rpms
/
nvme-cli
7
0
Fork 0
Code Issues Pull Requests Releases Activity

Fix error propagation from ioctl() calls 

nvme-cli was not correctly propagating errors from ioctl() calls,
resulting in dereferencing bad pointers.

$ nvme telemetry-log /dev/nvme0 --output-file=telemetry_log.bin
ERROR: get_telemetry_log: : write failed with error : Bad address

Resolves: RHEL-36139

Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
This commit is contained in:
Maurizio Lombardi 2024-06-19 08:48:39 +02:00
parent 4905e06746
commit d44ad57fb0
2 changed files with 82 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

76
0002-nvme-telemetry-report-the-correct-error-if-the-ioctl.patch Normal file
View File

@ -0,0 +1,76 @@
From 42f788edccb954fe8b54c43809523e603c99f63c Mon Sep 17 00:00:00 2001
From: Maurizio Lombardi <mlombard@redhat.com>
Date: Wed, 22 May 2024 15:06:18 +0200
Subject: [PATCH] nvme: telemetry: report the correct error if the ioctl()
fails.
It's wrong to assume that if the ioctl() returns a non-zero number
then the errno variable is set.
The ioctl() might return an NVMe Status error to inform the caller
that the requested log page is not supported, in that case errno is left
untouched.
The original code didn't handle this case and returned "-errno" even when
the latter was zero. The caller interpreted this as a successful operation
and this might lead to improperly dereferencing the log page pointer.
$ nvme telemetry-log /dev/nvme0 --output-file=telemetry_log.bin
ERROR: get_telemetry_log: : write failed with error : Bad address
Fix this bug by returning the NVMe status if errno is zero:
$ nvme telemetry-log /dev/nvme0 --output-file=telemetry_log.bin
NVMe status: Invalid Log Page: The log page indicated is invalid(0x109)
Failed to acquire telemetry log 265!
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
---
nvme.c | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/nvme.c b/nvme.c
index d722b91e5ed8..226413bc0eda 100644
--- a/nvme.c
+++ b/nvme.c
@@ -717,7 +717,10 @@ static int get_log_telemetry_ctrl(struct nvme_dev *dev, bool rae, size_t size,
err = nvme_cli_get_log_telemetry_ctrl(dev, rae, 0, size, log);
if (err) {
free(log);
- return -errno;
+ if (errno)
+ return -errno;
+ else
+ return err;
}
*buf = log;
@@ -737,7 +740,10 @@ static int get_log_telemetry_host(struct nvme_dev *dev, size_t size,
err = nvme_cli_get_log_telemetry_host(dev, 0, size, log);
if (err) {
free(log);
- return -errno;
+ if (errno)
+ return -errno;
+ else
+ return err;
}
*buf = log;
@@ -757,8 +763,12 @@ static int __create_telemetry_log_host(struct nvme_dev *dev,
return -ENOMEM;
err = nvme_cli_get_log_create_telemetry_host(dev, log);
- if (err)
- return -errno;
+ if (err) {
+ if (errno)
+ return -errno;
+ else
+ return err;
+ }
err = parse_telemetry_da(dev, da, log, size);
if (err)
--
2.43.0

7
nvme-cli.spec
View File

@ -3,7 +3,7 @@
Name: nvme-cli
Version: 2.9.1
Release: 2%{?dist}
Release: 3%{?dist}
Summary: NVMe management command line interface
License: GPL-2.0-only
@ -12,6 +12,7 @@ Source0: %{url}/archive/v%{version}/%{name}-%{version}.tar.gz
Source1: nvmf-connect-nbft.sh
Patch0: 0001-Revert-fabrics-Use-corresponding-hostid-when-hostnqn.patch
Patch1: 0002-nvme-telemetry-report-the-correct-error-if-the-ioctl.patch
BuildRequires: meson >= 0.50.0
BuildRequires: gcc gcc-c++
@ -34,6 +35,7 @@ nvme-cli provides NVM-Express user space tooling for Linux.
%setup -q
%patch0 -p1
%patch1 -p1
%build
%meson -Dudevrulesdir=%{_udevrulesdir} -Dsystemddir=%{_unitdir} -Ddocs=all -Ddocs-build=true -Dhtmldir=%{_pkgdocdir}
@ -94,6 +96,9 @@ if [ $1 -eq 1 ] || [ $1 -eq 2 ]; then
fi
%changelog
* Wed June 19 2024 Maurizio Lombardi <mlombard@redhat.com> - 2.9.1-3
- Fix RHEL-36139
* Tue May 07 2024 Tomas Bzatek <tbzatek@redhat.com> - 2.9.1-2
- Install custom nvmf-connect-nbft.sh NetworkManager hook (RHEL-18912)