Task #1127
openCheck that ETV shows skipped function usb_free_urb for af9005-fe.ko
0%
Description
Creating this task just to check that this issue will not coccur on the new version of the toolset.
The function usb_free_urb is skipped by fdepth option, but Error Trace Visualizer does not show it and misinterprets it as function having body
http://falcon.intra.ispras.ru:8999/stats/errortrace/name/rdb/user/user/host/localhost/password/passwd/page/Error+trace/value/18143/Task+id/4/Task+description/linux-2.6.39-devel/Environment+version/artificial_env_1/Rule+name/68_1/Module/af9005-fe.ko/Entry+point/ldv_main0_sequence_infinite_withcheck_stateful
pax attached
raw trace file:artificial_env_1/csd_deg_dscv/6/dscv_tempdir/dscv/cmdfiles/rcv/68_1/reports/af9005-fe.ko.ldv_main0_sequence_infinite_withcheck_stateful.trace contains the message:
FunctionCall(usb_free_urb(urb@usb_start_wait_urb))
LDV: skipping call to function due to stack overflow: usb_free_urb
This is the old message format, so this issue should not occur on the new toolset version.
Files
Updated by Alexey Khoroshilov over 13 years ago
I would say a problem to check is if the usb_free_urb() handled as a model function and is not truncated due to stack overflow.
Updated by Vadim Mutilin over 13 years ago
BLAST does not know that usb_free_urb() is a model function, because it does not have ldv_ prefix, so it may be truncated.
Updated by Alexey Khoroshilov over 13 years ago
Вадим Мутилин wrote:
BLAST does not know that usb_free_urb() is a model function, because it does not have ldv_ prefix, so it may be truncated.
It is the main source of the issue.