matthias
/
iproute2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
5,527 Commits 5 Branches 108 Tags 7.5 MiB
C 91.1%
C++ 5.9%
Shell 2.1%
Makefile 0.5%
Yacc 0.2%
Go to file
Jacob Keller 954a0077c8 devlink: fix infinite loop on flash update for drivers without status 
When processing device flash update, cmd_dev_flash function waits until
the flash process has completed. This requires the following two
conditions to both be true:

a) we've received an exit status from the child process
b) we've received the DEVLINK_CMD_FLASH_UPDATE_END *or*
   we haven't received any status notifications from the driver.

The original devlink flash status monitoring code in 9b13cddfe2
("devlink: implement flash status monitoring") was written assuming that
a driver will either send no status updates, or it will send at least
one DEVLINK_CMD_FLASH_UPDATE_STATUS before DEVLINK_CMD_FLASH_UPDATE_END.

Newer versions of the kernel since commit 52cc5f3a166a ("devlink: move flash
end and begin to core devlink") in v5.10 moved handling of the
DEVLINK_CMD_FLASH_UPDATE_END into the core stack, and will send this
regardless of whether or not the driver sends any of its own status
notifications.

The handling of DEVLINK_CMD_FLASH_UPDATE_END in cmd_dev_flash_status_cb
has an additional condition that it must not be the first message.
Otherwise, it falls back to treating it like
a DEVLINK_CMD_FLASH_UPDATE_STATUS.

This is wrong because it can lead to an infinite loop if a driver does
not send any status updates.

In this case, the kernel will send DEVLINK_CMD_FLASH_UPDATE_END without
any DEVLINK_CMD_FLASH_UPDATE_STATUS. The devlink application will see
that ctx->not_first is false, and will treat this like any other status
message. Thus, ctx->not_first will be set to 1.

The loop condition to exit flash update will thus never be true, since
we will wait forever, because ctx->not_first is true, and
ctx->received_end is false.

This leads to the application appearing to process the flash update, but
it will never exit.

Fix this by simply always treating DEVLINK_CMD_FLASH_UPDATE_END the same
regardless of whether its the first message or not.

This is obviously the correct thing to do: once we've received the
DEVLINK_CMD_FLASH_UPDATE_END the flash update must be finished. For new
kernels this is always true, because we send this message in the core
stack after the driver flash update routine finishes.

For older kernels, some drivers may not have sent any
DEVLINK_CMD_FLASH_UPDATE_STATUS or DEVLINK_CMD_FLASH_UPDATE_END. This is
handled by the while loop conditional that exits if we get a return
value from the child process without having received any status
notifications.

An argument could be made that we should exit immediately when we get
either the DEVLINK_CMD_FLASH_UPDATE_END or an exit code from the child
process. However, at a minimum it makes no sense to ever process
DEVLINK_CMD_FLASH_UPDATE_END as if it were a DEVLINK_CMD_FLASH_UPDATE_STATUS.

This is easy to test as it is triggered by the selftests for the
netdevsim driver, which has a test case for both with and without status
notifications.

Fixes: 9b13cddfe2 ("devlink: implement flash status monitoring")
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
 		2021-08-10 19:54:39 -07:00
bash-completion devlink: add support for port params get/set 2021-02-11 09:21:24 -07:00
bridge bridge: vlan: dump port only if there are any vlans 2021-04-26 02:32:46 +00:00
dcb dcb: fix memory leak 2021-05-06 14:48:02 +00:00
devlink devlink: fix infinite loop on flash update for drivers without status 2021-08-10 19:54:39 -07:00
doc/actions doc: remove trailing whitespace 2018-12-19 12:02:38 -08:00
etc/iproute2 Add Open/R to rt_protos 2021-03-30 03:04:09 +00:00
examples/bpf examples/bpf: add bpf examples with BTF defined maps 2020-11-24 22:14:08 -07:00
genl genl: ctrl: print op -> policy idx mapping 2020-10-11 20:10:09 -06:00
include uapi: headers update 2021-07-17 11:12:47 -07:00
ip ip route: ignore ENOENT during save if RT_TABLE_MAIN is being dumped 2021-07-07 07:32:56 -07:00
lib lib/bpf: Fix btf_load error lead to enable debug log 2021-08-10 19:53:54 -07:00
man tc/skbmod: Remove misinformation about the swap action 2021-07-22 15:14:29 -07:00
misc Merge git://git.kernel.org/pub/scm/network/iproute2/iproute2-next 2021-02-23 23:03:42 -08:00
netem doc: remove trailing whitespace 2018-12-19 12:02:38 -08:00
rdma rdma: Add copy-on-fork to get sys command 2021-05-03 14:43:13 +00:00
schema bridge: add json schema for bridge fdb show 2016-07-20 12:02:02 -07:00
tc tc/skbmod: Remove misinformation about the swap action 2021-07-22 15:14:29 -07:00
testsuite testsuite: Add mpls packet matching tests for tc flower 2020-12-16 04:14:26 +00:00
tipc tipc: bail out if key is abnormally long 2021-05-09 22:08:47 +00:00
vdpa vdpa: add .gitignore 2021-02-23 23:12:14 -08:00
.clang-format clang-format: add configuration file 2019-01-17 13:38:23 -08:00
.gitignore Remove leftovers from removed Latex documentation 2018-02-13 16:43:19 -08:00
.mailmap mailmap: map David's mail address 2019-05-15 09:50:42 -07:00
COPYING Update address of FSF in license 2008-03-08 13:31:03 -08:00
Makefile config.mk: Rerun configure when it is newer than config.mk 2021-05-17 02:13:56 +00:00
README README: remove doc instructions 2021-03-20 09:29:02 -07:00
README.devel README: update location of git repositories, remove broken info link 2018-02-13 16:42:51 -08:00
configure configure: convert LIBBPF environment variables to command-line options 2021-06-03 03:25:59 +00:00

README 

This is a set of utilities for Linux networking.

Information:
    https://wiki.linuxfoundation.org/networking/iproute2

Download:
    http://www.kernel.org/pub/linux/utils/net/iproute2/

Stable version repository:
    git://git.kernel.org/pub/scm/network/iproute2/iproute2.git

Development repository:
    git://git.kernel.org/pub/scm/network/iproute2/iproute2-next.git

How to compile this.
--------------------
1. libdbm

arpd needs to have the berkeleydb development libraries. For Debian
users this is the package with a name like libdbX.X-dev.
DBM_INCLUDE points to the directory with db_185.h which
is the include file used by arpd to get to the old format Berkeley
database routines.  Often this is in the db-devel package.

2. make

The makefile will automatically build a config.mk file which
contains definitions of libraries that may or may not be available
on the system such as: ATM, ELF, MNL, and SELINUX.

3. include/uapi

This package includes matching sanitized kernel headers because
the build environment may not have up to date versions. See Makefile
if you have special requirements and need to point at different
kernel include files.

Stephen Hemminger
stephen@networkplumber.org

Alexey Kuznetsov
kuznet@ms2.inr.ac.ru