- there seem to be another problem in the acpica resource handling depths if _CRS functions do not return anything. Strictly speaking it's illegal for _CRS to not return anything. Practically, perhaps we should repair this case to a "NULL" resource descriptor which perhaps could be defined as a resource descriptor with a single End Tag descriptor. From: Thomas Renninger [mailto:trenn@xxxxxxx] Sent: Tuesday, March 01, 2011 8:44 AM To: Bjorn Helgaas Cc: lenb@xxxxxxxxxx; Moore, Robert; linux-acpi@xxxxxxxxxxxxxxx Subject: Re: [PATCH] pnpacpi: Better format error message On Tuesday, March 01, 2011 05:25:04 PM Bjorn Helgaas wrote: > On Tue, Mar 01, 2011 at 03:24:50PM +0100, Thomas Renninger wrote: > > On a system with empty _CRS method I see: > > pnp 00:0c: can't evaluate _CRS: 12311 > > which is 0x3017, which would mean: AE_AML_INVALID_RESOURCE_TYPE > > > > This patch would directly show: AE_AML_INVALID_RESOURCE_TYPE > > The patch seems fine to me, but I question whether we need these > printks at all. It seems pointless to have an empty _CRS method, > but it does seem legal, or at least debatable, so I'm not sure > that the message is telling us anything useful. Theoretically the callback function passed to: acpi_walk_resources(handle, METHOD_NAME__CRS should get called with an empty resource string or not at all in case _CRS can return nothing. But in this case acpi_walk_resources fails which is worth reporting. I could imagine other _CRS functions are not processed anymore when acpi_walk_resources bails out with AE_AML_INVALID_RESOURCE_TYPE on a specific _CRS function, don't know. Summary: - acpi_walk_resources failures should be reported (and this patch is to do that more nicely) - there seem to be another problem in the acpica resource handling depths if _CRS functions do not return anything. Thomas > > Bjorn > > > Signed-off-by: Thomas Renninger <trenn@xxxxxxx> > > CC: bjorn@xxxxxxxxxxx > > CC: lenb@xxxxxxxxxx > > CC: robert.moore@xxxxxxxxx > > > > --- > > Corrected email address of Bjorn. > > > > drivers/pnp/pnpacpi/rsparser.c | 12 ++++++++---- > > 1 file changed, 8 insertions(+), 4 deletions(-) > > > > Index: linux-2.6.37-master/drivers/pnp/pnpacpi/rsparser.c > > =================================================================== > > --- linux-2.6.37-master.orig/drivers/pnp/pnpacpi/rsparser.c > > +++ linux-2.6.37-master/drivers/pnp/pnpacpi/rsparser.c > > @@ -498,7 +498,8 @@ int pnpacpi_parse_allocated_resource(str > > > > if (ACPI_FAILURE(status)) { > > if (status != AE_NOT_FOUND) > > - dev_err(&dev->dev, "can't evaluate _CRS: %d", status); > > + dev_err(&dev->dev, "can't evaluate _CRS: %s", > > + acpi_format_exception(status)); > > return -EPERM; > > } > > return 0; > > @@ -809,7 +810,8 @@ int __init pnpacpi_parse_resource_option > > > > if (ACPI_FAILURE(status)) { > > if (status != AE_NOT_FOUND) > > - dev_err(&dev->dev, "can't evaluate _PRS: %d", status); > > + dev_err(&dev->dev, "can't evaluate _PRS: %s", > > + acpi_format_exception(status)); > > return -EPERM; > > } > > return 0; > > @@ -876,7 +878,8 @@ int pnpacpi_build_resource_template(stru > > status = acpi_walk_resources(handle, METHOD_NAME__CRS, > > pnpacpi_count_resources, &res_cnt); > > if (ACPI_FAILURE(status)) { > > - dev_err(&dev->dev, "can't evaluate _CRS: %d\n", status); > > + dev_err(&dev->dev, "can't evaluate _CRS: %s\n", > > + acpi_format_exception(status)); > > return -EINVAL; > > } > > if (!res_cnt) > > @@ -891,7 +894,8 @@ int pnpacpi_build_resource_template(stru > > pnpacpi_type_resources, &resource); > > if (ACPI_FAILURE(status)) { > > kfree(buffer->pointer); > > - dev_err(&dev->dev, "can't evaluate _CRS: %d\n", status); > > + dev_err(&dev->dev, "can't evaluate _CRS: %s\n", > > + acpi_format_exception(status)); > > return -EINVAL; > > } > > /* resource will pointer the end resource now */ > -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
