[Qemu-devel] [PATCHv2] remove unused field from pci device.

[Qemu-devel] [PATCHv2] remove unused field from pci device.

[Gleb Natapov]

Rename pci_irq_index to pci_device_count because this is what it
actually does.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
diff --git a/hw/pci.c b/hw/pci.c
index 0ab5b94..d00041a 100644
--- a/hw/pci.c
+++ b/hw/pci.c
@@ -54,7 +54,7 @@ static void pci_set_irq(void *opaque, int irq_num, int level);
 target_phys_addr_t pci_mem_base;
 static uint16_t pci_default_sub_vendor_id = PCI_SUBVENDOR_ID_REDHAT_QUMRANET;
 static uint16_t pci_default_sub_device_id = PCI_SUBDEVICE_ID_QEMU;
-static int pci_irq_index;
+static int pci_device_count;
 static PCIBus *first_bus;
 
 static void pcibus_save(QEMUFile *f, void *opaque)
@@ -245,7 +245,7 @@ static PCIDevice *do_pci_register_device(PCIDevice *pci_dev, PCIBus *bus,
                                          PCIConfigReadFunc *config_read,
                                          PCIConfigWriteFunc *config_write)
 {
-    if (pci_irq_index >= PCI_DEVICES_MAX)
+    if (pci_device_count >= PCI_DEVICES_MAX)
         return NULL;
 
     if (devfn < 0) {
@@ -268,7 +268,7 @@ static PCIDevice *do_pci_register_device(PCIDevice *pci_dev, PCIBus *bus,
         config_write = pci_default_write_config;
     pci_dev->config_read = config_read;
     pci_dev->config_write = config_write;
-    pci_dev->irq_index = pci_irq_index++;
+    pci_device_count++;
     bus->devices[devfn] = pci_dev;
     pci_dev->irq = qemu_allocate_irqs(pci_set_irq, pci_dev, 4);
     return pci_dev;
@@ -322,7 +322,7 @@ int pci_unregister_device(PCIDevice *pci_dev)
     pci_unregister_io_regions(pci_dev);
 
     qemu_free_irqs(pci_dev->irq);
-    pci_irq_index--;
+    pci_device_count--;
     pci_dev->bus->devices[pci_dev->devfn] = NULL;
     qdev_free(&pci_dev->qdev);
     return 0;
diff --git a/hw/pci.h b/hw/pci.h
index 0405837..fb7b89a 100644
--- a/hw/pci.h
+++ b/hw/pci.h
@@ -154,8 +154,6 @@ struct PCIDevice {
     PCIConfigReadFunc *config_read;
     PCIConfigWriteFunc *config_write;
     PCIUnregisterFunc *unregister;
-    /* ??? This is a PC-specific hack, and should be removed.  */
-    int irq_index;
 
     /* IRQ objects for the INTA-INTD pins.  */
     qemu_irq *irq;
--
			Gleb.

Re: [Qemu-devel] [PATCHv2] remove unused field from pci device.

[Paul Brook]

On Tuesday 09 June 2009, Gleb Natapov wrote:
> Rename pci_irq_index to pci_device_count because this is what it
> actually does.

This whole thing looks bogus. AFAICS the limit is completely arbitrary and 
unnecessary.

Paul

Re: [Qemu-devel] [PATCHv2] remove unused field from pci device.

[Gleb Natapov]

On Tue, Jun 09, 2009 at 02:01:52PM +0100, Paul Brook wrote:
> On Tuesday 09 June 2009, Gleb Natapov wrote:
> > Rename pci_irq_index to pci_device_count because this is what it
> > actually does.
> 
> This whole thing looks bogus. AFAICS the limit is completely arbitrary and 
> unnecessary.
> 
So? Left it bogus with unused field and cryptic name instead?

--
			Gleb.

Re: [Qemu-devel] [PATCHv2] remove unused field from pci device.

[Paul Brook]

On Tuesday 09 June 2009, Gleb Natapov wrote:
> On Tue, Jun 09, 2009 at 02:01:52PM +0100, Paul Brook wrote:
> > On Tuesday 09 June 2009, Gleb Natapov wrote:
> > > Rename pci_irq_index to pci_device_count because this is what it
> > > actually does.
> >
> > This whole thing looks bogus. AFAICS the limit is completely arbitrary
> > and unnecessary.
>
> So? Left it bogus with unused field and cryptic name instead?

No. Remove the whole thing.

Paul

Re: [Qemu-devel] [PATCHv2] remove unused field from pci device.

[Gleb Natapov]

On Tue, Jun 09, 2009 at 02:11:58PM +0100, Paul Brook wrote:
> On Tuesday 09 June 2009, Gleb Natapov wrote:
> > On Tue, Jun 09, 2009 at 02:01:52PM +0100, Paul Brook wrote:
> > > On Tuesday 09 June 2009, Gleb Natapov wrote:
> > > > Rename pci_irq_index to pci_device_count because this is what it
> > > > actually does.
> > >
> > > This whole thing looks bogus. AFAICS the limit is completely arbitrary
> > > and unnecessary.
> >
> > So? Left it bogus with unused field and cryptic name instead?
> 
> No. Remove the whole thing.
Well, not with may understanding of that code, but if somebody wants to
do it...

--
			Gleb.