clang-format: re-format for clang-18

clang-format-18 isn't compatible with the clang-format-17 output, so we
need to reformat the code with the latest version. The way clang-18
handles lambda forma

clang-format: re-format for clang-18

clang-format-18 isn't compatible with the clang-format-17 output, so we
need to reformat the code with the latest version. The way clang-18
handles lambda formatting also changed, so we have made changes to the
organization default style format to better handle lambda formatting.

See I5e08687e696dd240402a2780158664b7113def0e for updated style.
See Iea0776aaa7edd483fa395e23de25ebf5a6288f71 for clang-18 enablement.

Change-Id: I01547e98d27910919e09ebf7907c86292a6c825d
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>

show more ...

user_channel: switch to lg2

Signed-off-by: George Liu <liuxiwei@ieisystem.com>
Change-Id: I80149836f196f01eb09cc874f74eec0d787b5ac5

Fix the bugs found in static analysis

This commit fixes the following static analyzer reported issues:

Operands don't affect result
some conditions are not required to check as its always true

Fix the bugs found in static analysis

This commit fixes the following static analyzer reported issues:

Operands don't affect result
some conditions are not required to check as its always true
Unsigned compared against 0
Unchecked return value from library
Uninitialized scalar variable

Change-Id: I0b1fd426794bb88f6eafcc817cef5dd2f655e1ba
Signed-off-by: PavanKumarIntel <pavanx.kumar.martha@intel.com>

show more ...

clang-format: copy latest and re-format

clang-format-16 has some backwards incompatible changes that require
additional settings for best compatibility and re-running the formatter.
Copy the latest

clang-format: copy latest and re-format

clang-format-16 has some backwards incompatible changes that require
additional settings for best compatibility and re-running the formatter.
Copy the latest .clang-format from the docs repository and reformat the
repository.

Change-Id: I44441096113929ce96eb1439e2932e6ff3c87f27
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>

show more ...

Revert "user_mgmt: Fix SEGV on 64bit"

This reverts commit 05703adfe2bf70945d2befa8a7beda4ce5e5182a.

Due to we found that root cause about SEGV on 64bit and after modify
in pam-ipmit then SEGV on 64

Revert "user_mgmt: Fix SEGV on 64bit"

This reverts commit 05703adfe2bf70945d2befa8a7beda4ce5e5182a.

Due to we found that root cause about SEGV on 64bit and after modify
in pam-ipmit then SEGV on 64bit will not happen anymore.

Thus, we submitted this commit in gerrit in pam-ipmi as below:
https://gerrit.openbmc.org/c/openbmc/pam-ipmi/+/58579

Root cause:
There is default ipmi_pass file be created and encrypted in 32bit
environment from pam-ipmi original design before.
According the failed message, we found that PAM_AUTHTOK_ERR return
from update_pass_special_file() function that cause ipmid got bad
authentication token. Seems this file cannot be used in 64bit
environment by default that will cause ipmitool user set password
got failed.

Solution:
According current pam-ipmi design that already consider if this file
does not exist then BMC will create it once a user in the ipmi group
has been added. There should not be any file there by default.
Thus, there is no need for this file.
Without install this file by default then ipmitool command successful.

Thus, this default ipmi_pass file that will cause SEGV in host-ipmid on
64 bit project. After remove this default file, SEGV crash issue in
host-ipmid symptom was gone.

BTW, there is another issue we found that
“ipmitool user set password 1 0penBmc0” command failed.
Due to pam-ipmi still use size_t structure, but host-ipmid already
change to use uint32_t structure for MetaPassStruct.

Thus, we need to revert your commit and remove ipmi_pass file in
pam-ipmi for fixing two issues.

Verified:
root@evb-npcm845:~# ipmitool user set password 1 0penBmc0
Set User Password command successful (user 1)

Signed-off-by: Tim Lee <timlee660101@gmail.com>
Change-Id: I03ba57a9006694a09b75c16d897377b1b57051e5

show more ...

user_mgmt: Fix SEGV on 64bit

The password management code takes a file off disk and interprets it
as a C struct. This is neither endian safe, 32/64 bit safe nor safe
against C struct alignment rules

user_mgmt: Fix SEGV on 64bit

The password management code takes a file off disk and interprets it
as a C struct. This is neither endian safe, 32/64 bit safe nor safe
against C struct alignment rules.

Fix the SEGV on 64bit by switching from size_t to uint32_t.

Signed-off-by: Anton Blanchard <anton@ozlabs.org>
Change-Id: Ibabd947284441ec141a9d2d3800f1ae95a0a7906

show more ...

Use SecureString where there is data to be cleansed

SecureString has quite a few places it should be used in the user
management code.

Tested: ran set password, test password, a

Use SecureString where there is data to be cleansed

SecureString has quite a few places it should be used in the user
management code.

Tested: ran set password, test password, and other commands

Change-Id: Ia53bc914d25f7965c3e72d5cf18346e0fa9339b9
Signed-off-by: Vernon Mauery <vernon.mauery@linux.intel.com>

show more ...

NULL check before pointer dereference

The function EVP_CIPHER_CTX_set_padding() is called with ctx.get()
before checking ctx for NULL. The NULL check is performed after the
function.

NULL check before pointer dereference

The function EVP_CIPHER_CTX_set_padding() is called with ctx.get()
before checking ctx for NULL. The NULL check is performed after the
function.
This commit moves the NULL check before Function call.

Signed-off-by: P Dheeraj Srujan Kumar <p.dheeraj.srujan.kumar@intel.com>
Change-Id: If8f3b78d2e33c216adac9da54bdc22079435aad4

show more ...

Fix Kloxwork Issue : NULL check after new

A unique pointer of EVP_CIPHER_CTX is created using EVP_CIPHER_CTX_new()
which could return NULL.
This Commit adds a NULL check to avoid NUL

Fix Kloxwork Issue : NULL check after new

A unique pointer of EVP_CIPHER_CTX is created using EVP_CIPHER_CTX_new()
which could return NULL.
This Commit adds a NULL check to avoid NULL pointer dereferencing and
returns memory error.

Signed-off-by: P Dheeraj Srujan Kumar <p.dheeraj.srujan.kumar@intel.com>
Change-Id: Icbc599210034feb6f7b3c35542bcfff205987628

show more ...

shadowlock: disable until proper unit tests are run

This code as-is is not doing anything other then declaring a function.
The solution is to utilize {} to tell the compiler it's an obje

shadowlock: disable until proper unit tests are run

This code as-is is not doing anything other then declaring a function.
The solution is to utilize {} to tell the compiler it's an object
instantiation. Given that this code has been a noop up until this point,
it's best to just comment it out until someone can come in enable and
test it properly.

The latest yocto update which brings in a newer compiler is erroring out
on this so we need something quick to be able to make progress there.

openbmc/phosphor-host-ipmid#170 is tracking this issue.

Signed-off-by: Andrew Geissler <geissonator@yahoo.com>
Change-Id: Idd4acd8e8541195f1b6e73c854866281832f1878

show more ...

Clear security sensitive data

As password is sensitive data Clear after use.

Tested:
Verified using ipmitool commands
Command: ipmitool user set password 5 asdf1234 //Set us

Clear security sensitive data

As password is sensitive data Clear after use.

Tested:
Verified using ipmitool commands
Command: ipmitool user set password 5 asdf1234 //Set user password
Response: Set User Password command successful (user 5)
Command: ipmitool raw 6 0x47 5 2 0x30 0x70 0x65 0x6e 0x42 0x6d 0x63
0x31 0 0 0 0 0 0 0 0 //set user password - set password
Response: //Success
Command: ipmitool raw 6 0x47 5 3 0x30 0x70 0x65 0x6e 0x42 0x6d 0x63
0x31 0 0 0 0 0 0 0 0 //set user password - test password
Response: //Success

Signed-off-by: Jayaprakash Mutyala <mutyalax.jayaprakash@intel.com>
Change-Id: I06196233ac5468534bd10fd34f99f7d35fd7b971

show more ...

[Fix]: Restrict password file permissions to 600

pam-ipmi is already updated restrict file permission of /etc/key_file
& /etc/ipmi_pass to 600 during creation. But this won't affect when

[Fix]: Restrict password file permissions to 600

pam-ipmi is already updated restrict file permission of /etc/key_file
& /etc/ipmi_pass to 600 during creation. But this won't affect when firmware
is updated with nv section preserved or without user credentials getting
changed. This commit will check the file permission on every boot and update
both the files to 600.

Tested:
1. Verified that RMCP+ connection still works after this change
2. Manually set file permission to 777 and restrating ipmid or BMC
will fix the same.

Change-Id: Icfe8af5af918792412fb42e8114fcf859848e1a8
Signed-off-by: Richard Marian Thomaiyar <richard.marian.thomaiyar@linux.intel.com>

show more ...

Use more restrictive permissions on /etc/ipmi-pass

This forces the permissions on /etc/ipmi-pass to be 0600 or RW only by
owner. This is to prevent non-owners from reading the file, even

Use more restrictive permissions on /etc/ipmi-pass

This forces the permissions on /etc/ipmi-pass to be 0600 or RW only by
owner. This is to prevent non-owners from reading the file, even though
it is obfuscated to make it harder for ipmi passwords to leak.

Tested: change ipmi passwords and see that the /etc/ipmi-pass file has
0600 permissions.

Change-Id: I4be0b8a65f98ced031493f7767879eb054e1ee84
Signed-off-by: Vernon Mauery <vernon.mauery@linux.intel.com>

show more ...

user_channel: passwd_mgr: minor cleanup from cppcheck

[user_channel/passwd_mgr.cpp:220]: (style) The scope of the
variable 'userEPos' can be reduced.
[user_channel/passwd_mgr.cpp:358

user_channel: passwd_mgr: minor cleanup from cppcheck

[user_channel/passwd_mgr.cpp:220]: (style) The scope of the
variable 'userEPos' can be reduced.
[user_channel/passwd_mgr.cpp:358]: (style) The scope of the
variable 'userEPos' can be reduced.
[user_channel/passwd_mgr.cpp:444]: (style) Variable 'fd' is
assigned a value that is never used.

Tested: Not tested.
Signed-off-by: Patrick Venture <venture@google.com>
Change-Id: If02125c422ab1e9c15c6420dbf8da6a52fb68541

show more ...

Removal of excess 16 bytes padding

Cipher block padding is applied twice(once manually, and
once through EVP_CipherFinal()) causing excess 16 bytes of
data. Manual padding is removed

Removal of excess 16 bytes padding

Cipher block padding is applied twice(once manually, and
once through EVP_CipherFinal()) causing excess 16 bytes of
data. Manual padding is removed to fix the issue.

Unit-Test:
1. Verified that files are padded correctly without
excess 16 bytes. Also, verified the password update and
RMCP+ login using ipmitool
ipmitool user set password <userid> <password>
ipmitool -I lanplus -H 127.0.0.1 -U <userid> -P <password>
raw 6 1
2. Also verified renaming of user name and login
after that

Resolves openbmc/openbmc#3463

Change-Id: Idec14b100bf7558137ff14601eb590b50cdbd7bf
Signed-off-by: Richard Marian Thomaiyar <richard.marian.thomaiyar@linux.intel.com>

show more ...

Change struct names to UpperCamelCase

Change struct names to UpperCamelCase to align with
OpenBMC coding guidelines

Change-Id: Ib71768be45e57556a26b14b64adec98db3bc7762
Sign

Change struct names to UpperCamelCase

Change struct names to UpperCamelCase to align with
OpenBMC coding guidelines

Change-Id: Ib71768be45e57556a26b14b64adec98db3bc7762
Signed-off-by: Richard Marian Thomaiyar <richard.marian.thomaiyar@linux.intel.com>

show more ...

IPMI password entry cleanup

Whenever user is removed from the system, password map in IPMI
database has to be cleaned up. Similarly password entry must
be updated when user rename ha

IPMI password entry cleanup

Whenever user is removed from the system, password map in IPMI
database has to be cleaned up. Similarly password entry must
be updated when user rename happens. This commit add's API
changes for the same

Change-Id: I304c721b31fc7fbad019c85d8ca2ecc60ca398d8
Signed-off-by: Richard Marian Thomaiyar <richard.marian.thomaiyar@linux.intel.com>

show more ...

Handling delete password entry from ipmi-pass

API to Handle the delete password entry from
ipmi-pass encrypted file when user gets deleted
by any interface

Change-Id: I692a8

Handling delete password entry from ipmi-pass

API to Handle the delete password entry from
ipmi-pass encrypted file when user gets deleted
by any interface

Change-Id: I692a81b166b53d6fc981fdb85ce5d6980887560b
Signed-off-by: AppaRao Puli <apparao.puli@intel.com>
Signed-off-by: Richard Marian Thomaiyar <richard.marian.thomaiyar@linux.intel.com>

show more ...

Add first user layer support.

Exposes user layer library which will be used by both
apphandler and netipmid for RAKP. Password is decrypted
from the special file which is used by pam

Add first user layer support.

Exposes user layer library which will be used by both
apphandler and netipmid for RAKP. Password is decrypted
from the special file which is used by pam-ipmi.

Change-Id: Ia6b99d4359a03843f19cd0e18f7269ad2e3cdb44
Signed-off-by: Richard Marian Thomaiyar <richard.marian.thomaiyar@linux.intel.com>

show more ...