qemu/accel-tcg-Optimize-jump-cache-flush-during-tlb-range.patch
Jiabo Feng c4dab45526 QEMU update to version 6.2.0-76(master)
- qga/win32: Use rundll for VSS installation
- qga/win32: Remove change action from MSI installer
- ide: Increment BB in-flight counter for TRIM BH
- hw/pci-bridge/pxb: Fix missing swizzle
- host-vdpa: make notifiers _init()/_uninit() symmetric
- hw/virtio: vdpa: Fix leak of host-notifier memory-region
- accel/tcg/cpu-exec: Fix precise single-stepping after interrupt
- Allow setting up to 8 bytes with the generic loader
- hw/net/virtio-net: make some VirtIONet const
- accel/tcg: Optimize jump cache flush during tlb range flush
- 9pfs: prevent opening special files (CVE-2023-2861)
- tcg: Reduce tcg_assert_listed_vecop() scope
- gitlab: Disable plugins for cross-i386-tci
- vfio/pci: Fix a segfault in vfio_realize
- block/iscsi: fix double-free on BUSY or similar statuses
- tests/tcg: fix unused variable in linux-test
- hw/net/vmxnet3: allow VMXNET3_MAX_MTU itself as a value
- qga/vss-win32: fix warning for clang++-15
- vnc: avoid underflow when accessing user-provided address
- block/monitor: Fix crash when executing HMP commit
- virtio-gpu: add a FIXME for virtio_gpu_load()
- hw/ppc/Kconfig: MAC_NEWWORLD should always select USB_OHCI_PCI
- migration: report compress thread pid to libvirt

Signed-off-by: Jiabo Feng <fengjiabo1@huawei.com>
2023-08-07 16:46:33 +08:00

50 lines
1.7 KiB
Diff

From 28ca488c585c556ce04419f927d13d46771e1ea4 Mon Sep 17 00:00:00 2001
From: tangbinzy <tangbin_yewu@cmss.chinamobile.com>
Date: Tue, 18 Jul 2023 06:29:51 +0000
Subject: [PATCH] accel/tcg: Optimize jump cache flush during tlb range flush
mainline inclusion commit cfc2a2d69d59f02b32df3098ce17e10ab86d43c6 category:
bugfix
---------------------------------------------------------------
When the length of the range is large enough, clearing the whole cache is
faster than iterating over the (possibly extremely large) set of pages
contained in the range.
This mimics the pre-existing similar optimization done on the flush of the
tlb itself.
Signed-off-by: Idan Horowitz <idan.horowitz@gmail.com>
Message-Id: <20220110164754.1066025-1-idan.horowitz@gmail.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: tangbinzy <tangbin_yewu@cmss.chinamobile.com>
---
accel/tcg/cputlb.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index b69a953447..03526fa1ab 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -783,6 +783,15 @@ static void tlb_flush_range_by_mmuidx_async_0(CPUState *cpu,
}
qemu_spin_unlock(&env_tlb(env)->c.lock);
+ /*
+ * If the length is larger than the jump cache size, then it will take
+ * longer to clear each entry individually than it will to clear it all.
+ */
+ if (d.len >= (TARGET_PAGE_SIZE * TB_JMP_CACHE_SIZE)) {
+ cpu_tb_jmp_cache_clear(cpu);
+ return;
+ }
+
for (target_ulong i = 0; i < d.len; i += TARGET_PAGE_SIZE) {
tb_flush_jmp_cache(cpu, d.addr + i);
}
--
2.41.0.windows.1