]> git.ipfire.org Git - thirdparty/kernel/linux.git/commitdiff
x86/hyperv: Update comment in hyperv_cleanup()
authorMichael Kelley <mhklinux@outlook.com>
Mon, 2 Feb 2026 16:48:39 +0000 (08:48 -0800)
committerWei Liu <wei.liu@kernel.org>
Wed, 4 Feb 2026 06:50:00 +0000 (06:50 +0000)
The comment in hyperv_cleanup() became out-of-date as a result of
commit c8ed0812646e ("x86/hyperv: Use direct call to hypercall-page").

Update the comment. No code or functional change.

Signed-off-by: Michael Kelley <mhklinux@outlook.com>
Signed-off-by: Wei Liu <wei.liu@kernel.org>
arch/x86/hyperv/hv_init.c

index 6b00842ddda8fb14241e9d3e4cd9e85ec77e5092..5dbe9bd67891dc3db938a48d48cac59d7f993c1f 100644 (file)
@@ -635,9 +635,13 @@ void hyperv_cleanup(void)
        hv_ivm_msr_write(HV_X64_MSR_GUEST_OS_ID, 0);
 
        /*
-        * Reset hypercall page reference before reset the page,
-        * let hypercall operations fail safely rather than
-        * panic the kernel for using invalid hypercall page
+        * Reset hv_hypercall_pg before resetting it in the hypervisor.
+        * hv_set_hypercall_pg(NULL) is not used because at this point in the
+        * panic path other CPUs have been stopped, causing static_call_update()
+        * to hang. So resetting hv_hypercall_pg to cause hypercalls to fail
+        * cleanly is only operative on 32-bit builds. But this is OK as it is
+        * just a preventative measure to ease detecting a hypercall being made
+        * after this point, which shouldn't be happening anyway.
         */
        hv_hypercall_pg = NULL;