[PATCH v2 4/4] x86/shadow: Don't leave trace record field uninitialized

Andrew Cooper posted 4 patches 6 months ago
There is a newer version of this series
[PATCH v2 4/4] x86/shadow: Don't leave trace record field uninitialized
Posted by Andrew Cooper 6 months ago
From: Jan Beulich <jbeulich@suse.com>

The emulation_count field is set only conditionally right now. Convert
all field setting to an initializer, thus guaranteeing that field to be
set to 0 (default initialized) when GUEST_PAGING_LEVELS != 3.

Rework trace_shadow_emulate() to be consistent with the other trace helpers.

Coverity-ID: 1598430
Fixes: 9a86ac1aa3d2 ("xentrace 5/7: Additional tracing for the shadow code")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
---
CC: Jan Beulich <JBeulich@suse.com>
CC: Roger Pau Monné <roger.pau@citrix.com>
CC: George Dunlap <george.dunlap@citrix.com>

v2:
 * Rebase over packing/sh_trace() cleanup.
---
 xen/arch/x86/mm/shadow/multi.c | 31 +++++++++++++++----------------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/xen/arch/x86/mm/shadow/multi.c b/xen/arch/x86/mm/shadow/multi.c
index d2fe4e148fe0..47dd1cc626b2 100644
--- a/xen/arch/x86/mm/shadow/multi.c
+++ b/xen/arch/x86/mm/shadow/multi.c
@@ -2063,30 +2063,29 @@ static void cf_check trace_emulate_write_val(
 #endif
 }
 
-static inline void trace_shadow_emulate(guest_l1e_t gl1e, unsigned long va)
+static inline void sh_trace_emulate(guest_l1e_t gl1e, unsigned long va)
 {
     if ( tb_init_done )
     {
-        struct __packed {
-            /* for PAE, guest_l1e may be 64 while guest_va may be 32;
-               so put it first for alignment sake. */
+        struct {
+            /*
+             * For GUEST_PAGING_LEVELS=3 (PAE paging), guest_l1e is 64 while
+             * guest_va is 32.  Put it first to avoid padding.
+             */
             guest_l1e_t gl1e, write_val;
             guest_va_t va;
             uint32_t flags:29, emulation_count:3;
-        } d;
-        u32 event;
-
-        event = TRC_SHADOW_EMULATE | ((GUEST_PAGING_LEVELS-2)<<8);
-
-        d.gl1e = gl1e;
-        d.write_val.l1 = this_cpu(trace_emulate_write_val);
-        d.va = va;
+        } d = {
+            .gl1e = gl1e,
+            .write_val.l1 = this_cpu(trace_emulate_write_val),
+            .va = va,
 #if GUEST_PAGING_LEVELS == 3
-        d.emulation_count = this_cpu(trace_extra_emulation_count);
+            .emulation_count = this_cpu(trace_extra_emulation_count),
 #endif
-        d.flags = this_cpu(trace_shadow_path_flags);
+            .flags = this_cpu(trace_shadow_path_flags),
+        };
 
-        trace(event, sizeof(d), &d);
+        sh_trace(TRC_SHADOW_EMULATE, sizeof(d), &d);
     }
 }
 #endif /* CONFIG_HVM */
@@ -2815,7 +2814,7 @@ static int cf_check sh_page_fault(
     }
 #endif /* PAE guest */
 
-    trace_shadow_emulate(gw.l1e, va);
+    sh_trace_emulate(gw.l1e, va);
  emulate_done:
     SHADOW_PRINTK("emulated\n");
     return EXCRET_fault_fixed;
-- 
2.30.2


Re: [PATCH v2 4/4] x86/shadow: Don't leave trace record field uninitialized
Posted by Jan Beulich 6 months ago
On 22.05.2024 15:17, Andrew Cooper wrote:
> From: Jan Beulich <jbeulich@suse.com>
> 
> The emulation_count field is set only conditionally right now. Convert
> all field setting to an initializer, thus guaranteeing that field to be
> set to 0 (default initialized) when GUEST_PAGING_LEVELS != 3.
> 
> Rework trace_shadow_emulate() to be consistent with the other trace helpers.
> 
> Coverity-ID: 1598430
> Fixes: 9a86ac1aa3d2 ("xentrace 5/7: Additional tracing for the shadow code")
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
> Acked-by: Roger Pau Monné <roger.pau@citrix.com>

Your additional changes look pretty much independent of what my original
patch did. I don't mind the folding though, yet I think you need to add
your own S-o-b as well. Then in turn
Acked-by: Jan Beulich <jbeulich@suse.com>

Jan