Re: [PATCH v2 2/3] tools: build qemu-vmsr-helper

[Daniel P . Berrangé]

On Tue, Oct 31, 2023 at 03:46:02PM +0100, Anthony Harivel wrote:
> Introduce a privileged helper to access RAPL MSR.
> 
> The privileged helper tool, qemu-vmsr-helper, is designed to provide
> virtual machines with the ability to read specific RAPL (Running Average
> Power Limit) MSRs without requiring CAP_SYS_RAWIO privileges or relying
> on external, out-of-tree patches.
> 
> The helper tool leverages Unix permissions and SO_PEERCRED socket
> options to enforce access control, ensuring that only processes
> explicitly requesting read access via readmsr() from a valid Thread ID
> can access these MSRs.
> 
> The list of RAPL MSRs that are allowed to be read by the helper tool is
> defined in rapl-msr-index.h. This list corresponds to the RAPL MSRs that
> will be supported in the next commit titled "Add support for RAPL MSRs
> in KVM/QEMU."
> 
> Signed-off-by: Anthony Harivel <aharivel@redhat.com>
> ---
>  docs/tools/index.rst            |   1 +
>  docs/tools/qemu-vmsr-helper.rst |  89 ++++++
>  meson.build                     |   5 +
>  tools/i386/qemu-vmsr-helper.c   | 507 ++++++++++++++++++++++++++++++++
>  tools/i386/rapl-msr-index.h     |  28 ++
>  5 files changed, 630 insertions(+)
>  create mode 100644 docs/tools/qemu-vmsr-helper.rst
>  create mode 100644 tools/i386/qemu-vmsr-helper.c
>  create mode 100644 tools/i386/rapl-msr-index.h
> 


> +/*
> + * Check if the TID that request the MSR read
> + * belongs to the peer. It should a TID of a vCPU.
> + */
> +static bool is_tid_present(pid_t pid, pid_t tid)
> +{
> +    char pidStr[20];
> +    char tidStr[20];
> +
> +    snprintf(pidStr, sizeof(pidStr), "%d", pid);
> +    snprintf(tidStr, sizeof(tidStr), "%d", tid);
> +
> +    char pidPath[256];
> +    char tidPath[256];
> +
> +    snprintf(pidPath, sizeof(pidPath), "/proc/%s", pidStr);
> +    snprintf(tidPath, sizeof(tidPath), "/proc/%s/task/%s", pidStr, tidStr);
> +
> +    /* Check if the TID directory exists within the PID directory */
> +    if (access(tidPath, F_OK) == 0) {
> +        return true;
> +    }
> +
> +    return false;
> +}
> +
> +/*
> + * Only the RAPL MSR in target/i386/cpu.h are allowed
> + */
> +static bool is_msr_allowed(uint32_t reg)
> +{
> +    switch (reg) {
> +    case MSR_RAPL_POWER_UNIT:
> +    case MSR_PKG_POWER_LIMIT:
> +    case MSR_PKG_ENERGY_STATUS:
> +    case MSR_PKG_POWER_INFO:
> +        return true;
> +    default:
> +        return false;
> +    }
> +}

> +    /*
> +     * Check peer credentials
> +     * Only QEMU PID/TID are allowed

This says only QEMU is allowed

> +     */
> +    qio_channel_get_peercred(QIO_CHANNEL(client->ioc), &cred, &local_err);
> +
> +    if (cred.pid == 0) {
> +        if (local_err != NULL) {
> +            error_report_err(local_err);
> +        }
> +        error_report("Failed to get peer credentials");
> +        goto out;
> +    }
> +
> +    /*
> +     * Read the requested MSR
> +     * Only RAPL MSR in rapl-msr-index.h is allowed
> +     */
> +    r = qio_channel_read_all(QIO_CHANNEL(client->ioc),
> +                             (char *) &request, sizeof(request), NULL);
> +    if (!is_msr_allowed(request[0]) || r < 0) {
> +        error_report("Read request fail: %d, %d", request[0], request[1]);
> +        goto out;
> +    }
> +
> +    vmsr = vmsr_read_msr(request[0], request[1]);
> +
> +    if (!is_tid_present(cred.pid, request[2])) {
> +        error_report("requested TID not in peer PID");
> +        vmsr = 0;
> +    }

This check is merely validating the the thread ID in the message
is a child of the process ID connected to the socket. Any process
on the entire host can satisfy this requirement.

I don't see what is limiting this to only QEMU as claimed earlier,
unless you're expecting the UNIX socket permissions to be such
that only processes under the qemu:qemu user:group pair can
access to the socket ? That would be a libvirt based permissions
assumption though.



> +
> +    r = qio_channel_write_all(QIO_CHANNEL(client->ioc),
> +                         (char *) &vmsr, sizeof(vmsr), NULL);
> +    if (r < 0) {
> +        error_report("write vmsr failed");
> +        goto out;
> +    }
> +
> +out:
> +    object_unref(OBJECT(client->ioc));
> +    g_free(client);
> +}

With regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|