summaryrefslogtreecommitdiffstats
path: root/meta/recipes-devtools/qemu/qemu/mmap2.patch
diff options
context:
space:
mode:
Diffstat (limited to 'meta/recipes-devtools/qemu/qemu/mmap2.patch')
-rw-r--r--meta/recipes-devtools/qemu/qemu/mmap2.patch26
1 files changed, 26 insertions, 0 deletions
diff --git a/meta/recipes-devtools/qemu/qemu/mmap2.patch b/meta/recipes-devtools/qemu/qemu/mmap2.patch
new file mode 100644
index 0000000000..9d40565938
--- /dev/null
+++ b/meta/recipes-devtools/qemu/qemu/mmap2.patch
@@ -0,0 +1,26 @@
1When using qemu-i386 to build qemux86 webkitgtk on musl, it sits in an
2infinite loop of mremap calls of ever decreasing/increasing addresses.
3
4I suspect something in the musl memory allocation code loops indefinitely
5if it only sees ENOMEM and only exits when it hits EFAULT.
6
7According to the docs, trying to mremap outside the address space
8can/should return EFAULT and changing this allows the build to succeed.
9Whether this should be fixed in qemu and/or musl, not sure.
10
11Upstream-Status: Pending
12Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org
13
14Index: qemu-5.2.0/linux-user/mmap.c
15===================================================================
16--- qemu-5.2.0.orig/linux-user/mmap.c
17+++ qemu-5.2.0/linux-user/mmap.c
18@@ -727,7 +727,7 @@ abi_long target_mremap(abi_ulong old_add
19 !guest_range_valid(new_addr, new_size)) ||
20 ((flags & MREMAP_MAYMOVE) == 0 &&
21 !guest_range_valid(old_addr, new_size))) {
22- errno = ENOMEM;
23+ errno = EFAULT;
24 return -1;
25 }
26