From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Zenghui Yu <yuzenghui@huawei.com>
Subject: [PULL 26/45] kvm: Take into account the unaligned section size when preparing bitmap
Date: Tue, 15 Dec 2020 12:54:26 -0500 [thread overview]
Message-ID: <20201215175445.1272776-27-pbonzini@redhat.com> (raw)
In-Reply-To: <20201215175445.1272776-1-pbonzini@redhat.com>
From: Zenghui Yu <yuzenghui@huawei.com>
The kernel KVM_CLEAR_DIRTY_LOG interface has align requirement on both the
start and the size of the given range of pages. We have been careful to
handle the unaligned cases when performing CLEAR on one slot. But it seems
that we forget to take the unaligned *size* case into account when
preparing bitmap for the interface, and we may end up clearing dirty status
for pages outside of [start, start + size).
If the size is unaligned, let's go through the slow path to manipulate a
temp bitmap for the interface so that we won't bother with those unaligned
bits at the end of bitmap.
I don't think this can happen in practice since the upper layer would
provide us with the alignment guarantee. I'm not sure if kvm-all could rely
on it. And this patch is mainly intended to address correctness of the
specific algorithm used inside kvm_log_clear_one_slot().
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Message-Id: <20201208114013.875-1-yuzenghui@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
accel/kvm/kvm-all.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 666b9ab96c..389eaace72 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -745,7 +745,7 @@ static int kvm_log_clear_one_slot(KVMSlot *mem, int as_id, uint64_t start,
assert(bmap_start % BITS_PER_LONG == 0);
/* We should never do log_clear before log_sync */
assert(mem->dirty_bmap);
- if (start_delta) {
+ if (start_delta || bmap_npages - size / psize) {
/* Slow path - we need to manipulate a temp bitmap */
bmap_clear = bitmap_new(bmap_npages);
bitmap_copy_with_src_offset(bmap_clear, mem->dirty_bmap,
@@ -758,7 +758,10 @@ static int kvm_log_clear_one_slot(KVMSlot *mem, int as_id, uint64_t start,
bitmap_clear(bmap_clear, 0, start_delta);
d.dirty_bitmap = bmap_clear;
} else {
- /* Fast path - start address aligns well with BITS_PER_LONG */
+ /*
+ * Fast path - both start and size align well with BITS_PER_LONG
+ * (or the end of memory slot)
+ */
d.dirty_bitmap = mem->dirty_bmap + BIT_WORD(bmap_start);
}
--
2.26.2
next prev parent reply other threads:[~2020-12-15 18:05 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-15 17:54 [PULL 00/45] Misc patches for 2020-12-15 Paolo Bonzini
2020-12-15 17:54 ` [PULL 01/45] remove preconfig state Paolo Bonzini
2020-12-15 17:54 ` [PULL 02/45] vl: remove separate preconfig main_loop Paolo Bonzini
2020-12-19 21:30 ` Laurent Vivier
2020-12-20 8:52 ` Paolo Bonzini
2020-12-15 17:54 ` [PULL 03/45] vl: allow -incoming defer with -preconfig Paolo Bonzini
2020-12-15 17:54 ` [PULL 04/45] vl: extract softmmu/runstate.c Paolo Bonzini
2020-12-15 17:54 ` [PULL 05/45] vl: extract softmmu/globals.c Paolo Bonzini
2020-12-15 17:54 ` [PULL 06/45] vl: move all generic initialization out of vl.c Paolo Bonzini
2020-12-15 17:54 ` [PULL 07/45] chardev: do not use machine_init_done Paolo Bonzini
2020-12-15 17:54 ` [PULL 08/45] machine: introduce MachineInitPhase Paolo Bonzini
2020-12-15 17:54 ` [PULL 09/45] ppc/spapr: cleanup -machine pseries,nvdimm=X handling Paolo Bonzini
2020-12-15 17:54 ` [PULL 10/45] vl: make qemu_get_machine_opts static Paolo Bonzini
2020-12-15 17:54 ` [PULL 11/45] plugin: propagate errors Paolo Bonzini
2020-12-15 17:54 ` [PULL 12/45] memory: allow creating MemoryRegions before accelerators Paolo Bonzini
2020-12-15 17:54 ` [PULL 13/45] monitor: allow quitting while in preconfig state Paolo Bonzini
2020-12-15 17:54 ` [PULL 14/45] qmp: generalize watchdog-set-action to -no-reboot/-no-shutdown Paolo Bonzini
2020-12-15 17:54 ` [PULL 15/45] vl: Add an -action option specifying response to guest events Paolo Bonzini
2020-12-15 17:54 ` [PULL 16/45] vl: Add option to avoid stopping VM upon guest panic Paolo Bonzini
2021-01-19 21:34 ` Peter Maydell
2021-01-20 5:28 ` Alejandro Jimenez
2021-01-20 13:47 ` Peter Maydell
2021-01-20 13:54 ` Daniel P. Berrangé
2021-01-20 14:47 ` Paolo Bonzini
2021-01-20 13:58 ` Paolo Bonzini
2020-12-15 17:54 ` [PULL 17/45] qtest/pvpanic: Test panic option that allows VM to continue Paolo Bonzini
2020-12-15 17:54 ` [PULL 18/45] msix: assert that accesses are within bounds Paolo Bonzini
2020-12-15 17:54 ` [PULL 19/45] memory: clamp cached translation in case it points to an MMIO region Paolo Bonzini
2021-01-13 13:27 ` Michael S. Tsirkin
2020-12-15 17:54 ` [PULL 20/45] accel/tcg: Remove deprecated '-tb-size' option Paolo Bonzini
2020-12-15 17:54 ` [PULL 21/45] docs/system: Move the list of removed features to a separate file Paolo Bonzini
2020-12-15 17:54 ` [PULL 22/45] Remove the deprecated -realtime option Paolo Bonzini
2020-12-15 17:54 ` [PULL 23/45] Remove the deprecated -show-cursor option Paolo Bonzini
2020-12-15 17:54 ` [PULL 24/45] icount: improve exec nocache usage Paolo Bonzini
2020-12-15 17:54 ` [PULL 25/45] scsi: fix device removal race vs IO restart callback on resume Paolo Bonzini
2020-12-15 17:54 ` Paolo Bonzini [this message]
2020-12-15 17:54 ` [PULL 27/45] qemu-option: simplify search for end of key Paolo Bonzini
2020-12-15 17:54 ` [PULL 28/45] qemu-option: pass QemuOptsList to opts_accepts_any Paolo Bonzini
2020-12-15 17:54 ` [PULL 29/45] vl: rename local variable in configure_accelerators Paolo Bonzini
2020-12-15 17:54 ` [PULL 30/45] docs: set CONFDIR when running sphinx Paolo Bonzini
2020-12-15 17:54 ` [PULL 31/45] hw/core: Restrict 'fw-path-provider.c' to system mode emulation Paolo Bonzini
2020-12-15 17:54 ` [PULL 32/45] qemu/atomic: Drop special case for unsupported compiler Paolo Bonzini
2020-12-15 17:54 ` [PULL 33/45] accel/tcg: Remove special case for GCC < 4.6 Paolo Bonzini
2020-12-15 17:54 ` [PULL 34/45] compiler.h: remove GCC < 3 __builtin_expect fallback Paolo Bonzini
2020-12-15 17:54 ` [PULL 35/45] qemu-plugin.h: remove GCC < 4 Paolo Bonzini
2020-12-15 17:54 ` [PULL 36/45] tests: remove GCC < 4 fallbacks Paolo Bonzini
2020-12-15 17:54 ` [PULL 37/45] virtiofsd: replace _Static_assert with QEMU_BUILD_BUG_ON Paolo Bonzini
2020-12-15 17:54 ` [PULL 38/45] compiler.h: explicit case for Clang printf attribute Paolo Bonzini
2020-12-15 17:54 ` [PULL 39/45] poison: remove GNUC check Paolo Bonzini
2020-12-15 17:54 ` [PULL 40/45] xen: " Paolo Bonzini
2020-12-15 17:54 ` [PULL 41/45] compiler: " Paolo Bonzini
2020-12-15 17:54 ` [PULL 42/45] linux-user: " Paolo Bonzini
2020-12-15 17:54 ` [PULL 43/45] compiler.h: remove QEMU_GNUC_PREREQ Paolo Bonzini
2020-12-15 17:54 ` [PULL 44/45] scripts/git.orderfile: Keep files with .inc extension sorted Paolo Bonzini
2020-12-15 17:54 ` [PULL 45/45] build: -no-pie is no functional linker flag Paolo Bonzini
2020-12-16 10:55 ` [PULL 00/45] Misc patches for 2020-12-15 Peter Maydell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20201215175445.1272776-27-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=yuzenghui@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).