Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Signed-off-by: Hannes Reinecke hare@suse.de --- src/megasas.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/src/megasas.c b/src/megasas.c index f4eeba0..167fc32 100644 --- a/src/megasas.c +++ b/src/megasas.c @@ -353,6 +353,10 @@ init_megasas(struct pci_device *pci) u32 iobase = pci_config_readl(pci->bdf, PCI_BASE_ADDRESS_2) & PCI_BASE_ADDRESS_IO_MASK;
+ if (!iobase) + iobase = pci_config_readl(pci->bdf, PCI_BASE_ADDRESS_0) + & PCI_BASE_ADDRESS_IO_MASK; + dprintf(1, "found MegaRAID SAS at %02x:%02x.%x, io @ %x\n", pci_bdf_to_bus(bdf), pci_bdf_to_dev(bdf), pci_bdf_to_fn(bdf), iobase);
On 06/11/2014 14:31, Hannes Reinecke wrote:
Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Signed-off-by: Hannes Reinecke hare@suse.de
Thanks---Gerd, can we somehow add this patch to QEMU 2.2? O:-)
Paolo
src/megasas.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/src/megasas.c b/src/megasas.c index f4eeba0..167fc32 100644 --- a/src/megasas.c +++ b/src/megasas.c @@ -353,6 +353,10 @@ init_megasas(struct pci_device *pci) u32 iobase = pci_config_readl(pci->bdf, PCI_BASE_ADDRESS_2) & PCI_BASE_ADDRESS_IO_MASK;
- if (!iobase)
iobase = pci_config_readl(pci->bdf, PCI_BASE_ADDRESS_0)& PCI_BASE_ADDRESS_IO_MASK;- dprintf(1, "found MegaRAID SAS at %02x:%02x.%x, io @ %x\n", pci_bdf_to_bus(bdf), pci_bdf_to_dev(bdf), pci_bdf_to_fn(bdf), iobase);
On Do, 2014-11-06 at 14:44 +0100, Paolo Bonzini wrote:
On 06/11/2014 14:31, Hannes Reinecke wrote:
Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Signed-off-by: Hannes Reinecke hare@suse.de
Thanks---Gerd, can we somehow add this patch to QEMU 2.2? O:-)
cherry-pick into stable-1.7.5, then roll out 1.7.5.1 ...
/me can handle that. While being at it: other important bugfixes in master which should be picked up?
cheers, Gerd
On 11/07/14 09:33, Gerd Hoffmann wrote:
On Do, 2014-11-06 at 14:44 +0100, Paolo Bonzini wrote:
On 06/11/2014 14:31, Hannes Reinecke wrote:
Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Signed-off-by: Hannes Reinecke hare@suse.de
Thanks---Gerd, can we somehow add this patch to QEMU 2.2? O:-)
cherry-pick into stable-1.7.5, then roll out 1.7.5.1 ...
/me can handle that. While being at it: other important bugfixes in master which should be picked up?
I had Markus's %x fixes in mind (for rom and scsi lun numbering, for the boot order), but you've already picked those.
Thanks Laszlo
On Fri, Nov 07, 2014 at 09:33:13AM +0100, Gerd Hoffmann wrote:
On Do, 2014-11-06 at 14:44 +0100, Paolo Bonzini wrote:
On 06/11/2014 14:31, Hannes Reinecke wrote:
Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Signed-off-by: Hannes Reinecke hare@suse.de
Thanks---Gerd, can we somehow add this patch to QEMU 2.2? O:-)
cherry-pick into stable-1.7.5, then roll out 1.7.5.1 ...
/me can handle that. While being at it: other important bugfixes in master which should be picked up?
I see the following changes that look like simple bug fixes:
c58799cb - vgabios: Return from handle_1011() if handler found dd7527c6 - Don't enable thread preemption during S3 resume vga option rom execution ec44fac1 - build: Avoid absolute paths during "whole-program" compiling 68354a6c - romlayout: Use "rep ; nop" not "rep nop" 67d1fbef - ehci: Fix bug in hub port assignment
I'll leave it up to you to decide which patches should go into the stable release though.
-Kevin
On Fri, Nov 07, 2014 at 11:02:32AM -0500, Kevin O'Connor wrote:
On Fri, Nov 07, 2014 at 09:33:13AM +0100, Gerd Hoffmann wrote:
On Do, 2014-11-06 at 14:44 +0100, Paolo Bonzini wrote:
On 06/11/2014 14:31, Hannes Reinecke wrote:
Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Signed-off-by: Hannes Reinecke hare@suse.de
Thanks---Gerd, can we somehow add this patch to QEMU 2.2? O:-)
cherry-pick into stable-1.7.5, then roll out 1.7.5.1 ...
/me can handle that. While being at it: other important bugfixes in master which should be picked up?
I see the following changes that look like simple bug fixes:
c58799cb - vgabios: Return from handle_1011() if handler found dd7527c6 - Don't enable thread preemption during S3 resume vga option rom execution ec44fac1 - build: Avoid absolute paths during "whole-program" compiling 68354a6c - romlayout: Use "rep ; nop" not "rep nop"
Oops - 68354a6c was a fix for a build problem introduced after rel-1.7.5, so ignore that one.
-Kevin
67d1fbef - ehci: Fix bug in hub port assignment
I'll leave it up to you to decide which patches should go into the stable release though.
-Kevin
On Do, 2014-11-06 at 14:31 +0100, Hannes Reinecke wrote:
Some newer cards have the I/O bar at BAR 0, so we need to check that, too, if we cannot get an address for BAR 2. Without this patch the new 'megasas-gen2' emulation in qemu is not detected.
Applied to master.
thanks, Gerd
-
Gerd Hoffmann -
Hannes Reinecke -
Kevin O'Connor -
Laszlo Ersek -
Paolo Bonzini