From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Eduardo Habkost <ehabkost@redhat.com>,
Peter Xu <peterx@redhat.com>,
David Hildenbrand <david@redhat.com>
Subject: [PULL 26/33] util/mmap-alloc: Support RAM_NORESERVE via MAP_NORESERVE under Linux
Date: Tue, 15 Jun 2021 15:38:48 +0200 [thread overview]
Message-ID: <20210615133855.775687-27-pbonzini@redhat.com> (raw)
In-Reply-To: <20210615133855.775687-1-pbonzini@redhat.com>
From: David Hildenbrand <david@redhat.com>
Let's support RAM_NORESERVE via MAP_NORESERVE on Linux. The flag has no
effect on most shared mappings - except for hugetlbfs and anonymous memory.
Linux man page:
"MAP_NORESERVE: Do not reserve swap space for this mapping. When swap
space is reserved, one has the guarantee that it is possible to modify
the mapping. When swap space is not reserved one might get SIGSEGV
upon a write if no physical memory is available. See also the discussion
of the file /proc/sys/vm/overcommit_memory in proc(5). In kernels before
2.6, this flag had effect only for private writable mappings."
Note that the "guarantee" part is wrong with memory overcommit in Linux.
Also, in Linux hugetlbfs is treated differently - we configure reservation
of huge pages from the pool, not reservation of swap space (huge pages
cannot be swapped).
The rough behavior is [1]:
a) !Hugetlbfs:
1) Without MAP_NORESERVE *or* with memory overcommit under Linux
disabled ("/proc/sys/vm/overcommit_memory == 2"), the following
accounting/reservation happens:
For a file backed map
SHARED or READ-only - 0 cost (the file is the map not swap)
PRIVATE WRITABLE - size of mapping per instance
For an anonymous or /dev/zero map
SHARED - size of mapping
PRIVATE READ-only - 0 cost (but of little use)
PRIVATE WRITABLE - size of mapping per instance
2) With MAP_NORESERVE, no accounting/reservation happens.
b) Hugetlbfs:
1) Without MAP_NORESERVE, huge pages are reserved.
2) With MAP_NORESERVE, no huge pages are reserved.
Note: With "/proc/sys/vm/overcommit_memory == 0", we were already able
to configure it for !hugetlbfs globally; this toggle now allows
configuring it more fine-grained, not for the whole system.
The target use case is virtio-mem, which dynamically exposes memory
inside a large, sparse memory area to the VM.
[1] https://www.kernel.org/doc/Documentation/vm/overcommit-accounting
Reviewed-by: Peter Xu <peterx@redhat.com>
Acked-by: Eduardo Habkost <ehabkost@redhat.com> for memory backend and machine core
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210510114328.21835-10-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
include/qemu/osdep.h | 3 ++
softmmu/physmem.c | 1 +
util/mmap-alloc.c | 69 ++++++++++++++++++++++++++++++++++++++++++--
3 files changed, 71 insertions(+), 2 deletions(-)
diff --git a/include/qemu/osdep.h b/include/qemu/osdep.h
index 38a3327db9..488d8a4955 100644
--- a/include/qemu/osdep.h
+++ b/include/qemu/osdep.h
@@ -195,6 +195,9 @@ extern "C" {
#ifndef MAP_FIXED_NOREPLACE
#define MAP_FIXED_NOREPLACE 0
#endif
+#ifndef MAP_NORESERVE
+#define MAP_NORESERVE 0
+#endif
#ifndef ENOMEDIUM
#define ENOMEDIUM ENODEV
#endif
diff --git a/softmmu/physmem.c b/softmmu/physmem.c
index a7725d8474..bf5493eab0 100644
--- a/softmmu/physmem.c
+++ b/softmmu/physmem.c
@@ -2251,6 +2251,7 @@ void qemu_ram_remap(ram_addr_t addr, ram_addr_t length)
flags = MAP_FIXED;
flags |= block->flags & RAM_SHARED ?
MAP_SHARED : MAP_PRIVATE;
+ flags |= block->flags & RAM_NORESERVE ? MAP_NORESERVE : 0;
if (block->fd >= 0) {
area = mmap(vaddr, length, PROT_READ | PROT_WRITE,
flags, block->fd, offset);
diff --git a/util/mmap-alloc.c b/util/mmap-alloc.c
index d0cf4aaee5..838e286ce5 100644
--- a/util/mmap-alloc.c
+++ b/util/mmap-alloc.c
@@ -20,6 +20,7 @@
#include "qemu/osdep.h"
#include "qemu/mmap-alloc.h"
#include "qemu/host-utils.h"
+#include "qemu/cutils.h"
#include "qemu/error-report.h"
#define HUGETLBFS_MAGIC 0x958458f6
@@ -83,6 +84,70 @@ size_t qemu_mempath_getpagesize(const char *mem_path)
return qemu_real_host_page_size;
}
+#define OVERCOMMIT_MEMORY_PATH "/proc/sys/vm/overcommit_memory"
+static bool map_noreserve_effective(int fd, uint32_t qemu_map_flags)
+{
+#if defined(__linux__)
+ const bool readonly = qemu_map_flags & QEMU_MAP_READONLY;
+ const bool shared = qemu_map_flags & QEMU_MAP_SHARED;
+ gchar *content = NULL;
+ const char *endptr;
+ unsigned int tmp;
+
+ /*
+ * hugeltb accounting is different than ordinary swap reservation:
+ * a) Hugetlb pages from the pool are reserved for both private and
+ * shared mappings. For shared mappings, all mappers have to specify
+ * MAP_NORESERVE.
+ * b) MAP_NORESERVE is not affected by /proc/sys/vm/overcommit_memory.
+ */
+ if (qemu_fd_getpagesize(fd) != qemu_real_host_page_size) {
+ return true;
+ }
+
+ /*
+ * Accountable mappings in the kernel that can be affected by MAP_NORESEVE
+ * are private writable mappings (see mm/mmap.c:accountable_mapping() in
+ * Linux). For all shared or readonly mappings, MAP_NORESERVE is always
+ * implicitly active -- no reservation; this includes shmem. The only
+ * exception is shared anonymous memory, it is accounted like private
+ * anonymous memory.
+ */
+ if (readonly || (shared && fd >= 0)) {
+ return true;
+ }
+
+ /*
+ * MAP_NORESERVE is globally ignored for applicable !hugetlb mappings when
+ * memory overcommit is set to "never". Sparse memory regions aren't really
+ * possible in this system configuration.
+ *
+ * Bail out now instead of silently committing way more memory than
+ * currently desired by the user.
+ */
+ if (g_file_get_contents(OVERCOMMIT_MEMORY_PATH, &content, NULL, NULL) &&
+ !qemu_strtoui(content, &endptr, 0, &tmp) &&
+ (!endptr || *endptr == '\n')) {
+ if (tmp == 2) {
+ error_report("Skipping reservation of swap space is not supported:"
+ " \"" OVERCOMMIT_MEMORY_PATH "\" is \"2\"");
+ return false;
+ }
+ return true;
+ }
+ /* this interface has been around since Linux 2.6 */
+ error_report("Skipping reservation of swap space is not supported:"
+ " Could not read: \"" OVERCOMMIT_MEMORY_PATH "\"");
+ return false;
+#endif
+ /*
+ * E.g., FreeBSD used to define MAP_NORESERVE, never implemented it,
+ * and removed it a while ago.
+ */
+ error_report("Skipping reservation of swap space is not supported");
+ return false;
+}
+
/*
* Reserve a new memory region of the requested size to be used for mapping
* from the given fd (if any).
@@ -131,13 +196,13 @@ static void *mmap_activate(void *ptr, size_t size, int fd,
int flags = MAP_FIXED;
void *activated_ptr;
- if (noreserve) {
- error_report("Skipping reservation of swap space is not supported");
+ if (noreserve && !map_noreserve_effective(fd, qemu_map_flags)) {
return MAP_FAILED;
}
flags |= fd == -1 ? MAP_ANONYMOUS : 0;
flags |= shared ? MAP_SHARED : MAP_PRIVATE;
+ flags |= noreserve ? MAP_NORESERVE : 0;
if (shared && sync) {
map_sync_flags = MAP_SYNC | MAP_SHARED_VALIDATE;
}
--
2.31.1
next prev parent reply other threads:[~2021-06-15 13:57 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-15 13:38 [PULL 00/33] Misc patches for 2021-06-15 Paolo Bonzini
2021-06-15 13:38 ` [PULL 01/33] vnc: avoid deprecation warnings for SASL on OS X Paolo Bonzini
2021-06-15 13:38 ` [PULL 02/33] vl: Fix an assert failure in error path Paolo Bonzini
2021-06-15 13:38 ` [PULL 03/33] qemu-config: use qemu_opts_from_qdict Paolo Bonzini
2021-06-15 13:38 ` [PULL 04/33] block/scsi: correctly emulate the VPD block limits page Paolo Bonzini
2021-06-15 13:38 ` [PULL 05/33] runstate: Initialize Error * to NULL Paolo Bonzini
2021-06-15 13:38 ` [PULL 06/33] esp: only assert INTR_DC interrupt flag if selection fails Paolo Bonzini
2021-06-15 13:38 ` [PULL 07/33] esp: only set ESP_RSEQ at the start of the select sequence Paolo Bonzini
2021-06-15 13:38 ` [PULL 08/33] esp: allow non-DMA callback in esp_transfer_data() initial transfer Paolo Bonzini
2021-06-15 13:38 ` [PULL 09/33] esp: handle non-DMA transfers from the target one byte at a time Paolo Bonzini
2021-06-15 13:38 ` [PULL 10/33] esp: ensure PDMA write transfers are flushed from the FIFO to the target immediately Paolo Bonzini
2021-06-15 13:38 ` [PULL 11/33] esp: revert 75ef849696 "esp: correctly fill bus id with requested lun" Paolo Bonzini
2021-06-15 13:38 ` [PULL 12/33] esp: correctly accumulate extended messages for PDMA Paolo Bonzini
2021-06-15 13:38 ` [PULL 13/33] esp: fix migration version check in esp_is_version_5() Paolo Bonzini
2021-06-15 13:38 ` [PULL 14/33] esp: store lun coming from the MESSAGE OUT phase Paolo Bonzini
2021-06-15 13:38 ` [PULL 15/33] softmmu/physmem: Mark shared anonymous memory RAM_SHARED Paolo Bonzini
2021-06-15 13:38 ` [PULL 16/33] softmmu/physmem: Fix ram_block_discard_range() to handle shared anonymous memory Paolo Bonzini
2021-06-15 13:38 ` [PULL 17/33] softmmu/physmem: Fix qemu_ram_remap() " Paolo Bonzini
2021-06-15 13:38 ` [PULL 18/33] util/mmap-alloc: Factor out calculation of the pagesize for the guard page Paolo Bonzini
2021-06-15 13:38 ` [PULL 19/33] util/mmap-alloc: Factor out reserving of a memory region to mmap_reserve() Paolo Bonzini
2021-06-15 13:38 ` [PULL 20/33] util/mmap-alloc: Factor out activating of memory to mmap_activate() Paolo Bonzini
2021-06-15 13:38 ` [PULL 21/33] softmmu/memory: Pass ram_flags to qemu_ram_alloc_from_fd() Paolo Bonzini
2021-06-15 13:38 ` [PULL 22/33] softmmu/memory: Pass ram_flags to memory_region_init_ram_shared_nomigrate() Paolo Bonzini
2021-06-15 13:38 ` [PULL 23/33] softmmu/memory: Pass ram_flags to qemu_ram_alloc() and qemu_ram_alloc_internal() Paolo Bonzini
2021-06-15 13:38 ` [PULL 24/33] util/mmap-alloc: Pass flags instead of separate bools to qemu_ram_mmap() Paolo Bonzini
2021-06-15 13:38 ` [PULL 25/33] memory: Introduce RAM_NORESERVE and wire it up in qemu_ram_mmap() Paolo Bonzini
2021-06-15 13:38 ` Paolo Bonzini [this message]
2021-06-15 13:38 ` [PULL 27/33] hostmem: Wire up RAM_NORESERVE via "reserve" property Paolo Bonzini
2021-06-15 13:38 ` [PULL 28/33] qmp: Clarify memory backend properties returned via query-memdev Paolo Bonzini
2021-06-15 13:38 ` [PULL 29/33] qmp: Include "share" property of memory backends Paolo Bonzini
2021-06-15 13:38 ` [PULL 30/33] hmp: Print "share" property of memory backends with "info memdev" Paolo Bonzini
2021-06-15 13:38 ` [PULL 31/33] qmp: Include "reserve" property of memory backends Paolo Bonzini
2021-06-15 13:38 ` [PULL 32/33] hmp: Print "reserve" property of memory backends with "info memdev" Paolo Bonzini
2021-06-15 13:38 ` [PULL 33/33] configure: map x32 to cpu_family x86_64 for meson Paolo Bonzini
2021-06-15 14:17 ` [PULL 00/33] Misc patches for 2021-06-15 Peter Maydell
2021-06-15 16:52 ` Philippe Mathieu-Daudé
2021-06-17 10:06 ` Alex Bennée
2021-06-17 12:55 ` Paolo Bonzini
2021-06-17 15:34 ` Thomas Huth
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=20210615133855.775687-27-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=david@redhat.com \
--cc=ehabkost@redhat.com \
--cc=peterx@redhat.com \
--cc=qemu-devel@nongnu.org \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).