]> git.proxmox.com Git - mirror_qemu.git/commit
projects / mirror_qemu.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: f5e6786) | patch
tcg/tci: Adjust passing of MemOpIdx
authorRichard Henderson <richard.henderson@linaro.org>
Wed, 7 Jun 2023 00:52:41 +0000 (17:52 -0700)
committerRichard Henderson <richard.henderson@linaro.org>
Wed, 7 Jun 2023 15:33:45 +0000 (08:33 -0700)
commitab64da79774060450046ce8c800eef000024dc8c
treefd95944eee2b242ec515d5ecfcf4f87509ea6e57tree
parentf5e6786de4815751b0a3d2235c760361f228ea48commit | diff
tcg/tci: Adjust passing of MemOpIdx

Since adding MO_ATOM_MASK, the maximum MemOpIdx requires 15 bits,
which overflows the 12 bit field allocated for TCI memory ops.
Expand the field to 16 bits for 2-operand memory ops, and place
the value in TCG_REG_TMP for 3-operand memory ops (same as we
already do for 4-operand memory ops).

Cures a debug assert for aarch64, with FEAT_LSE2 enabled.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
tcg/tci.c diff | blob | blame | history
tcg/tci/tcg-target.c.inc diff | blob | blame | history
QEMU mirror