On Mon, 22 Nov 2010, Nikanth Karthikesan wrote:
> On Wednesday 17 November 2010 03:13:41 Anil Ravindranath wrote:
> > The following are the fixes in this patch:
>
> Fixes for what? Can you include the bugs caused by enabling msi-x inthe
> changelog.
>
> > 1. Disable msi-x for now
> > 2. Change device configuration entry size
> >
>
> Why?
>
The reasons are very much controller and FW specific and not driver bugs.
The product does not want to have MSI-X support for now. Maybe in future it will enabled.
Regarding device configuration entry size change again its FW
requirement.
> So this patch kind of disables c20c426732a5a5d21e99b36286f79c2024115341?
> Would it be better to revert it?
>
The commit you are referring is quite old and there are lot other changes
after that. This patch just disables MSI-X.
> > Signed-off-by: Anil Ravindranath<anil_ravindranath@xxxxxxxxxxxxxx>
> >
> > pmcraid.c | 4 +++-
> > pmcraid.h | 6 ++----
> > 2 files changed, 5 insertions(+), 5 deletions(-)
> >
> > --- scsi-misc-2.6.orig/drivers/scsi/pmcraid.h 2010-11-04 12:03:46.000000000
> > -0700 +++ scsi-misc-2.6/drivers/scsi/pmcraid.h 2010-11-04
> > 11:15:06.000000000 -0700 @@ -42,7 +42,7 @@
> > */
> > #define PMCRAID_DRIVER_NAME "PMC MaxRAID"
> > #define PMCRAID_DEVFILE "pmcsas"
> > -#define PMCRAID_DRIVER_VERSION "2.0.3"
> > +#define PMCRAID_DRIVER_VERSION "1.0.3"
>
> hm.. version jump back!
>
One of main reasons we bumped from 1.0. to 2.0 was support for MSI-X and
since we have disabled for now, the version is back to 1.0.3.
> > #define PMCRAID_DRIVER_DATE __DATE__
> >
> > #define PMCRAID_FW_VERSION_1 0x002
> > @@ -333,11 +333,9 @@
> > __u8 lun[PMCRAID_LUN_LEN];
> > } __attribute__((packed, aligned(4)));
> >
> > -/* extended configuration table sizes are of 64 bytes in size */
> > -#define PMCRAID_CFGTE_EXT_SIZE 32
> > +/* extended configuration table sizes are also of 32 bytes in size */
> > struct pmcraid_config_table_entry_ext {
> > struct pmcraid_config_table_entry cfgte;
> > - __u8 cfgte_ext[PMCRAID_CFGTE_EXT_SIZE];
> > };
> >
> > /* resource types (config_table_entry.resource_type values) */
> > --- scsi-misc-2.6.orig/drivers/scsi/pmcraid.c 2010-11-04 12:03:38.000000000
> > -0700 +++ scsi-misc-2.6/drivers/scsi/pmcraid.c 2010-11-04
> > 12:11:21.000000000 -0700 @@ -62,6 +62,7 @@
> > static unsigned int pmcraid_debug_log;
> > static unsigned int pmcraid_disable_aen;
> > static unsigned int pmcraid_log_level = IOASC_LOG_LEVEL_MUST;
> > +static unsigned int pmcraid_enable_msix;
> >
> > /*
> > * Data structures to support multiple adapters by the LLD.
> > @@ -4689,7 +4690,8 @@
> > int rc;
> > struct pci_dev *pdev = pinstance->pdev;
> >
> > - if (pci_find_capability(pdev, PCI_CAP_ID_MSIX)) {
> > + if ((pmcraid_enable_msix) &&
> > + (pci_find_capability(pdev, PCI_CAP_ID_MSIX))) {
>
> coding style nitpick: Unnecessary parenthesis.
>
For readiblity sake the paranthesis are kept. If it was an coding
style issue checkpatch would have picked, I guess, correct?.
> Thanks
> Nikanth
>
> > int num_hrrq = PMCRAID_NUM_MSIX_VECTORS;
> > struct msix_entry entries[PMCRAID_NUM_MSIX_VECTORS];
> > int i;
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
>
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
