Hi Andrew,
On 12/02/2021 15:39, Andrew Cooper wrote:
Various version of gcc, when compiling with -Og, complain:
xg_dom_arm.c: In function 'meminit':
xg_dom_arm.c:420:19: error: 'p2m_size' may be used uninitialized in this
function [-Werror=maybe-uninitialized]
420 | dom->p2m_size = p2m_size;
| ~~~~~~~~~~~~~~^~~~~~~~~~
Signed-off-by: Andrew Cooper <[email protected]>
This was reported nearly 3 years ago (see [1]) and it is pretty sad this
was never merged :(.
---
CC: Ian Jackson <[email protected]>
CC: Wei Liu <[email protected]>
CC: Stefano Stabellini <[email protected]>
CC: Julien Grall <[email protected]>
Julien/Stefano: I can't work out how this variable is supposed to work, and
the fact that it isn't a straight accumulation across the RAM banks looks
suspect.
It looks buggy, but the P2M is never used on Arm. In fact, you sent a
patch a year ago to drop it (see [2]). It would be nice to revive it.
---
tools/libs/guest/xg_dom_arm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/libs/guest/xg_dom_arm.c b/tools/libs/guest/xg_dom_arm.c
index 94948d2b20..f1b8d06f75 100644
--- a/tools/libs/guest/xg_dom_arm.c
+++ b/tools/libs/guest/xg_dom_arm.c
@@ -373,7 +373,7 @@ static int meminit(struct xc_dom_image *dom)
const uint64_t modsize = dtb_size + ramdisk_size;
const uint64_t ram128mb = bankbase[0] + (128<<20);
- xen_pfn_t p2m_size;
+ xen_pfn_t p2m_size = 0;
uint64_t bank0end;
assert(dom->rambase_pfn << XC_PAGE_SHIFT == bankbase[0]);
If your original series is too risky for 4.15, I would consider to
remote p2m_size completely and always 0 dom->p2m_size.
Cheers,
[1]
https://lore.kernel.org/xen-devel/[email protected]/
[2]
https://patchwork.kernel.org/project/xen-devel/patch/[email protected]/
--
Julien Grall
