_info *fi)
+static void update_open_flags(int writeback, int cache_mode,
+ struct fuse_file_info *fi)
{
/*
* With writeback cache, kernel may send read requests even
@@ -1702,10 +1703,11 @@ static void update_open_flags(int writeback, struct fuse_file_info *fi)
/*
* O_DIRECT in guest should not necessarily mean bypassing page
- * cache on host as well. If somebody needs that behavior, it
- * probably should be a configuration knob in daemon.
+ * cache on host as well. If cache=none, keep the flag unchanged
*/
- fi->flags &= ~O_DIRECT;
+ if (cache_mode != CACHE_NONE) {
+ fi->flags &= ~O_DIRECT;
+ }
}
static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name,
@@ -1737,7 +1739,7 @@ static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name,
goto out;
}
- update_open_flags(lo->writeback, fi);
+ update_open_flags(lo->writeback, lo->cache, fi);
fd = openat(parent_inode->fd, name, (fi->flags | O_CREAT) & ~O_NOFOLLOW,
mode);
@@ -1947,7 +1949,7 @@ static void lo_open(fuse_req_t req, fuse_ino_t ino, struct fuse_file_info *fi)
fuse_log(FUSE_LOG_DEBUG, "lo_open(ino=%" PRIu64 ", flags=%d)\n", ino,
fi->flags);
- update_open_flags(lo->writeback, fi);
+ update_open_flags(lo->writeback, lo->cache, fi);
sprintf(buf, "%i", lo_fd(req, ino));
fd = openat(lo->proc_self_fd, buf, fi->flags & ~O_NOFOLLOW);
--
1.7.1
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH v2] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none
2020-04-11 8:35 ` [PATCH v2] " Catherine Ho
@ 2020-04-21 1:55 ` Catherine Ho
2020-04-30 8:35 ` [Virtio-fs] " Stefan Hajnoczi
1 sibling, 0 replies; 5+ messages in thread
From: Catherine Ho @ 2020-04-21 1:55 UTC (permalink / raw)
To: Dr. David Alan Gilbert, Stefan Hajnoczi, QEMU Developers
Ping :)
---
B.R.
Catherine
On Sat, 11 Apr 2020 at 16:41, Catherine Ho <catherine.hecx@gmail.com> wrote:
>
> cache=none means to bypass host cache. So we can't remove O_DIRECT flag in
> unconditionally in update_open_flags();
>
> Signed-off-by: Catherine Ho <catherine.hecx@gmail.com>
> ---
> v2: Fix to keep flags unchanged if cache=none, otherwise changed the file
> without O_DIRECT incorrectly.
> tools/virtiofsd/passthrough_ll.c | 14 ++++++++------
> 1 files changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/tools/virtiofsd/passthrough_ll.c b/tools/virtiofsd/passthrough_ll.c
> index 4c35c95..59e64dd 100644
> --- a/tools/virtiofsd/passthrough_ll.c
> +++ b/tools/virtiofsd/passthrough_ll.c
> @@ -1677,7 +1677,8 @@ static void lo_releasedir(fuse_req_t req, fuse_ino_t ino,
> fuse_reply_err(req, 0);
> }
>
> -static void update_open_flags(int writeback, struct fuse_file_info *fi)
> +static void update_open_flags(int writeback, int cache_mode,
> + struct fuse_file_info *fi)
> {
> /*
> * With writeback cache, kernel may send read requests even
> @@ -1702,10 +1703,11 @@ static void update_open_flags(int writeback, struct fuse_file_info *fi)
>
> /*
> * O_DIRECT in guest should not necessarily mean bypassing page
> - * cache on host as well. If somebody needs that behavior, it
> - * probably should be a configuration knob in daemon.
> + * cache on host as well. If cache=none, keep the flag unchanged
> */
> - fi->flags &= ~O_DIRECT;
> + if (cache_mode != CACHE_NONE) {
> + fi->flags &= ~O_DIRECT;
> + }
> }
>
> static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name,
> @@ -1737,7 +1739,7 @@ static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name,
> goto out;
> }
>
> - update_open_flags(lo->writeback, fi);
> + update_open_flags(lo->writeback, lo->cache, fi);
>
> fd = openat(parent_inode->fd, name, (fi->flags | O_CREAT) & ~O_NOFOLLOW,
> mode);
> @@ -1947,7 +1949,7 @@ static void lo_open(fuse_req_t req, fuse_ino_t ino, struct fuse_file_info *fi)
> fuse_log(FUSE_LOG_DEBUG, "lo_open(ino=%" PRIu64 ", flags=%d)\n", ino,
> fi->flags);
>
> - update_open_flags(lo->writeback, fi);
> + update_open_flags(lo->writeback, lo->cache, fi);
>
> sprintf(buf, "%i", lo_fd(req, ino));
> fd = openat(lo->proc_self_fd, buf, fi->flags & ~O_NOFOLLOW);
> --
> 1.7.1
>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v2] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none
2020-04-11 8:35 ` [PATCH v2] " Catherine Ho
@ 2020-04-30 8:35 ` Stefan Hajnoczi
2020-04-30 8:35 ` [Virtio-fs] " Stefan Hajnoczi
1 sibling, 0 replies; 5+ messages in thread
From: Stefan Hajnoczi @ 2020-04-30 8:35 UTC (permalink / raw)
To: Catherine Ho; +Cc: virtio-fs, Dr. David Alan Gilbert, qemu-devel
[-- Attachment #1: Type: text/plain, Size: 1452 bytes --]
On Sat, Apr 11, 2020 at 04:35:44AM -0400, Catherine Ho wrote:
> @@ -1702,10 +1703,11 @@ static void update_open_flags(int writeback, struct fuse_file_info *fi)
>
> /*
> * O_DIRECT in guest should not necessarily mean bypassing page
> - * cache on host as well. If somebody needs that behavior, it
> - * probably should be a configuration knob in daemon.
> + * cache on host as well. If cache=none, keep the flag unchanged
> */
> - fi->flags &= ~O_DIRECT;
> + if (cache_mode != CACHE_NONE) {
> + fi->flags &= ~O_DIRECT;
> + }
> }
>
Thanks for the patch! I have CCed the virtio-fs mailing list so more
people see it.
Please add a new command-line option to control O_DIRECT behavior.
There are two cases:
1. O_DIRECT bypasses the guest page cache but not the host page cache.
This makes sense when the DAX feature is enabled.
2. O_DIRECT bypasses both the guest and host page cache. This make
sense for non-DAX and for I/O performance benchmarking.
Today only #1 is supported.
Your patch makes the behavior dependent on the cache mode option, but
the cache mode doesn't necessarily determine how O_DIRECT should be
handled. For example, in the DAX case the guest page cache is bypassed
and cache=none can be used, but we do want to use the host page cache.
You can add a new option so that O_DIRECT handling is configurable for
all cases.
Stefan
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [Virtio-fs] [PATCH v2] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none
@ 2020-04-30 8:35 ` Stefan Hajnoczi
0 siblings, 0 replies; 5+ messages in thread
From: Stefan Hajnoczi @ 2020-04-30 8:35 UTC (permalink / raw)
To: Catherine Ho; +Cc: virtio-fs, qemu-devel
[-- Attachment #1: Type: text/plain, Size: 1452 bytes --]
On Sat, Apr 11, 2020 at 04:35:44AM -0400, Catherine Ho wrote:
> @@ -1702,10 +1703,11 @@ static void update_open_flags(int writeback, struct fuse_file_info *fi)
>
> /*
> * O_DIRECT in guest should not necessarily mean bypassing page
> - * cache on host as well. If somebody needs that behavior, it
> - * probably should be a configuration knob in daemon.
> + * cache on host as well. If cache=none, keep the flag unchanged
> */
> - fi->flags &= ~O_DIRECT;
> + if (cache_mode != CACHE_NONE) {
> + fi->flags &= ~O_DIRECT;
> + }
> }
>
Thanks for the patch! I have CCed the virtio-fs mailing list so more
people see it.
Please add a new command-line option to control O_DIRECT behavior.
There are two cases:
1. O_DIRECT bypasses the guest page cache but not the host page cache.
This makes sense when the DAX feature is enabled.
2. O_DIRECT bypasses both the guest and host page cache. This make
sense for non-DAX and for I/O performance benchmarking.
Today only #1 is supported.
Your patch makes the behavior dependent on the cache mode option, but
the cache mode doesn't necessarily determine how O_DIRECT should be
handled. For example, in the DAX case the guest page cache is bypassed
and cache=none can be used, but we do want to use the host page cache.
You can add a new option so that O_DIRECT handling is configurable for
all cases.
Stefan
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-04-30 8:39 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-04-11 6:19 [PATCH] virtiofsd/passthrough_ll: don't remove O_DIRECT when cache=none Catherine Ho
2020-04-11 8:35 ` [PATCH v2] " Catherine Ho
2020-04-21 1:55 ` Catherine Ho
2020-04-30 8:35 ` Stefan Hajnoczi
2020-04-30 8:35 ` [Virtio-fs] " Stefan Hajnoczi
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.