Date: Fri, 19 Apr 2013 11:31:49 -0400 From: John Baldwin <jhb@freebsd.org> To: freebsd-acpi@freebsd.org Subject: Re: panic: acpi_pci_link_srs_from_crs: can't put non-ISA IRQ 20 in legacy IRQ resource type) Message-ID: <201304191131.49433.jhb@freebsd.org> In-Reply-To: <20130418124940.47e3618a@b1c1l1.com> References: <20130418124940.47e3618a@b1c1l1.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday, April 18, 2013 3:49:40 pm Benjamin Lee wrote:
> I have a system that panics on boot with 10-CURRENT and boots with many
> ACPI error messages and non-functional devices with 9.1-RELEASE.
>
> Motherboard is Foxconn C51XEM2AA (NVIDIA nForce 590) desktop board.
>
> Here is the panic on 10-CURRENT:
>
> pcib0: allocated type 3 (0xdffff000-0xdfffffff) for rid 10 of pci0:0:10:0
> pcib0: matched entry for 0.10.INTA (src \_SB_.PCI0.AUBA:0)
> pci_link26: Picked IRQ 20 with weight 0
> panic: acpi_pci_link_srs_from_crs: can't put non-ISA IRQ 20 in legacy IRQ resource type
> cpuid = 0
> KDB: enter: panic
> [ thread pid 0 tid 100000 ]
> Stopped at kdb_enter+0x3e: movq $0,kdb_why
> db> bt
> Tracing pid 0 tid 100000 td 0xffffffff814f9810
> kdb_enter() at kdb_enter+0x3e/frame 0xffffffff818344a0
> vpanic() at vpanic+0x146/frame 0xffffffff818344e0
> kassert_panic() at kassert_panic+0x136/frame 0xffffffff81834550
> acpi_pci_link_route_irqs() at acpi_pci_link_route_irqs+0x2b6/frame 0xffffffff81834600
> acpi_pci_link_route_interrupt() at acpi_pci_link_route_interrupt+0x422/frame 0xffffffff818346a0
> acpi_pcib_route_interrupt() at acpi_pcib_route_interrupt+0x4cd/frame 0xffffffff81834700
> pci_assign_interrupt() at pci_assign_interrupt+0xef/frame 0xffffffff81834790
> pci_add_resources() at pci_add_resources+0x653/frame 0xffffffff81834840
> pci_add_children() at pci_add_children+0x18a/frame 0xffffffff818348a0
> acpi_pci_attach() at acpi_pci_attach+0xd5/frame 0xffffffff818348f0
> device_attach() at device_attach+0x396/frame 0xffffffff81834940
> bus_generic_attach() at bus_generic_attach+0x4a/frame 0xffffffff81834960
> acpi_pcib_attach() at acpi_pcib_attach+0x24d/frame 0xffffffff818349b0
> acpi_pcib_acpi_attach() at acpi_pcib_acpi_attach+0x299/frame 0xffffffff81834a00
> device_attach() at device_attach+0x396/frame 0xffffffff81834a50
> bus_generic_attach() at bus_generic_attach+0x4a/frame 0xffffffff81834a70
> acpi_attach() at acpi_attach+0xdd6/frame 0xffffffff81834b30
> device_attach() at device_attach+0x396/frame 0xffffffff81834b80
> bus_generic_attach() at bus_generic_attach+0x4a/frame 0xffffffff81834ba0
> nexus_acpi_attach() at nexus_acpi_attach+0x76/frame 0xffffffff81834bd0
> device_attach() at device_attach+0x396/frame 0xffffffff81834c20
> bus_generic_new_pass() at bus_generic_new_pass+0x116/frame 0xffffffff81834c50
> bus_set_pass() at bus_set_pass+0x8f/frame 0xffffffff81834c80
> configure() at configure+0xa/frame 0xffffffff81834c90
> mi_startup() at mi_startup+0x118/frame 0xffffffff81834cb0
> btext() at btext+0x2c
>
> And here is an example of the ACPI error messages on 9.1-RELEASE:
>
> pcib0: allocated type 3 (0xdffff000-0xdfffffff) for rid 10 of pci0:0:10:0
> pcib0: matched entry for 0.10.INTA (src \_SB_.PCI0.AUBA:0)
> pci_link26: Picked IRQ 20 with weight 0
> ACPI Error: Field [INZ6] at 56 exceeds Buffer [NULL] size 48 (bits) (20110527/ds
> opcode-254)
> ACPI Error: Method parse/execution failed [\_SB_.PCI0.SRSA] (Node 0xfffffe0002d9
> 8b00), AE_AML_BUFFER_LIMIT (20110527/psparse-560)
> ACPI Error: Method parse/execution failed [\_SB_.PCI0.AUBA._SRS] (Node 0xfffffe0
> 002d9d400), AE_AML_BUFFER_LIMIT (20110527/psparse-560)
> pci_link26: Unable to route IRQs: AE_AML_BUFFER_LIMIT
>
> Even though 9.1-RELEASE boots successfully, devices such as the ehci USB
> controller and SATA controller do not work.
Ugh, your BIOS does unexpected things. It uses a _CRS for these pci link
devices that uses a "short" IRQ resource, but uses an extended IRQ resource in
_PRS (and expects an extended one in _SRS). We use _CRS as a template for the
resource to build.
Try this patch. It's a bit hackish, but it forces us to not use _CRS as a
template if _CRS uses a "short" IRQ resource, but the link supports non-ISA
IRQs.
Index: /home/jhb/work/freebsd/svn/head/sys/dev/acpica/acpi_pci_link.c
===================================================================
--- /home/jhb/work/freebsd/svn/head/sys/dev/acpica/acpi_pci_link.c (revision 249553)
+++ /home/jhb/work/freebsd/svn/head/sys/dev/acpica/acpi_pci_link.c (working copy)
@@ -99,6 +99,7 @@
uint8_t l_bios_irq;
uint8_t l_irq;
uint8_t l_initial_irq;
+ UINT32 l_crs_type;
int l_res_index;
int l_num_irqs;
int *l_irqs;
@@ -236,6 +237,7 @@
("%s: array boundary violation", __func__));
link = &req->sc->pl_links[req->link_index];
link->l_res_index = req->res_index;
+ link->l_crs_type = res->Type;
req->link_index++;
req->res_index++;
@@ -364,6 +366,14 @@
link->l_isa_irq = FALSE;
}
}
+
+ /*
+ * If this is not an ISA IRQ but _CRS used a non-extended
+ * IRQ descriptor, don't use _CRS as a template for _SRS.
+ */
+ if (!req->sc->pl_crs_bad && !link->l_isa_irq &&
+ link->l_crs_type == ACPI_RESOURCE_TYPE_IRQ)
+ req->sc->pl_crs_bad = TRUE;
break;
default:
if (req->in_dpf == DPF_IGNORE)
--
John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201304191131.49433.jhb>