From: Philip J Kelleher
This patch contains a software workaround for a firmware bug that
can cause the pcie adapter to train to a width below the desired
width of x8.
It will reset the adapter 3 times before the driver gives up and
informs the user that the link width has been trained
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
This patch contains a software workaround for a firmware bug that
can cause the pcie adapter to train to a width below the desired
width of x8.
It will reset the adapter 3 times before the driver gives up and
informs the user that the link width
From: Philip J Kelleher
This patch fixes a possible Kernel Panic on driver load if
the configuration on the card is messed up or not yet set.
The driver could possible give a 32 bit unsigned all Fs to
the kernel as the device's block size.
Now we only write the block size to the kernel
From: Philip J Kelleher
This patch fixes a bug in which discards were always
calling pci_unmap_page. Discards should never call the
pci_unmap_page function call because they are never mapped.
This caused a race condition on PowerPC systems when issuing
discards, writes, and reads all
The incoming patches are for the rsxx driver (drivers/block/rsxx)
Patch1:
This patch fixes a bug in which discards were always
calling pci_unmap_page. Discards should never call the
pci_unmap_page function call because they are never mapped.
This caused a race condition on PowerPC systems when
The incoming patches are for the rsxx driver (drivers/block/rsxx)
Patch1:
This patch fixes a bug in which discards were always
calling pci_unmap_page. Discards should never call the
pci_unmap_page function call because they are never mapped.
This caused a race condition on PowerPC systems when
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
This patch fixes a bug in which discards were always
calling pci_unmap_page. Discards should never call the
pci_unmap_page function call because they are never mapped.
This caused a race condition on PowerPC systems when issuing
discards, writes
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
This patch fixes a possible Kernel Panic on driver load if
the configuration on the card is messed up or not yet set.
The driver could possible give a 32 bit unsigned all Fs to
the kernel as the device's block size.
Now we only write the block
From: Philip J Kelleher
This fixes a kernel panic injected by commit id
8d26750143341831bc312f61c5ed141eeb75b8d0 where discards
are getting mapped through the pci_map_page function call.
The driver will now start verifying that a dma is not a
discard before issuing a the pci_map_page function
The incoming patch is for the rsxx driver (drivers/block/rsxx)
This patch fixes a kernel panic injected by patch id
8d26750143341831bc312f61c5ed141eeb75b8d0 where discards
are getting mapped through the pci_map_page function call.
This patch relies on git commit
The incoming patch is for the rsxx driver (drivers/block/rsxx)
This patch fixes a kernel panic injected by patch id
8d26750143341831bc312f61c5ed141eeb75b8d0 where discards
are getting mapped through the pci_map_page function call.
This patch relies on git commit
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
This fixes a kernel panic injected by commit id
8d26750143341831bc312f61c5ed141eeb75b8d0 where discards
are getting mapped through the pci_map_page function call.
The driver will now start verifying that a dma is not a
discard before issuing
From: Philip J Kelleher
The pci_map_page function has been moved into our
issued workqueue to prevent an us running out of
mappable addresses on non-HWWD PCIe x8 slots. The
maximum amount that can possible be mapped at one
time now is: 255 dmas X 4 dma channels X 4096 Bytes.
Signed-off
From: Philip J Kelleher
The rsxx driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free if the bio was returned before it was
broken up into indiviadual 4k dmas, that is also
fixed here.
Signed-off-by: Philip J Kelleher
The incoming patch series contains 2 patches for the
rsxx driver (drivers/block/rsxx).
Version 1:
A patch series containing 5 patches was sent out a few
weeks ago.
Version 2:
o This version decreased the number of patches for the patch
series based on importance.
o Started using the function
The incoming patch series contains 2 patches for the
rsxx driver (drivers/block/rsxx).
Version 1:
A patch series containing 5 patches was sent out a few
weeks ago.
Version 2:
o This version decreased the number of patches for the patch
series based on importance.
o Started using the function
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The pci_map_page function has been moved into our
issued workqueue to prevent an us running out of
mappable addresses on non-HWWD PCIe x8 slots. The
maximum amount that can possible be mapped at one
time now is: 255 dmas X 4 dma channels X 4096
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free if the bio was returned before it was
broken up into indiviadual 4k dmas, that is also
fixed here.
Signed-off-by: Philip J
From: Philip J Kelleher
The pci_map_page function has been moved into our
issued workqueue to prevent an us running out of
mappable addresses on non-HWWD PCIe x8 slots. The
maximum amount that can possible be mapped at one
time now is: 255 dmas X 4 dma channels X 4096 Bytes.
Signed-off
From: Philip J Kelleher
The rsxx driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free if the bio was returned before it was
broken up into indiviadual 4k dmas, that is also
fixed here.
Signed-off-by: Philip J Kelleher
The incoming patch series contains 2 patches for the
rsxx driver (drivers/block/rsxx). I sent a patch series
containing 5 patches a couple weeks ago, which contained
version 1 of the this series.
Version 2 changes:
o Patches 3 thru 5 were removed because of importance.
o pci_dma_mapping_error was
The incoming patch series contains 2 patches for the
rsxx driver (drivers/block/rsxx). I sent a patch series
containing 5 patches a couple weeks ago, which contained
version 1 of the this series.
Version 2 changes:
o Patches 3 thru 5 were removed because of importance.
o pci_dma_mapping_error was
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free if the bio was returned before it was
broken up into indiviadual 4k dmas, that is also
fixed here.
Signed-off-by: Philip J
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The pci_map_page function has been moved into our
issued workqueue to prevent an us running out of
mappable addresses on non-HWWD PCIe x8 slots. The
maximum amount that can possible be mapped at one
time now is: 255 dmas X 4 dma channels X 4096
From: Philip J Kelleher
The rsxx driver is returning the wrong values during
copy_from_user and copy_to_user. It, also, is not
setting ENOMEM during a failed workqueue allocation.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
The rsxx driver was not releasing the IDA entry during
driver unload. This would be a big problem on a DLPAR
remove where the card is remove from the LPAR but the
driver is still loaded. This patch fixes that issue.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
Changed the device name to IBM Flash Adapter 90 (PCIe2 0.9TB).
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla/Documentation/dontdiff
linux-block-vanilla/drivers/block
From: Philip J Kelleher
The pci_map_page function has been moved into our
issued workqueue to prevent us running out of
mappable addresses on non-HWWD PCIe x8 slots. The
maximum amount that can possible be mapped at one
time now is: 255 dmas X 4 dma channels X 4096 Bytes.
Signed-off-by: Philip
From: Philip J Kelleher
The rsxx driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free if the bio was returned before it was
broken up into indiviadual 4k dmas, that is also
fixed here.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free if the bio was returned before it was
broken up into indiviadual 4k dmas, that is also
fixed here.
Signed-off-by: Philip J
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The pci_map_page function has been moved into our
issued workqueue to prevent us running out of
mappable addresses on non-HWWD PCIe x8 slots. The
maximum amount that can possible be mapped at one
time now is: 255 dmas X 4 dma channels X 4096
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Changed the device name to IBM Flash Adapter 90 (PCIe2 0.9TB).
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block-vanilla
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver was not releasing the IDA entry during
driver unload. This would be a big problem on a DLPAR
remove where the card is remove from the LPAR but the
driver is still loaded. This patch fixes that issue.
Signed-off-by: Philip J
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver is returning the wrong values during
copy_from_user and copy_to_user. It, also, is not
setting ENOMEM during a failed workqueue allocation.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher
The rsxx driver was not checking the correct value during a
pci_map_page failure for PowerPC. Fixing this also uncovered a
double free if the bio was returned before it was broken up
into indiviadual 4k dmas, that is also fixed here.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
The rsxx driver is returning the wrong values during
copy_from_user and copy_to_user. It, also, is not
setting ENOMEM during a failed workqueue allocation.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
The rsxx driver was not releasing the IDA entry during
driver unload. This would be a big problem on a DLPAR
remove where the card is remove from the LPAR but the
driver is still loaded. This patch fixes that issue.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
Changed the device name to IBM Flash Adapter 90 (PCIe2 0.9TB).
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla/Documentation/dontdiff
linux-block-vanilla/drivers/block
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Changed the device name to IBM Flash Adapter 90 (PCIe2 0.9TB).
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block-vanilla
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver was not releasing the IDA entry during
driver unload. This would be a big problem on a DLPAR
remove where the card is remove from the LPAR but the
driver is still loaded. This patch fixes that issue.
Signed-off-by: Philip J
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver is returning the wrong values during
copy_from_user and copy_to_user. It, also, is not
setting ENOMEM during a failed workqueue allocation.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The rsxx driver was not checking the correct value during a
pci_map_page failure for PowerPC. Fixing this also uncovered a
double free if the bio was returned before it was broken up
into indiviadual 4k dmas, that is also fixed here.
Signed-off
Jens,
The incoming patch series contains 4 patches for the
rsxx driver (drivers/block/rsxx). They are all bug
fixes.
Patch1:
Our adapter is going through a name change.
Patch2:
The driver isn't removing the IDA entry during
driver unload. This becomes a problem during
From: Philip J Kelleher
The driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free, that is also fixed here.
Signed-off-by: Philip J Kelleher
---
diff -uprN
From: Philip J Kelleher
The driver is returning the wrong values during
copy_from_user and copy_to_user. It, also, is not
setting ENOMEM during a failed workqueue allocation.
Signed-off-by: Philip J Kelleher
---
diff
From: Philip J Kelleher
The driver was not releasing the IDA entry during
driver unload.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla/Documentation/dontdiff
linux-block-vanilla/drivers
From: Philip J Kelleher
Changed the device name to IBM Flash Adapter 90 (PCIe2 0.9TB).
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla/Documentation/dontdiff
linux-block-vanilla/drivers/block
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Changed the device name to IBM Flash Adapter 90 (PCIe2 0.9TB).
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block-vanilla
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The driver was not releasing the IDA entry during
driver unload.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block-vanilla
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The driver is returning the wrong values during
copy_from_user and copy_to_user. It, also, is not
setting ENOMEM during a failed workqueue allocation.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The driver was not checking the correct value during a
pci_map_page failure. Fixing this also uncovered a
double free, that is also fixed here.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Jens,
The incoming patch series contains 4 patches for the
rsxx driver (drivers/block/rsxx). They are all bug
fixes.
Patch1:
Our adapter is going through a name change.
Patch2:
The driver isn't removing the IDA entry during
driver unload. This becomes a problem during
Hello all,
I have a question regarding a DLPAR remove operations with a mounted
partition.
I am writing a block level device driver that will support DLPAR remove
operations and I was doing some testing with a partition from the adapter
still mounted while I ran a DLPAR remove.
The remove
Hello all,
I have a question regarding a DLPAR remove operations with a mounted
partition.
I am writing a block level device driver that will support DLPAR remove
operations and I was doing some testing with a partition from the adapter
still mounted while I ran a DLPAR remove.
The remove
On Fri, Jun 21, 2013 at 09:21:11AM +0300, Dan Carpenter wrote:
> copy_to/from_user() returns the number of bytes remaining to be copied.
> We should return -EFAULT here instead.
>
> Signed-off-by: Dan Carpenter
>
> diff --git a/drivers/block/rsxx/core.c b/drivers/block/rsxx/core.c
> index
On Fri, Jun 21, 2013 at 10:46:38AM +0800, Wei Yongjun wrote:
> From: Wei Yongjun
>
> Fix to return -ENOMEM in the create_singlethread_workqueue() error handling
> case instead of 0, as done elsewhere in this function.
>
> Introduced by commit a3299ab18591d36ad5622f5064619123c439b779.
> (rsxx:
On Fri, Jun 21, 2013 at 10:46:38AM +0800, Wei Yongjun wrote:
From: Wei Yongjun yongjun_...@trendmicro.com.cn
Fix to return -ENOMEM in the create_singlethread_workqueue() error handling
case instead of 0, as done elsewhere in this function.
Introduced by commit
On Fri, Jun 21, 2013 at 09:21:11AM +0300, Dan Carpenter wrote:
copy_to/from_user() returns the number of bytes remaining to be copied.
We should return -EFAULT here instead.
Signed-off-by: Dan Carpenter dan.carpen...@oracle.com
diff --git a/drivers/block/rsxx/core.c
From: Philip J Kelleher
Adding debugfs entries to help with debugging and testing and
testing code.
pci_regs:
This entry will spit out all of the data stored on the BAR.
stats:
This entry will display all of the driver stats for each
DMA channel.
cram
From: Philip J Kelleher
Fixing incorrect stats calculation during read retries.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla/Documentation/dontdiff
linux-block-vanilla/drivers/block/rsxx
From: Philip J Kelleher
Unfortunaly, our CPU register path does not do any kind of
EEH error checking. So to fix this issue, an ioread32 was
added to the CPU register timeout code. This way, the
driver can check to see if the timeout was caused by an EEH
error or not. This is a dummy read
From: Philip J Kelleher
Adding a sanity check to guarentee that DMAs outside of the device's
address space will be errored out right away.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla
From: Philip J Kelleher
A kernel panic would occur on a DLPAR add if there was a partition
still mounted during the DLPAR remove. This bug fix will allow the
user to unmount the partition and bring the driver back into a
good state after the DLPAR add.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
Changing the adapter name from FlashSystem-80 to the official
name: Flash Adapter 900GB Full Height.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla/Documentation
From: Philip J Kelleher
Before, the partition table would have to be reread because our
card was attached before it transistioned out of it's 'starting'
state.
This change will cause the driver to wait to attach the device
until the adapter is ready.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
Previously, the block size was determined by whether or not
our Hardware could handle 512 byte accesses. Now, all of our
Hardware can handle 512 and 4096 block sizes.
This fix allows it to be user configurable.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
The workqueue mechanism has been reworked to prevent soft
lockup issues from occuring by adding in mutex sychronization.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block-vanilla
From: Philip J Kelleher
Before, DMAs would never be cancelled if there was a data stall
or an EEH Permenant failure which would cause an unrecoverable
I/O hang.
The DMA cancellation mechanism has been modified to fix
these issues and allows DMAs to be cancelled during the
above mentioned
From: Philip J Kelleher
Giving all interrupt based events their own workqueue to complete
tasks on. This fixes a bug that would cause creg commands to timeout
if too many are issued at once.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Giving all interrupt based events their own workqueue to complete
tasks on. This fixes a bug that would cause creg commands to timeout
if too many are issued at once.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Before, DMAs would never be cancelled if there was a data stall
or an EEH Permenant failure which would cause an unrecoverable
I/O hang.
The DMA cancellation mechanism has been modified to fix
these issues and allows DMAs to be cancelled during
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The workqueue mechanism has been reworked to prevent soft
lockup issues from occuring by adding in mutex sychronization.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Previously, the block size was determined by whether or not
our Hardware could handle 512 byte accesses. Now, all of our
Hardware can handle 512 and 4096 block sizes.
This fix allows it to be user configurable.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Before, the partition table would have to be reread because our
card was attached before it transistioned out of it's 'starting'
state.
This change will cause the driver to wait to attach the device
until the adapter is ready.
Signed-off
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Changing the adapter name from FlashSystem-80 to the official
name: Flash Adapter 900GB Full Height.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
A kernel panic would occur on a DLPAR add if there was a partition
still mounted during the DLPAR remove. This bug fix will allow the
user to unmount the partition and bring the driver back into a
good state after the DLPAR add.
Signed-off
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Adding a sanity check to guarentee that DMAs outside of the device's
address space will be errored out right away.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Fixing incorrect stats calculation during read retries.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block-vanilla/Documentation
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Unfortunaly, our CPU register path does not do any kind of
EEH error checking. So to fix this issue, an ioread32 was
added to the CPU register timeout code. This way, the
driver can check to see if the timeout was caused by an EEH
error
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Adding debugfs entries to help with debugging and testing and
testing code.
pci_regs:
This entry will spit out all of the data stored on the BAR.
stats:
This entry will display all of the driver stats for each
DMA
From: Philip J Kelleher
Giving all interrupt based events their own workqueue to complete
tasks on.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block/Documentation/dontdiff
linux-block-vanilla/drivers
From: Philip J Kelleher
Before the partition table would have to be reread because our
card was attached before it transistioned out of it's 'starting'
state.
This change will wait until the card has transitioned from it's
'starting' state before it will attach the device. This way
From: Philip J Kelleher
Adding debugfs entries to help with debugging and testing and
testing code.
pci_regs:
This entry will spit out all of the data stored on the BAR.
stats:
This entry will display all of the driver stats for each
DMA channel.
Signed-off-by: Philip
From: Philip J Kelleher
The block size was determined by whether or not our HW could handle
512 accesses or not. So our configuration utility was not able to
change the block size because all of our newer HW can handle 512
accesses.
This patch allows block sizes to be determined by what
From: Philip J Kelleher
The RHEL 6.x version of the driver is using the legacy
workqueue create function, 'create_singlethread_workqueue'.
This would cause heavy usage of one CPU if data was being
thrashed pretty hard. So, in order to fix it, workqueues
are now being created
From: Philip J Kelleher
Unfortunaly, our CPU register path does not do any kind of
EEH error checking. So to fix this issue, an ioread32 was
added to the CPU register timeout code. This way, the
driver can check to see if the timeout was caused by an EEH
error or not. This is a dummy read
From: Philip J Kelleher
Before DMAs would never be cancelled if the data stall or
EEH Permenant failure occured.
The DMA cancellation mechanism has been modified to fix
these issues and allows DMAs to be cancelled whenever they
should be.
Signed-off-by: Philip J Kelleher
From: Philip J Kelleher
Fixing incorrect stats calculation during read retries.
Signed-off-by: Philip J Kelleher
---
diff -uprN -X linux-block/Documentation/dontdiff
linux-block-vanilla/drivers/block/rsxx/dma.c
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Fixing incorrect stats calculation during read retries.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block/Documentation
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Before DMAs would never be cancelled if the data stall or
EEH Permenant failure occured.
The DMA cancellation mechanism has been modified to fix
these issues and allows DMAs to be cancelled whenever they
should be.
Signed-off-by: Philip J
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Unfortunaly, our CPU register path does not do any kind of
EEH error checking. So to fix this issue, an ioread32 was
added to the CPU register timeout code. This way, the
driver can check to see if the timeout was caused by an EEH
error
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The RHEL 6.x version of the driver is using the legacy
workqueue create function, 'create_singlethread_workqueue'.
This would cause heavy usage of one CPU if data was being
thrashed pretty hard. So, in order to fix it, workqueues
are now being
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
The block size was determined by whether or not our HW could handle
512 accesses or not. So our configuration utility was not able to
change the block size because all of our newer HW can handle 512
accesses.
This patch allows block sizes
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Adding debugfs entries to help with debugging and testing and
testing code.
pci_regs:
This entry will spit out all of the data stored on the BAR.
stats:
This entry will display all of the driver stats for each
DMA
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Before the partition table would have to be reread because our
card was attached before it transistioned out of it's 'starting'
state.
This change will wait until the card has transitioned from it's
'starting' state before it will attach
From: Philip J Kelleher pjk1...@linux.vnet.ibm.com
Giving all interrupt based events their own workqueue to complete
tasks on.
Signed-off-by: Philip J Kelleher pjk1...@linux.vnet.ibm.com
---
diff -uprN -X linux-block
I will look into all of the suggestion that you submitted to me.
Thank you for taking the time to review the patches.
Regards,
-Philip Kelleher
On Wed, May 29, 2013 at 09:18:12PM +0300, Andy Shevchenko wrote:
> On Wed, May 29, 2013 at 12:08 AM, Philip J. Kelleher
> wrote:
> > Fr
I will look into all of the suggestion that you submitted to me.
Thank you for taking the time to review the patches.
Regards,
-Philip Kelleher
On Wed, May 29, 2013 at 09:18:12PM +0300, Andy Shevchenko wrote:
On Wed, May 29, 2013 at 12:08 AM, Philip J. Kelleher
pjk1...@linux.vnet.ibm.com
From: Philip J Kelleher
The RHEL 6.x version of the driver is using the legacy
worqueue create function, 'create_singlethread_workqueue'.
This would cause heavy usage of one CPU if data was being
thrashed pretty hard. So, in order to fix it, workqueues
are now being created
From: Philip J Kelleher
Unfortunaly, our CPU register path does not do any kind of
EEH error checking. So to fix this issue, an ioread32 was
added to the CPU register timeout code. This way, the
driver can check to see if the timeout was caused by an EEH
error or not.
Signed-off-by: Philip J
1 - 100 of 194 matches
Mail list logo
