Skip to contentSkip to conversation

exofs: 3 changes to exofs & osd

GIT PULL Boaz Harrosh 2 participants 5 messages Dec 17, 2012 – Dec 18, 2012

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

Hi Linus.

Please pull the following changes since commit [ddffeb8c] Linux 3.7-rc1 They are available in the git repository at:

git://git.open-osd.org/linux-open-osd.git for-linus

for you to fetch changes up to

	[861d6660] exofs: don't leak io_state and pages on read error (2012-12-14 12:17:32 +0200)

These are just 3 patches, the last two are bug fixes on the error paths in exofs.

The important patch is the one to osd_uld which adds sysfs info to osd devices for use by user-mode clustering discovery software. I'm already sitting on this patch since before February this year, It is important for some of the big installation cluster systems, who's been compiling their own kernel just for that patch.

Thanks in advance
Boaz

----------------------------------------------------------------
Boaz Harrosh (1):
exofs: don't leak io_state and pages on read error

Idan Kedar (1):
exofs: clean up the correct page collection on write error

Sachin Bhamare (1):
osduld: Add osdname & systemid sysfs at scsi_osd class

 drivers/scsi/osd/osd_uld.c | 28 ++++++++++++++++++++++++++++
 fs/exofs/inode.c           | 16 +++++++++-------
 2 files changed, 37 insertions(+), 7 deletions(-)
maintainer
#1 lore raw
Quoted 10 lines · Boaz Harrosh

Hi Linus.

Please pull the following changes since commit [ddffeb8c] Linux 3.7-rc1 They are available in the git repository at:

git://git.open-osd.org/linux-open-osd.git for-linus

for you to fetch changes up to

	[861d6660] exofs: don't leak io_state and pages on read error (2012-12-14 12:17:32 +0200)
Quoted 8 lines

These are just 3 patches, the last two are bug fixes on the error paths in exofs.

The important patch is the one to osd_uld which adds sysfs info to osd devices for use by user-mode clustering discovery software. I'm already sitting on this patch since before February this year, It is important for some of the big installation cluster systems, who's been compiling their own kernel just for that patch.

I'm a bit perplexed by this. You got notice when it was added to the SCSI tree and now it's already upstream:

commit 51976a8c85cec0c62e410bc38b8a11dbc690764d

Author: Boaz Harrosh <[email protected]>
Date:   Wed Oct 24 14:51:41 2012 -0700

[SCSI] osd_uld: Add osdname & systemid sysfs at scsi_osd class

But the authorship info differs ... it looks like you forgot to include the From: tag in your original patch send.

James

#2 lore raw
Quoted 23 lines · James Bottomley
Quoted 10 lines · Boaz Harrosh

Hi Linus.

Please pull the following changes since commit [ddffeb8c] Linux 3.7-rc1 They are available in the git repository at:

git://git.open-osd.org/linux-open-osd.git for-linus

for you to fetch changes up to

	[861d6660] exofs: don't leak io_state and pages on read error (2012-12-14 12:17:32 +0200)
Quoted 8 lines

These are just 3 patches, the last two are bug fixes on the error paths in exofs.

The important patch is the one to osd_uld which adds sysfs info to osd devices for use by user-mode clustering discovery software. I'm already sitting on this patch since before February this year, It is important for some of the big installation cluster systems, who's been compiling their own kernel just for that patch.

I'm a bit perplexed by this. You got notice when it was added to the SCSI tree and now it's already upstream:

Quoted 9 lines

commit 51976a8c85cec0c62e410bc38b8a11dbc690764d

Author: Boaz Harrosh <[email protected]>
Date:   Wed Oct 24 14:51:41 2012 -0700

[SCSI] osd_uld: Add osdname & systemid sysfs at scsi_osd class

But the authorship info differs ... it looks like you forgot to include the From: tag in your original patch send.

I'm so sorry, I completely goofed on this one. It's what happens when you are swamped with other work and are doing things without thinking. I totally forgot that I need to remove this patch.

Both these patches where in linux-next for a long time. So I believe the merge will go just fine. Lets leave it like this, or I can rebase and remove it?

Quoted 3 lines

James

Thanks
Boaz

maintainer
#3 lore raw
Quoted 34 lines · Boaz Harrosh
Quoted 23 lines · James Bottomley
Quoted 10 lines · Boaz Harrosh

Hi Linus.

Please pull the following changes since commit [ddffeb8c] Linux 3.7-rc1 They are available in the git repository at:

git://git.open-osd.org/linux-open-osd.git for-linus

for you to fetch changes up to

	[861d6660] exofs: don't leak io_state and pages on read error (2012-12-14 12:17:32 +0200)
Quoted 8 lines

These are just 3 patches, the last two are bug fixes on the error paths in exofs.

The important patch is the one to osd_uld which adds sysfs info to osd devices for use by user-mode clustering discovery software. I'm already sitting on this patch since before February this year, It is important for some of the big installation cluster systems, who's been compiling their own kernel just for that patch.

I'm a bit perplexed by this. You got notice when it was added to the SCSI tree and now it's already upstream:

Quoted 9 lines

commit 51976a8c85cec0c62e410bc38b8a11dbc690764d

Author: Boaz Harrosh <[email protected]>
Date:   Wed Oct 24 14:51:41 2012 -0700

[SCSI] osd_uld: Add osdname & systemid sysfs at scsi_osd class

But the authorship info differs ... it looks like you forgot to include the From: tag in your original patch send.

Quoted 7 lines

I'm so sorry, I completely goofed on this one. It's what happens when you are swamped with other work and are doing things without thinking. I totally forgot that I need to remove this patch.

Both these patches where in linux-next for a long time. So I believe the merge will go just fine. Lets leave it like this, or I can rebase and remove it?

If it merges OK, I'd just leave it as is. It wouldn't be anywhere close to the first time we've had the same patch via different trees.

James

#4 lore raw

On 12/18/2012 01:28 PM, James Bottomley wrote:
<>

Quoted 7 lines
Quoted 3 lines

Both these patches where in linux-next for a long time. So I believe the merge will go just fine. Lets leave it like this, or I can rebase and remove it?

If it merges OK, I'd just leave it as is. It wouldn't be anywhere close to the first time we've had the same patch via different trees.

Thanks James, it will not happen again. I just had a look on linux-next of Dec 17 they both where merged in and I did not see any complains.

Quoted 2 lines

James

Boaz