RE: [PATCH v2 1/1] x86/hyperv: Use Hyper-V entropy to seed guest random number generator

From: Michael Kelley
Date: Wed Mar 13 2024 - 01:29:21 EST


From: Long Li <longli@xxxxxxxxxxxxx> Sent: Tuesday, March 12, 2024 9:51 PM
>
> > +void __init ms_hyperv_late_init(void)
> > +{
> > + struct acpi_table_header *header;
> > + acpi_status status;
> > + u8 *randomdata;
> > + u32 length, i;
> > +
> > + /*
> > + * Seed the Linux random number generator with entropy provided by
> > + * the Hyper-V host in ACPI table OEM0. It would be nice to do this
> > + * even earlier in ms_hyperv_init_platform(), but the ACPI subsystem
> > + * isn't set up at that point. Skip if booted via EFI as generic EFI
> > + * code has already done some seeding using the EFI RNG protocol.
> > + */
> > + if (!IS_ENABLED(CONFIG_ACPI) || efi_enabled(EFI_BOOT))
> > + return;
> > +
> > + status = acpi_get_table("OEM0", 0, &header);
> > + if (ACPI_FAILURE(status) || !header)
> > + return;
>
> Should we call acpi_put_table() if header == 0?

No. acpi_get_table() setting header to NULL is equivalent to
returning a failure status, per a comment in the code for
acpi_get_table(). So checking header for NULL is probably
redundant, but it doesn't hurt.

> It will also be helpful doing a
> pr_info() here so user knows that hyper-v random number is not used.

In v1 of the patch, I had such a pr_info(), but Wei Liu recommended
removing it, and I agreed it wasn't really necessary. See the brief
discussion here:

https://lore.kernel.org/linux-hyperv/SN6PR02MB4157B61CA09C0DAF0BB994E1D4212@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/

Michael