[systemd-devel] [PATCH] units: split fsck.target from sysinit.target

I need this patch in order to avoid cycles during boot. It looks to me that others should be hit by the same problem (but I have not heard anyone complain, so I might be wrong): A .mount unit (e.g. home.mount) will have dependencies After=fsck.target and Before=local-fs.target, this will give the

Re: [systemd-devel] [PATCH] var-run.service: create console dir needed by pam

Kay Sievers ([email protected]) said: > On Fri, Sep 24, 2010 at 16:26, wrote: > > +ExecStart=/bin/mkdir /var/run/console ; /bin/chmod 0755 /var/run/console ; > > -/sbin/restorecon /var/run/console > > Please let the broken pam_console finally die. It was never correct do that. I'd love to.

Re: [systemd-devel] [PATCH] var-run.service: create console dir needed by pam

On 09/24/2010 05:35 PM, Kay Sievers wrote: > On Fri, Sep 24, 2010 at 16:26, wrote: >> +ExecStart=/bin/mkdir /var/run/console ; /bin/chmod 0755 /var/run/console ; >> -/sbin/restorecon /var/run/console > > Please let the broken pam_console finally die. It was never correct do that. > > The last

Re: [systemd-devel] [PATCH] var-run.service: create console dir needed by pam

On Fri, Sep 24, 2010 at 16:26, wrote: > +ExecStart=/bin/mkdir /var/run/console ; /bin/chmod 0755 /var/run/console ; > -/sbin/restorecon /var/run/console Please let the broken pam_console finally die. It was never correct do that. The last user might be the conceptually broken at_console in D-B

[systemd-devel] [PATCH] var-run.service: create console dir needed by pam

From: Harald Hoyer --- units/var-run.service |1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/units/var-run.service b/units/var-run.service index 2cb861e..16acb5a 100644 --- a/units/var-run.service +++ b/units/var-run.service @@ -17,3 +17,4 @@ Type=oneshot RemainAfterExi

[systemd-devel] [PATCH 1/6] readahead-replay: use posix_fadvise instead of readahead

From: Harald Hoyer --- src/readahead-replay.c |8 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/src/readahead-replay.c b/src/readahead-replay.c index 1f1ec52..c93f147 100644 --- a/src/readahead-replay.c +++ b/src/readahead-replay.c @@ -85,8 +85,8 @@ static int unpac

Re: [systemd-devel] [PATCH 1/6] readahead-replay: use posix_fadvise instead of readahead

On Fri, Sep 24, 2010 at 12:05:59PM +0200, [email protected] wrote: > From: Harald Hoyer > > --- > src/readahead-replay.c |4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/src/readahead-replay.c b/src/readahead-replay.c > index b886857..f4d252b 100644 > --- a/src/r

Re: [systemd-devel] [PATCH 2/6] readahead-common:fs_on_ssd() stat might not get the real device name

2010/9/24 Tomasz Torcz : > On Fri, Sep 24, 2010 at 12:22:34PM +0200, Kay Sievers wrote: >> On Fri, Sep 24, 2010 at 12:06,   wrote: >> > From: Harald Hoyer >> > >> > btrfs returns a major(0) for its device, so try to find the mountpoint and >> > real device. >> >> > +        fp = fopen("/proc/self

Re: [systemd-devel] [PATCH 2/6] readahead-common:fs_on_ssd() stat might not get the real device name

On Fri, Sep 24, 2010 at 12:22:34PM +0200, Kay Sievers wrote: > On Fri, Sep 24, 2010 at 12:06, wrote: > > From: Harald Hoyer > > > > btrfs returns a major(0) for its device, so try to find the mountpoint and > > real device. > > > +        fp = fopen("/proc/self/mountinfo", "r"); > > +        i

Re: [systemd-devel] [PATCH 2/6] readahead-common:fs_on_ssd() stat might not get the real device name

On Fri, Sep 24, 2010 at 12:06, wrote: > From: Harald Hoyer > > btrfs returns a major(0) for its device, so try to find the mountpoint and > real device. > +        fp = fopen("/proc/self/mountinfo", "r"); > +        if (fp == NULL) > +                return NULL; > +        while (fscanf(fp, "

[systemd-devel] [PATCH 6/6] readahead-collect: typo, free the correct pointer

From: Harald Hoyer --- src/readahead-collect.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/readahead-collect.c b/src/readahead-collect.c index f7f0f14..9fb4771 100644 --- a/src/readahead-collect.c +++ b/src/readahead-collect.c @@ -434,7 +434,7 @@ finish:

[systemd-devel] [PATCH 5/6] readahead-collect: handle btrfs FIEMAP

From: Harald Hoyer Some files on btrfs do not have a physical extent. Just return an increasing number. --- src/readahead-collect.c |9 - 1 files changed, 8 insertions(+), 1 deletions(-) diff --git a/src/readahead-collect.c b/src/readahead-collect.c index 7983b31..f7f0f14 100644 --

[systemd-devel] [PATCH 4/6] readahead-collect: check for negative return codes of fs_on_ssd()

From: Harald Hoyer --- src/readahead-collect.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/readahead-collect.c b/src/readahead-collect.c index 5ff3711..7983b31 100644 --- a/src/readahead-collect.c +++ b/src/readahead-collect.c @@ -326,7 +326,7 @@ static int coll

[systemd-devel] [PATCH 3/6] hashmap.h: HASHMAP_FOREACH* iterate until ITERATOR_LAST

From: Harald Hoyer --- src/hashmap.h |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/hashmap.h b/src/hashmap.h index ac5a8ae..64a468d 100644 --- a/src/hashmap.h +++ b/src/hashmap.h @@ -77,12 +77,12 @@ void* hashmap_first(Hashmap *h); void* hashmap_last(Hashmap

[systemd-devel] [PATCH 2/6] readahead-common:fs_on_ssd() stat might not get the real device name

From: Harald Hoyer btrfs returns a major(0) for its device, so try to find the mountpoint and real device. --- src/readahead-common.c | 36 ++-- 1 files changed, 34 insertions(+), 2 deletions(-) diff --git a/src/readahead-common.c b/src/readahead-common.c inde

[systemd-devel] [PATCH 1/6] readahead-replay: use posix_fadvise instead of readahead

From: Harald Hoyer --- src/readahead-replay.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/readahead-replay.c b/src/readahead-replay.c index b886857..f4d252b 100644 --- a/src/readahead-replay.c +++ b/src/readahead-replay.c @@ -85,7 +85,7 @@ static int unpack_fi