diff options
author | Luke Shumaker <lukeshu@lukeshu.com> | 2018-05-28 21:18:43 -0400 |
---|---|---|
committer | Luke Shumaker <lukeshu@lukeshu.com> | 2018-05-29 11:10:42 -0400 |
commit | e1ec62c46a0be4c40414df02f9b6fccfca94006e (patch) | |
tree | abf63f0d50d6c724e90511f8baf4c4a9971b03b4 /libre/qemu-user-static/0004-linux-user-init_guest_space-Correctly-handle-guest_s.patch | |
parent | 06f09684205a5674088113b098f5fd2a6176772c (diff) | |
download | abslibre-e1ec62c46a0be4c40414df02f9b6fccfca94006e.tar.gz abslibre-e1ec62c46a0be4c40414df02f9b6fccfca94006e.tar.bz2 abslibre-e1ec62c46a0be4c40414df02f9b6fccfca94006e.zip |
Update libre/qemu-user-static and dependencies
Fighting with glib2 was no fun. But hey, at least I got a GCC bug report
out of it! https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85957
To get around this, I enabled SSE2 for glib2-static. This means it won't
run on pre-Pentium 4 (2001) i686 processors; which will cause problems on
the Pentium Pro/2 and the Pentium 3, and their AMD competitors: K6 and K7
(the latter of which was branded as Athlon, Athlon XP, Duron, and Sempron).
https://lists.parabola.nu/pipermail/dev/2018-May/006748.html
If someone digs out a Pentium 2 and gets Parabola to boot on it, I'll
figure out a different way to get around glib2's disagreement with x87
floating point math; and turn SSE2 back off.
Diffstat (limited to 'libre/qemu-user-static/0004-linux-user-init_guest_space-Correctly-handle-guest_s.patch')
-rw-r--r-- | libre/qemu-user-static/0004-linux-user-init_guest_space-Correctly-handle-guest_s.patch | 47 |
1 files changed, 0 insertions, 47 deletions
diff --git a/libre/qemu-user-static/0004-linux-user-init_guest_space-Correctly-handle-guest_s.patch b/libre/qemu-user-static/0004-linux-user-init_guest_space-Correctly-handle-guest_s.patch deleted file mode 100644 index 7f9cfc7a8..000000000 --- a/libre/qemu-user-static/0004-linux-user-init_guest_space-Correctly-handle-guest_s.patch +++ /dev/null @@ -1,47 +0,0 @@ -From bc26025ccc64a2a63fc58776c93e2e4392445abe Mon Sep 17 00:00:00 2001 -From: Luke Shumaker <lukeshu@parabola.nu> -Date: Wed, 27 Dec 2017 19:44:00 -0500 -Subject: [PATCH 04/10] linux-user: init_guest_space: Correctly handle - guest_start in commpage initialization - -init_guest_commpage needs to check if the mapped space, which ends at -real_start+real_size overlaps with where it needs to put the commpage, -which is (assuming sane qemu_host_page_size) guest_base + 0xffff000, where -guest_base is real_start - guest_start. - - [guest_base][ 0xffff0000 ][commpage] - [guest_base][guest_start][real_size] [commpage] - [ real_start ][real_size] [commpage] - ^ - fail if this gap < 0 - -Since init_guest_commpage wants to do everything relative to guest_base -(rather than real_start), it obviously needs to be comparing 0xffff0000 -against guest_start+real_size, not just real_size. - -This bug has been present since 806d102141b99d4f1e55a97d68b7ea8c8ba3129f in -2012, but guest_start is usually 0, and prior to v2.11 real_size was -usually much smaller than 0xffff0000, so it was uncommon for it to have -made a difference. - -Signed-off-by: Luke Shumaker <lukeshu@parabola.nu> ---- - linux-user/elfload.c | 2 +- - 1 file changed, 1 insertion(+), 1 deletion(-) - -diff --git a/linux-user/elfload.c b/linux-user/elfload.c -index 1a6b660b25..f41cecc3cb 100644 ---- a/linux-user/elfload.c -+++ b/linux-user/elfload.c -@@ -1882,7 +1882,7 @@ unsigned long init_guest_space(unsigned long host_start, - #if defined(TARGET_ARM) && !defined(TARGET_AARCH64) - /* On 32-bit ARM, we need to also be able to map the commpage. */ - int valid = init_guest_commpage(real_start - guest_start, -- real_size); -+ real_size + guest_start); - if (valid == 1) { - break; - } else if (valid == -1) { --- -2.15.1 - |