[DPDK/ethdev Bug 1580] Nthw: array overrun
bugzilla at dpdk.org
bugzilla at dpdk.org
Fri Nov 15 05:10:51 CET 2024
https://bugs.dpdk.org/show_bug.cgi?id=1580
Bug ID: 1580
Summary: Nthw: array overrun
Product: DPDK
Version: 24.11
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
Priority: Normal
Component: ethdev
Assignee: dev at dpdk.org
Reporter: stephen at networkplumber.org
Target Milestone: ---
In the article:
https://pvs-studio.com/en/blog/posts/cpp/1183/
Fix is not obvious, either offset the bus_type_id or change checks.
Bug 35: strange check and possible array overrun
static const char *const sa_nthw_fpga_bus_type_str[] = {
"ERR", /* NTHW_FPGA_BUS_TYPE_UNKNOWN, */
"BAR", /* NTHW_FPGA_BUS_TYPE_BAR, */
"PCI", /* NTHW_FPGA_BUS_TYPE_PCI, */
"CCIP", /* NTHW_FPGA_BUS_TYPE_CCIP, */
"RAB0", /* NTHW_FPGA_BUS_TYPE_RAB0, */
"RAB1", /* NTHW_FPGA_BUS_TYPE_RAB1, */
"RAB2", /* NTHW_FPGA_BUS_TYPE_RAB2, */
"NMB", /* NTHW_FPGA_BUS_TYPE_NMB, */
"NDM", /* NTHW_FPGA_BUS_TYPE_NDM, */
};
static const char *get_bus_name(int n_bus_type_id)
{
if (n_bus_type_id >= 1 &&
n_bus_type_id <= (int)ARRAY_SIZE(sa_nthw_fpga_bus_type_str))
return sa_nthw_fpga_bus_type_str[n_bus_type_id];
else
return "ERR";
}
PVS-Studio warnings:
V557 Array overrun is possible. The value of 'n_bus_type_id' index could reach
9. nthw_fpga_model.c 32
The n_bus_type_id index is checked before extracting a row from an array. There
are two questions to this check:
Why is an index starting with 1 considered valid?
Why is the right boundary checked using the <= operator? If the index is equal
to the number of elements in the array, an Off-by-one Error will occur.
I would venture to guess that the ID values in the n_bus_type_id variable start
with 1. This way, the mistake is that one forgot to subtract 1 before
extracting an element from the array. In this case, the correct code will look
like this:
static const char *get_bus_name(int n_bus_type_id)
{
if (n_bus_type_id >= 1 &&
n_bus_type_id <= (int)ARRAY_SIZE(sa_nthw_fpga_bus_type_str))
return sa_nthw_fpga_bus_type_str[n_bus_type_id - 1];
else
return "ERR";
}
I'm not sure, though. It's strange that no one noticed that the function
returns the wrong lines. Perhaps the indexes are numbered from 0 after all.
Then the check should be rewritten:
static const char *get_bus_name(int n_bus_type_id)
{
if (n_bus_type_id >= 0 &&
n_bus_type_id < (int)ARRAY_SIZE(sa_nthw_fpga_bus_type_str))
return sa_nthw_fpga_bus_type_str[n_bus_type_id];
else
return "ERR";
}
Please forgive my uncertainty. It's the first time when I see this code. The
code is obviously incorrect, but unfortunately, I am limited in time to study
each found error in more depth. There are dozens of them, and one of me.
--
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mails.dpdk.org/archives/dev/attachments/20241115/5dae0010/attachment.htm>
More information about the dev
mailing list