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>
Let's introduce RAM_NORESERVE, allowing mmap'ing with MAP_NORESERVE. The
new flag has the following semantics:
"
RAM is mmap-ed with MAP_NORESERVE. When set, reserving swap space (or huge
pages if applicable) is skipped: will bail out if not supported. When not
set, the OS will do the reservation, if supported for the memory type.
"
Allow passing it into:
- memory_region_init_ram_nomigrate()
- memory_region_init_resizeable_ram()
- memory_region_init_ram_from_file()
... and teach qemu_ram_mmap() and qemu_anon_ram_alloc() about the flag.
Bail out if the flag is not supported, which is the case right now for
both, POSIX and win32. We will add Linux support next and allow specifying
RAM_NORESERVE via memory backends.
The target use case is virtio-mem, which dynamically exposes memory
inside a large, sparse memory area to the VM.
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
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: <20210510114328.21835-9-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Let's pass flags instead of bools to prepare for passing other flags and
update the documentation of qemu_ram_mmap(). Introduce new QEMU_MAP_
flags that abstract the mmap() PROT_ and MAP_ flag handling and simplify
it.
We expose only flags that are currently supported by qemu_ram_mmap().
Maybe, we'll see qemu_mmap() in the future as well that can implement these
flags.
Note: We don't use MAP_ flags as some flags (e.g., MAP_SYNC) are only
defined for some systems and we want to always be able to identify
these flags reliably inside qemu_ram_mmap() -- for example, to properly
warn when some future flags are not available or effective on a system.
Also, this way we can simplify PROT_ handling as well.
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
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: <20210510114328.21835-8-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We want to activate memory within a reserved memory region, to make it
accessible. Let's factor that out.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
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: <20210510114328.21835-4-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We want to reserve a memory region without actually populating memory.
Let's factor that out.
Reviewed-by: Igor Kotrasinski <i.kotrasinsk@partner.samsung.com>
Acked-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
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: <20210510114328.21835-3-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Let's factor out calculating the size of the guard page and rename the
variable to make it clearer that this pagesize only applies to the
guard page.
Reviewed-by: Peter Xu <peterx@redhat.com>
Acked-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Acked-by: Eduardo Habkost <ehabkost@redhat.com> for memory backend and machine core
Cc: Igor Kotrasinski <i.kotrasinsk@partner.samsung.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210510114328.21835-2-david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Allow RAM MemoryRegion to be created from an offset in a file, instead
of allocating at offset of 0 by default. This is needed to synchronize
RAM between QEMU & remote process.
Signed-off-by: Jagannathan Raman <jag.raman@oracle.com>
Signed-off-by: John G Johnson <john.g.johnson@oracle.com>
Signed-off-by: Elena Ufimtseva <elena.ufimtseva@oracle.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 609996697ad8617e3b01df38accc5c208c24d74e.1611938319.git.jag.raman@oracle.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
There is currently no way to open(O_RDONLY) and mmap(PROT_READ) when
creating a memory region from a file. This functionality is needed since
the underlying host file may not allow writing.
Add a bool readonly argument to memory_region_init_ram_from_file() and
the APIs it calls.
Extend memory_region_init_ram_from_file() rather than introducing a
memory_region_init_rom_from_file() API so that callers can easily make a
choice between read/write and read-only at runtime without calling
different APIs.
No new RAMBlock flag is introduced for read-only because it's unclear
whether RAMBlocks need to know that they are read-only. Pass a bool
readonly argument instead.
Both of these design decisions can be changed in the future. It just
seemed like the simplest approach to me.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20210104171320.575838-2-stefanha@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
There are three page size in qemu:
real host page size
host page size
target page size
All of them have dedicate variable to represent. For the last two, we
use the same form in the whole qemu project, while for the first one we
use two forms: qemu_real_host_page_size and getpagesize().
qemu_real_host_page_size is defined to be a replacement of
getpagesize(), so let it serve the role.
[Note] Not fully tested for some arch or device.
Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
Message-Id: <20191013021145.16011-3-richardw.yang@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When a file supporting DAX is used as vNVDIMM backend, mmap it with
MAP_SYNC flag in addition which can ensure file system metadata
synced in each guest writes to the backend file, without other QEMU
actions (e.g., periodic fsync() by QEMU).
Current, We have below different possible use cases:
1. pmem=on is set, shared=on is set, MAP_SYNC supported:
a: backend is a dax supporting file.
- MAP_SYNC will active.
b: backend is not a dax supporting file.
- mmap will trigger a warning. then MAP_SYNC flag will be ignored
2. The rest of cases:
- we will never pass the MAP_SYNC to mmap2
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@linux.intel.com>
[ehabkost: Rebased patch to latest code on master]
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
Tested-by: Wei Yang <richardw.yang@linux.intel.com>
Message-Id: <20190422004849.26463-2-richardw.yang@linux.intel.com>
[ehabkost: squashed documentation patch]
Message-Id: <20190422004849.26463-3-richardw.yang@linux.intel.com>
[ehabkost: documentation fixup]
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Pankaj Gupta <pagupta@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
besides the existing 'shared' flags, we are going to add
'is_pmem' to qemu_ram_mmap(), which indicated the memory backend
file is a persist memory.
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@linux.intel.com>
Reviewed-by: Pankaj Gupta <pagupta@redhat.com>
Message-Id: <786c46862cfeb253ee0ea2f44d62ffe76edb7fa4.1549555521.git.yi.z.zhang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Pankaj Gupta <pagupta@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The commit 7197fb4058 ("util/mmap-alloc:
fix hugetlb support on ppc64") fixed Huge TLB mappings on ppc64.
However, we still need to consider the underlying huge page size
during munmap() because it requires that both address and length be a
multiple of the underlying huge page size for Huge TLB mappings.
Quote from "Huge page (Huge TLB) mappings" paragraph under NOTES
section of the munmap(2) manual:
"For munmap(), addr and length must both be a multiple of the
underlying huge page size."
On ppc64, the munmap() in qemu_ram_munmap() does not work for Huge TLB
mappings because the mapped segment can be aligned with the underlying
huge page size, not aligned with the native system page size, as
returned by getpagesize().
This has the side effect of not releasing huge pages back to the pool
after a hugetlbfs file-backed memory device is hot-unplugged.
This patch fixes the situation in qemu_ram_mmap() and
qemu_ram_munmap() by considering the underlying page size on ppc64.
After this patch, memory hot-unplug releases huge pages back to the
pool.
Fixes: 7197fb4058
Signed-off-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Unfold parts of qemu_ram_mmap() for the sake of understanding, moving
declarations to the top, and keeping architecture-specifics in the
ifdef-else blocks. No changes in the function behaviour.
Give ptr and ptr1 meaningful names:
ptr -> guardptr : pointer to the PROT_NONE guard region
ptr1 -> ptr : pointer to the mapped memory returned to caller
Signed-off-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
qemu_mempath_getpagesize() gets the effective (host side) page size for
a block of memory backed by an mmap()ed file on the host. It requires
the mem_path parameter to be non-NULL.
This ends up meaning all the callers need a different case for handling
anonymous memory (for memory-backend-ram or default memory with -mem-path
is not specified).
We can make all those callers a little simpler by having
qemu_mempath_getpagesize() accept NULL, and treat that as the anonymous
memory case.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Greg Kurz <groug@kaod.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
SPARC Linux has an oddity that it insists that mmap()
of MAP_FIXED memory must be at an alignment defined by
SHMLBA, which is more aligned than the page size
(typically, SHMLBA alignment is to 16K, and pages are 8K).
This is a relic of ancient hardware that had cache
aliasing constraints, but even on modern hardware the
kernel still insists on the alignment.
To ensure that we get mmap() alignment sufficient to
make the kernel happy, change QEMU_VMALLOC_ALIGN,
qemu_fd_getpagesize() and qemu_mempath_getpagesize()
to use the maximum of getpagesize() and SHMLBA.
In particular, this allows 'make check' to pass on Sparc:
we were previously failing the ivshmem tests.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 1512752248-17857-1-git-send-email-peter.maydell@linaro.org
getrampagesize() returns the largest supported page size and mainly
used to know if huge pages are enabled.
However is implemented in target-ppc/kvm.c and not available
in TCG or other architectures.
This renames and moves gethugepagesize() to mmap-alloc.c where
fd-based analog of it is already implemented. This renames and moves
getrampagesize() to exec.c as it seems to be the common place for
helpers like this.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
1st mmap returns *ptr* which aligns to host page size,
| size + align |
------------------------------------------
ptr
input param *align* could be 1M, or 2M, or host page size. After
QEMU_ALIGN_UP, offset will >= 0
2nd mmap use flag MAP_FIXED, then it return ptr+offset, or else fail.
If it success, then we will have something like:
| offset | size |
--------------------------------------
ptr ptr1
*ptr1* is what we really want to return, it equals ptr+offset.
Signed-off-by: Cao jin <caoj.fnst@cn.fujitsu.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Tracked down with an ugly, brittle and probably buggy Perl script.
Also move includes converted to <...> up so they get included before
ours where that's obviously okay.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
qemu/osdep.h checks whether MAP_ANONYMOUS is defined, but this check
is bogus without a previous inclusion of sys/mman.h. Include it in
sysemu/os-posix.h and remove it from everywhere else.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Clean up includes so that osdep.h is included first and headers
which it implies are not included manually.
This commit was created with scripts/clean-includes.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1454089805-5470-6-git-send-email-peter.maydell@linaro.org
The comment I put in mmap-alloc to document the ppc64 rules
refers to the previous revision of the patch:
we don't look at memory alignment anymore, we check
the fs from which the fd is mapped, instead.
It's also not clear what does "in this case" refer
to, rearrange text to make it clearer.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Since commit 8561c9244d "exec: allocate PROT_NONE pages on top of
RAM", it is no longer possible to back guest RAM with hugepages on ppc64
hosts:
mmap(NULL, 285212672, PROT_NONE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x3fff57000000
mmap(0x3fff57000000, 268435456, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED, 19, 0) = -1 EBUSY (Device or resource busy)
This is because on ppc64, Linux fixes a page size for a virtual address
at mmap time, so we can't switch a range of memory from anonymous
small pages to hugetlbs with MAP_FIXED.
See commit d0f13e3c20b6fb73ccb467bdca97fa7cf5a574cd
("[POWERPC] Introduce address space "slices"") in Linux
history for the details.
Detect this and create the PROT_NONE mapping using the same fd.
Naturally, this makes the guard page bigger with hugetlbfs.
Based on patch by Greg Kurz.
Acked-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Tested-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Existing callers are checking for MAP_FAILED,
so we should return that on error.
Reported-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Anonymous and file-backed RAM allocation are now almost exactly the same.
Reduce code duplication by moving RAM mmap code out of oslib-posix.c and
exec.c.
Reported-by: Marc-André Lureau <mlureau@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Thibaut Collet <thibaut.collet@6wind.com>
