While they should be unique, boot certainly shouldn't fail if the HEST
table isn't sane. The problem is that platform_device_release() wants
to kfree() ->dev.platform_data, which gets set to point into the
middle of some (allocated or mapped) enclosing table.
It appears safe to simply create a copy of the table (no data in there
should change post-boot); if it isn't, the pointer would need to be
cleared before dropping the device.
Signed-off-by: Jan Beulich <jbeulich@xxxxxxxxxx>
---
drivers/acpi/apei/hest.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- linux-2.6.36-rc3/drivers/acpi/apei/hest.c
+++ 2.6.36-rc3-acpi-hest-error-path/drivers/acpi/apei/hest.c
@@ -150,7 +150,9 @@ static int hest_parse_ghes(struct acpi_h
ghes_dev = platform_device_alloc("GHES", hest_hdr->source_id);
if (!ghes_dev)
return -ENOMEM;
- ghes_dev->dev.platform_data = generic;
+ rc = platform_device_add_data(ghes_dev, generic, sizeof(*generic));
+ if (rc)
+ goto err;
rc = platform_device_add(ghes_dev);
if (rc)
goto err;
--
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
