Chris Leech wrote:
> Christoph Hellwig wrote:
>
>> I just did a very quick glance over the tree. Some extremly highlevel
>> comments to start with before actually starting the source review:
>>
>
> Thanks for taking a look Christoph
>
>
>> - why do you need your own libcrc? lib/crc32.
From: Joe Eykholt <[EMAIL PROTECTED]>
Date: Mon, 26 Nov 2007 18:51:01 -0800
If expecting pre-T11 frames, a T11 frame caused a data fault.
This is because the T11 frames have zeros where the old length/SOF
field was, and a length of less than the size of the FC header was
not checked for.
Add a missing spin_unlock() found when a BLS frame was mis-delivered.
---
drivers/scsi/ofc/openfc/openfc_scsi.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/scsi/ofc/openfc/openfc_scsi.c
b/drivers/scsi/ofc/openfc/openfc_scsi.c
index 86e6895..5fa0ad6 100644
--
Change the default FCoE protocol version to the T11 standard-track one.
The old version can still be selected using 'fcconf ethX fcoe_version old'.
---
usr/ofc/tools/fcconf/src/fcc_adapt.c |1 +
usr/ofc/tools/fcconf/src/fcc_eth.c |1 +
2 files changed, 2 insertions(+), 0 deletions(-)
d
Don't deliver BLS frames to the fast path.
---
usr/common/libfc/src/fc/fc_local_port.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/usr/common/libfc/src/fc/fc_local_port.c
b/usr/common/libfc/src/fc/fc_local_port.c
index ba909d0..a798765 100644
--- a/usr/common/libfc
Don't deliver BLS frames to the fast path.
---
drivers/scsi/ofc/libfc/fc_local_port.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/scsi/ofc/libfc/fc_local_port.c
b/drivers/scsi/ofc/libfc/fc_local_port.c
index 55de5d3..66a1d47 100644
--- a/drivers/scsi/ofc/li
Change the default FCoE protocol version to the T11 (Aug. 2007) version.
Use environment to override: export FCOE_OLD_ALL=1 to use the old protocol
on all interfaces, or FCOE_OLD_eth0=1 to use it on eth0.
---
usr/common/libfc/src/encaps/fcoe.c | 12 ++--
1 files changed, 6 insertions(+
Performance improvement, combine received data copy with CRC.
---
drivers/scsi/ofc/openfc/openfc_scsi.c | 130 +++--
1 files changed, 90 insertions(+), 40 deletions(-)
diff --git a/drivers/scsi/ofc/openfc/openfc_scsi.c
b/drivers/scsi/ofc/openfc/openfc_scsi.c
index 5
fcoe_create() fails without destroying the kmem slab.
This causes the next create to panic.
The destroy should be done in openfc_put_dev.
---
drivers/scsi/ofc/openfc/openfc_if.c |5 +
1 files changed, 1 insertions(+), 4 deletions(-)
diff --git a/drivers/scsi/ofc/openfc/openfc_if.c
b/dri
Add a missing spin_unlock() found when a BLS frame was mis-delivered.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
drivers/scsi/ofc/openfc/openfc_scsi.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/scsi/ofc/openfc/openfc_scsi.c
b/drivers/sc
Don't deliver BLS frames to the fast path.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
drivers/scsi/ofc/libfc/fc_local_port.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/scsi/ofc/libfc/fc_local_port.c
b/drivers/scsi/ofc/libfc/fc_l
fcoe_create() fails without destroying the kmem slab.
This causes the next create to panic.
The destroy should be done in openfc_put_dev.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
drivers/scsi/ofc/openfc/openfc_if.c |5 +
1 files changed, 1 insertions(+), 4 deletions(-)
If expecting pre-T11 frames, a T11 frame caused a data fault.
This is because the T11 frames have zeros where the old length/SOF
field was, and a length of less than the size of the FC header was
not checked for.
Also limit length error messages to frames (per-CPU actually).
Signed-off-by: Joe
[PATCH] Performance improvement, combine received data copy with CRC.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
drivers/scsi/ofc/openfc/openfc_scsi.c | 130 +++--
1 files changed, 90 insertions(+), 40 deletions(-)
diff --git a/drivers/scsi/ofc/
pause on. If not, they are treated as if
there is no link.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
drivers/scsi/ofc/fcoe/fcoe_def.h |1
drivers/scsi/ofc/fcoe/fcoe_if.c | 90 ++
drivers/scsi/ofc/fcoe/fcoein
Joe Eykholt wrote:
> Add a missing spin_unlock() found when a BLS frame was mis-delivered.
Please disregard this patch. It got merged into the wrong function. A better
one is coming.
Joe
> Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
>
> ---
> dr
Add a missing spin_unlock() found when a BLS frame was mis-delivered.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
drivers/scsi/ofc/openfc/openfc_scsi.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/scsi/ofc/openfc/openfc_scsi.c
b/drivers/sc
Don't deliver BLS frames to the fast path.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
usr/common/libfc/src/fc/fc_local_port.c |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/usr/common/libfc/src/fc/fc_local_port.c
b/usr/common/libfc/src/fc/fc_l
Change the default FCoE protocol version to the T11 standard-track one.
The old version can still be selected using 'fcconf ethX fcoe_version old'.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
usr/ofc/tools/fcconf/src/fcc_adapt.c |1 +
usr/ofc/tools/fcconf/src/fcc_eth
Change the default FCoE protocol version to the T11 (Aug. 2007) version.
Use environment to override: export FCOE_OLD_ALL=1 to use the old protocol
on all interfaces, or FCOE_OLD_eth0=1 to use it on eth0.
Signed-off-by: Joe Eykholt <[EMAIL PROTECTED]>
---
usr/common/libfc/src/encaps/
Christoph Hellwig wrote:
> Just wondering:
>
>>From my quick reading of the code these pre-T11 frames are basically
> a different frame-level protocol. Given that T11 has standardized
> on a different one what's the rationale for supporting the old frames?
There is still some use in lab environm
Rob Love wrote:
> Joe Eykholt wrote:
>> [PATCH] Performance improvement, combine received data copy with CRC.
>>
> Shouldn't we remove openfc_cp_to_user() if we're moving that
> functionality into openfc_scsi_recv_data()?
Yes. That was an oversight. I did int
22 matches
Mail list logo