On 23/10/2019 15.07, Konstantin Khlebnikov wrote:
commit 1ed95e52d902035e39a715ff3a314a893a96e5b7 upstream.
Commit d96d87834d5b870402a4a5b565706a4869ebc020 in v4.4.190 which is
backport of upstream commit 1ed95e52d902035e39a715ff3a314a893a96e5b7
removed only HPET access from vdso but leaved HPET mapped in "vvar".
So userspace still could read HPET directly and confuse hardware.
This patch removes mapping HPET page into userspace.
Fixes: d96d87834d5b ("x86/vdso: Remove direct HPET access through the vDSO") # v4.4.190
Signed-off-by: Konstantin Khlebnikov <khlebnikov@xxxxxxxxxxxxxx>
Link: https://lore.kernel.org/lkml/6fd42b2b-e29a-1fd6-03d1-e9da9192e6c5@xxxxxxxxxxxxxx/
---
arch/x86/entry/vdso/vma.c | 14 --------------
1 file changed, 14 deletions(-)
diff --git a/arch/x86/entry/vdso/vma.c b/arch/x86/entry/vdso/vma.c
index 6b46648588d8..cc0a3c16a95d 100644
--- a/arch/x86/entry/vdso/vma.c
+++ b/arch/x86/entry/vdso/vma.c
@@ -18,7 +18,6 @@
#include <asm/vdso.h>
#include <asm/vvar.h>
#include <asm/page.h>
-#include <asm/hpet.h>
#include <asm/desc.h>
#include <asm/cpufeature.h>
@@ -159,19 +158,6 @@ static int map_vdso(const struct vdso_image *image, bool calculate_addr)
if (ret)
goto up_fail;
-#ifdef CONFIG_HPET_TIMER
- if (hpet_address && image->sym_hpet_page) {
Probably this patch is not required.
It seems after removing symbol "hpet_page" from vdso code
image->sym_hpet_page always is NULL and this branch never executed.
- ret = io_remap_pfn_range(vma,
- text_start + image->sym_hpet_page,
- hpet_address >> PAGE_SHIFT,
- PAGE_SIZE,
- pgprot_noncached(PAGE_READONLY));
-
- if (ret)
- goto up_fail;
- }
-#endif
-
pvti = pvclock_pvti_cpu0_va();
if (pvti && image->sym_pvclock_page) {
ret = remap_pfn_range(vma,
