From: Philipp Reisner <philipp.reisner@linbit.com>
To: Joe Perches <joe@perches.com>, Jens Axboe <axboe@kernel.dk>,
linux-kernel@vger.kernel.org
Cc: drbd-dev@lists.linbit.com
Subject: Re: [Drbd-dev] [PATCH 068/118] drbd: conn_printk() a dev_printk() alike for drbd's connections
Date: Mon, 29 Aug 2011 13:42:08 +0200 [thread overview]
Message-ID: <201108291342.08926.philipp.reisner@linbit.com> (raw)
In-Reply-To: <1314296185.15882.75.camel@Joe-Laptop>
Am Donnerstag, 25. August 2011, 20:16:25 schrieb Joe Perches:
> On Thu, 2011-08-25 at 17:08 +0200, Philipp Reisner wrote:
> > Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
> > Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
> > ---
> >
> > drivers/block/drbd/drbd_int.h | 9 +++++++++
> > drivers/block/drbd/drbd_main.c | 16 +++++++++++++++-
> > 2 files changed, 24 insertions(+), 1 deletions(-)
>
> Couple of issues with this patch.
>
> > diff --git a/drivers/block/drbd/drbd_int.h
> > b/drivers/block/drbd/drbd_int.h
>
> []
>
> > @@ -102,6 +102,15 @@ struct drbd_tconn;
>
> []
>
> > +extern void conn_printk(const char *level, struct drbd_tconn *tconn,
> > const char *fmt, ...);
>
> This should be
>
> extern __attribute__((format (printf, 3, 4)))
> void conn_printk(const char *level, struct drbd_tconn *tconn, const char
> *fmt, ...);
>
> to have gcc validate the printf format and arguments.
>
> > diff --git a/drivers/block/drbd/drbd_main.c
> > b/drivers/block/drbd/drbd_main.c
>
> []
>
> > @@ -170,6 +170,18 @@ int _get_ldev_if_state(struct drbd_conf *mdev, enum
> > drbd_disk_state mins)
> >
> > #endif
> >
> > +/* printk functions for connections
> > + */
> > +void conn_printk(const char *level, struct drbd_tconn *tconn, const char
> > *fmt, ...) +{
> > + va_list args;
> > +
> > + printk("%sd-con %s: ", level, tconn->name);
> > + va_start(args, fmt);
> > + vprintk(fmt, args);
> > + va_end(args);
>
> And using printk then vprintk is susceptible
> to another thread interleaving a different
> message between the printk and the vprintk.
>
> Using struct va_format and %pV is better
> because no message interleaving is possible.
>
[...]
Actually we came across the interleaving issue as
well and solved it at a later point in time by
converting it into a macro. I folded that into this
patch, which gives us:
>From 3fd752956c255f85429f29603fd080e41ca2d90c Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Mon, 7 Feb 2011 14:01:51 +0100
Subject: [PATCH 068/118] drbd: conn_printk() a dev_printk() alike for drbd's connections
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
drivers/block/drbd/drbd_int.h | 10 ++++++++++
drivers/block/drbd/drbd_main.c | 4 +++-
2 files changed, 13 insertions(+), 1 deletions(-)
diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 927bf77..4d3320d 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -102,6 +102,16 @@ struct drbd_tconn;
/* to shorten dev_warn(DEV, "msg"); and relatives statements */
#define DEV (disk_to_dev(mdev->vdisk))
+#define conn_printk(LEVEL, TCONN, FMT, ARGS...) \
+ printk(LEVEL "d-con %s: " FMT, TCONN->name , ## ARGS)
+#define conn_alert(TCONN, FMT, ARGS...) conn_printk(KERN_ALERT, TCONN, FMT, ## ARGS)
+#define conn_crit(TCONN, FMT, ARGS...) conn_printk(KERN_CRIT, TCONN, FMT, ## ARGS)
+#define conn_err(TCONN, FMT, ARGS...) conn_printk(KERN_ERR, TCONN, FMT, ## ARGS)
+#define conn_warn(TCONN, FMT, ARGS...) conn_printk(KERN_WARNING, TCONN, FMT, ## ARGS)
+#define conn_notice(TCONN, FMT, ARGS...) conn_printk(KERN_NOTICE, TCONN, FMT, ## ARGS)
+#define conn_info(TCONN, FMT, ARGS...) conn_printk(KERN_INFO, TCONN, FMT, ## ARGS)
+#define conn_dbg(TCONN, FMT, ARGS...) conn_printk(KERN_DEBUG, TCONN, FMT, ## ARGS)
+
#define D_ASSERT(exp) if (!(exp)) \
dev_err(DEV, "ASSERT( " #exp " ) in %s:%d\n", __FILE__, __LINE__)
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index dbd6d72..59a3166 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -2217,12 +2217,14 @@ struct drbd_conf *drbd_new_device(unsigned int minor)
struct drbd_conf *mdev;
struct gendisk *disk;
struct request_queue *q;
+ char conn_name[9]; /* drbd1234N */
/* GFP_KERNEL, we are outside of all write-out paths */
mdev = kzalloc(sizeof(struct drbd_conf), GFP_KERNEL);
if (!mdev)
return NULL;
- mdev->tconn = drbd_new_tconn("dummy");
+ sprintf(conn_name, "drbd%d", minor);
+ mdev->tconn = drbd_new_tconn(conn_name);
if (!mdev->tconn)
goto out_no_tconn;
--
1.7.4.1
--
: Dipl-Ing Philipp Reisner
: LINBIT | Your Way to High Availability
: Tel: +43-1-8178292-50, Fax: +43-1-8178292-82
: http://www.linbit.com
DRBD(R) and LINBIT(R) are registered trademarks of LINBIT, Austria.
next prev parent reply other threads:[~2011-08-29 11:42 UTC|newest]
Thread overview: 132+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-08-25 15:06 [RFC 000/118] drbd: part 1 of adding multiple volume support to drbd Philipp Reisner
2011-08-25 15:06 ` [PATCH 001/118] drbd: Get rid of req_validator_fn typedef Philipp Reisner
2011-08-25 15:06 ` [PATCH 002/118] drbd: Remove superfluous declaration Philipp Reisner
2011-08-25 15:06 ` [PATCH 003/118] drbd: Consistently use block_id == ID_SYNCER for checksum based resync and online verify Philipp Reisner
2011-08-25 15:07 ` [PATCH 004/118] drbd: Get rid of BE_DRBD_MAGIC and BE_DRBD_MAGIC_BIG Philipp Reisner
2011-08-25 15:07 ` [PATCH 005/118] drbd: Endianness convert the constants instead of the variables Philipp Reisner
2011-08-25 15:07 ` [PATCH 006/118] drbd: Magic reserved block_id value cleanup Philipp Reisner
2011-08-25 15:07 ` [PATCH 007/118] drbd: Move drbd_free_tl_hash() to drbd_main() Philipp Reisner
2011-08-25 15:07 ` [PATCH 008/118] drbd: Update outdated comment Philipp Reisner
2011-08-25 15:07 ` [PATCH 009/118] drbd: Request lookup code cleanup (1) Philipp Reisner
2011-08-25 15:07 ` [PATCH 010/118] drbd: Request lookup code cleanup (2) Philipp Reisner
2011-08-25 15:07 ` [PATCH 011/118] drbd: Request lookup code cleanup (3) Philipp Reisner
2011-08-25 15:07 ` [PATCH 012/118] drbd: Request lookup code cleanup (4) Philipp Reisner
2011-08-25 15:07 ` [PATCH 013/118] drbd: Add interval tree data structure Philipp Reisner
2011-08-25 15:07 ` [PATCH 014/118] drbd: Put sector and size in struct drbd_request into struct drbd_interval Philipp Reisner
2011-08-25 15:07 ` [PATCH 015/118] drbd: Use interval tree for overlapping write request detection Philipp Reisner
2011-08-25 15:07 ` [PATCH 016/118] drbd: Add read_requests tree Philipp Reisner
2011-08-25 18:02 ` [Drbd-dev] " Pawel Jakub Dawidek
2011-08-29 11:42 ` Philipp Reisner
2011-08-25 15:07 ` [PATCH 017/118] drbd: Use the read and write request trees for request lookups Philipp Reisner
2011-08-25 15:07 ` [PATCH 018/118] drbd: Put sector and size in struct drbd_epoch_entry into struct drbd_interval Philipp Reisner
2011-08-25 15:07 ` [PATCH 019/118] drbd: Use interval tree for overlapping epoch entry detection Philipp Reisner
2011-08-25 15:07 ` [PATCH 020/118] drbd: Remove the unused hash tables Philipp Reisner
2011-08-25 15:07 ` [PATCH 021/118] drbd: Convert all constants in enum drbd_req_event to upper case Philipp Reisner
2011-08-25 15:07 ` [PATCH 022/118] drbd: Convert all constants in enum drbd_thread_state " Philipp Reisner
2011-08-25 15:07 ` [PATCH 023/118] drbd: Replace the ERR_IF macro with an assert-like macro Philipp Reisner
2011-08-25 15:07 ` [PATCH 024/118] drbd: Remove some useless paranoia code Philipp Reisner
2011-08-25 15:07 ` [PATCH 025/118] drbd: Inline function overlaps() is now unused Philipp Reisner
2011-08-25 15:07 ` [PATCH 026/118] drbd: Interval tree bugfix Philipp Reisner
2011-08-25 15:07 ` [PATCH 027/118] idr: idr_for_each_entry() macro Philipp Reisner
2011-08-25 15:07 ` [PATCH 028/118] drbd: Minimal struct drbd_tconn Philipp Reisner
2011-08-25 15:07 ` [PATCH 029/118] drbd: moved net_conf from mdev to tconn Philipp Reisner
2011-08-25 15:07 ` [PATCH 030/118] drbd: moved net_cont and net_cnt_wait " Philipp Reisner
2011-08-25 15:07 ` [PATCH 031/118] drbd: moved data and meta " Philipp Reisner
2011-08-25 15:07 ` [PATCH 032/118] drbd: moved receiver, worker and asender " Philipp Reisner
2011-08-25 15:07 ` [PATCH 033/118] drbd: moved agreed_pro_version, last_received and ko_count " Philipp Reisner
2011-08-25 15:07 ` [PATCH 034/118] drbd: moved req_lock and transfer log from mdev " Philipp Reisner
2011-08-25 15:07 ` [PATCH 035/118] drbd: moved crypto transformations and friends " Philipp Reisner
2011-08-25 15:07 ` [PATCH 036/118] drbd: Made drbd_flush_workqueue() to take a tconn instead of an mdev Philipp Reisner
2011-08-25 15:07 ` [PATCH 037/118] drbd: Preparing to use p_header96 for all packets Philipp Reisner
2011-08-25 15:07 ` [PATCH 038/118] drbd: Replaced all p_header80 with a generic p_header Philipp Reisner
2011-08-25 15:07 ` [PATCH 039/118] drbd: Use new header layout, and send volume IOs Philipp Reisner
2011-08-25 15:07 ` [PATCH 040/118] drbd: Implemented receiving of new style packets on meta socket Philipp Reisner
2011-08-25 15:07 ` [PATCH 041/118] drbd: Do not access tconn after it was freed Philipp Reisner
2011-08-25 15:07 ` [PATCH 042/118] drbd: Move cmdname() out of drbd_int.h Philipp Reisner
2011-08-25 15:07 ` [PATCH 043/118] drbd: Rename "enum drbd_packets" to "enum drbd_packet" Philipp Reisner
2011-08-25 15:07 ` [PATCH 044/118] drbd: Remove redundant initialization Philipp Reisner
2011-08-25 15:07 ` [PATCH 045/118] drbd: Initialize the sequence number sent over the network even when not used Philipp Reisner
2011-08-25 15:07 ` [PATCH 046/118] drbd: Move sequence number logic into drbd_receiver.c and simplify it Philipp Reisner
2011-08-25 15:07 ` [PATCH 047/118] drbd: Move some functions to where they are used Philipp Reisner
2011-08-25 15:07 ` [PATCH 048/118] drbd: struct drbd_request: Introduce a new collision flag Philipp Reisner
2011-08-25 15:07 ` [PATCH 049/118] drbd: Remove redundant check from drbd_contains_interval() Philipp Reisner
2011-08-25 15:07 ` [PATCH 050/118] drbd: Allow to wait for the completion of an epoch entry as well Philipp Reisner
2011-08-25 15:07 ` [PATCH 051/118] drbd: _req_conflicts(): Get rid of the epoch_entries tree Philipp Reisner
2011-08-25 15:07 ` [PATCH 052/118] drbd: Remove unnecessary reference counting left-over Philipp Reisner
2011-08-25 15:07 ` [PATCH 053/118] drbd: Defer new writes when detecting conflicting writes Philipp Reisner
2011-08-25 15:07 ` [PATCH 054/118] drbd: Make the peer_seq updating code more obvious Philipp Reisner
2011-08-25 15:07 ` [PATCH 055/118] drbd: Improve the drbd_find_overlap() documentation Philipp Reisner
2011-08-25 15:07 ` [PATCH 056/118] drbd: Remove unused variable in struct drbd_conf Philipp Reisner
2011-08-25 15:07 ` [PATCH 057/118] drbd: Rename struct drbd_epoch_entry to struct drbd_peer_request Philipp Reisner
2011-08-25 15:07 ` [PATCH 058/118] drbd: Clean up some left-overs Philipp Reisner
2011-08-25 15:07 ` [PATCH 059/118] drbd: Update some comments Philipp Reisner
2011-08-25 15:07 ` [PATCH 060/118] drbd: Local variable renames: e -> peer_req Philipp Reisner
2011-08-25 15:07 ` [PATCH 061/118] drbd: Moved the state functions into its own source file Philipp Reisner
2011-08-25 15:07 ` [PATCH 062/118] drbd: Moved the thread name into the data structure Philipp Reisner
2011-08-25 15:07 ` [PATCH 063/118] drbd: Eliminated the user of drbd_task_to_thread() Philipp Reisner
2011-08-25 15:08 ` [PATCH 064/118] drbd: Moved code Philipp Reisner
2011-08-25 15:08 ` [PATCH 065/118] drbd: Do no sleep long in drbd_start_resync Philipp Reisner
2011-08-25 15:08 ` [PATCH 066/118] drbd: Revert "Make sure we dont send state if a cluster wide state change is in progress" Philipp Reisner
2011-08-25 15:08 ` [PATCH 067/118] drbd: Moving state related macros to drbd_state.h Philipp Reisner
2011-08-25 15:08 ` [PATCH 068/118] drbd: conn_printk() a dev_printk() alike for drbd's connections Philipp Reisner
2011-08-25 18:16 ` Joe Perches
2011-08-29 11:42 ` Philipp Reisner [this message]
2011-08-25 15:08 ` [PATCH 069/118] drbd: Converted drbd_try_connect() from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 070/118] drbd: Converted drbd_wait_for_connect() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 071/118] drbd: Started to separated connection flags (tconn) from block device flags (mdev) Philipp Reisner
2011-08-25 15:08 ` [PATCH 072/118] drbd: Moved DISCARD_CONCURRENT to the per connection (tconn) flags Philipp Reisner
2011-08-25 15:08 ` [PATCH 073/118] drbd: Moved SEND_PING " Philipp Reisner
2011-08-25 15:08 ` [PATCH 074/118] drbd: Moved SIGNAL_ASENDER " Philipp Reisner
2011-08-25 15:08 ` [PATCH 075/118] drbd: Converted wake_asender() and request_ping() from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 076/118] drbd: Converted helper functions for drbd_send() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 077/118] drbd: Converted drbd_send() from mdev " Philipp Reisner
2011-08-25 15:08 ` [PATCH 078/118] drbd: Converted drbd_send_fp() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 079/118] drbd: Removed unused mdev argument from drbd_recv_short() and drbd_socket_okay() Philipp Reisner
2011-08-25 15:08 ` [PATCH 080/118] drbd: Converted drbd_recv_fp() from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 081/118] drbd: Converted drbd_send_handshake() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 082/118] drbd: Converted drbd_recv() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 083/118] drbd: struct packet_info to hold information of decoded packets Philipp Reisner
2011-08-25 15:08 ` [PATCH 084/118] drbd: Converted decode_header() from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 085/118] drbd: Converted drbd_recv_header() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 086/118] drbd: Converted drbd_do_handshake() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 087/118] drbd: Converted drbd_(get|put)_data_sock() and drbd_send_cmd2() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 088/118] drbd: Converted drbd_do_auth() from mdev " Philipp Reisner
2011-08-25 15:08 ` [PATCH 089/118] drbd: Converted drbd_send_protocol() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 090/118] drbd: Use and idr data structure to map volume numbers to mdev pointers Philipp Reisner
2011-08-25 15:08 ` [PATCH 091/118] drbd: Converted drbd_connect() from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 092/118] drbd: Converted drbd_calc_cpu_mask() and drbd_thread_current_set_cpu() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 093/118] drbd: Converted drbdd() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 094/118] drbd: Converted drbd_free_sock() and drbd_disconnect() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 095/118] drbd: Moved the mdev member into drbd_work (from drbd_request and drbd_peer_request) Philipp Reisner
2011-08-25 15:08 ` [PATCH 096/118] drbd: Consolidated the setup of the thread name into the framework Philipp Reisner
2011-08-25 15:08 ` [PATCH 097/118] drbd: Converted drbdd_init() from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 098/118] drbd: Converted drbd_asender() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 099/118] drbd: Converted drbd_worker() " Philipp Reisner
2011-08-25 15:08 ` [PATCH 100/118] drbd: drbd_thread has now a pointer to a tconn instead of to a mdev Philipp Reisner
2011-08-25 15:08 ` [PATCH 101/118] drbd: Moved some initializing code into drbd_new_tconn() Philipp Reisner
2011-08-25 15:08 ` [PATCH 102/118] drbd: Generalized the work callbacks Philipp Reisner
2011-08-25 15:08 ` [PATCH 103/118] drbd: Converted drbd_send_ping() and related functions from mdev to tconn Philipp Reisner
2011-08-25 15:08 ` [PATCH 104/118] drbd: Extracted after_conn_state_ch() out of after_state_ch() Philipp Reisner
2011-08-25 15:08 ` [PATCH 105/118] drbd: Renamed is_valid_state_transition() to is_valid_soft_transition() Philipp Reisner
2011-08-25 15:08 ` [PATCH 106/118] drbd: Extracted is_valid_transition() out of sanitize_state() Philipp Reisner
2011-08-25 15:08 ` [PATCH 107/118] drbd: Extracted is_valid_conn_transition() out of is_valid_transition() Philipp Reisner
2011-08-25 15:08 ` [PATCH 108/118] drbd: Removed the os parameter form sanitize_state() Philipp Reisner
2011-08-25 15:08 ` [PATCH 109/118] drbd: Code de-duplication; new function apply_mask_val() Philipp Reisner
2011-08-25 15:08 ` [PATCH 110/118] drbd: Killed volume0; last step of multi-volume-enablement Philipp Reisner
2011-08-25 15:08 ` [PATCH 111/118] drbd: Removed drbd_state_lock() and drbd_state_unlock() Philipp Reisner
2011-08-25 15:08 ` [PATCH 112/118] drbd: Introduced tconn->cstate_mutex Philipp Reisner
2011-08-25 15:08 ` [PATCH 113/118] drbd: Implemented conn_send_state_req() Philipp Reisner
2011-08-25 15:08 ` [PATCH 114/118] drbd: Global_state_lock not necessary here Philipp Reisner
2011-08-25 15:08 ` [PATCH 115/118] drbd: Implemented conn_send_sr_reply() Philipp Reisner
2011-08-25 15:08 ` [PATCH 116/118] drbd: Implemented receiving of P_CONN_ST_CHG_REPLY Philipp Reisner
2011-08-25 15:08 ` [PATCH 117/118] drbd: implemented receiving of P_CONN_ST_CHG_REQ Philipp Reisner
2011-08-25 15:08 ` [PATCH 118/118] drbd: Implemented connection wide state changes Philipp Reisner
2011-08-26 1:30 ` [RFC 000/118] drbd: part 1 of adding multiple volume support to drbd David Miller
[not found] ` <CAGpXXZKem-uEdF4SyUgpQFV6=d5Vo3v9p3wmcaygAKa4w4x01w@mail.gmail.com>
2011-08-26 13:54 ` [Drbd-dev] " Lars Ellenberg
2011-08-26 15:18 ` Kyle Moffett
2011-08-29 11:51 ` Philipp Reisner
2011-08-29 12:52 ` Pekka Enberg
2011-08-29 15:38 ` Philipp Reisner
2011-08-29 11:46 ` Philipp Reisner
2011-08-29 13:26 ` Greg Freemyer
2011-08-29 16:07 ` Philipp Reisner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201108291342.08926.philipp.reisner@linbit.com \
--to=philipp.reisner@linbit.com \
--cc=axboe@kernel.dk \
--cc=drbd-dev@lists.linbit.com \
--cc=joe@perches.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox