In article <[email protected]>,
Kengo NAKAHARA  <[email protected]> wrote:

>Thus, here is the implementation of above specification (include man)
>    http://netbsd.org/~knakahara/unify-alloc-api/unify-alloc-api.patch
>furthermore, here is if_wm usage example by [email protected]

This has a problem; in the man and code page, you declare pci_intr_alloc as:

+.Ft int
+.Fn pci_intr_alloc  "struct pci_attach_args *pa" \
+"pci_intr_handle_t **ihp" "pci_intr_type_t *counts" \
+"pci_intr_type_t max_type"

And then in the examples you are passing it "int counts[x];":

+       int counts[PCI_INTR_TYPE_SIZE];
+       counts[PCI_INTR_TYPE_MSIX] = 5;
+       counts[PCI_INTR_TYPE_MSI] = 1;
+       counts[PCI_INTR_TYPE_INTX] = 1;
+       error = pci_intr_alloc(pa, ihps, counts,

I think you want int *counts in the declaration; this also avoids the problem
of the enum being possibly unsigned (depending on the implementation).

>http://netbsd.org/~knakahara/unify-alloc-api/unify-alloc-api-wm-example.patch

I like the patch. There is a typo "interrput".

>Could you comment this patch?

I like it! Thanks for working on this...

christos

Reply via email to