Bart,
> The two patches in this series make the qla2xxx driver source code
> easier to read without changing the driver functionality. Please
> consider these patches for kernel v4.21.
I applied patch #1. #2 had conflicts, please rebase.
Thanks!
--
Martin K. Petersen Oracle Linux
Nilesh,
> Please consider below patch set for next 'scsi-fixes' submission.
Some of these smelled more like features than bug fixes. So I applied
the series to 4.21/scsi-queue.
--
Martin K. Petersen Oracle Linux Engineering
Steffen,
As I said, I don't have a problem with having module parameters.
> There's one more important thing that has performance impact: We need to
> pack payload and protection data into the same queue of limited
> length. So for the worst case with DIX, we have to use half the size for
>
On 11/28/18 5:01 PM, David Disseldorp wrote:
> From: Bart Van Assche
>
> The existing for loops step over null-terminators for right-padding.
> Padding can be achieved in a much simpler way using printf width
> specifiers.
>
> Signed-off-by: David Disseldorp
> ---
>
On 11/28/18 5:01 PM, David Disseldorp wrote:
> Use the value stored in t10_wwn.vendor, which defaults to "LIO-ORG", but
> can be reconfigured via the vendor_id ConfigFS attribute.
>
> Signed-off-by: David Disseldorp
> ---
> drivers/target/target_core_spc.c | 8 +---
> 1 file changed, 5
On 11/28/18 5:01 PM, David Disseldorp wrote:
> The vendor_id attribute will allow for the modification of the T10
> Vendor Identification string returned in inquiry responses. Its value
> can be viewed and modified via the ConfigFS path at:
> target/core/$backstore/$name/wwn/vendor_id
>
>
On 11/28/18 5:01 PM, David Disseldorp wrote:
> The pscsi_set_inquiry_info() codepath doesn't currently explicitly
> null-terminate t10_wwn.revision.
> Add an extra byte to the t10_wwn.model buffer and perform null string
> termination in all cases.
>
> Signed-off-by: David Disseldorp
> ---
>
On 11/28/18 5:01 PM, David Disseldorp wrote:
> The pscsi_set_inquiry_info() and emulate_model_alias_store() codepaths
> don't currently explicitly null-terminate t10_wwn.model.
> Add an extra byte to the t10_wwn.model buffer and perform null string
> termination in all cases.
>
>
On 11/28/18 5:01 PM, David Disseldorp wrote:
> In preparation for supporting user provided vendor strings, add an extra
> byte to t10_wwn.vendor which will ensure null string termination when an
> eight byte vendor string is provided by the user.
>
> Signed-off-by: David Disseldorp
> ---
>
> On Nov 28, 2018, at 7:01 PM, David Disseldorp wrote:
>
> In preparation for supporting user provided vendor strings, add an extra
> byte to t10_wwn.vendor which will ensure null string termination when an
> eight byte vendor string is provided by the user.
>
> Signed-off-by: David
> On Nov 28, 2018, at 7:01 PM, David Disseldorp wrote:
>
> spc5r17.pdf specifies:
> 4.3.1 ASCII data field requirements
> ASCII data fields shall contain only ASCII printable characters (i.e.,
> code values 20h to 7Eh) and may be terminated with one or more ASCII
> null (00h) characters.
On 11/28/18 5:01 PM, David Disseldorp wrote:
> spc5r17.pdf specifies:
> 4.3.1 ASCII data field requirements
> ASCII data fields shall contain only ASCII printable characters (i.e.,
> code values 20h to 7Eh) and may be terminated with one or more ASCII
> null (00h) characters.
> ASCII
> On Nov 28, 2018, at 7:01 PM, David Disseldorp wrote:
>
> Use the value stored in t10_wwn.vendor, which defaults to "LIO-ORG", but
> can be reconfigured via the vendor_id ConfigFS attribute.
>
> Signed-off-by: David Disseldorp
> ---
> drivers/target/target_core_spc.c | 8 +---
> 1 file
> On Nov 28, 2018, at 7:01 PM, David Disseldorp wrote:
>
> The vendor_id attribute will allow for the modification of the T10
> Vendor Identification string returned in inquiry responses. Its value
> can be viewed and modified via the ConfigFS path at:
>
> On Nov 28, 2018, at 7:01 PM, David Disseldorp wrote:
>
> The pscsi_set_inquiry_info() and emulate_model_alias_store() codepaths
> don't currently explicitly null-terminate t10_wwn.model.
> Add an extra byte to the t10_wwn.model buffer and perform null string
> termination in all cases.
>
>
> On Nov 28, 2018, at 7:01 PM, David Disseldorp wrote:
>
> The pscsi_set_inquiry_info() codepath doesn't currently explicitly
> null-terminate t10_wwn.revision.
> Add an extra byte to the t10_wwn.model buffer and perform null string
> termination in all cases.
>
> Signed-off-by: David
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> From: Manish Rangankar
>
> 1. Removed logic to update HW producer index in interrupt context.
> 2. Update HW producer index after UIO ring and buffer gets initialized.
>
> Signed-off-by: Manish Rangankar
> ---
> drivers/scsi/qedi/qedi_main.c | 31
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> Default packet size is 0x400.
> For jumbo packets set to 0x2400.
>
> Signed-off-by: Nilesh Javali
> ---
> drivers/scsi/qedi/qedi.h | 1 -
> drivers/scsi/qedi/qedi_main.c | 13 +
> 2 files changed, 9 insertions(+), 5 deletions(-)
>
>
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> From: Manish Rangankar
>
> Add packet filter to avoid unnecessary packet processing in iscsiuio.
>
> Signed-off-by: Manish Rangankar
> ---
> drivers/scsi/qedi/qedi_main.c | 24
> 1 file changed, 24 insertions(+)
>
> diff
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> From: Manish Rangankar
>
> The kernel panic was observed after switch side perturbation,
>
> BUG: unable to handle kernel NULL pointer dereference at (null)
> IP: [] strcmp+0x20/0x40
> PGD 0 Oops: [#1] SMP
> CPU: 8 PID: 647 Comm:
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> Use QEDI_PAGE_SIZE for enablement of module on systems with 64K page size.
>
> Signed-off-by: Nilesh Javali
> ---
> drivers/scsi/qedi/qedi_main.c | 16 +---
> 1 file changed, 9 insertions(+), 7 deletions(-)
>
> diff --git
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> Fix trivial spelling mistake within macro definition.
>
> Signed-off-by: Nilesh Javali
> ---
> drivers/scsi/qedi/qedi.h | 4 ++--
> drivers/scsi/qedi/qedi_main.c | 2 +-
> 2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> Remove redundant macro definition.
>
> Signed-off-by: Nilesh Javali
> ---
> drivers/scsi/qedi/qedi.h | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/drivers/scsi/qedi/qedi.h b/drivers/scsi/qedi/qedi.h
> index a6f96b3..e966855 100644
> ---
On 11/21/18 1:25 AM, Nilesh Javali wrote:
> The driver load on some systems failed with error,
> [0004:01:00.5]:[qedi_request_msix_irq:2524]:8: request_irq failed.
>
> Allocate the IRQs based on MSIX count obtained from qed module
> instead of number of queues.
>
> Signed-off-by: Nilesh Javali
The pscsi_set_inquiry_info() and emulate_model_alias_store() codepaths
don't currently explicitly null-terminate t10_wwn.model.
Add an extra byte to the t10_wwn.model buffer and perform null string
termination in all cases.
dev_set_t10_wwn_model_alias() continues to truncate at the same length
to
From: Bart Van Assche
The existing for loops step over null-terminators for right-padding.
Padding can be achieved in a much simpler way using printf width
specifiers.
Signed-off-by: David Disseldorp
---
drivers/target/target_core_device.c | 35 ---
Use the value stored in t10_wwn.vendor, which defaults to "LIO-ORG", but
can be reconfigured via the vendor_id ConfigFS attribute.
Signed-off-by: David Disseldorp
---
drivers/target/target_core_spc.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git
The pscsi_set_inquiry_info() codepath doesn't currently explicitly
null-terminate t10_wwn.revision.
Add an extra byte to the t10_wwn.model buffer and perform null string
termination in all cases.
Signed-off-by: David Disseldorp
---
drivers/target/target_core_device.c | 6 --
The vendor_id attribute will allow for the modification of the T10
Vendor Identification string returned in inquiry responses. Its value
can be viewed and modified via the ConfigFS path at:
target/core/$backstore/$name/wwn/vendor_id
"LIO-ORG" remains the default value, which is set when the
In preparation for supporting user provided vendor strings, add an extra
byte to t10_wwn.vendor which will ensure null string termination when an
eight byte vendor string is provided by the user.
Signed-off-by: David Disseldorp
---
drivers/target/target_core_device.c | 6 --
This patchset allows for the modification of the T10 Vendor
Identification string returned in the SCSI INQUIRY response, via the
target/core/$backstore/$name/wwn/vendor_id ConfigFS path.
Changes since v3:
- perform explicit null termination of t10_wwn vendor, model and
revision fields.
-
spc5r17.pdf specifies:
4.3.1 ASCII data field requirements
ASCII data fields shall contain only ASCII printable characters (i.e.,
code values 20h to 7Eh) and may be terminated with one or more ASCII
null (00h) characters.
ASCII data fields described as being left-aligned shall have any
David,
> This patchset removes unneeded se_dev_attrib.pi_prot_format and
> fabric_ops.get_fabric_name() struct members. Removal of the latter
> allowed for further cleanup due to the fact that all fabric modules
> except iscsi_target_mod provide matching strings for fabric_ops.name
> and
> On Nov 27, 2018, at 3:04 PM, Bart Van Assche wrote:
>
> External Email
>
> Nesting in __qla2x00_abort_all_cmds() is way too deep. Reduce the nesting
> level by introducing a helper function. This patch does not change any
> functionality.
>
> Cc: Himanshu Madhani
> Signed-off-by: Bart Van
> On Nov 27, 2018, at 3:04 PM, Bart Van Assche wrote:
>
> External Email
>
> This patch improves code readability but does not change any
> functionality.
>
> Cc: Himanshu Madhani
> Signed-off-by: Bart Van Assche
> ---
> drivers/scsi/qla2xxx/qla_target.c | 14 +++---
> 1 file
Martin,
> Since f44ac12f1dcc, BG enablement is tracked with the
> LPFC_SLI3_BG_ENABLED bit, which is set in lpfc_get_cfgparam before
> lpfc_sli_config_sli_port() is called. The bit shouldn't be cleared
> before checking the feature. Based on problem analysis by David Bond.
Applied to
On Wed, 2018-11-28 at 17:44 +0100, David Disseldorp wrote:
> Hi Bart,
>
> On Wed, 28 Nov 2018 08:36:19 -0800, Bart Van Assche wrote:
>
> > Maybe I'm missing something, but why is zeroing of unused bytes in these
> > functions
> > necessary? Would the following be correct if all strings in
Hi Bart,
On Wed, 28 Nov 2018 08:36:19 -0800, Bart Van Assche wrote:
> Maybe I'm missing something, but why is zeroing of unused bytes in these
> functions
> necessary? Would the following be correct if all strings in struct t10_wwn
> would be
> '\0'-terminated?
Your patch looks good to me.
On Wed, 2018-11-28 at 17:28 +0100, David Disseldorp wrote:
> On Wed, 28 Nov 2018 08:08:30 -0800, Bart Van Assche wrote:
>
> > > Just a follow up here. I think it's safer to retain strncpy() in this
> > > function for the purpose of zero filling. scsi_dump_inquiry() and
> > >
On Wed, 28 Nov 2018 08:08:30 -0800, Bart Van Assche wrote:
> > Just a follow up here. I think it's safer to retain strncpy() in this
> > function for the purpose of zero filling. scsi_dump_inquiry() and
> > target_stat_lu_vend_show() walk the entire length of the t10_wwn.vendor
> > buffer.
>
>
On Wed, 2018-11-28 at 16:37 +0100, David Disseldorp wrote:
> On Tue, 20 Nov 2018 19:00:57 +0100, David Disseldorp wrote:
>
> > > > + strncpy(buf, page, sizeof(buf));
> > >
> > > Isn't strncpy() deprecated? How about using strlcpy() instead?
> >
> > Will change to use strlcpy in the
On Tue, 20 Nov 2018 19:00:57 +0100, David Disseldorp wrote:
> > > + strncpy(buf, page, sizeof(buf));
> >
> > Isn't strncpy() deprecated? How about using strlcpy() instead?
>
> Will change to use strlcpy in the next round.
Just a follow up here. I think it's safer to retain strncpy() in
Assalamu alaikum,
I came across your e-mail contact prior a private search while in need
of a
trusted person. My name is Mrs. Aisha Gaddafi, a single Mother and a
Widow
with three Children. I am the only biological Daughter of late Libyan
President (Late Colonel Muammar Gaddafi)I have a
43 matches
Mail list logo