Git-commit: b987718bbb
References: bsc#1207205 (CVE-2023-0330)
We cannot use the generic reentrancy guard in the LSI code, so
we have to manually prevent endless reentrancy here. The problematic
lsi_execute_script() function has already a way to detect whether
too many instructions have been executed - we just have to slightly
change the logic here that it also takes into account if the function
has been called too often in a reentrant way.
The code in fuzz-lsi53c895a-test.c has been taken from an earlier
patch by Mauro Matteo Cascella.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1563
Message-Id: <20230522091011.1082574-1-thuth@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
For symmetric algorithms, the length of ciphertext must be as same
as the plaintext.
The missing verification of the src_len and the dst_len in
virtio_crypto_sym_op_helper() may lead buffer overflow/divulged.
This patch is originally written by Yiming Tao for QEMU-SECURITY,
resend it(a few changes of error message) in qemu-devel.
Fixes: CVE-2023-3180
Fixes: 04b9b37edda("virtio-crypto: add data queue processing handler")
Cc: Gonglei <arei.gonglei@huawei.com>
Cc: Mauro Matteo Cascella <mcascell@redhat.com>
Cc: Yiming Tao <taoym@zju.edu.cn>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20230803024314.29962-2-pizhenwei@bytedance.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 9d38a84347)
References: bsc#1213925
References: CVE-2023-3180
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
References: bsc#1190011 (CVE-2021-3750)
40p and g3beige will hang forever with these messages:
qemu-system-ppc: warning: Blocked re-entrant IO on MemoryRegion: pci-conf-idx at addr: 0x0
qemu-system-ppc64: warning: Blocked re-entrant IO on MemoryRegion: lpc-hc at addr: 0x34
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Git-commit: 790762e548
References: bsc#1172033, CVE-2020-13253
Only move the state machine to ReceivingData if there is no
pending error. This avoids later OOB access while processing
commands queued.
"SD Specifications Part 1 Physical Layer Simplified Spec. v3.01"
4.3.3 Data Read
Read command is rejected if BLOCK_LEN_ERROR or ADDRESS_ERROR
occurred and no data transfer is performed.
4.3.4 Data Write
Write command is rejected if BLOCK_LEN_ERROR or ADDRESS_ERROR
occurred and no data transfer is performed.
WP_VIOLATION errors are not modified: the error bit is set, we
stay in receive-data state, wait for a stop command. All further
data transfer is ignored. See the check on sd->card_status at the
beginning of sd_read_data() and sd_write_data().
Fixes: CVE-2020-13253
Cc: qemu-stable@nongnu.org
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Buglink: https://bugs.launchpad.net/qemu/+bug/1880822
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20200630133912.9428-6-f4bug@amsat.org>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: effaf5a240
References: bsc#1180207, CVE-2020-14394
The loop condition in xhci_ring_chain_length() is under control of
the guest, and additionally the code does not check for failed DMA
transfers (e.g. if reaching the end of the RAM), so the loop there
could run for a very long time or even forever. Fix it by checking
the return value of dma_memory_read() and by introducing a maximum
loop length.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/646
Message-Id: <20220804131300.96368-1-thuth@redhat.com>
Reviewed-by: Mauro Matteo Cascella <mcascell@redhat.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: 4367a20cc4
References: bsc#1198038, CVE-2022-0216
Set current_req to NULL, not current_req->req, to prevent reusing a free'd
buffer in case of repeated SCSI cancel requests. Also apply the fix to
CLEAR QUEUE and BUS DEVICE RESET messages as well, since they also cancel
the request.
Thanks to Alexander Bulekov for providing a reproducer.
Fixes: CVE-2022-0216
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/972
Signed-off-by: Mauro Matteo Cascella <mcascell@redhat.com>
Tested-by: Alexander Bulekov <alxndr@bu.edu>
Message-Id: <20220711123316.421279-1-mcascell@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
[DF: Kill the hunk that patches the test as it's not there]
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: 736b01642d
References: bsc#1193880, CVE-2021-3929
This fixes CVE-2021-3929 "locally" by denying DMA to the iomem of the
device itself. This still allows DMA to MMIO regions of other devices
(e.g. doing P2P DMA to the controller memory buffer of another NVMe
device).
Fixes: CVE-2021-3929
Reported-by: Qiuhao Li <Qiuhao.Li@outlook.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>41
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
Signed-off-by: Lin Ma <lma@suse.com>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: 6dbbf05514
References: bsc#1205808, CVE-2022-4144
Have qxl_get_check_slot_offset() return false if the requested
buffer size does not fit within the slot memory region.
Similarly qxl_phys2virt() now returns NULL in such case, and
qxl_dirty_one_surface() aborts.
This avoids buffer overrun in the host pointer returned by
memory_region_get_ram_ptr().
Fixes: CVE-2022-4144 (out-of-bounds read)
Reported-by: Wenxu Yin (@awxylitol)
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1336
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-5-philmd@linaro.org>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: 8efec0ef8b)
References: bsc#1205808, CVE-2022-4144
Currently qxl_phys2virt() doesn't check for buffer overrun.
In order to do so in the next commit, pass the buffer size
as argument.
For QXLCursor in qxl_render_cursor() -> qxl_cursor() we
verify the size of the chunked data ahead, checking we can
access 'sizeof(QXLCursor) + chunk->data_size' bytes.
Since in the SPICE_CURSOR_TYPE_MONO case the cursor is
assumed to fit in one chunk, no change are required.
In SPICE_CURSOR_TYPE_ALPHA the ahead read is handled in
qxl_unpack_chunks().
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-4-philmd@linaro.org>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: 61c34fc194
References: bac#1205808, CVE-2022-4144
Only 3 command types are logged: no need to call qxl_phys2virt()
for the other types. Using different cases will help to pass
different structure sizes to qxl_phys2virt() in a pair of commits.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-2-philmd@linaro.org>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: 979f7ef896
References: bsc#1198035, CVE-2021-4206
When processing monitor config from guest store head0 width and height
for single-head configurations. Use these when creating the
DisplaySurface in the local renderer.
This fixes a rendering issue with wayland. Wayland rounds up the
framebuffer width and height to a multiple of 64, so with odd
resolutions (800x600 for example) the framebuffer is larger than the
actual screen. The monitor config has the actual screen size though.
This fixes guest display for anything using the local renderer
(non-spice UI, screendump monitor command).
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20180919103057.9666-1-kraxel@redhat.com
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: defac5e2fb
References: bsc#1185000, CVE-2021-3507
Per the 82078 datasheet, if the end-of-track (EOT byte in
the FIFO) is more than the number of sectors per side, the
command is terminated unsuccessfully:
* 5.2.5 DATA TRANSFER TERMINATION
The 82078 supports terminal count explicitly through
the TC pin and implicitly through the underrun/over-
run and end-of-track (EOT) functions. For full sector
transfers, the EOT parameter can define the last
sector to be transferred in a single or multisector
transfer. If the last sector to be transferred is a par-
tial sector, the host can stop transferring the data in
mid-sector, and the 82078 will continue to complete
the sector as if a hardware TC was received. The
only difference between these implicit functions and
TC is that they return "abnormal termination" result
status. Such status indications can be ignored if they
were expected.
* 6.1.3 READ TRACK
This command terminates when the EOT specified
number of sectors have been read. If the 82078
does not find an I D Address Mark on the diskette
after the second· occurrence of a pulse on the
INDX# pin, then it sets the IC code in Status Regis-
ter 0 to "01" (Abnormal termination), sets the MA bit
in Status Register 1 to "1", and terminates the com-
mand.
* 6.1.6 VERIFY
Refer to Table 6-6 and Table 6-7 for information
concerning the values of MT and EC versus SC and
EOT value.
* Table 6·6. Result Phase Table
* Table 6-7. Verify Command Result Phase Table
Fix by aborting the transfer when EOT > # Sectors Per Side.
Cc: qemu-stable@nongnu.org
Cc: Hervé Poussineau <hpoussin@reactos.org>
Fixes: baca51faff ("floppy driver: disk geometry auto detect")
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/339
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20211118115733.4038610-2-philmd@redhat.com>
Reviewed-by: Hanna Reitz <hreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: b263d8f928
References: bsc#1175144
At the end of sdhci_send_command(), it starts a data transfer if the
command register indicates data is associated. But the data transfer
should only be initiated when the command execution has succeeded.
With this fix, the following reproducer:
outl 0xcf8 0x80001810
outl 0xcfc 0xe1068000
outl 0xcf8 0x80001804
outw 0xcfc 0x7
write 0xe106802c 0x1 0x0f
write 0xe1068004 0xc 0x2801d10101fffffbff28a384
write 0xe106800c 0x1f 0x9dacbbcad9e8f7061524334251606f7e8d9cabbac9d8e7f60514233241505f
write 0xe1068003 0x28 0x80d000251480d000252280d000253080d000253e80d000254c80d000255a80d000256880d0002576
write 0xe1068003 0x1 0xfe
cannot be reproduced with the following QEMU command line:
$ qemu-system-x86_64 -nographic -M pc-q35-5.0 \
-device sdhci-pci,sd-spec-version=3 \
-drive if=sd,index=0,file=null-co://,format=raw,id=mydrive \
-device sd-card,drive=mydrive \
-monitor none -serial none -qtest stdio
Cc: qemu-stable@nongnu.org
Fixes: CVE-2020-17380
Fixes: CVE-2020-25085
Fixes: CVE-2021-3409
Fixes: d7dfca0807 ("hw/sdhci: introduce standard SD host controller")
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Reported-by: Cornelius Aschermann (Ruhr-Universität Bochum)
Reported-by: Sergej Schumilo (Ruhr-Universität Bochum)
Reported-by: Simon Wörner (Ruhr-Universität Bochum)
Buglink: https://bugs.launchpad.net/qemu/+bug/1892960
Buglink: https://bugs.launchpad.net/qemu/+bug/1909418
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1928146
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Tested-by: Alexander Bulekov <alxndr@bu.edu>
Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Message-Id: <20210303122639.20004-2-bmeng.cn@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Dario Faggioli <dfaggioli@suse.com>
Git-commit: ffd205ef29
References: bsc#1192463
This partially reverts commit bbd2d5a812.
This commit was misguided and broke using --disable-pie on any distro
that enables PIE by default in their compiler driver, including Debian
and its derivatives. Whilst -no-pie is not a linker flag, it is a
compiler driver flag that ensures -pie is not automatically passed by it
to the linker. Without it, all compile_prog checks will fail as any code
built with the explicit -fno-pie will fail to link with the implicit
default -pie due to trying to use position-dependent relocations. The
only bug that needed fixing was LDFLAGS_NOPIE being used as a flag for
the linker itself in pc-bios/optionrom/Makefile.
Note this does not reinstate exporting LDFLAGS_NOPIE, as it is unused,
since the only previous use was the one that should not have existed. I
have also updated the comment for the -fno-pie and -no-pie checks to
reflect what they're actually needed for.
Fixes: bbd2d5a812
Cc: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Jessica Clarke <jrtc27@jrtc27.com>
Message-Id: <20210805192545.38279-1-jrtc27@jrtc27.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Liang Yan <lyan@suse.com>
Git-commit: bbd2d5a812
References: bsc#1192463
Recent binutils changes dropping unsupported options [1] caused a build
issue in regard to the optionroms.
ld -m elf_i386 -T /<<PKGBUILDDIR>>/pc-bios/optionrom//flat.lds -no-pie \
-s -o multiboot.img multiboot.o
ld.bfd: Error: unable to disambiguate: -no-pie (did you mean --no-pie ?)
This isn't really a regression in ld.bfd, filing the bug upstream
revealed that this never worked as a ld flag [2] - in fact it seems we
were by accident setting --nmagic).
Since it never had the wanted effect this usage of LDFLAGS_NOPIE, should be
droppable without any effect. This also is the only use-case of LDFLAGS_NOPIE
in .mak, therefore we can also remove it from being added there.
[1]: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=983d925d
[2]: https://sourceware.org/bugzilla/show_bug.cgi?id=27050#c5
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Message-Id: <20201214150938.1297512-1-christian.ehrhardt@canonical.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Liang Yan <lyan@suse.com>
Git-commit: bedd7e93d0
References: bsc#1189938 CVE-2021-3748
When mergeable buffer is enabled, we try to set the num_buffers after
the virtqueue elem has been unmapped. This will lead several issues,
E.g a use after free when the descriptor has an address which belongs
to the non direct access region. In this case we use bounce buffer
that is allocated during address_space_map() and freed during
address_space_unmap().
Fixing this by storing the elems temporarily in an array and delay the
unmap after we set the the num_buffers.
This addresses CVE-2021-3748.
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Fixes: fbe78f4f55 ("virtio-net support")
Cc: qemu-stable@nongnu.org
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 05a40b172e
References: bsc#1186012, CVE-2021-3527
usb-host and usb-redirect try to batch bulk transfers by combining many
small usb packets into a single, large transfer request, to reduce the
overhead and improve performance.
This patch adds a size limit of 1 MiB for those combined packets to
restrict the host resources the guest can bind that way.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-Id: <20210503132915.2335822-6-kraxel@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 000000000000000000000000000000000000000000000
References: bsc#1182651, CVE-2021-20255
Patch based on discussion:
https://lists.gnu.org/archive/html/qemu-devel/2021-02/msg06098.html
While processing controller commands, eepro100 emulator gets
command unit(CU) base address OR receive unit (RU) base address
OR command block (CB) address from guest. If these values are not
checked, it may lead to an infinite loop kind of issues. Add checks
to avoid it.
Reported-by: Ruhr-University Bochum <bugs-syssec@rub.de>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Acked-By: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 00000000000000000000000000000000000000000000
References: bsc#1180432, CVE-2020-35503
Ensure that 'cmd->frame' is not NULL before accessing the 'header' field.
This check prevents a potential NULL pointer dereference issue.
RHBZ: https://bugzilla.redhat.com/show_bug.cgi?id=1910346
Signed-off-by: Mauro Matteo Cascella <mcascell@redhat.com>
Reported-by: Cheolwoo Myung <cwmyung@snu.ac.kr>
Acked-By: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: d7fb54218424c3b2517aee5b391ced0f75386a5d
References: bsc#1187364, CVE-2021-3592
RFC2131 suggests that the options field may be at least 312 bytes.
Some DHCP clients seem to assume that it has to be at least 312 bytes.
Fixes#51
Fixes: f13cad45b25d92760bb0ad67bec0300a4d7d5275 ("bootp: limit
vendor-specific area to input packet memory buffer")
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: f13cad45b25d92760bb0ad67bec0300a4d7d5275
References: bsc#1187364, CVE-2021-3592
sizeof(bootp_t) currently holds DHCP_OPT_LEN. Remove this optional field
from the structure, to help with the following patch checking for
minimal header size. Modify the bootp_reply() function to take the
buffer boundaries and avoiding potential buffer overflow.
Related to CVE-2021-3592.
https://gitlab.freedesktop.org/slirp/libslirp/-/issues/44
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git commit: 6fd057f7960bc7b3a69f3c53de5c9d0d5d34a79c
References: bsc#1187364, CVE-2021-3592
When user provides a long domainname or hostname that doesn't fit in the
DHCP packet, we mustn't overflow the response packet buffer. Instead,
report errors, following the g_warning() in the slirp->vdnssearch
branch.
Also check the strlen against 256 when initializing slirp, which limit
is also from the protocol where one byte represents the string length.
This gives an early error before the warning which is harder to notice
or diagnose.
Reported-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Fam Zheng <famz@redhat.com>
Tested-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 93e645e72a056ec0b2c16e0299fc5c6b94e4ca17
References: bsc#1187364, CVE-2021-3592
bsc#1187367, CVE-2021-3594
Recent security issues demonstrate the lack of safety care when casting
a mbuf to a particular structure type. At least, it should check that
the buffer is large enough. The following patches will make use of this
function.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 1bf8b88f14
References: bsc#1187529 CVE-2021-3611
Object property insertion code iterates over an integer to get an unused
index that can be used as an unique name for an object property. This loop
increments the integer value indefinitely. Although very unlikely, this can
still cause an integer overflow.
In this change, we fix the above code by checking against INT16_MAX and making
sure that the interger index does not overflow beyond that value. If no
available index is found, the code would cause an assertion failure. This
assertion failure is necessary because the callers of the function do not check
the return value for NULL.
Signed-off-by: Ani Sinha <ani@anisinha.ca>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20200921093325.25617-1-ani@anisinha.ca>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Cho, Yu-Chen <acho@suse.com>
Git-commit: 3059344f01
References: bsc#1172382 CVE-2020-13754
The "BCM2835 ARM Peripherals" datasheet [*] chapter 2
("Auxiliaries: UART1 & SPI1, SPI2"), list the register
sizes as 3/8/16/32 bits. We assume this means this
peripheral allows 8-bit accesses.
This was not an issue until commit 5d971f9e67 which reverted
("memory: accept mismatching sizes in memory_region_access_valid").
The model is implemented as 32-bit accesses (see commit 97398d900c,
all registers are 32-bit) so replace MemoryRegionOps.valid as
MemoryRegionOps.impl, and re-introduce MemoryRegionOps.valid
with a 8/32-bit range.
[*] https://www.raspberrypi.org/app/uploads/2012/02/BCM2835-ARM-Peripherals.pdf
Fixes: 97398d900c ("bcm2835_aux: add emulation of BCM2835 AUX (aka UART1) block")
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20201002181032.1899463-1-f4bug@amsat.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 8e67fda2dd
References: bsc#1172382 CVE-2020-13754
QEMU XHCI advertises AC64 (64-bit addressing) but doesn't allow
64-bit mode access in "runtime" and "operational" MemoryRegionOps.
Set the max_access_size based on sizeof(dma_addr_t) as AC64 is set.
XHCI specs:
"If the xHC supports 64-bit addressing (AC64 = ‘1’), then software
should write 64-bit registers using only Qword accesses. If a
system is incapable of issuing Qword accesses, then writes to the
64-bit address fields shall be performed using 2 Dword accesses;
low Dword-first, high-Dword second. If the xHC supports 32-bit
addressing (AC64 = ‘0’), then the high Dword of registers containing
64-bit address fields are unused and software should write addresses
using only Dword accesses"
The problem has been detected with SLOF, as linux kernel always accesses
registers using 32-bit access even if AC64 is set and revealed by
5d971f9e67 ("memory: Revert "memory: accept mismatching sizes in memory_region_access_valid"")
Suggested-by: Alexey Kardashevskiy <aik@au1.ibm.com>
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Message-id: 20200721083322.90651-1-lvivier@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: f22a57ac09
References: bsc#1172382 CVE-2020-13754
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1737400
Fixed setting max_queue_num if there are no peers in
NICConf. qemu_new_nic() creates NICState with 1 NetClientState(index
0) without peers, set max_queue_num to 0 - It prevents undefined
behavior and possible crashes, especially during pcie hotplug.
Fixes: 6f3fbe4ed0 ("net: Introduce e1000e device emulation")
Signed-off-by: Andrew Melnychenko <andrew@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Dmitry Fleytman <dmitry.fleytman@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 5d971f9e67
References: bsc#1172382 CVE-2020-13754
Memory API documentation documents valid .min_access_size and .max_access_size
fields and explains that any access outside these boundaries is blocked.
This is what devices seem to assume.
However this is not what the implementation does: it simply
ignores the boundaries unless there's an "accepts" callback.
Naturally, this breaks a bunch of devices.
Revert to the documented behaviour.
Devices that want to allow any access can just drop the valid field,
or add the impl field to have accesses converted to appropriate
length.
Cc: qemu-stable@nongnu.org
Reviewed-by: Richard Henderson <rth@twiddle.net>
Fixes: CVE-2020-13754
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1842363
Fixes: a014ed07bd ("memory: accept mismatching sizes in memory_region_access_valid")
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20200610134731.1514409-1-mst@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jose R. Ziviani <jose.ziviani@suse.com>
Git-commit: 89ed83d8b2
References: bsc#1172382 CVE-2020-13754
The memory region ops have min_access_size == 4 so obey it.
Tested-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: 4b7c06837a
References: bsc#1172382 CVE-2020-13754
The memory region ops have min_access_size == 4 so obey it.
Tested-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: c7ede54cbd2e2b25385325600958ba0124e31cc0
References: bsc#1172380 CVE-2020-10756
Drop IPv6 message shorter than what's mentioned in the payload
length header (+ the size of the IPv6 header). They're invalid an could
lead to data leakage in icmp6_send_echoreply().
Signed-off-by: Jose R Ziviani <jose.ziviani@suse.com>
Git-commit: edfe2eb436
References: bsc#1181933 CVE-2021-20221
Per the ARM Generic Interrupt Controller Architecture specification
(document "ARM IHI 0048B.b (ID072613)"), the SGIINTID field is 4 bit,
not 10:
- 4.3 Distributor register descriptions
- 4.3.15 Software Generated Interrupt Register, GICD_SG
- Table 4-21 GICD_SGIR bit assignments
The Interrupt ID of the SGI to forward to the specified CPU
interfaces. The value of this field is the Interrupt ID, in
the range 0-15, for example a value of 0b0011 specifies
Interrupt ID 3.
Correct the irq mask to fix an undefined behavior (which eventually
lead to a heap-buffer-overflow, see [Buglink]):
$ echo 'writel 0x8000f00 0xff4affb0' | qemu-system-aarch64 -M virt,accel=qtest -qtest stdio
[I 1612088147.116987] OPENED
[R +0.278293] writel 0x8000f00 0xff4affb0
../hw/intc/arm_gic.c:1498:13: runtime error: index 944 out of bounds for type 'uint8_t [16][8]'
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior ../hw/intc/arm_gic.c:1498:13
This fixes a security issue when running with KVM on Arm with
kernel-irqchip=off. (The default is kernel-irqchip=on, which is
unaffected, and which is also the correct choice for performance.)
Cc: qemu-stable@nongnu.org
Fixes: CVE-2021-20221
Fixes: 9ee6e8bb85 ("ARMv7 support.")
Buglink: https://bugs.launchpad.net/qemu/+bug/1913916
Buglink: https://bugs.launchpad.net/qemu/+bug/1913917
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20210131103401.217160-1-f4bug@amsat.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: b946434f26
References: bsc#1175441, CVE-2020-14364
Store calculated setup_len in a local variable, verify it, and only
write it to the struct (USBDevice->setup_len) in case it passed the
sanity checks.
This prevents other code (do_token_{in,out} functions specifically)
from working with invalid USBDevice->setup_len values and overrunning
the USBDevice->setup_buf[] buffer.
Fixes: CVE-2020-14364
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Tested-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-id: 20200825053636.29648-1-kraxel@redhat.com
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 035e69b063
References: bsc#1174641, CVE-2020-16092
An assertion failure issue was found in the code that processes network packets
while adding data fragments into the packet context. It could be abused by a
malicious guest to abort the QEMU process on the host. This patch replaces the
affected assert() with a conditional statement, returning false if the current
data fragment exceeds max_raw_frags.
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Reported-by: Ziming Zhang <ezrakiez@gmail.com>
Reviewed-by: Dmitry Fleytman <dmitry.fleytman@gmail.com>
Signed-off-by: Mauro Matteo Cascella <mcascell@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 5519724a13
References: bsc#1174386 CVE-2020-15863
A buffer overflow issue was reported by Mr. Ziming Zhang, CC'd here. It
occurs while sending an Ethernet frame due to missing break statements
and improper checking of the buffer size.
Reported-by: Ziming Zhang <ezrakiez@gmail.com>
Signed-off-by: Mauro Matteo Cascella <mcascell@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 0000000000000000000000000000000000000000
References: bsc#1181639
While activating device in vmxnet3_acticate_device(), it does not
validate guest supplied configuration values against predefined
minimum - maximum limits. This may lead to integer overflow or
OOB access issues. Add checks to avoid it.
Fixes: CVE-2021-20203
Buglink: https://bugs.launchpad.net/qemu/+bug/1913873
Reported-by: Gaoning Pan <pgn@zju.edu.cn>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: ff0507c239
References: bsc#1180523, CVE-2020-11947
There is an overflow, the source 'datain.data[2]' is 100 bytes,
but the 'ss' is 252 bytes.This may cause a security issue because
we can access a lot of unrelated memory data.
The len for sbp copy data should take the minimum of mx_sb_len and
sb_len_wr, not the maximum.
If we use iscsi device for VM backend storage, ASAN show stack:
READ of size 252 at 0xfffd149dcfc4 thread T0
#0 0xaaad433d0d34 in __asan_memcpy (aarch64-softmmu/qemu-system-aarch64+0x2cb0d34)
#1 0xaaad45f9d6d0 in iscsi_aio_ioctl_cb /qemu/block/iscsi.c:996:9
#2 0xfffd1af0e2dc (/usr/lib64/iscsi/libiscsi.so.8+0xe2dc)
#3 0xfffd1af0d174 (/usr/lib64/iscsi/libiscsi.so.8+0xd174)
#4 0xfffd1af19fac (/usr/lib64/iscsi/libiscsi.so.8+0x19fac)
#5 0xaaad45f9acc8 in iscsi_process_read /qemu/block/iscsi.c:403:5
#6 0xaaad4623733c in aio_dispatch_handler /qemu/util/aio-posix.c:467:9
#7 0xaaad4622f350 in aio_dispatch_handlers /qemu/util/aio-posix.c:510:20
#8 0xaaad4622f350 in aio_dispatch /qemu/util/aio-posix.c:520
#9 0xaaad46215944 in aio_ctx_dispatch /qemu/util/async.c:298:5
#10 0xfffd1bed12f4 in g_main_context_dispatch (/lib64/libglib-2.0.so.0+0x512f4)
#11 0xaaad46227de0 in glib_pollfds_poll /qemu/util/main-loop.c:219:9
#12 0xaaad46227de0 in os_host_main_loop_wait /qemu/util/main-loop.c:242
#13 0xaaad46227de0 in main_loop_wait /qemu/util/main-loop.c:518
#14 0xaaad43d9d60c in qemu_main_loop /qemu/softmmu/vl.c:1662:9
#15 0xaaad4607a5b0 in main /qemu/softmmu/main.c:49:5
#16 0xfffd1a460b9c in __libc_start_main (/lib64/libc.so.6+0x20b9c)
#17 0xaaad43320740 in _start (aarch64-softmmu/qemu-system-aarch64+0x2c00740)
0xfffd149dcfc4 is located 0 bytes to the right of 100-byte region [0xfffd149dcf60,0xfffd149dcfc4)
allocated by thread T0 here:
#0 0xaaad433d1e70 in __interceptor_malloc (aarch64-softmmu/qemu-system-aarch64+0x2cb1e70)
#1 0xfffd1af0e254 (/usr/lib64/iscsi/libiscsi.so.8+0xe254)
#2 0xfffd1af0d174 (/usr/lib64/iscsi/libiscsi.so.8+0xd174)
#3 0xfffd1af19fac (/usr/lib64/iscsi/libiscsi.so.8+0x19fac)
#4 0xaaad45f9acc8 in iscsi_process_read /qemu/block/iscsi.c:403:5
#5 0xaaad4623733c in aio_dispatch_handler /qemu/util/aio-posix.c:467:9
#6 0xaaad4622f350 in aio_dispatch_handlers /qemu/util/aio-posix.c:510:20
#7 0xaaad4622f350 in aio_dispatch /qemu/util/aio-posix.c:520
#8 0xaaad46215944 in aio_ctx_dispatch /qemu/util/async.c:298:5
#9 0xfffd1bed12f4 in g_main_context_dispatch (/lib64/libglib-2.0.so.0+0x512f4)
#10 0xaaad46227de0 in glib_pollfds_poll /qemu/util/main-loop.c:219:9
#11 0xaaad46227de0 in os_host_main_loop_wait /qemu/util/main-loop.c:242
#12 0xaaad46227de0 in main_loop_wait /qemu/util/main-loop.c:518
#13 0xaaad43d9d60c in qemu_main_loop /qemu/softmmu/vl.c:1662:9
#14 0xaaad4607a5b0 in main /qemu/softmmu/main.c:49:5
#15 0xfffd1a460b9c in __libc_start_main (/lib64/libc.so.6+0x20b9c)
#16 0xaaad43320740 in _start (aarch64-softmmu/qemu-system-aarch64+0x2c00740)
Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Chen Qun <kuhn.chenqun@huawei.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 20200418062602.10776-1-kuhn.chenqun@huawei.com
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 2e1dcbc0c2af64fcb17009eaf2ceedd81be2b27f
References: bsc#1179467
While processing ARP/NCSI packets in 'arp_input' or 'ncsi_input'
routines, ensure that pkt_len is large enough to accommodate the
respective protocol headers, lest it should do an OOB access.
Add check to avoid it.
CVE-2020-29129 CVE-2020-29130
QEMU: slirp: out-of-bounds access while processing ARP/NCSI packets
-> https://www.openwall.com/lists/oss-security/2020/11/27/1
Reported-by: Qiuhao Li <Qiuhao.Li@outlook.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-Id: <20201126135706.273950-1-ppandit@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
[BR: no NCSI code, so only ARP patched]
Git-commit: 705df5466c
References: bsc#1182968, CVE-2021-3416
Some NIC supports loopback mode and this is done by calling
nc->info->receive() directly which in fact suppresses the effort of
reentrancy check that is done in qemu_net_queue_send().
Unfortunately we can't use qemu_net_queue_send() here since for
loopback there's no sender as peer, so this patch introduce a
qemu_receive_packet() which is used for implementing loopback mode
for a NIC with this check.
NIC that supports loopback mode will be converted to this helper.
This is intended to address CVE-2021-3416.
Cc: Prasad J Pandit <ppandit@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 3de46e6fc4
References: bsc#1182577, CVE-2021-20257
During procss_tx_desc(), driver can try to chain data descriptor with
legacy descriptor, when will lead underflow for the following
calculation in process_tx_desc() for bytes:
if (tp->size + bytes > msh)
bytes = msh - tp->size;
This will lead a infinite loop. So check and fail early if tp->size if
greater or equal to msh.
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Reported-by: Cheolwoo Myung <cwmyung@snu.ac.kr>
Reported-by: Ruhr-University Bochum <bugs-syssec@rub.de>
Cc: Prasad J Pandit <ppandit@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 8132122889
References: bsc#1181108, CVE-2020-29443
A case was reported where s->io_buffer_index can be out of range.
The report skimped on the details but it seems to be triggered
by s->lba == -1 on the READ/READ CD paths (e.g. by sending an
ATAPI command with LBA = 0xFFFFFFFF). For now paper over it
with assertions. The first one ensures that there is no overflow
when incrementing s->io_buffer_index, the second checks for the
buffer overrun.
Note that the buffer overrun is only a read, so I am not sure
if the assertion failure is actually less harmful than the overrun.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20201201120926.56559-1-pbonzini@redhat.com
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: c2cb511634
References: bsc#1179468, CVE-2020-28916
While receiving packets via e1000e_write_packet_to_guest() routine,
'desc_offset' is advanced only when RX descriptor is processed. And
RX descriptor is not processed if it has NULL buffer address.
This may lead to an infinite loop condition. Increament 'desc_offset'
to process next descriptor in the ring to avoid infinite loop.
Reported-by: Cheol-woo Myung <330cjfdn@gmail.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 7564bf7701
References: bsc#1178174, CVE-2020-27617
eth_get_gso_type() routine returns segmentation offload type based on
L3 protocol type. It calls g_assert_not_reached if L3 protocol is
unknown, making the following return statement unreachable. Remove the
g_assert call, it maybe triggered by a guest user.
Reported-by: Gaoning Pan <pgn@zju.edu.cn>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 1be90ebecc
References: bsc#1176684, CVE-2020-25625
While servicing OHCI transfer descriptors(TD), ohci_service_iso_td
retires a TD if it has passed its time frame. It does not check if
the TD was already processed once and holds an error code in TD_CC.
It may happen if the TD list has a loop. Add check to avoid an
infinite loop condition.
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-id: 20200915182259.68522-3-ppandit@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: f50ab86a26
References: bsc#1172383, CVE-2020-13362
A guest user may set 'reply_queue_head' field of MegasasState to
a negative value. Later in 'megasas_lookup_frame' it is used to
index into s->frames[] array. Use unsigned type to avoid OOB
access issue.
Also check that 'index' value stays within s->frames[] bounds
through the while() loop in 'megasas_lookup_frame' to avoid OOB
access.
Reported-by: Ren Ding <rding@gatech.edu>
Reported-by: Hanqing Zhao <hanqing@gatech.edu>
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Acked-by: Alexander Bulekov <alxndr@bu.edu>
Message-Id: <20200513192540.1583887-2-ppandit@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
References: bsc#1172385, CVE-2020-12829
This is in place of commit 4a1f253adb which added the log.h
reference, but did other changes we're not interested in.
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: f3a60058c9
References: bsc#1172385, CVE-2020-12829
The value from twoD_foreground (which is in host endian format) must
be converted to the endianness of the framebuffer (currently always
little endian) before it can be used to perform the fill operation.
Signed-off-by: Marcus Comstedt <marcus@mc.pp.se>
Reviewed-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: eb76243c9d
References: bsc#1172385, CVE-2020-12829
Set the changed memory region dirty after performed a 2D operation to
ensure that the screen is updated properly.
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 33159dd7ce
References: bsc#1172385, CVE-2020-12829
Display updates and drawing hardware cursor did not work when frame
buffer address was non-zero. Fix this by taking the frame buffer
address into account in these cases. This fixes screen dragging on
AmigaOS. Based on patch by Sebastian Bauer.
Signed-off-by: Sebastian Bauer <mail@sebastianbauer.info>
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 06cb926aaa
References: bsc#1172385, CVE-2020-12829
The sm501 currently implements only a very limited set of raster operation
modes. After this change, unknown raster operation modes are logged so
these can be easily spotted.
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: debc7e7dad
References: bsc#1172385, CVE-2020-12829
Add support for the negated destination operation mode. This is used e.g.
by AmigaOS for the INVERSEVID drawing mode. With this change, the cursor
in the shell and non-immediate window adjustment are working now.
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 54b2a4339c
References: bsc#1172385, CVE-2020-12829
Before, crt_h_total was used for src_width and dst_width. This is a
property of the current display setting and not relevant for the 2D
operation that also can be done off-screen. The pitch register's purpose
is to describe line pitch relevant of the 2D operation.
Signed-off-by: Sebastian Bauer <mail@sebastianbauer.info>
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 5690d9ecef
References: bsc#1172385, CVE-2020-12829
These are not really implemented (just return zero or default values)
but add these so guests accessing them can run.
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 6a2a5aae02
References: bsc#1172385, CVE-2020-12829
Rework HWC handling to simplify it and fix cursor not updating on
screen as needed. Previously cursor was not updated because checking
for changes in a line overrode the update flag set for the cursor but
fixing this is not enough because the cursor should also be updated if
its shape or location changes. Introduce hwc_invalidate() function to
handle that similar to other display controller models.
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Tested-by: Aurelien Jarno <aurelien@aurel32.net>
Message-id: 6970a5e9868b7246656c1d02038dc5d5fa369507.1492787889.git.balaton@eik.bme.hu
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: e423455c4f
References: bsc#1172478, CVE-2020-13765
Both, "rom->addr" and "addr" are derived from the binary image
that can be loaded with the "-kernel" paramer. The code in
rom_copy() then calculates:
d = dest + (rom->addr - addr);
and uses "d" as destination in a memcpy() some lines later. Now with
bad kernel images, it is possible that rom->addr is smaller than addr,
thus "rom->addr - addr" gets negative and the memcpy() then tries to
copy contents from the image to a bad memory location. This could
maybe be used to inject code from a kernel image into the QEMU binary,
so we better fix it with an additional sanity check here.
Cc: qemu-stable@nongnu.org
Reported-by: Guangming Liu
Buglink: https://bugs.launchpad.net/qemu/+bug/1844635
Message-Id: <20190925130331.27825-1-thuth@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 369ff955a8
References: bsc#1172384, CVE-2020-13361
A guest user may set channel frame count via es1370_write()
such that, in es1370_transfer_audio(), total frame count
'size' is lesser than the number of frames that are processed
'cnt'.
int cnt = d->frame_cnt >> 16;
int size = d->frame_cnt & 0xffff;
if (size < cnt), it results in incorrect calculations leading
to OOB access issue(s). Add check to avoid it.
Reported-by: Ren Ding <rding@gatech.edu>
Reported-by: Hanqing Zhao <hanqing@gatech.edu>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-id: 20200514200608.1744203-1-ppandit@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Bruce Rogers <brogers@suse.com>
When executing script in lsi_execute_script(), the LSI scsi adapter
emulator advances 's->dsp' index to read next opcode. This can lead
to an infinite loop if the next opcode is empty. Move the existing
loop exit after 10k iterations so that it covers no-op opcodes as
well.
Reported-by: Bugs SysSec <bugs-syssec@rub.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit de594e4765)
[BR: BSC#1146873 CVE-2019-12068 - minor trace related tweak applied]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 693fd2acdf
References: bsc#1166240
When querying an iSCSI server for the provisioning status of blocks (via
GET LBA STATUS), Qemu only validates that the response descriptor zero's
LBA matches the one requested. Given the SCSI spec allows servers to
respond with the status of blocks beyond the end of the LUN, Qemu may
have its heap corrupted by clearing/setting too many bits at the end of
its allocmap for the LUN.
A malicious guest in control of the iSCSI server could carefully program
Qemu's heap (by selectively setting the bitmap) and then smash it.
This limits the number of bits that iscsi_co_block_status() will try to
update in the allocmap so it can't overflow the bitmap.
Fixes: CVE-2020-1711
Cc: qemu-stable@nongnu.org
Signed-off-by: Felipe Franciosi <felipe@nutanix.com>
Signed-off-by: Peter Turschmid <peter.turschm@nutanix.com>
Signed-off-by: Raphael Norwitz <raphael.norwitz@nutanix.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
[BR: Converted patch from byte based to sector based]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 2faae0f778f818fadc873308f983289df697eb93
References: bsc#1170940, CVE-2020-1983
The q pointer is updated when the mbuf data is moved from m_dat to
m_ext.
m_ext buffer may also be realloc()'ed and moved during m_cat():
q should also be updated in this case.
Reported-by: Aviv Sasson <asasson@paloaltonetworks.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit 9bd6c5913271eabcb7768a58197ed3301fe19f2d)
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 68ccb8021a838066f0951d4b2817eb6b6f10a843
References: bsc#1163018, CVE-2020-8608
Various calls to snprintf() assume that snprintf() returns "only" the
number of bytes written (excluding terminating NUL).
https://pubs.opengroup.org/onlinepubs/9699919799/functions/snprintf.html#tag_16_159_04
"Upon successful completion, the snprintf() function shall return the
number of bytes that would be written to s had n been sufficiently
large excluding the terminating null byte."
Before patch ce131029, if there isn't enough room in "m_data" for the
"DCC ..." message, we overflow "m_data".
After the patch, if there isn't enough room for the same, we don't
overflow "m_data", but we set "m_len" out-of-bounds. The next time an
access is bounded by "m_len", we'll have a buffer overflow then.
Use slirp_fmt*() to fix potential OOB memory access.
Reported-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Message-Id: <20200127092414.169796-7-marcandre.lureau@redhat.com>
[Since porting from a quite different now libslirp, quite a bit changed]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 82ebe9c370a0e2970fb5695aa19aa5214a6a1c80
References: bsc#1161066, CVE-2020-7039, bsc#1163018, CVE-2020-8608
While emulating services in tcp_emu(), it uses 'mbuf' size
'm->m_size' to write commands via snprintf(3). Use M_FREEROOM(m)
size to avoid possible OOB access.
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Message-Id: <20200109094228.79764-3-ppandit@redhat.com>
[Since porting from a quite different now libslirp, quite a bit changed]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: ce131029d6d4a405cb7d3ac6716d03e58fb4a5d9
References: bsc#1161066, CVE-2020-7039, bsc#1163018, CVE-2020-8608
While emulating IRC DCC commands, tcp_emu() uses 'mbuf' size
'm->m_size' to write DCC commands via snprintf(3). This may
lead to OOB write access, because 'bptr' points somewhere in
the middle of 'mbuf' buffer, not at the start. Use M_FREEROOM(m)
size to avoid OOB access.
Reported-by: Vishnu Dev TJ <vishnudevtj@gmail.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Message-Id: <20200109094228.79764-2-ppandit@redhat.com>
[Since porting from a quite different now libslirp, quite a bit changed]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 2655fffed7a9e765bcb4701dd876e9dab975f289
References: bsc#1161066, CVE2020-7039, bsc#1161066, CVS-2020-7039
The main loop only checks for one available byte, while we sometimes
need two bytes.
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 30648c03b27fb8d9611b723184216cd3174b6775
References: bsc#1163018, CVE-2020-8608
Various calls to snprintf() in libslirp assume that snprintf() returns
"only" the number of bytes written (excluding terminating NUL).
https://pubs.opengroup.org/onlinepubs/9699919799/functions/snprintf.html#tag_16_159_04
"Upon successful completion, the snprintf() function shall return the
number of bytes that would be written to s had n been sufficiently
large excluding the terminating null byte."
Introduce slirp_fmt() that handles several pathological cases the
way libslirp usually expect:
- treat error as fatal (instead of silently returning -1)
- fmt0() will always \0 end
- return the number of bytes actually written (instead of what would
have been written, which would usually result in OOB later), including
the ending \0 for fmt0()
- warn if truncation happened (instead of ignoring)
Other less common cases can still be handled with strcpy/snprintf() etc.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Message-Id: <20200127092414.169796-2-marcandre.lureau@redhat.com>
[Since porting from a quite different now libslirp, quite a bit changed]
Signed-off-by: Bruce Rogers <brogers@suse.com>
For some reason, EMU_IDENT is not like other "emulated" protocols and
tries to reconstitute the original buffer, if it came in multiple
packets. Unfortunately, it does so wrongly, as it doesn't respect the
sbuf circular buffer appending rules, nor does it maintain some of the
invariants (rptr is incremented without bounds, etc): this leads to
further memory corruption revealed by ASAN or various malloc
errors. Furthermore, the so_rcv buffer is regularly flushed, so there
is no guarantee that buffer reconstruction will do what is expected.
Instead, do what the function comment says: "XXX Assumes the whole
command came in one packet", and don't touch so_rcv.
Related to: https://bugzilla.redhat.com/show_bug.cgi?id=1664205
Cc: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Using ip_deq after m_free might read pointers from an allocation reuse.
This would be difficult to exploit, but that is still related with
CVE-2019-14378 which generates fragmented IP packets that would trigger this
issue and at least produce a DoS.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(from libslirp.git commit c59279437eda91841b9d26079c70b8a540d41204)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
[BR: BSC#1149811 CVE-2019-15890]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Under certain circumstances normal xen-mapcache functioning may be broken
by guest's actions. This may lead to either QEMU performing exit() due to
a caught bad pointer (and with QEMU process gone the guest domain simply
appears hung afterwards) or actual use of the incorrect pointer inside
QEMU address space -- a write to unmapped memory is possible. The bug is
hard to reproduce on a i440 machine as multiple DMA sources are required
(though it's possible in theory, using multiple emulated devices), but can
be reproduced somewhat easily on a Q35 machine using an emulated AHCI
controller -- each NCQ queue command slot may be used as an independent
DMA source ex. using READ FPDMA QUEUED command, so a single storage
device on the AHCI controller port will be enough to produce multiple DMAs
(up to 32). The detailed description of the issue follows.
Xen-mapcache provides an ability to map parts of a guest memory into
QEMU's own address space to work with.
There are two types of cache lookups:
- translating a guest physical address into a pointer in QEMU's address
space, mapping a part of guest domain memory if necessary (while trying
to reduce a number of such (re)mappings to a minimum)
- translating a QEMU's pointer back to its physical address in guest RAM
These lookups are managed via two linked-lists of structures.
MapCacheEntry is used for forward cache lookups, while MapCacheRev -- for
reverse lookups.
Every guest physical address is broken down into 2 parts:
address_index = phys_addr >> MCACHE_BUCKET_SHIFT;
address_offset = phys_addr & (MCACHE_BUCKET_SIZE - 1);
MCACHE_BUCKET_SHIFT depends on a system (32/64) and is equal to 20 for
a 64-bit system (which assumed for the further description). Basically,
this means that we deal with 1 MB chunks and offsets within those 1 MB
chunks. All mappings are created with 1MB-granularity, i.e. 1MB/2MB/3MB
etc. Most DMA transfers typically are less than 1MB, however, if the
transfer crosses any 1MB border(s) - than a nearest larger mapping size
will be used, so ex. a 512-byte DMA transfer with the start address
700FFF80h will actually require a 2MB range.
Current implementation assumes that MapCacheEntries are unique for a given
address_index and size pair and that a single MapCacheEntry may be reused
by multiple requests -- in this case the 'lock' field will be larger than
1. On other hand, each requested guest physical address (with 'lock' flag)
is described by each own MapCacheRev. So there may be multiple MapCacheRev
entries corresponding to a single MapCacheEntry. The xen-mapcache code
uses MapCacheRev entries to retrieve the address_index & size pair which
in turn used to find a related MapCacheEntry. The 'lock' field within
a MapCacheEntry structure is actually a reference counter which shows
a number of corresponding MapCacheRev entries.
The bug lies in ability for the guest to indirectly manipulate with the
xen-mapcache MapCacheEntries list via a special sequence of DMA
operations, typically for storage devices. In order to trigger the bug,
guest needs to issue DMA operations in specific order and timing.
Although xen-mapcache is protected by the mutex lock -- this doesn't help
in this case, as the bug is not due to a race condition.
Suppose we have 3 DMA transfers, namely A, B and C, where
- transfer A crosses 1MB border and thus uses a 2MB mapping
- transfers B and C are normal transfers within 1MB range
- and all 3 transfers belong to the same address_index
In this case, if all these transfers are to be executed one-by-one
(without overlaps), no special treatment necessary -- each transfer's
mapping lock will be set and then cleared on unmap before starting
the next transfer.
The situation changes when DMA transfers overlap in time, ex. like this:
|===== transfer A (2MB) =====|
|===== transfer B (1MB) =====|
|===== transfer C (1MB) =====|
time --->
In this situation the following sequence of actions happens:
1. transfer A creates a mapping to 2MB area (lock=1)
2. transfer B (1MB) tries to find available mapping but cannot find one
because transfer A is still in progress, and it has 2MB size + non-zero
lock. So transfer B creates another mapping -- same address_index,
but 1MB size.
3. transfer A completes, making 1st mapping entry available by setting its
lock to 0
4. transfer C starts and tries to find available mapping entry and sees
that 1st entry has lock=0, so it uses this entry but remaps the mapping
to a 1MB size
5. transfer B completes and by this time
- there are two locked entries in the MapCacheEntry list with the SAME
values for both address_index and size
- the entry for transfer B actually resides farther in list while
transfer C's entry is first
6. xen_ram_addr_from_mapcache() for transfer B gets correct address_index
and size pair from corresponding MapCacheRev entry, but then it starts
looking for MapCacheEntry with these values and finds the first entry
-- which belongs to transfer C.
At this point there may be following possible (bad) consequences:
1. xen_ram_addr_from_mapcache() will use a wrong entry->vaddr_base value
in this statement:
raddr = (reventry->paddr_index << MCACHE_BUCKET_SHIFT) +
((unsigned long) ptr - (unsigned long) entry->vaddr_base);
resulting in an incorrent raddr value returned from the function. The
(ptr - entry->vaddr_base) expression may produce both positive and negative
numbers and its actual value may differ greatly as there are many
map/unmap operations take place. If the value will be beyond guest RAM
limits then a "Bad RAM offset" error will be triggered and logged,
followed by exit() in QEMU.
2. If raddr value won't exceed guest RAM boundaries, the same sequence
of actions will be performed for xen_invalidate_map_cache_entry() on DMA
unmap, resulting in a wrong MapCacheEntry being unmapped while DMA
operation which uses it is still active. The above example must
be extended by one more DMA transfer in order to allow unmapping as the
first mapping in the list is sort of resident.
The patch modifies the behavior in which MapCacheEntry's are added to the
list, avoiding duplicates.
Signed-off-by: Alexey Gerasimenko <x1917x@gmail.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 7fb394ad8a)
[BSC#1160024]
If we have a system with xenforeignmemory_map2() implemented
we don't need to save/restore physmap on suspend/restore
anymore. In case we resume a VM without physmap - try to
recreate the physmap during memory region restore phase and
remap map cache entries accordingly. The old code is left
for compatibility reasons.
Signed-off-by: Igor Druzhinin <igor.druzhinin@citrix.com>
Reviewed-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 331b5189d7)
(add only code within XEN_COMPAT_PHYSMAP)
[BSC#1160024]
This new call is trying to update a requested map cache entry
according to the changes in the physmap. The call is searching
for the entry, unmaps it and maps again at the same place using
a new guest address. If the mapping is dummy this call will
make it real.
This function makes use of a new xenforeignmemory_map2() call
with an extended interface that was recently introduced in
libxenforeignmemory [1].
[1] https://www.mail-archive.com/xen-devel@lists.xen.org/msg113007.html
Signed-off-by: Igor Druzhinin <igor.druzhinin@citrix.com>
Reviewed-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 5ba3d75645)
(adjusted to use xenforeignmemory_map instead of xenforeignmemory_map2)
[BSC#1160024]
Dummys are simple anonymous mappings that are placed instead
of regular foreign mappings in certain situations when we need
to postpone the actual mapping but still have to give a
memory region to QEMU to play with.
This is planned to be used for restore on Xen.
Signed-off-by: Igor Druzhinin <igor.druzhinin@citrix.com>
Reviewed-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 759235653d)
[BSC#1160024]
The default cache-clean-interval is set to 10 minutes, in order to lower
the overhead of the qcow2 caches (before the default was 0, i.e.
disabled).
* For non-Linux platforms the default is kept at 0, because
cache-clean-interval is not supported there yet.
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit e957b50b8d)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
The upper limit on the L2 cache size is increased from 1 MB to 32 MB
on Linux platforms, and to 8 MB on other platforms (this difference is
caused by the ability to set intervals for cache cleaning on Linux
platforms only).
This is done in order to allow default full coverage with the L2 cache
for images of up to 256 GB in size (was 8 GB). Note, that only the
needed amount to cover the full image is allocated. The value which is
changed here is just the upper limit on the L2 cache size, beyond which
it will not grow, even if the size of the image will require it to.
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 80668d0fb7)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
Sufficient L2 cache can noticeably improve the performance when using
large images with frequent I/O.
Previously, unless 'cache-size' was specified and was large enough, the
L2 cache was set to a certain size without taking the virtual image size
into account.
Now, the L2 cache assignment is aware of the virtual size of the image,
and will cover the entire image, unless the cache size needed for that is
larger than a certain maximum. This maximum is set to 1 MB by default
(enough to cover an 8 GB image with the default cluster size) but can
be increased or decreased using the 'l2-cache-size' option. This option
was previously documented as the *maximum* L2 cache size, and this patch
makes it behave as such, instead of as a constant size. Also, the
existing option 'cache-size' can limit the sum of both L2 and refcount
caches, as previously.
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit b749562d98)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
The refcount cache size does not need to be set to its minimum value in
read_cache_sizes(), as it is set to at least its minimum value in
qcow2_update_options_prepare().
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 657ada52ab)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
Adding a lookup table for the powers of two, with the appropriate size
prefixes. This is needed when a size has to be stringified, in which
case something like '(1 * KiB)' would become a literal '(1 * (1L << 10))'
string. Powers of two are used very often for sizes, so such a table
will also make it easier and more intuitive to write them.
This table is generatred using the following AWK script:
BEGIN {
suffix="KMGTPE";
for(i=10; i<64; i++) {
val=2**i;
s=substr(suffix, int(i/10), 1);
n=2**(i%10);
pad=21-int(log(n)/log(10));
printf("#define S_%d%siB %*d\n", n, s, pad, val);
}
}
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 540b849261)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
MIN_REFCOUNT_CACHE_SIZE is 4 and the cluster size is guaranteed to be
at most 2MB, so the minimum refcount cache size (in bytes) is always
going to fit in a 32-bit integer.
Coverity doesn't know that, and since we're storing the result in a
uint64_t (*refcount_cache_size) it thinks that we need the 64 bits and
that we probably want to do a 64-bit multiplication to prevent the
result from being truncated.
This is a false positive in this case, but it's a fair warning.
We could do a 64-bit multiplication to get rid of it, but since we
know that a 32-bit variable is enough to store this value let's simply
reuse min_refcount_cache, make it a normal int and stop doing casts.
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 7af5eea9b3)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
The L2 and refcount caches have default sizes that can be overridden
using the l2-cache-size and refcount-cache-size (an additional
parameter named cache-size sets the combined size of both caches).
Unless forced by one of the aforementioned parameters, QEMU will set
the unspecified sizes so that the L2 cache is 4 times larger than the
refcount cache.
This is based on the premise that the refcount metadata needs to be
only a fourth of the L2 metadata to cover the same amount of disk
space. This is incorrect for two reasons:
a) The amount of disk covered by an L2 table depends solely on the
cluster size, but in the case of a refcount block it depends on
the cluster size *and* the width of each refcount entry.
The 4/1 ratio is only valid with 16-bit entries (the default).
b) When we talk about disk space and L2 tables we are talking about
guest space (L2 tables map guest clusters to host clusters),
whereas refcount blocks are used for host clusters (including
L1/L2 tables and the refcount blocks themselves). On a fully
populated (and uncompressed) qcow2 file, image size > virtual size
so there are more refcount entries than L2 entries.
Problem (a) could be fixed by adjusting the algorithm to take into
account the refcount entry width. Problem (b) could be fixed by
increasing a bit the refcount cache size to account for the clusters
used for qcow2 metadata.
However this patch takes a completely different approach and instead
of keeping a ratio between both cache sizes it assigns as much as
possible to the L2 cache and the remainder to the refcount cache.
The reason is that L2 tables are used for every single I/O request
from the guest and the effect of increasing the cache is significant
and clearly measurable. Refcount blocks are however only used for
cluster allocation and internal snapshots and in practice are accessed
sequentially in most cases, so the effect of increasing the cache is
negligible (even when doing random writes from the guest).
So, make the refcount cache as small as possible unless the user
explicitly asks for a larger one.
Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 9695182c2eb11b77cb319689a1ebaa4e7c9d6591.1523968389.git.berto@igalia.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 52253998ec)
[LM: BSC#1139926]
Signed-off-by: Lin Ma <lma@suse.com>
When the first fragment does not fit in the preallocated buffer, q will
already be pointing to the ext buffer, so we mustn't try to update it.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit 126c04acbabd7ad32c2b018fe10dfac2a3bc1210)
[LY: CVE-2019-14378 BSC#1143794]
Signed-off-by: Liang Yan <lyan@suse.com>
The interface names in qemu-bridge-helper are defined to be
of size IFNAMSIZ(=16), including the terminating null('\0') byte.
The same is applied to interface names read from 'bridge.conf'
file to form ACLs rules. If user supplied '--br=bridge' name
is not restricted to the same length, it could lead to ACL bypass
issue. Restrict bridge name to IFNAMSIZ, including null byte.
Reported-by: Riccardo Schirone <rschiron@redhat.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
[LY: BSC#1140402 CVE-2019-13164]
Signed-off-by: Liang Yan <lyan@suse.com>
When releasing spice resources in release_resource() routine,
if release info object 'ext.info' is null, it leads to null
pointer dereference. Add check to avoid it.
Reported-by: Bugs SysSec <bugs-syssec@rub.de>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-id: 20190425063534.32747-1-ppandit@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit d52680fc93)
[LY: BSC#1135902 CVE-2019-12155]
Signed-off-by: Liang Yan <lyan@suse.com>
This reverts commit 4c011c37ec.
The commit causes speed setting of migration doesn't work while vm uses
hugepage.
Dropping this commit is harmless due to postcopy needs kernel 4.11+ support.
(bsc#1127077)
md-clear is a new CPUID bit which is set when microcode provides the
mechanism to invoke a flush of various exploitable CPU buffers by invoking
the VERW instruction. Add the new feature.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[BR: BSC#1111331 CVE-2018-12126 CVE-2018-12127 CVE-2018-12130
CVE-2019-11091]
Signed-off-by: Bruce Rogers <brogers@suse.com>
27461d69a0 "ppc: add host-serial and host-model machine attributes
(CVE-2019-8934)" introduced 'host-serial' and 'host-model' machine
properties for spapr to explicitly control the values advertised to the
guest in device tree properties with the same names.
The previous behaviour on KVM was to unconditionally populate the device
tree with the real host serial number and model, which leaks possibly
sensitive information about the host to the guest.
To maintain compatibility for old machine types, we allowed those props
to be set to "passthrough" to take the value from the host as before. Or
they could be set to "none" to explicitly omit the device tree items.
Special casing specific values on what's otherwise a user supplied string
is very ugly. So, this patch simplifies things by implementing the
backwards compatibility in a different way: we have a machine class flag
set for the older machines, and we only load the host values into the
device tree if A) they're not set by the user and B) we have that flag set.
This does mean that the "passthrough" functionality is no longer available
with the current machine type. That's ok though: if a user or management
layer really wants the information passed through they can read it
themselves (OpenStack Nova already does something similar for x86).
It also means the user can't explicitly ask for the values to be omitted
on the old machine types. I think that's an acceptable trade-off: if you
care enough about not leaking the host information you can either move to
the new machine type, or use a dummy value for the properties.
For the new machine type, this also removes an odd inconsistency
between running on a POWER and non-POWER (or non-Linux) hosts: if the
host information couldn't be read from where we expect (in the host's
device tree as exposed by Linux), we'd fallback to omitting the guest
device tree items.
While we're there, improve some poorly worded comments, and the help text
for the properties.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Tested-by: Greg Kurz <groug@kaod.org>
(cherry picked from commit 0a794529bd)
[LM: BSC#1126455 CVE-2019-03812]
Signed-off-by: Lin Ma <lma@suse.com>
Since Linux switched to blk-mq as the default in Linux commit
5c279bd9e406 ("scsi: default to scsi-mq"), virtio-scsi LUNs consume
about 10x as much guest kernel memory.
This commit allows you to choose the virtqueue size for each
virtio-scsi-pci controller like this:
-device virtio-scsi-pci,id=scsi,virtqueue_size=16
The default is still 128 as before. Using smaller virtqueue_size
allows many more disks to be added to small memory virtual machines.
For a 1 vCPU, 500 MB, no swap VM I observed:
With scsi-mq enabled (upstream kernel): 175 disks
-"- ditto -"- virtqueue_size=64: 318 disks
-"- ditto -"- virtqueue_size=16: 775 disks
With scsi-mq disabled (kernel before 5c279bd9e406): 1755 disks
Note that to have any effect, this requires a kernel patch:
https://lkml.org/lkml/2017/8/10/689
Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
Message-Id: <20170810165255.20865-1-rjones@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 5c0919d020)
[BR: FATE#327255. Fate was for SLE12-SP2, but we need to be able to
forward migrate this to SLE12-SP3 (and beyond)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
The new MSR IA32_SPEC_CTRL MSR was introduced by a recent Intel
microcode updated and can be used by OSes to mitigate
CVE-2017-5715. Unfortunately we can't change the existing CPU
models without breaking existing setups, so users need to
explicitly update their VM configuration to use the new *-IBRS
CPU model if they want to expose IBRS to guests.
The new CPU models are simple copies of the existing CPU models,
with just CPUID_7_0_EDX_SPEC_CTRL added and model_id updated.
Cc: Jiri Denemark <jdenemar@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20180109154519.25634-6-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit ac96c41354)
[BR: FATE#327261]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Introduce Skylake-Server cpu mode which inherits the features from
Skylake-Client and supports some additional features that are: AVX512,
CLWB and PGPE1GB.
Signed-off-by: Boqun Feng (Intel) <boqun.feng@gmail.com>
Message-Id: <20170621052935.20715-1-boqun.feng@gmail.com>
[ehabkost: copied comment about XSAVES from Skylake-Client]
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit 53f9a6f45f)
[BR: FATE#327261]
Signed-off-by: Bruce Rogers <brogers@suse.com>
depth == 0 is used to indicate 256 color modes. Our region calculation
goes wrong in that case. So detect that and just take the safe code
path we already have for the wraparound case.
While being at it also catch depth == 15 (where our region size
calculation goes wrong too). And make the comment more verbose,
explaining what is going on here.
Without this windows guest install might trigger an assert due to trying
to check dirty bitmap outside the snapshot region.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1575541
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20180514103117.21059-1-kraxel@redhat.com
(cherry picked from commit a89fe6c329)
Signed-off-by: Bruce Rogers <brogers@suse.com>
Typically the scanline length and the line offset are identical. But
in case they are not our calculation for region_end is incorrect. Using
line_offset is fine for all scanlines, except the last one where we have
to use the actual scanline length.
Fixes: CVE-2018-7550
Reported-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Prasad J Pandit <pjp@fedoraproject.org>
Tested-by: Ross Lagerwall <ross.lagerwall@citrix.com>
Message-id: 20180309143704.13420-1-kraxel@redhat.com
(cherry picked from commit 7cdc61becd)
[BR: BSC#1084604 CVE-2018-7858 (it appears the above CVE ref. is wrong)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Commit "3d90c62548 vga: stop passing pointers to vga_draw_line*
functions" is incomplete. It doesn't handle the case that the vga
rendering code tries to create a shared surface, i.e. a pixman image
backed by vga video memory. That can not work in case the guest display
wraps from end of video memory to the start. So force shadowing in that
case. Also adjust the snapshot region calculation.
Can trigger with cirrus only, when programming vbe modes using the bochs
api (stdvga, also qxl and virtio-vga in vga compat mode) wrap arounds
can't happen.
Fixes: CVE-2017-13672
Fixes: 3d90c62548
Cc: P J P <ppandit@redhat.com>
Reported-by: David Buchanan <d@vidbuchanan.co.uk>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20171010141323.14049-3-kraxel@redhat.com
(cherry picked from commit 28f77de26a)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 362f811793)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
vga display update mis-calculated the region for the dirty bitmap
snapshot in case split screen mode is used. This can trigger an
assert in cpu_physical_memory_snapshot_get_dirty().
Impact: DoS for privileged guest users.
Fixes: CVE-2017-13673
Fixes: fec5e8c92b
Cc: P J P <ppandit@redhat.com>
Reported-by: David Buchanan <d@vidbuchanan.co.uk>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170828123307.15392-1-kraxel@redhat.com
(cherry picked from commit e65294157d)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
vga display update mis-calculated the region for the dirty bitmap
snapshot in case the scanlines are padded. This can triggere an
assert in cpu_physical_memory_snapshot_get_dirty().
Fixes: fec5e8c92b
Reported-by: Kevin Wolf <kwolf@redhat.com>
Reported-by: 李强 <liqiang6-s@360.cn>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170509104839.19415-1-kraxel@redhat.com
(cherry picked from commit bfc56535f7)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
The vga code clears the dirty bits *after* reading the framebuffer
memory. So if the guest framebuffer updates hits the race window
between vga reading the framebuffer and vga clearing the dirty bits
vga will miss that update
Fix it by using the new memory_region_copy_and_clear_dirty()
memory_region_copy_get_dirty() functions. That way we clear the
dirty bitmap before reading the framebuffer. Any guest display
updates happening in parallel will be properly tracked in the
dirty bitmap then and the next display refresh will pick them up.
Problem triggers with mttcg only. Before mttcg was merged tcg
never ran in parallel to vga emulation. Using kvm will hide the
problem too, due to qemu operating on a userspace copy of the
kernel's dirty bitmap.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170421091632.30900-5-kraxel@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit fec5e8c92b)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Add vga_scanline_invalidated helper to check whenever a scanline was
invalidated. Add a sanity check to fix OOB read access for display
heights larger than 2048.
Only cirrus uses this, for hardware cursor rendering, so having this
work properly for the first 2048 scanlines only shouldn't be a problem
as the cirrus can't handle large resolutions anyway. Also changing the
invalidated_y_table size would break live migration.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170421091632.30900-4-kraxel@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit f3289f6f0f)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
This patch adds support for getting and using a local copy of the dirty
bitmap.
memory_region_snapshot_and_clear_dirty() will create a snapshot of the
dirty bitmap for the specified range, clear the dirty bitmap and return
the copy. The returned bitmap can be a bit larger than requested, the
range is expanded so the code can copy unsigned longs from the bitmap
and avoid atomic bit update operations.
memory_region_snapshot_get_dirty() will return the dirty status of
pages, pretty much like memory_region_get_dirty(), but using the copy
returned by memory_region_copy_and_clear_dirty().
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170421091632.30900-3-kraxel@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 8deaf12ca1)
[BR: Support for BSC#1084604 (and other useful vga fixes)]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Recent commit 5b76ef50f6 fixed a race where v9fs_co_open2() could
possibly overwrite a fid path with v9fs_path_copy() while it is being
accessed by some other thread, ie, use-after-free that can be detected
by ASAN with a custom 9p client.
It turns out that the same can happen at several locations where
v9fs_path_copy() is used to set the fid path. The fix is again to
take the write lock.
Fixes CVE-2018-19364.
Cc: P J P <ppandit@redhat.com>
Reported-by: zhibin hu <noirfate@gmail.com>
Reviewed-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Greg Kurz <groug@kaod.org>
(cherry picked from commit 5b3c77aa58)
[BR: BSC#1116717 CVE-2018-19364]
Signed-off-by: Bruce Rogers <brogers@suse.com>
The assumption that the fid cannot be used by any other operation is
wrong. At least, nothing prevents a misbehaving client to create a
file with a given fid, and to pass this fid to some other operation
at the same time (ie, without waiting for the response to the creation
request). The call to v9fs_path_copy() performed by the worker thread
after the file was created can race with any access to the fid path
performed by some other thread. This causes use-after-free issues that
can be detected by ASAN with a custom 9p client.
Unlike other operations that only read the fid path, v9fs_co_open2()
does modify it. It should hence take the write lock.
Cc: P J P <ppandit@redhat.com>
Reported-by: zhibin hu <noirfate@gmail.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
(cherry picked from commit 5b76ef50f6)
[BR: BSC#1116717 CVE-2018-19364]
Signed-off-by: Bruce Rogers <brogers@suse.com>
When using the 9P2000.u version of the protocol, the following shell
command line in the guest can cause QEMU to crash:
while true; do rm -rf aa; mkdir -p a/b & touch a/b/c & mv a aa; done
With 9P2000.u, file renaming is handled by the WSTAT command. The
v9fs_wstat() function calls v9fs_complete_rename(), which calls
v9fs_fix_path() for every fid whose path is affected by the change.
The involved calls to v9fs_path_copy() may race with any other access
to the fid path performed by some worker thread, causing a crash like
shown below:
Thread 12 "qemu-system-x86" received signal SIGSEGV, Segmentation fault.
0x0000555555a25da2 in local_open_nofollow (fs_ctx=0x555557d958b8, path=0x0,
flags=65536, mode=0) at hw/9pfs/9p-local.c:59
59 while (*path && fd != -1) {
(gdb) bt
path=0x0, flags=65536, mode=0) at hw/9pfs/9p-local.c:59
path=0x0) at hw/9pfs/9p-local.c:92
fs_path=0x555556b56858, stbuf=0x7fff84830ef0) at hw/9pfs/9p-local.c:185
path=0x555556b56858, stbuf=0x7fff84830ef0) at hw/9pfs/cofile.c:53
at hw/9pfs/9p.c:1083
at util/coroutine-ucontext.c:116
(gdb)
The fix is to take the path write lock when calling v9fs_complete_rename(),
like in v9fs_rename().
Impact: DoS triggered by unprivileged guest users.
Fixes: CVE-2018-19489
Cc: P J P <ppandit@redhat.com>
Reported-by: zhibin hu <noirfate@gmail.com>
Reviewed-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Greg Kurz <groug@kaod.org>
(cherry picked from commit 1d20398694)
[BR: BSC#1117275]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Open files and directories with O_NOFOLLOW to avoid symlinks attacks.
While being at it also add O_CLOEXEC.
usb-mtp only handles regular files and directories and ignores
everything else, so users should not see a difference.
Because qemu ignores symlinks, carrying out a successful symlink attack
requires swapping an existing file or directory below rootdir for a
symlink and winning the race against the inotify notification to qemu.
Fixes: CVE-2018-16872
Cc: Prasad J Pandit <ppandit@redhat.com>
Cc: Bandan Das <bsd@redhat.com>
Reported-by: Michael Hanselmann <public@hansmi.ch>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Michael Hanselmann <public@hansmi.ch>
Message-id: 20181213122511.13853-1-kraxel@redhat.com
(cherry picked from commit bab9df35ce)
[BR: BSC#1119493]
Signed-off-by: Bruce Rogers <brogers@suse.com>
While emulating identification protocol, tcp_emu() does not check
available space in the 'sc_rcv->sb_data' buffer. It could lead to
heap buffer overflow issue. Add check to avoid it.
Reported-by: Kira <864786842@qq.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit a7104eda7d)
[BR: BSC#1123156 CVE-2019-6778, modify patch to use spaces instead of tabs]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Do an update of system_time_msr address every time before reading
the value of tsc_timestamp from guest's kvmclock page.
There is no other code paths which ensure that qemu has an up-to-date
value of system_time_msr. So, force this update on guest's tsc_timestamp
reading.
This bug causes effect on those nested setups which turn off TPR access
interception for L2 guests and that access being intercepted by L0 doesn't
show up in L1.
Linux bootstrap initiate kvmclock before APIC initializing causing TPR access.
That's why on L1 guests, having TPR interception turned on for L2, the effect
of the bug is not revealed.
This patch fixes this problem by making sure it knows the correct
system_time_msr address every time it is needed.
Signed-off-by: Denis Plotnikov <dplotnikov@virtuozzo.com>
Message-Id: <1496054944-25623-1-git-send-email-dplotnikov@virtuozzo.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit e2b6c1712e)
[BR: BSC#1113231]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Bring linux-user write(2) handling into line with linux for the case
of a 0-byte write with a NULL buffer. Based on a patch originally
written by Zhuowei Zhang.
Addresses https://bugs.launchpad.net/qemu/+bug/1716292.
>From Zhuowei Zhang's patch (https://lists.gnu.org/archive/html/qemu-devel/2017-09/msg08073.html):
Linux returns success for the special case of calling write with a
zero-length NULL buffer: compiling and running
int main() {
ssize_t ret = write(STDOUT_FILENO, NULL, 0);
fprintf(stderr, "write returned %ld\n", ret);
return 0;
}
gives "write returned 0" when run directly, but "write returned
-1" in QEMU.
This commit checks for this situation and returns success if
found.
Subsequent discussion raised the following questions (and my answers):
- Q. Should TARGET_NR_read pass through to safe_read in this
situation too?
A. I'm wary of changing unrelated code to the specific problem I'm
addressing. TARGET_NR_read is already consistent with Linux for
this case.
- Q. Do pread64/pwrite64 need to be changed similarly?
A. Experiment suggests not: both linux and linux-user yield -1 for
NULL 0-length reads/writes.
Signed-off-by: Tony Garnock-Jones <tonygarnockjones@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20180908182205.GB409@mornington.dcs.gla.ac.uk>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
(cherry picked from commit 58cfa6c2e6)
[LY: BSC#1121600]
Signed-off-by: Liang Yan <lyan@suse.com>
xen_disk currently allocates memory to hold the data for each ioreq
as that ioreq is used, and frees it afterwards. Because it requires
page-aligned blocks, this interacts poorly with non-page-aligned
allocations and balloons the heap.
Instead, allocate the maximum possible requirement, which is
BLKIF_MAX_SEGMENTS_PER_REQUEST pages (currently 11 pages) when
the ioreq is created, and keep that allocation until it is destroyed.
Since the ioreqs themselves are re-used via a free list, this
should actually improve memory usage.
Signed-off-by: Tim Smith <tim.smith@citrix.com>
[BSC#1100408]
While writing a message in 'lsi_do_msgin', message length value
in 'msg_len' could be invalid due to an invalid migration stream.
Add an assertion to avoid an out of bounds access, and reject
the incoming migration data if it contains an invalid message
length.
Discovered by Deja vu Security. Reported by Oracle.
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-Id: <20181026194314.18663-1-ppandit@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit e58ccf0396)
[BR: BSC#1114422 CVE-2018-18849]
Signed-off-by: Bruce Rogers <brogers@suse.com>
There should not be a reason for passing a packet size greater than
INT_MAX. It's usually a hint of bug somewhere, so ignore packet size
greater than INT_MAX in qemu_deliver_packet_iov()
CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit 1592a99470)
[LD: BSC#1111013 CVE-2018-17963]
Signed-off-by: Larry Dewey <ldewey@suse.com>
In pcnet_receive(), we try to assign size_ to size which converts from
size_t to integer. This will cause troubles when size_ is greater
INT_MAX, this will lead a negative value in size and it can then pass
the check of size < MIN_BUF_SIZE which may lead out of bound access
for both buf and buf1.
Fixing by converting the type of size to size_t.
CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit b1d80d12c5)
[LD: BSC#1111010 CVE-2018-17962]
Signed-off-by: Larry Dewey <ldewey@suse.com>
In rtl8139_do_receive(), we try to assign size_ to size which converts
from size_t to integer. This will cause troubles when size_ is greater
INT_MAX, this will lead a negative value in size and it can then pass
the check of size < MIN_BUF_SIZE which may lead out of bound access of
for both buf and buf1.
Fixing by converting the type of size to size_t.
CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit 1a326646fe)
[LD: BSC#1111006 CVE-2018-17958]
Signed-off-by: Larry Dewey <ldewey@suse.com>
In ne2000_receive(), we try to assign size_ to size which converts
from size_t to integer. This will cause troubles when size_ is greater
INT_MAX, this will lead a negative value in size and it can then pass
the check of size < MIN_BUF_SIZE which may lead out of bound access of
for both buf and buf1.
Fixing by converting the type of size to size_t.
CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit fdc89e90fa)
[LD: BSC#1110910 CVE-2018-10839]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Since xen_disk now always copies data to and from a guest there is no need
to maintain a vector entry corresponding to every page of a request.
This means there is less per-request state to maintain so the ioreq
structure can shrink significantly.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony Perard <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 5ebf962652)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Now that the (native or emulated) xen_be_copy_grant_refs() helper is
always available, the xen_disk code can be significantly simplified by
removing direct use of grant map and unmap operations.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony Perard <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 06454c24ad)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Not all Xen environments support the xengnttab_grant_copy() operation.
E.g. where the OS is FreeBSD or Xen is older than 4.8.0.
This patch introduces an emulation of that operation using
xengnttab_map_domain_grant_refs() and memcpy() for those environments.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 3fe12b8403)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Now that helpers are available in xen_backend, use them throughout all
Xen PV backends.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony Perard <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 58560f2ae7)
[LD: BSC#1100408 - Removed xen_9p* file because it doesn't exist]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Now that helpers are present in xen_backend, this patch removes open-coded
calls to libxengnttab from the xen_disk code.
This patch also fixes one whitspace error in the assignment of the
XenDevOps initialise method.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony Perard <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 5ee1d99913)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch adds grant table helper functions to the xen_backend code to
localize error reporting and use of xen_domid.
The patch also defers the call to xengnttab_open() until just before the
initialise method in XenDevOps is invoked. This method is responsible for
mapping the shared ring. No prior method requires access to the grant table.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 9838824aff)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Currently the xen_disk source has to carry #ifdef exclusions to compile
against Xen older then 4.8. This is a bit messy so this patch lifts the
definition of struct xengnttab_grant_copy_segment and adds it into the
pre-4.8 compat area in xen_common.h, which allows xen_disk to be cleaned
up.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 5c0d914a9b)
[LD: BSC#1100408 - Modified as the Xen interface versioning has changed]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Trying to call xengnttab_set_max_grants() with the same file handle
might fail on some kernels, as this operation is allowed only once.
This is a problem for the qdisk backend as blk_connect() can be
called multiple times for a domain, e.g. in case grub-xen is being
used to boot it.
So instead of letting the generic backend code open the gnttab device
do it in blk_connect() and close it again in blk_disconnect.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit e38c3e86df)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
The Xen qdisk backend needs to test whether grant copy operations is
available in the kernel. Unfortunately this collides with using
xengnttab_set_max_grants() on some kernels as this operation has to
be the first one after opening the gnttab device.
In order to solve this problem test for the availability of grant copy
in xen_be_init() opening the gnttab device just for that purpose and
closing it again afterwards. Advertise the availability via a global
flag and use that flag in the qdisk backend.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit b5e397a79e)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
g_malloc0_n is available since glib-2.24. To allow build with older glib
versions use the generic g_new0, which is already used in many other
places in the code.
Fixes commit 3284fad728 ("xen-disk: add support for multi-page shared rings")
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit a3fd781f65)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
The blkif protocol has had provision for negotiation of multi-page shared
rings for some time now and many guest OS have support in their frontend
drivers.
This patch makes the necessary modifications to xen-disk support a shared
ring up to order 4 (i.e. 16 pages).
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 3284fad728)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
If grant copy is available then it will always be used in preference to
persistent maps. In this case feature-persistent should not be advertized
to the frontend, otherwise it may needlessly copy data into persistently
granted buffers.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 976eba1c88)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch modifies the wrapper functions in xen_common.h to use the
new xendevicemodel interface if it is available along with compatibility
code to use the old libxenctrl interface if it is not.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
Reviewed-by: Anthony Perard <anthony.perard@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit d655f34e6d)
[LD: BSC#1100408 - Commit combined with 8f25e7544 for simplification]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Doing this will make the transition to using the new libxendevicemodel
interface less intrusive on the callers of these functions, since using
the new library will require a change of handle.
NOTE: The patch also moves the 'externs' for xen_xc and xen_fmem from
xen_backend.h to xen_common.h, and the declarations from
xen_backend.c to xen-common.c, which is where they belong.
Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Anthony Perard <anthony.perard@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
(cherry picked from commit 260cabed71)
[LD: BSC#1100408]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Remove -sandbox option if the host is not capable of TSYNC, since the
sandbox will fail at setup time otherwise. This will help libvirt, for
ex, to figure out if -sandbox will work.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
Acked-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 5780760f5e)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
If CONFIG_SECCOMP is undefined, the option 'elevatedprivileges' remains
compiled. This would make libvirt set the corresponding capability and
then trigger failure during guest startup. This patch moves the code
regarding seccomp command line options to qemu-seccomp.c file and
wraps qemu_opts_foreach finding sandbox option with CONFIG_SECCOMP.
Because parse_sandbox() is moved into qemu-seccomp.c file, change
seccomp_start() to static function.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Tested-by: Ján Tomko <jtomko@redhat.com>
Acked-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 9d0fdecbad)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
When using "-seccomp on", the seccomp policy is only applied to the
main thread, the vcpu worker thread and other worker threads created
after seccomp policy is applied; the seccomp policy is not applied to
e.g. the RCU thread because it is created before the seccomp policy is
applied and SECCOMP_FILTER_FLAG_TSYNC isn't used.
This can be verified with
for task in /proc/`pidof qemu`/task/*; do cat $task/status | grep Secc ; done
Seccomp: 2
Seccomp: 0
Seccomp: 0
Seccomp: 2
Seccomp: 2
Seccomp: 2
Starting with libseccomp 2.2.0 and kernel >= 3.17, we can use
seccomp_attr_set(ctx, > SCMP_FLTATR_CTL_TSYNC, 1) to update the policy
on all threads.
libseccomp requirement was bumped to 2.2.0 in previous patch.
libseccomp should fail to set the filter if it can't honour
SCMP_FLTATR_CTL_TSYNC (untested), and thus -sandbox will now fail on
kernel < 3.17.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Acked-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 70dfabeaa7)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
The following patch is going to require TSYNC, which is only available
since libseccomp 2.2.0.
libseccomp 2.2.0 was released February 12, 2015.
According to repology, libseccomp version in different distros:
RHEL-7: 2.3.1
Debian (Stretch): 2.3.1
OpenSUSE Leap 15: 2.3.2
Ubuntu (Xenial): 2.3.1
This will drop support for -sandbox on:
Debian (Jessie): 2.1.1 (but 2.2.3 in backports)
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Acked-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit d0699bd37c)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
The upcoming libseccomp release should have SCMP_ACT_KILL_PROCESS
action (https://github.com/seccomp/libseccomp/issues/96).
SCMP_ACT_KILL_PROCESS is preferable to immediately terminate the
offending process, rather than having the SIGSYS handler running.
Use SECCOMP_GET_ACTION_AVAIL to check availability of kernel support,
as libseccomp will fallback on SCMP_ACT_KILL otherwise, and we still
prefer SCMP_ACT_TRAP.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit bda08a5764)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
Current and upcoming mesa releases rely on a shader disk cash. It uses
a thread job queue with low priority, set with
sched_setscheduler(SCHED_IDLE). However, that syscall is rejected by
the "resourcecontrol" seccomp qemu filter.
Since it should be safe to allow lowering thread priority, let's allow
scheduling thread to idle policy.
Related to:
https://bugzilla.redhat.com/show_bug.cgi?id=1594456
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Acked-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 056de1e894)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch adds [,resourcecontrol=deny] to `-sandbox on' option. It
blacklists all process affinity and scheduler priority system calls to
avoid any bigger of the process.
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 24f8cdc572)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch adds [,spawn=deny] argument to `-sandbox on' option. It
blacklists fork and execve system calls, avoiding Qemu to spawn new
threads or processes.
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 995a226f88)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch introduces the new argument
[,elevateprivileges=allow|deny|children] to the `-sandbox on'. It allows
or denies Qemu process to elevate its privileges by blacklisting all
set*uid|gid system calls. The 'children' option will let forks and
execves run unprivileged.
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 73a1e64725)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch introduces the argument [,obsolete=allow] to the `-sandbox on'
option. It allows Qemu to run safely on old system that still relies on
old system calls.
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 2b716fa6d6)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
This patch changes the default behavior of the seccomp filter from
whitelist to blacklist. By default now all system calls are allowed and
a small black list of definitely forbidden ones was created.
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
(cherry picked from commit 1bd6152ae2)
[LD: BSC#1106222 CVE-2018-15746]
Signed-off-by: Larry Dewey <ldewey@suse.com>
AMD Zen expose the Intel equivalant to Speculative Store Bypass Disable
via the 0x80000008_EBX[25] CPUID feature bit.
This needs to be exposed to guest OS to allow them to protect
against CVE-2018-3639.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20180521215424.13520-3-berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit 403503b162)
[BR: BSC#1092885 CVE-2018-3639]
Signed-off-by: Bruce Rogers <brogers@suse.com>
"Some AMD processors only support a non-architectural means of enabling
speculative store bypass disable (SSBD). To allow a simplified view of
this to a guest, an architectural definition has been created through a new
CPUID bit, 0x80000008_EBX[25], and a new MSR, 0xc001011f. With this, a
hypervisor can virtualize the existence of this definition and provide an
architectural method for using SSBD to a guest.
Add the new CPUID feature, the new MSR and update the existing SSBD
support to use this MSR when present." (from x86/speculation: Add virtualized
speculative store bypass disable support in Linux).
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20180521215424.13520-4-berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit cfeea0c021)
[BR: BSC#1092885 CVE-2018-3639]
Signed-off-by: Bruce Rogers <brogers@suse.com>
While reading file content via 'guest-file-read' command,
'qmp_guest_file_read' routine allocates buffer of count+1
bytes. It could overflow for large values of 'count'.
Add check to avoid it.
Reported-by: Fakhri Zulkifli <mohdfakhrizulkifli@gmail.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Cc: qemu-stable@nongnu.org
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
(cherry picked from commit 141b197408)
[FL: BSC#1098735 CVE-2018-12617]
Signed-off-by: Fei Li <fli@suse.com>
While reassembling incoming fragmented datagrams, 'm_cat' routine
extends the 'mbuf' buffer, if it has insufficient room. It computes
a wrong buffer size, which leads to overwriting adjacent heap buffer
area. Correct this size computation in m_cat.
Reported-by: ZDI Disclosures <zdi-disclosures@trendmicro.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit 864036e251)
[FL: BSC#1096223 CVE-2018-11806]
Signed-off-by: Fei Li <fli@suse.com>
Provide monitor naming of xen disks, and plumb guest driver
notification through xenstore of resizing instigated via the
monitor.
[BR: FATE#325467]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Since Linux commit 313d21eeab9282e, tpm devices have their own device
class "tpm" and the cancel path must be looked up under
/sys/class/tpm/ instead of /sys/class/misc/.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
(cherry picked from commit 05b71fb207)
[LY: BSC#1070615]
Signed-off-by: Liang Yan <lyan@suse.com>
As an attempt to help the user do the right thing, warn if we
detect spec_ctrl data in the migration stream, but where the
cpu defined doesn't have the feature. This would indicate the
migration is from the quick and dirty qemu produced in January
2018 to handle Spectre v2. That qemu version exposed the IBRS
cpu feature to all vcpu types, which helped in the short term
but wasn't a well designed approach.
Warn the user that the now migrated guest needs to be restarted
as soon as possible, using the spec_ctrl cpu feature flag or a
*-IBRS vcpu model specified as appropriate.
Signed-off-by: Bruce Rogers <brogers@suse.com>
The new MSR IA32_SPEC_CTRL MSR was introduced by a recent Intel
microcode updated and can be used by OSes to mitigate
CVE-2017-5715. Unfortunately we can't change the existing CPU
models without breaking existing setups, so users need to
explicitly update their VM configuration to use the new *-IBRS
CPU model if they want to expose IBRS to guests.
The new CPU models are simple copies of the existing CPU models,
with just CPUID_7_0_EDX_SPEC_CTRL added and model_id updated.
Cc: Jiri Denemark <jdenemar@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20180109154519.25634-6-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit ac96c41354)
[BR: BSC#1068032 CVE-2017-5715 CPU models are reduced as appropriate
to match this QEMU version]
Signed-off-by: Bruce Rogers <brogers@suse.com>
This is perhaps the easiest way to handle the longer model names
introduced with the -IBRS models. This is in lew of backporting
commit 807e9869b8 and other commits
which that would require.
[BR: BSC#1068032 CVE-2017-5715]
Signed-off-by: Bruce Rogers <brogers@suse.com>
The previous patches fix problems with throttling of forced framebuffer updates
and audio data capture that would cause the QEMU output buffer size to grow
without bound. Those fixes are graceful in that once the client catches up with
reading data from the server, everything continues operating normally.
There is some data which the server sends to the client that is impractical to
throttle. Specifically there are various pseudo framebuffer update encodings to
inform the client of things like desktop resizes, pointer changes, audio
playback start/stop, LED state and so on. These generally only involve sending
a very small amount of data to the client, but a malicious guest might be able
to do things that trigger these changes at a very high rate. Throttling them is
not practical as missed or delayed events would cause broken behaviour for the
client.
This patch thus takes a more forceful approach of setting an absolute upper
bound on the amount of data we permit to be present in the output buffer at
any time. The previous patch set a threshold for throttling the output buffer
by allowing an amount of data equivalent to one complete framebuffer update and
one seconds worth of audio data. On top of this it allowed for one further
forced framebuffer update to be queued.
To be conservative, we thus take that throttling threshold and multiply it by
5 to form an absolute upper bound. If this bound is hit during vnc_write() we
forceably disconnect the client, refusing to queue further data. This limit is
high enough that it should never be hit unless a malicious client is trying to
exploit the sever, or the network is completely saturated preventing any sending
of data on the socket.
This completes the fix for CVE-2017-15124 started in the previous patches.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-12-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit f887cf165d)
[LY: BSC#1073489 CVE-2017-15124]
Signed-off-by: Liang Yan <lyan@suse.com>
The VNC server must throttle data sent to the client to prevent the 'output'
buffer size growing without bound, if the client stops reading data off the
socket (either maliciously or due to stalled/slow network connection).
The current throttling is very crude because it simply checks whether the
output buffer offset is zero. This check is disabled if the client has requested
a forced update, because we want to send these as soon as possible.
As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM.
They can first start something in the guest that triggers lots of framebuffer
updates eg play a youtube video. Then repeatedly send full framebuffer update
requests, but never read data back from the server. This can easily make QEMU's
VNC server send buffer consume 100MB of RAM per second, until the OOM killer
starts reaping processes (hopefully the rogue QEMU process, but it might pick
others...).
To address this we make the throttling more intelligent, so we can throttle
full updates. When we get a forced update request, we keep track of exactly how
much data we put on the output buffer. We will not process a subsequent forced
update request until this data has been fully sent on the wire. We always allow
one forced update request to be in flight, regardless of what data is queued
for incremental updates or audio data. The slight complication is that we do
not initially know how much data an update will send, as this is done in the
background by the VNC job thread. So we must track the fact that the job thread
has an update pending, and not process any further updates until this job is
has been completed & put data on the output buffer.
This unbounded memory growth affects all VNC server configurations supported by
QEMU, with no workaround possible. The mitigating factor is that it can only be
triggered by a client that has authenticated with the VNC server, and who is
able to trigger a large quantity of framebuffer updates or audio samples from
the guest OS. Mostly they'll just succeed in getting the OOM killer to kill
their own QEMU process, but its possible other processes can get taken out as
collateral damage.
This is a more general variant of the similar unbounded memory usage flaw in
the websockets server, that was previously assigned CVE-2017-15268, and fixed
in 2.11 by:
commit a7b20a8efa
Author: Daniel P. Berrange <berrange@redhat.com>
Date: Mon Oct 9 14:43:42 2017 +0100
io: monitor encoutput buffer size from websocket GSource
This new general memory usage flaw has been assigned CVE-2017-15124, and is
partially fixed by this patch.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-11-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit ada8d2e436)
[LY: BSC#1073489 CVE-2017-15124]
Signed-off-by: Liang Yan <lyan@suse.com>
The VNC server must throttle data sent to the client to prevent the 'output'
buffer size growing without bound, if the client stops reading data off the
socket (either maliciously or due to stalled/slow network connection).
The current throttling is very crude because it simply checks whether the
output buffer offset is zero. This check must be disabled if audio capture is
enabled, because when streaming audio the output buffer offset will rarely be
zero due to queued audio data, and so this would starve framebuffer updates.
As a result, the VNC client can cause QEMU to allocate arbitrary amounts of RAM.
They can first start something in the guest that triggers lots of framebuffer
updates eg play a youtube video. Then enable audio capture, and simply never
read data back from the server. This can easily make QEMU's VNC server send
buffer consume 100MB of RAM per second, until the OOM killer starts reaping
processes (hopefully the rogue QEMU process, but it might pick others...).
To address this we make the throttling more intelligent, so we can throttle
when audio capture is active too. To determine how to throttle incremental
updates or audio data, we calculate a size threshold. Normally the threshold is
the approximate number of bytes associated with a single complete framebuffer
update. ie width * height * bytes per pixel. We'll send incremental updates
until we hit this threshold, at which point we'll stop sending updates until
data has been written to the wire, causing the output buffer offset to fall
back below the threshold.
If audio capture is enabled, we increase the size of the threshold to also
allow for upto 1 seconds worth of audio data samples. ie nchannels * bytes
per sample * frequency. This allows the output buffer to have a mixture of
incremental framebuffer updates and audio data queued, but once the threshold
is exceeded, audio data will be dropped and incremental updates will be
throttled.
This unbounded memory growth affects all VNC server configurations supported by
QEMU, with no workaround possible. The mitigating factor is that it can only be
triggered by a client that has authenticated with the VNC server, and who is
able to trigger a large quantity of framebuffer updates or audio samples from
the guest OS. Mostly they'll just succeed in getting the OOM killer to kill
their own QEMU process, but its possible other processes can get taken out as
collateral damage.
This is a more general variant of the similar unbounded memory usage flaw in
the websockets server, that was previously assigned CVE-2017-15268, and fixed
in 2.11 by:
commit a7b20a8efa
Author: Daniel P. Berrange <berrange@redhat.com>
Date: Mon Oct 9 14:43:42 2017 +0100
io: monitor encoutput buffer size from websocket GSource
This new general memory usage flaw has been assigned CVE-2017-15124, and is
partially fixed by this patch.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-10-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit e2b72cb6e0)
[LY: BSC#1073489 CVE-2017-15124]
Signed-off-by: Liang Yan <lyan@suse.com>
According to the RFB protocol, a client sends one or more framebuffer update
requests to the server. The server can reply with a single framebuffer update
response, that covers all previously received requests. Once the client has
read this update from the server, it may send further framebuffer update
requests to monitor future changes. The client is free to delay sending the
framebuffer update request if it needs to throttle the amount of data it is
reading from the server.
The QEMU VNC server, however, has never correctly handled the framebuffer
update requests. Once QEMU has received an update request, it will continue to
send client updates forever, even if the client hasn't asked for further
updates. This prevents the client from throttling back data it gets from the
server. This change fixes the flawed logic such that after a set of updates are
sent out, QEMU waits for a further update request before sending more data.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-8-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 728a7ac954)
[LY: BSC#1073489]
Signed-off-by: Liang Yan <lyan@suse.com>
When we encode data for writing with SASL, we encode the entire pending output
buffer. The subsequent write, however, may not be able to send the full encoded
data in one go though, particularly with a slow network. So we delay setting the
output buffer offset back to zero until all the SASL encoded data is sent.
Between encoding the data and completing sending of the SASL encoded data,
however, more data might have been placed on the pending output buffer. So it
is not valid to set offset back to zero. Instead we must keep track of how much
data we consumed during encoding and subtract only that amount.
With the current bug we would be throwing away some pending data without having
sent it at all. By sheer luck this did not previously cause any serious problem
because appending data to the send buffer is always an atomic action, so we
only ever throw away complete RFB protocol messages. In the case of frame buffer
updates we'd catch up fairly quickly, so no obvious problem was visible.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-6-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 8f61f1c5a6)
[LY: BSC#1073489]
Signed-off-by: Liang Yan <lyan@suse.com>
The vnc_update_client() method checks the 'has_dirty' flag to see if there are
dirty regions that are pending to send to the client. Regardless of this flag,
if a forced update is requested, updates must be sent. For unknown reasons
though, the code also tries to sent updates if audio capture is enabled. This
makes no sense as audio capture state does not impact framebuffer contents, so
this check is removed.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-5-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 3541b08475)
[LY: BSC#1073489]
Signed-off-by: Liang Yan <lyan@suse.com>
A previous commit:
commit 5a8be0f73d
Author: Gerd Hoffmann <kraxel@redhat.com>
Date: Wed Jul 13 12:21:20 2016 +0200
vnc: make sure we finish disconnect
Added a check for vs->disconnecting at the very start of the
vnc_update_client method. This means that the very next "if"
statement check for !vs->disconnecting always evaluates true,
and is thus redundant. This in turn means the vs->disconnecting
check at the very end of the method never evaluates true, and
is thus unreachable code.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20171218191228.31018-3-berrange@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit c53df96161)
[LY: BSC#1073489]
Signed-off-by: Liang Yan <lyan@suse.com>
While loading kernel via multiboot-v1 image, (flags & 0x00010000)
indicates that multiboot header contains valid addresses to load
the kernel image. In that, end of the data segment address
'mh_load_end_addr' should be less than the bss segment address
'mh_bss_end_addr'. Add check to validate that.
Reported-by: CERT CC <cert.cc@orange.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
[LY: BSC#1083291 CVE-2018-7550]
Signed-off-by: Liang Yan <lyan@suse.com>
Start a vm with qemu-kvm -enable-kvm -vnc :66 -smp 1 -m 1024 -hda
redhat_5.11.qcow2 -device pcnet -vga cirrus,
then use VNC client to connect to VM, and excute the code below in guest
OS will lead to qemu crash:
int main()
{
iopl(3);
srand(time(NULL));
int a,b;
while(1){
a = rand()%0x100;
b = 0x3c0 + (rand()%0x20);
outb(a,b);
}
return 0;
}
The above code is writing the registers of VGA randomly.
We can write VGA CRT controller registers index 0x0C or 0x0D
(which is the start address register) to modify the
the display memory address of the upper left pixel
or character of the screen. The address may be out of the
range of vga ram. So we should check the validation of memory address
when reading or writing it to avoid segfault.
Signed-off-by: linzhecheng <linzhecheng@huawei.com>
Message-id: 20180111132724.13744-1-linzhecheng@huawei.com
Fixes: CVE-2018-5683
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 191f59dc17)
[LY: BSC#1076114 CVE-2018-5683]
Signed-off-by: Liang Yan <lyan@suse.com>
When using bit-wise operations that exploit the power-of-two
nature of the second argument of ROUND_UP(), we still need to
ensure that the mask is as wide as the first argument (done
by using a ternary to force proper arithmetic promotion).
Unpatched, ROUND_UP(2ULL*1024*1024*1024*1024, 512U) produces 0,
instead of the intended 2TiB, because negation of an unsigned
32-bit quantity followed by widening to 64-bits does not
sign-extend the mask.
Broken since its introduction in commit 292c8e50 (v1.5.0).
Callers that passed the same width type to both macro parameters,
or that had other code to ensure the first parameter's maximum
runtime value did not exceed the second parameter's width, are
unaffected, but I did not audit to see which (if any) existing
clients of the macro could trigger incorrect behavior (I found
the bug while adding a new use of the macro).
While preparing the patch, checkpatch complained about poor
spacing, so I also fixed that here and in the nearby DIV_ROUND_UP.
CC: qemu-trivial@nongnu.org
CC: qemu-stable@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
(cherry picked from commit 2098b073f3)
[LY: BSC#1076775 CVE-2017-18043]
Signed-off-by: Liang Yan <lyan@suse.com>
The NBD spec gives us permission to abruptly disconnect on clients
that send outrageously large option requests, rather than having
to spend the time reading to the end of the option. No real
option request requires that much data anyways; and meanwhile, we
already have the practice of abruptly dropping the connection on
any client that sends NBD_CMD_WRITE with a payload larger than 32M.
For comparison, nbdkit drops the connection on any request with
more than 4096 bytes; however, that limit is probably too low
(as the NBD spec states an export name can theoretically be up
to 4096 bytes, which means a valid NBD_OPT_INFO could be even
longer) - even if qemu doesn't permit exports longer than 256
bytes.
It could be argued that a malicious client trying to get us to
read nearly 4G of data on a bad request is a form of denial of
service. In particular, if the server requires TLS, but a client
that does not know the TLS credentials sends any option (other
than NBD_OPT_STARTTLS or NBD_OPT_EXPORT_NAME) with a stated
payload of nearly 4G, then the server was keeping the connection
alive trying to read all the payload, tying up resources that it
would rather be spending on a client that can get past the TLS
handshake. Hence, this warranted a CVE.
Present since at least 2.5 when handling known options, and made
worse in 2.6 when fixing support for NBD_FLAG_C_FIXED_NEWSTYLE
to handle unknown options.
CC: qemu-stable@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
(cherry picked from commit fdad35ef6c)
[LY: BSC#1070144 CVE-2017-15119]
Signed-off-by: Liang Yan <lyan@suse.com>
During Qemu guest migration, a destination process invokes ps2
post_load function. In that, if 'rptr' and 'count' values were
invalid, it could lead to OOB access or infinite loop issue.
Add check to avoid it.
Reported-by: Cyrille Chatras <cyrille.chatras@orange.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-id: 20171116075155.22378-1-ppandit@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 802cbcb730)
[LY: BSC#1068613 CVE-2017-16845]
Signed-off-by: Liang Yan <lyan@suse.com>
A guest could attempt to use an uninitialised VirtQueue object
or unset Vring.align leading to a arithmetic exception. Add check
to avoid it.
Reported-by: Zhangboxian <zhangboxian@huawei.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
(cherry picked from commit 758ead31c7)
[LY: BSC#1071228 CVE-2017-17381]
Signed-off-by: Liang Yan <lyan@suse.com>
Commit id f0c9d64a avoids adding ACPI_PCIHP_PROP_BSEL twice for
machine type 1.7, but also removes the adding for 1.6 and older
and xen, which causes an error when hotplugging a pci device using
qemu v2.9.* and v2.10.0.
There are two chances for intializing ACPI_PCIHP_PROP_BSEL before
commit f0c9d64a, one is in acpi_pcihp_init and the other is in
acpi_set_bsel. This patch restores the first chance, and adds a check
for the second chance: if the property has already been intialized,
just skip.
[FL: BSC#1074572]
Signed-off-by: Fei Li <fli@suse.com>
stfle.81 (ppa15) is a transparent facility that can be passed to the
guest without the need to implement hypervisor support. As this feature
can be provided by firmware we add it to all full models.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
(cherry picked from commit 9f0d13f4f1)
[LY: BSC#1076813]
Signed-off-by: Liang Yan <lyan@suse.com>
We need to handle the bpb control on reset and migration. Normally
stfle.82 is transparent (and the normal guest part works without
hypervisor activity). To prevent any issues we require full
host kernel support for this feature.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
(cherry picked from commit b073c87517)
[LY: BSC#1076813]
Signed-off-by: Liang Yan <lyan@suse.com>
replace with proper header sync
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
(cherry picked from commit 9cbb636270)
[LY: BSC#1076813]
Signed-off-by: Liang Yan <lyan@suse.com>
Add a new base CPU model called 'EPYC' to model processors from AMD EPYC
family (which includes EPYC 76xx,75xx,74xx, 73xx and 72xx).
The following features bits have been added/removed compare to Opteron_G5
Added: monitor, movbe, rdrand, mmxext, ffxsr, rdtscp, cr8legacy, osvw,
fsgsbase, bmi1, avx2, smep, bmi2, rdseed, adx, smap, clfshopt, sha
xsaveopt, xsavec, xgetbv1, arat
Removed: xop, fma4, tbm
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Message-Id: <20170815170051.127257-1-brijesh.singh@amd.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit 2e2efc7dbe)
[BR: BSC#1052825 FATE#324038]
Signed-off-by: Bruce Rogers <brogers@suse.com>
9p back-end first queries the size of an extended attribute,
allocates space for it via g_malloc() and then retrieves its
value into allocated buffer. Race between querying attribute
size and retrieving its could lead to memory bytes disclosure.
Use g_malloc0() to avoid it.
Reported-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Greg Kurz <groug@kaod.org>
(cherry picked from commit 7bd9275630)
[BR: BSC#1062069 CVE-2017-15038]
Signed-off-by: Bruce Rogers <brogers@suse.com>
The websocket GSource is monitoring the size of the rawoutput
buffer to determine if the channel can accepts more writes.
The rawoutput buffer, however, is merely a temporary staging
buffer before data is copied into the encoutput buffer. Thus
its size will always be zero when the GSource runs.
This flaw causes the encoutput buffer to grow without bound
if the other end of the underlying data channel doesn't
read data being sent. This can be seen with VNC if a client
is on a slow WAN link and the guest OS is sending many screen
updates. A malicious VNC client can act like it is on a slow
link by playing a video in the guest and then reading data
very slowly, causing QEMU host memory to expand arbitrarily.
This issue is assigned CVE-2017-15268, publically reported in
https://bugs.launchpad.net/qemu/+bug/1718964
(cherry picked from commit a7b20a8efa)
Reviewed-by: Eric Blake <eblake@redhat.com>
[Dan: Added extra checks to deal with code refactored in master but
not stable 2.10]
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
[BR: BSC#1062942 CVE-2017-15268]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Instead pass around the address (aka offset into vga memory).
Add vga_read_* helper functions which apply vbe_size_mask to
the address, to make sure the address stays within the valid
range, similar to the cirrus blitter fixes (commits ffaf857778
and 026aeffcb4).
Impact: DoS for privileged guest users. qemu crashes with
a segfault, when hitting the guard page after vga memory
allocation, while reading vga memory for display updates.
Fixes: CVE-2017-13672
Cc: P J P <ppandit@redhat.com>
Reported-by: David Buchanan <d@vidbuchanan.co.uk>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170828122906.18993-1-kraxel@redhat.com
(cherry picked from commit 3d90c62548)
[FL: BSC#1056334 CVE-2017-13672]
Signed-off-by: Fei Li <fli@suse.com>
While loading kernel via multiboot-v1 image, (flags & 0x00010000)
indicates that multiboot header contains valid addresses to load
the kernel image. These addresses are used to compute kernel
size and kernel text offset in the OS image. Validate these
address values to avoid an OOB access issue.
This is CVE-2017-14167.
Reported-by: Thomas Garnier <thgarnie@google.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-Id: <20170907063256.7418-1-ppandit@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit ed4f86e8b6)
[FL: BSC#1057585 CVE-2017-14167]
Signed-off-by: Fei Li <fli@suse.com>
Rather than constructing a local structure instance on the stack, fill
the fields directly on the shared ring, just like other (Linux)
backends do. Build on the fact that all response structure flavors are
actually identical (aside from alignment and padding at the end).
This is XSA-216.
Reported by: Anthony Perard <anthony.perard@citrix.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
(cherry picked from commit b0ac694fdb)
[FL: BSC#1057378 CVE-2017-10911]
Signed-off-by: Fei Li <fli@suse.com>
While parsing dhcp options string in 'dhcp_decode', if an options'
length 'len' appeared towards the end of 'bp_vend' array, ensuing
read could lead to an OOB memory access issue. Add check to avoid it.
This is CVE-2017-11434.
Reported-by: Reno Robert <renorobert@gmail.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit 413d463f43)
[BR: BSC#1049381 CVE-2017-11434]
Signed-off-by: Bruce Rogers <brogers@suse.com>
This ensures that the request is unref'ed properly, and avoids a
segmentation fault in the new qtest testcase that is added.
Reported-by: Zhangyanyu <zyy4013@stu.ouc.edu.cn>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[BR: BSC#1043296 CVE-2017-9503, dropped testcase from patch]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Use sizeof instead of ARRAY_SIZE, fixing -Wmemset-elt-size with recent
GCC versions.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[LY: BSC#1040228]
Signed-off-by: Liang Yan <lyan@suse.com>
gcc 7 (on fedora 26) objects to many of the snprintf's
in the smb path and command creation because it can't
figure out that the smb_dir (i.e. the /tmp dir for the configuration)
is known to be short.
Replace all these fixed length buffers by g_str* functions that dynamically
allocate and use g_dir_make_tmp to make the directory.
(It's fairly new glib but we have a compat function for it).
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit f95cc8b6cc)
[LY: BSC#1040228]
Signed-off-by: Liang Yan <lyan@suse.com>
Detected by GCC 7's -Wformat-truncation. snprintf writes at most
2 bytes here including the terminating NUL, so the result is
truncated. In addition, the newline at the end is pointless.
Fix the buffer size and the format string.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
(cherry picked from commit e9c6ab62c7)
[LY: BSC#1040228]
Signed-off-by: Liang Yan <lyan@suse.com>
KVM has a feature bitmap of CPUID bits that it knows works for guests.
QEMU removes bits that are not part of that bitmap automatically on VM
start.
However, some times we just don't list features in that list because
they don't make sense for normal scenarios, but may be useful in specific,
targeted workloads.
For that purpose, add a new =force option to all CPUID feature flags in
the CPU property. With that we can override the accel filtering and give
users full control over the CPUID feature bits exposed into guests.
Signed-off-by: Alexander Graf <agraf@suse.de>
When driving QEMU from the outside, we have basically no chance to
determine how quickly the guest OS picks up key events, so we usually
have to limit ourselves to very slow keyboard presses to make sure
the guest always has enough chance to pick them up.
This patch adds a trace events when the keyboarde queue is drained.
An external driver can use that as hint that new keys can be pressed.
Signed-off-by: Alexander Graf <agraf@suse.de>
Message-id: 1490883775-94658-1-git-send-email-agraf@suse.de
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
[BR: BSC#1031692]
Signed-off-by: Bruce Rogers <brogers@suse.com>
When running with KVM enabled, you can choose between emulating the
gic in kernel or user space. If the kernel supports in-kernel virtualization
of the interrupt controller, it will default to that. If not, if will
default to user space emulation.
Unfortunately when running in user mode gic emulation, we miss out on
timer events which are only available from kernel space. This patch leverages
the new kernel/user space pending line synchronization for those timer events.
Signed-off-by: Alexander Graf <agraf@suse.de>
Get ioctl number and definitions for KVM_CAP_ARM_USER_IRQ.
Signed-off-by: Christoffer Dall <cdall@linaro.org>
[agraf: change cap to indicate downstream status]
Signed-off-by: Alexander Graf <agraf@suse.de>
Test scsi-{disk,hd,cd} wwn properties for correct 64-bit parsing.
For now piggyback on virtio-scsi.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
All integers would get parsed by strtoll(), not handling the case of
UINT64 properties with the most significient bit set.
Implement a .type_uint64 visitor callback, reusing the existing
parse_str() code through a new argument, using strtoull().
As this is a bug fix, it intentionally ignores checkpatch warnings to
prefer the use of qemu_strto[u]ll() over strto[u]ll().
Cc: qemu-stable@nongnu.org
Signed-off-by: Andreas Färber <afaerber@suse.de>
Since commit a1666142: acpi-build: make ROMs RAM blocks resizeable,
xen HVM direct kernel boot failed. Xen HVM direct kernel boot will
insert a linuxboot.bin or multiboot.bin to /genroms, before this
commit, in acpi_setup, for rom linuxboot.bin/multiboot.bin, it
only needs 0x20000 size; after the commit, it will reserve x16
size for resize, that is 0x200000 size. It causes xen_ram_alloc
failed due to running out of memory.
To resolve it, either:
1. keep using original rom size instead of max size, don't reserve x16 size.
2. guest maxmem needs to be increased. (commit c1d322e6 "xen-hvm: increase
maxmem before calling xc_domain_populate_physmap" solved the problem for
a time, by accident. But then it is reverted in commit ffffbb369 due to
other problem.)
For 2, more discussion is needed about howto. So this patch tries 1, to
use unresizable rom size in xen case in rom_set_mr.
[CYL: BSC#970791]
Signed-off-by: Chunyan Liu <cyliu@suse.com>
Using xen tools 'xl vncviewer' with tigervnc (default on SLE-12),
found that: the display of the guest is unexpected while keep
pressing a key. We expect the same character multiple times, but
it prints only one time. This happens on a PV guest in text mode.
After debugging, found that tigervnc sends repeated key down events
in this case, to differentiate from user pressing the same key many
times. Vnc server only prints the character when it finally receives
key up event.
To solve this issue, this patch tries to add additional key up event
before the next repeated key down event (if the key is not a control
key).
[CYL: BSC#882405]
Signed-off-by: Chunyan Liu <cyliu@suse.com>
qemu-kvm 0.15 uses the same GPE format as qemu 1.4, but as version 2
rather than 3.
Addresses part of BNC#812836.
Signed-off-by: Andreas Färber <afaerber@suse.de>
qemu-kvm 0.15 had a VMSTATE_UINT32(flags, PITState) field that
qemu 1.4 does not have.
Addresses part of BNC#812836.
Signed-off-by: Andreas Färber <afaerber@suse.de>
qemu-kvm.git commit a7fe0297840908a4fd65a1cf742481ccd45960eb
(Extend vram size to 16MB) deviated from qemu.git since kvm-61, and only
in commit 9e56edcf8d (vga: raise default
vgamem size) did qemu.git adjust the VRAM size for v1.2.
Add compatibility properties so that up to and including pc-0.15 we
maintain migration compatibility with qemu-kvm rather than QEMU and
from pc-1.0 on with QEMU (last qemu-kvm release was 1.2).
Addresses part of BNC#812836.
Signed-off-by: Andreas Färber <afaerber@suse.de>
[BR: adjust comma position in list in macro for v2.5.0 compat]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Allow for guests with higher amounts of ram. The current thought
is that 2TB specified on qemu commandline would be an appropriate
limit. Note that this requires the next higher bit value since
the highest address is actually more than 2TB due to the pci
memory hole.
Signed-off-by: Bruce Rogers <brogers@suse.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
For SLES we want users to be able to use large memory configurations
with KVM without fiddling with ulimit -Sv.
Signed-off-by: Andreas Färber <afaerber@suse.de>
[BR: add include for sys/resource.h]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Certain rom subpackages build from qemu git-submodules call the date
program to include date information in the packaged binaries. This
causes repeated builds of the package to be different, wkere the only
real difference is due to the fact that time build timestamp has
changed. To promote reproducible builds and avoid customers being
prompted to update packages needlessly, we'll use the timestamp of the
VERSION file as the packaging timestamp for all packages that build in a
timestamp for whatever reason.
[BR: BSC#1011213]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Instead of post-processing the real contents use the remembered target
argv. That removes all traces of qemu, including command line options,
and handles QEMU_ARGV0.
Signed-off-by: Andreas Schwab <schwab@suse.de>
Signed-off-by: Andreas Färber <afaerber@suse.de>
After "linux-user: use target_ulong" the poll syscall was no longer
handling infinite timeout.
/home/abuild/rpmbuild/BUILD/qemu-2.7.0-rc5/linux-user/syscall.c:9773:26: warning: comparison of unsigned expression >= 0 is always true [-Wtype-limits]
if (arg3 >= 0) {
^~
Signed-off-by: Andreas Schwab <schwab@suse.de>
Change from using glib alloc and free routines to those
from libc. Also perform safety measure of dropping privs
to user if configured no-caps.
[BR: BOO#988279]
Signed-off-by: Bruce Rogers <brogers@suse.com>
[AF: Rebased for v2.7.0-rc2]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Add code to read the suse specific suse-diskcache-disable-flush flag out
of xenstore, and set the equivalent flag within QEMU.
Patch taken from Xen's patch queue, Olaf Hering being the original author.
[bsc#879425]
Signed-off-by: Bruce Rogers <brogers@suse.com>
Signed-off-by: Olaf Hering <olaf@aepfle.de>
The dictzip code in SLE11 received some treatment over time to support
running on big endian hosts. Somewhere in the transition to SLE12 this
support got lost. Add it back in again from the SLE11 code base.
Furthermore while at it, fix up the debug prints to not emit warnings.
[AG: BSC#937572]
Signed-off-by: Alexander Graf <agraf@suse.de>
[AF: Rebased for v2.7.0-rc2]
Signed-off-by: Andreas Färber <afaerber@suse.de>
On hosts with limited virtual address space (32bit pointers), we can very
easily run out of virtual memory with big thread pools.
Instead, we should limit ourselves to small pools to keep memory footprint
low on those systems.
This patch fixes random VM stalls like
(process:25114): GLib-ERROR **: gmem.c:103: failed to allocate 1048576 bytes
on 32bit ARM systems for me.
Signed-off-by: Alexander Graf <agraf@suse.de>
When doing lseek, SEEK_SET indicates that the offset is an unsigned variable.
Other seek types have parameters that can be negative.
When converting from 32bit to 64bit parameters, we need to take this into
account and enable SEEK_END and SEEK_CUR to be negative, while SEEK_SET stays
absolute positioned which we need to maintain as unsigned.
Signed-off-by: Alexander Graf <agraf@suse.de>
Virtio-Console can only process one character at a time. Using it on S390
gave me strage "lags" where I got the character I pressed before when
pressing one. So I typed in "abc" and only received "a", then pressed "d"
but the guest received "b" and so on.
While the stdio driver calls a poll function that just processes on its
queue in case virtio-console can't take multiple characters at once, the
muxer does not have such callbacks, so it can't empty its queue.
To work around that limitation, I introduced a new timer that only gets
active when the guest can not receive any more characters. In that case
it polls again after a while to check if the guest is now receiving input.
This patch fixes input when using -nographic on s390 for me.
[AF: Rebased for v2.7.0-rc2]
Some termcaps (found using SLES11SP1) use [? sequences. According to man
console_codes (http://linux.die.net/man/4/console_codes) the question mark
is a nop and should simply be ignored.
This patch does exactly that, rendering screen output readable when
outputting guest serial consoles to the graphical console emulator.
Signed-off-by: Alexander Graf <agraf@suse.de>
Tar is a very widely used format to store data in. Sometimes people even put
virtual machine images in there.
So it makes sense for qemu to be able to read from tar files. I implemented a
written from scratch reader that also knows about the GNU sparse format, which
is what pigz creates.
This version checks for filenames that end on well-known extensions. The logic
could be changed to search for filenames given on the command line, but that
would require changes to more parts of qemu.
The tar reader in conjunctiuon with dzip gives us the chance to download
tar'ed up virtual machine images (even via http) and instantly make use of
them.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Bruce Rogers <brogers@novell.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
[TH: Use bdrv_open options instead of filename]
Signed-off-by: Tim Hardeck <thardeck@suse.de>
[AF: bdrv_file_open got an Error **errp argument, bdrv_delete -> brd_unref]
[AF: qemu_opts_create_nofail() -> qemu_opts_create(),
bdrv_file_open() -> bdrv_open(), based on work by brogers]
[AF: error_is_set() dropped for v2.1.0-rc0]
[AF: BlockDriverAIOCB -> BlockAIOCB,
BlockDriverCompletionFunc -> BlockCompletionFunc,
qemu_aio_release() -> qemu_aio_unref(),
drop tar_aio_cancel()]
[AF: common-obj-y -> block-obj-y, drop probe hook (bsc#945778)]
[AF: Drop bdrv_open() drv parameter for 2.5]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Bruce Rogers <brogers@suse.com>
[AF: Changed bdrv_open() bs parameter and return value for v2.7.0-rc2,
for bdrv_pread() and bdrv_aio_readv() s/s->hd/s->hd->file/]
Signed-off-by: Andreas Färber <afaerber@suse.de>
DictZip is an extension to the gzip format that allows random seeks in gzip
compressed files by cutting the file into pieces and storing the piece offsets
in the "extra" header of the gzip format.
Thanks to that extension, we can use gzip compressed files as block backend,
though only in read mode.
This makes a lot of sense when stacked with tar files that can then be shipped
to VM users. If a VM image is inside a tar file that is inside a DictZip
enabled gzip file, the user can run the tar.gz file as is without having to
extract the image first.
Tar patch follows.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Bruce Rogers <brogers@novell.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
[TH: Use bdrv_open options instead of filename]
Signed-off-by: Tim Hardeck <thardeck@suse.de>
[AF: Error **errp added for bdrv_file_open, bdrv_delete -> bdrv_unref]
[AF: qemu_opts_create_nofail() -> qemu_opts_create(),
bdrv_file_open() -> bdrv_open(), based on work by brogers]
[AF: error_is_set() dropped for v2.1.0-rc0]
[AF: BlockDriverAIOCB -> BlockAIOCB,
BlockDriverCompletionFunc -> BlockCompletionFunc,
qemu_aio_release() -> qemu_aio_unref(),
drop dictzip_aio_cancel()]
[AF: common-obj-y -> block-obj-y, drop probe hook (bsc#945778)]
[AF: Drop bdrv_open() drv parameter for 2.5]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Bruce Rogers <brogers@suse.com>
[AF: Drop bdrv_open() bs parameter and change return value for v2.7.0-rc2,
for bdrv_pread() and bdrv_aio_readv() do s/s->hd/s->hd->file/]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Linux syscalls pass pointers or data length or other information of that sort
to the kernel. This is all stuff you don't want to have sign extended.
Otherwise a host 64bit variable parameter with a size parameter will extend
it to a negative number, breaking lseek for example.
Pass syscall arguments as ulong always.
Signed-off-by: Alexander Graf <agraf@suse.de>
Fedora 17 for ARM reads /proc/cpuinfo and fails if it doesn't contain
ARM related contents. This patch implements a quick hack to expose real
/proc/cpuinfo data taken from a real world machine.
The real fix would be to generate at least the flags automatically based
on the selected CPU. Please do not submit this patch upstream until this
has happened.
Signed-off-by: Alexander Graf <agraf@suse.de>
[AF: Rebased for v1.6 and v1.7]
Signed-off-by: Andreas Färber <afaerber@suse.de>
When we have a working host binary equivalent for the guest binary we're
trying to run, let's just use that instead as it will be a lot faster.
Signed-off-by: Alexander Graf <agraf@suse.de>
When entering the guest we take a lock to ensure that nobody else messes
with our TB chaining while we're doing it. If we get a segfault inside that
code, we manage to work on, but will not unlock the lock.
This patch forces unlocking of that lock in the segv handler. I'm not sure
this is the right approach though. Maybe we should rather make sure we don't
segfault in the code? I would greatly appreciate someone more intelligible
than me to look at this :).
Example code to trigger this is at: http://csgraf.de/tmp/conftest.c
Reported-by: Fabio Erculiani <lxnay@sabayon.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
[AF: Drop spinlock_safe_unlock() and switch to tb_lock_reset() (bonzini)]
Signed-off-by: Andreas Färber <afaerber@suse.de>
When using hugetlbfs (which is required for HV mode KVM on 970), we
check for MMU notifiers that on 970 can not be implemented properly.
So disable the check for mmu notifiers on PowerPC guests, making
KVM guests work there, even if possibly racy in some odd circumstances.
When using qemu's linux-user binaries through binfmt, argv[0] gets lost
along the execution because qemu only gets passed in the full file name
to the executable while argv[0] can be something completely different.
This breaks in some subtile situations, such as the grep and make test
suites.
This patch adds a wrapper binary called qemu-$TARGET-binfmt that can be
used with binfmt's P flag which passes the full path _and_ argv[0] to
the binfmt handler.
The binary would be smart enough to be versatile and only exist in the
system once, creating the qemu binary path names from its own argv[0].
However, this seemed like it didn't fit the make system too well, so
we're currently creating a new binary for each target archictecture.
CC: Reinhard Max <max@suse.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
[AF: Rebased onto new Makefile infrastructure, twice]
[AF: Updated for aarch64 for v2.0.0-rc1]
[AF: Rebased onto Makefile changes for v2.1.0-rc0]
[AF: Rebased onto script rewrite for v2.7.0-rc2 - to be fixed]
Signed-off-by: Andreas Färber <afaerber@suse.de>
the direction given in the ioctl should be correct so we can assume the
communication is uni-directional. The alsa developers did not like this
concept though and declared ioctls IOC_R and IOC_W even though they were
IOC_RW.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Ulrich Hecht <uli@suse.de>
Hack to prevent ALSA from using mmap() interface to simplify emulation.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Ulrich Hecht <uli@suse.de>
Implements ALSA ioctls on PPC hosts.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Ulrich Hecht <uli@suse.de>
[AF: Rebased for v2.7.0-rc2]
[BR: Rebased for v2.9.0-rc0: removed timespec ref. from syscall_types_alsa.h]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Bruce Rogers <brogers@suse.com>
Git-commit: 89fbea8737
References: bsc#1182137
Depending on the client activity, the server can be asked to open a huge
number of file descriptors and eventually hit RLIMIT_NOFILE. This is
currently mitigated using a reclaim logic : the server closes the file
descriptors of idle fids, based on the assumption that it will be able
to re-open them later. This assumption doesn't hold of course if the
client requests the file to be unlinked. In this case, we loop on the
entire fid list and mark all related fids as unreclaimable (the reclaim
logic will just ignore them) and, of course, we open or re-open their
file descriptors if needed since we're about to unlink the file.
This is the purpose of v9fs_mark_fids_unreclaim(). Since the actual
opening of a file can cause the coroutine to yield, another client
request could possibly add a new fid that we may want to mark as
non-reclaimable as well. The loop is thus restarted if the re-open
request was actually transmitted to the backend. This is achieved
by keeping a reference on the first fid (head) before traversing
the list.
This is wrong in several ways:
- a potential clunk request from the client could tear the first
fid down and cause the reference to be stale. This leads to a
use-after-free error that can be detected with ASAN, using a
custom 9p client
- fids are added at the head of the list : restarting from the
previous head will always miss fids added by a some other
potential request
All these problems could be avoided if fids were being added at the
end of the list. This can be achieved with a QSIMPLEQ, but this is
probably too much change for a bug fix. For now let's keep it
simple and just restart the loop from the current head.
Fixes: CVE-2021-20181
Buglink: https://bugs.launchpad.net/qemu/+bug/1911666
Reported-by: Zero Day Initiative <zdi-disclosures@trendmicro.com>
Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Message-Id: <161064025265.1838153.15185571283519390907.stgit@bahia.lan>
Signed-off-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Bruce Rogers <brogers@suse.com>
We internally convert -virtfs to -fsdev/-device. If the user doesn't
provide the path or security_model suboptions, and the fsdev backend
requires them, we hit an assertion when populating the internal -fsdev
option:
util/qemu-option.c:547: opt_set: Assertion `opt->str' failed.
Aborted (core dumped)
Let's test the suboption presence on the command line before trying
to set it in the internal -fsdev option, and let the backend code
error out gracefully (ie, like it already does when the user passes
-fsdev on the command line).
Reported-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 32b6943699)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Commit 04bf2526ce (exec: use
qemu_ram_ptr_length to access guest ram) start using qemu_ram_ptr_length
instead of qemu_map_ram_ptr, but when used with Xen, the behavior of
both function is different. They both call xen_map_cache, but one with
"lock", meaning the mapping of guest memory is never released
implicitly, and the second one without, which means, mapping can be
release later, when needed.
In the context of address_space_{read,write}_continue, the ptr to those
mapping should not be locked because it is used immediatly and never
used again.
The lock parameter make it explicit in which context qemu_ram_ptr_length
is called.
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Message-Id: <20170726165326.10327-1-anthony.perard@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit f5aa69bdc3)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The Xen mapcache is able to create long term mappings, they are called
"locked" mappings. The third parameter of the xen_map_cache call
specifies if a mapping is a "locked" mapping.
>From the QEMU point of view there are two kinds of long term mappings:
[a] device memory mappings, such as option roms and video memory
[b] dma mappings, created by dma_memory_map & friends
After certain operations, ballooning a VM in particular, Xen asks QEMU
kindly to destroy all mappings. However, certainly [a] mappings are
present and cannot be removed. That's not a problem as they are not
affected by balloonning. The *real* problem is that if there are any
mappings of type [b], any outstanding dma operations could fail. This is
a known shortcoming. In other words, when Xen asks QEMU to destroy all
mappings, it is an error if any [b] mappings exist.
However today we have no way of distinguishing [a] from [b]. Because of
that, we cannot even print a decent warning.
This patch introduces a new "dma" bool field to MapCacheRev entires, to
remember if a given mapping is for dma or is a long term device memory
mapping. When xen_invalidate_map_cache is called, we print a warning if
any [b] mappings exist. We ignore [a] mappings.
Mappings created by qemu_map_ram_ptr are assumed to be [a], while
mappings created by address_space_map->qemu_ram_ptr_length are assumed
to be [b].
The goal of the patch is to make debugging and system understanding
easier.
Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
(cherry picked from commit 1ff7c5986a)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
There's a rare exit seg if the guest is accessing
IO during exit.
It's always hitting the atomic_inc(&bs->in_flight) with a NULL
bs. This was added recently in 99723548 but I don't see it
as the cause.
Flip vl.c around so we pause the cpus before closing the block devices,
that way we shouldn't have anything trying to access them when
they're gone.
This was originally Red Hat bz https://bugzilla.redhat.com/show_bug.cgi?id=1451015
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reported-by: Cong Li <coli@redhat.com>
--
This is a very rare race, I'll leave it running in a loop to see if
we hit anything else and to check this really fixes it.
I do worry if there are other cases that can trigger this - e.g.
hot-unplug or ejecting a CD.
Message-Id: <20170713190116.21608-1-dgilbert@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 452589b6b4)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Previous submodule commit contained some unused files with comments
that made redistribution requirements unclear, and the submodule commit
should really match the blob we're shipped anyway.
Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The current VNC default keyboard delay is 1ms. With that we're constantly
typing faster than the guest receives keyboard events from an XHCI attached
USB HID device.
The default keyboard delay time in the input layer however is 10ms. I don't know
how that number came to be, but empirical tests on some OpenQA driven ARM
systems show that 10ms really is a reasonable default number for the delay.
This patch moves the VNC delay also to 10ms. That way our default is much
safer (good!) and also consistent with the input layer default (also good!).
Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 1499863425-103133-1-git-send-email-agraf@suse.de
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit d3b0db6dfe)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
qemu proper has done so for 13 years
(8a7ddc38a6), qemu-img and qemu-io have
done so for four years (526eda14a6).
Ignoring this signal is especially important in qemu-nbd because
otherwise a client can easily take down the qemu-nbd server by dropping
the connection when the server wants to send something, for example:
$ qemu-nbd -x foo -f raw -t null-co:// &
[1] 12726
$ qemu-io -c quit nbd://localhost/bar
can't open device nbd://localhost/bar: No export with name 'bar' available
[1] + 12726 broken pipe qemu-nbd -x foo -f raw -t null-co://
In this case, the client sends an NBD_OPT_ABORT and closes the
connection (because it is not required to wait for a reply), but the
server replies with an NBD_REP_ACK (because it is required to reply).
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20170611123714.31292-1-mreitz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 041e32b8d9)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Avoid TOC-TOU bugs by passing the frame_cmd down, and checking
cmd->dcmd_opcode instead of cmd->frame->header.frame_cmd.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 36c327a69d)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
When using the mapped-file security mode, we shouldn't let the client mess
with the metadata. The current code already tries to hide the metadata dir
from the client by skipping it in local_readdir(). But the client can still
access or modify it through several other operations. This can be used to
escalate privileges in the guest.
Affected backend operations are:
- local_mknod()
- local_mkdir()
- local_open2()
- local_symlink()
- local_link()
- local_unlinkat()
- local_renameat()
- local_rename()
- local_name_to_path()
Other operations are safe because they are only passed a fid path, which
is computed internally in local_name_to_path().
This patch converts all the functions listed above to fail and return
EINVAL when being passed the name of the metadata dir. This may look
like a poor choice for errno, but there's no such thing as an illegal
path name on Linux and I could not think of anything better.
This fixes CVE-2017-7493.
Reported-by: Leo Gaspard <leo@gaspard.io>
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
(cherry picked from commit 7a95434e0c)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
QEMU currently aborts unexpectedly when the user tries to add and
remove a "spapr-tce-table" device:
$ qemu-system-ppc64 -nographic -S -nodefaults -monitor stdio
QEMU 2.9.92 monitor - type 'help' for more information
(qemu) device_add spapr-tce-table,id=x
(qemu) device_del x
**
ERROR:qemu/qdev-monitor.c:872:qdev_unplug: assertion failed: (hotplug_ctrl)
Aborted (core dumped)
The device should not be accessable for the users at all, it's just
used internally, so mark it with user_creatable = false.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit 1f98e55385)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
QEMU currently aborts unexpectedly when a user tries to do something
like this:
$ qemu-system-ppc64 -nographic -S -nodefaults -monitor stdio
QEMU 2.9.92 monitor - type 'help' for more information
(qemu) device_add spapr-rtc,id=spapr-rtc
(qemu) device_del spapr-rtc
**
ERROR:qemu/qdev-monitor.c:872:qdev_unplug: assertion failed: (hotplug_ctrl)
Aborted (core dumped)
The RTC device is not meant to be hot-pluggable - it's an internal
device only and it even should not be possible to create it a
second time with the "-device" parameter, so let's mark this
with "user_creatable = false".
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit 8ccccff9dd)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
cannot_instantiate_with_device_add_yet was introduced by commit
efec3dd631 to replace no_user. It was
supposed to be a temporary measure.
When it was introduced, we had 54
cannot_instantiate_with_device_add_yet=true lines in the code.
Today (3 years later) this number has not shrunk: we now have
57 cannot_instantiate_with_device_add_yet=true lines. I think it
is safe to say it is not a temporary measure, and we won't see
the flag go away soon.
Instead of a long field name that misleads people to believe it
is temporary, replace it a shorter and less misleading field:
user_creatable.
Except for code comments, changes were generated using the
following Coccinelle patch:
@@
expression DC;
@@
(
-DC->cannot_instantiate_with_device_add_yet = false;
+DC->user_creatable = true;
|
-DC->cannot_instantiate_with_device_add_yet = true;
+DC->user_creatable = false;
)
@@
typedef ObjectClass;
expression dc;
identifier class, data;
@@
static void device_class_init(ObjectClass *class, void *data)
{
...
dc->hotpluggable = true;
+dc->user_creatable = true;
...
}
@@
@@
struct DeviceClass {
...
-bool cannot_instantiate_with_device_add_yet;
+bool user_creatable;
...
}
@@
expression DC;
@@
(
-!DC->cannot_instantiate_with_device_add_yet
+DC->user_creatable
|
-DC->cannot_instantiate_with_device_add_yet
+!DC->user_creatable
)
Cc: Alistair Francis <alistair.francis@xilinx.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Thomas Huth <thuth@redhat.com>
Acked-by: Alistair Francis <alistair.francis@xilinx.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Acked-by: Marcel Apfelbaum <marcel@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20170503203604.31462-2-ehabkost@redhat.com>
[ehabkost: kept "TODO remove once we're there" comment]
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit e90f2a8c3e)
Conflicts:
include/hw/qdev-core.h
* remove context dep on 08f00df
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Starting qemu-system-unicore32 without the -kernel parameter results in
an assert() returns false and aborts qemu. This patch replaces it with a
proper error message followed by exit(1).
Signed-off-by: Eduardo Otubo <otubo@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
(cherry picked from commit 36bed541ca)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
qemu-system-s390x currently crashes when it is started with a
virtio-scsi-pci device, e.g.:
qemu-system-s390x -nographic -enable-kvm -device virtio-scsi-pci \
-drive file=/tmp/disk.dat,if=none,id=d1,format=raw \
-device scsi-cd,drive=d1,bootindex=1
The problem is that the code in s390_gen_initial_iplb() currently assumes
that all SCSI devices are also CCW devices, which is not the case for
virtio-scsi-pci of course. Fix it by adding an appropriate check for
TYPE_CCW_DEVICE here.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <1493126327-13162-1-git-send-email-thuth@redhat.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
(cherry picked from commit 99efaa2696)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The if_fastq and if_batchq contain not only packets, but queues of packets
for the same socket. When sofree frees a socket, it thus has to clear ifq_so
from all the packets from the queues, not only the first.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Cc: qemu-stable@nongnu.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
(cherry picked from commit 1201d30851)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
ASAN detects an "unknown-crash" when running pxe-test:
/ppc64/pxe/spapr-vlan: =================================================================
==7143==ERROR: AddressSanitizer: unknown-crash on address 0x7f6dcd298d30 at pc 0x55e22218830d bp 0x7f6dcd2989e0 sp 0x7f6dcd2989d0
READ of size 128 at 0x7f6dcd298d30 thread T2
#0 0x55e22218830c in tftp_session_allocate /home/elmarco/src/qq/slirp/tftp.c:73
#1 0x55e22218a1f8 in tftp_handle_rrq /home/elmarco/src/qq/slirp/tftp.c:289
#2 0x55e22218b54c in tftp_input /home/elmarco/src/qq/slirp/tftp.c:446
#3 0x55e2221833fe in udp6_input /home/elmarco/src/qq/slirp/udp6.c:82
#4 0x55e222137b17 in ip6_input /home/elmarco/src/qq/slirp/ip6_input.c:67
Address 0x7f6dcd298d30 is located in stack of thread T2 at offset 96 in frame
#0 0x55e222182420 in udp6_input /home/elmarco/src/qq/slirp/udp6.c:13
This frame has 3 object(s):
[32, 48) '<unknown>'
[96, 124) 'lhost' <== Memory access at offset 96 partially overflows this variable
[160, 200) 'save_ip' <== Memory access at offset 96 partially underflows this variable
The sockaddr_storage pointer is the sockaddr_in6 lhost on the
stack. Copy only the source addr size.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
(cherry picked from commit 17eb587aeb)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
According to "CPU Signaling and Response", "Signal-Processor Orders",
the order field is bit position 56-63. Without this, the Linux
guest kernel is sometimes unable to stop emulation and enters
an infinite loop of "XXX unknown sigp: 0xffffffff00000005".
Signed-off-by: Philipp Kern <phil@philkern.de>
Reviewed-by: Thomas Huth <thuth@tuxfamily.org>
[agraf: add comment according to email]
Signed-off-by: Alexander Graf <agraf@suse.de>
(cherry picked from commit 601b9a9008)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
This function has to ensure it doesn't follow a symlink that could be used
to escape the virtfs directory. This could be easily achieved if fchmodat()
on linux honored the AT_SYMLINK_NOFOLLOW flag as described in POSIX, but
it doesn't. There was a tentative to implement a new fchmodat2() syscall
with the correct semantics:
https://patchwork.kernel.org/patch/9596301/
but it didn't gain much momentum. Also it was suggested to look at an O_PATH
based solution in the first place.
The current implementation covers most use-cases, but it notably fails if:
- the target path has access rights equal to 0000 (openat() returns EPERM),
=> once you've done chmod(0000) on a file, you can never chmod() again
- the target path is UNIX domain socket (openat() returns ENXIO)
=> bind() of UNIX domain sockets fails if the file is on 9pfs
The solution is to use O_PATH: openat() now succeeds in both cases, and we
can ensure the path isn't a symlink with fstat(). The associated entry in
"/proc/self/fd" can hence be safely passed to the regular chmod() syscall.
The previous behavior is kept for older systems that don't have O_PATH.
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Zhi Yong Wu <zhiyong.wu@ucloud.cn>
Acked-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
(cherry picked from commit 4751fd5328)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Commit c096358e74 introduced assertion
checks for when qemu_mutex() functions are called without the
corresponding qemu_mutex_init() having initialized the mutex.
This uncovered a latent bug in qemu's nfs driver - in
nfs_client_close(), the NFSClient structure is overwritten with zeros,
prior to the mutex being destroyed.
Go ahead and destroy the mutex in nfs_client_close(), and change where
we call qemu_mutex_init() so that it is correctly balanced.
There are also a couple of memory leaks obscured by the memset, so this
fixes those as well.
Finally, we should be able to get rid of the memset(), as it isn't
necessary.
Cc: qemu-stable@nongnu.org
Signed-off-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Peter Lieven <pl@kamp.de>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 113fe792fd)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
With pseries machine type a negative core-id is not managed properly:
-1 gives an inaccurate error message ("core -1 already populated"),
-2 crashes QEMU (core dump)
As it seems a negative value is invalid for any architecture,
instead of checking this in spapr_core_pre_plug() I think it's better
to check this in the generic part, core_prop_set_core_id()
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Message-Id: <20170802103259.25940-1-lvivier@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit be2960baae)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Commits 0db832f and 6cdbceb introduced the automatic insertion of filter
nodes above the top layer of mirror and commit block jobs. The
assumption made there was that since libvirt doesn't do node-level
management of the block layer yet, it shouldn't be affected by added
nodes.
This is true as far as commands issued by libvirt are concerned. It only
uses BlockBackend names to address nodes, so any operations it performs
still operate on the root of the tree as intended.
However, the assumption breaks down when you consider query commands,
which return data for the wrong node now. These commands also return
information on some child nodes (bs->file and/or bs->backing), which
libvirt does make use of, and which refer to the wrong nodes, too.
One of the consequences is that oVirt gets wrong information about the
image size and stops the VM in response as long as a mirror or commit
job is running:
https://bugzilla.redhat.com/show_bug.cgi?id=1470634
This patch fixes the problem by hiding the implicit nodes created
automatically by the mirror and commit block jobs in the output of
query-block and BlockBackend-based query-blockstats as long as the user
doesn't indicate that they are aware of those nodes by providing a node
name for them in the QMP command to start the block job.
The node-based commands query-named-block-nodes and query-blockstats
with query-nodes=true still show all nodes, including implicit ones.
This ensures that users that are capable of node-level management can
still access the full information; users that only know BlockBackends
won't use these commands.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Tested-by: Eric Blake <eblake@redhat.com>
(cherry picked from commit d3c8c67469)
Conflicts:
block/qapi.c
include/block/block_int.h
* fix context deps on 46eade7b and 5a9347c6
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Delays in the input layer are special cased input events. Every input
event is accounted for in a global intput queue count. The special cased
delays however did not get removed from the queue, leading to queue overruns
and thus silent key drops after typing quite a few characters.
Signed-off-by: Alexander Graf <agraf@suse.de>
Message-id: 1498117318-162102-1-git-send-email-agraf@suse.de
Fixes: be1a7176 ("input: add support for kbd delays")
Cc: qemu-stable@nongnu.org
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 77b0359bf4)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
In case of error, we must ensure the dynamically allocated base_core_type
is freed, like it is done everywhere else in this function.
This is a regression introduced in QEMU 2.9 by commit 8149e2992f.
Signed-off-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit df8658de43)
Conflicts:
hw/ppc/spapr.c
* fix context dep on 459264ef2
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
I can't see how overlay_bs could become NULL with the current code, but
other code in this function already checks it and we can make Coverity
happy with this check, so let's add it.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit b1e1fa0c3a)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
since commit 5c4537bd ("spapr: Fix 2.7<->2.8 migration of PCI host bridge"),
some migration fields are forged from the new ones in spapr_pci_pre_save().
It works well, except when the number of MSI devices is 0,
because in this case the function exits immediately.
This fix moves the migration code before the exit code.
The problem can be reproduced with these commands:
source qemu-2.9:
qemu-system-ppc64 -monitor stdio -M pseries-2.6 -nodefaults -S
destination qemu-2.6:
qemu-system-ppc64 -monitor stdio -M pseries-2.6 -nodefaults \
-incoming tcp:0:4444
on the source:
migrate tcp:localhost:4444
Destination fails with the following error:
qemu-system-ppc64: error while loading state for
instance 0x0 of device 'spapr_pci'
qemu-system-ppc64: load of migration failed: Invalid argument
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit e806b4db14)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Commit a0e640a8 introduced a path processing error.
Pass fstatat the dirpath based path component instead
of the entire path.
Signed-off-by: Bruce Rogers <brogers@suse.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
(cherry picked from commit 790db7efdb)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
uri_parse(...)->scheme may be NULL. In fact, probably every field may be
NULL, and the callers do test this for all of the other fields but not
for scheme (except for block/gluster.c; block/vxhs.c does not access
that field at all).
We can easily fix this by using g_strcmp0() instead of strcmp().
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170613205726.13544-1-mreitz@redhat.com
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit f69165a8fe)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
When attaching the NBD QIOChannel to an AioContext, the TLS channel should
be used, not the underlying socket channel. This is because, trivially,
the TLS channel will be the one that we read/write to and thus the one
that will get the qio_channel_yield() call.
Fixes: ff82911cd3
Cc: qemu-stable@nongnu.org
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Tested-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 96d06835dc)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
commit_complete() can't assume that after its block_job_completed() the
job is actually immediately freed; someone else may still be holding
references. In this case, the op blockers on the intermediate nodes make
the graph reconfiguration in the completion code fail.
Call block_job_remove_all_bdrv() manually so that we know for sure that
any blockers on intermediate nodes are given up.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 4f78a16fee)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Back in qemu 2.5, qemu-nbd was immune to port probes (a transient
server would not quit, regardless of how many probe connections
came and went, until a connection actually negotiated). But we
broke that in commit ee7d7aa when removing the return value to
nbd_client_new(), although that patch also introduced a bug causing
an assertion failure on a client that fails negotiation. We then
made it worse during refactoring in commit 1a6245a (a segfault
before we could even assert); the (masked) assertion was cleaned
up in d3780c2 (still in 2.6), and just recently we finally fixed
the segfault ("nbd: Fully intialize client in case of failed
negotiation"). But that still means that ever since we added
TLS support to qemu-nbd, we have been vulnerable to an ill-timed
port-scan being able to cause a denial of service by taking down
qemu-nbd before a real client has a chance to connect.
Since negotiation is now handled asynchronously via coroutines,
we no longer have a synchronous point of return by re-adding a
return value to nbd_client_new(). So this patch instead wires
things up to pass the negotiation status through the close_fn
callback function.
Simple test across two terminals:
$ qemu-nbd -f raw -p 30001 file
$ nmap 127.0.0.1 -p 30001 && \
qemu-io -c 'r 0 512' -f raw nbd://localhost:30001
Note that this patch does not change what constitutes successful
negotiation (thus, a client must enter transmission phase before
that client can be considered as a reason to terminate the server
when the connection ends). Perhaps we may want to tweak things
in a later patch to also treat a client that uses NBD_OPT_ABORT
as being a 'successful' negotiation (the client correctly talked
the NBD protocol, and informed us it was not going to use our
export after all), but that's a discussion for another day.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1451614
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20170608222617.20376-1-eblake@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 0c9390d978)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
If a non-NBD client connects to qemu-nbd, we would end up with
a SIGSEGV in nbd_client_put() because we were trying to
unregister the client's association to the export, even though
we skipped inserting the client into that list. Easy trigger
in two terminals:
$ qemu-nbd -p 30001 --format=raw file
$ nmap 127.0.0.1 -p 30001
nmap claims that it thinks it connected to a pago-services1
server (which probably means nmap could be updated to learn the
NBD protocol and give a more accurate diagnosis of the open
port - but that's not our problem), then terminates immediately,
so our call to nbd_negotiate() fails. The fix is to reorder
nbd_co_client_start() to ensure that all initialization occurs
before we ever try talking to a client in nbd_negotiate(), so
that the teardown sequence on negotiation failure doesn't fault
while dereferencing a half-initialized object.
While debugging this, I also noticed that nbd_update_server_watch()
called by nbd_client_closed() was still adding a channel to accept
the next client, even when the state was no longer RUNNING. That
is fixed by making nbd_can_accept() pay attention to the current
state.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1451614
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20170527030421.28366-1-eblake@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit df8ad9f128)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The final bdrv_set_backing_hd() could be working on already freed nodes
because the commit job drops its references (through BlockBackends) to
both overlay_bs and top already a bit earlier.
One way to trigger the bug is hot unplugging a disk for which
blockdev_mark_auto_del() cancels the block job.
Fix this by taking BDS-level references while we're still using the
nodes.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
(cherry picked from commit 19ebd13ed4)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Xtensa cores may have registers of types/sizes not supported by the
gdbstub accessors. Ignore writes to such registers and return zero on
read, but always return correct register size, so that gdb on the other
side is able to access all registers in the packet holding unsupported
registers in the middle. This fixes gdb interaction with cores that have
vector/custom TIE registers.
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
(cherry picked from commit dd7b952b79)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The string returned by object_property_get_str() is dynamically allocated.
(Spotted by Coverity, CID 1375942)
Signed-off-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit 8a9e0e7b89)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
This allows to manage errors before the memory
has started to be hotplugged. We already have
the function for the CPU cores.
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
[dwg: Fixed a couple of style nits]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit c871bc70bb)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
This function has three implementations. Two are stubs that do nothing
and the third one only passes the obj_path argument to:
Object *object_resolve_path(const char *path, bool *ambiguous);
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
(cherry picked from commit ec69355bef)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Since the automatic cpuid-level code was introduced in commit
c39c0edf9b ("target-i386: Automatically
set level/xlevel/xlevel2 when needed"), the CPU model tables just define
the default CPUID level code (set using "min-level"). Setting
"[x]level" forces CPUID level to a specific value and disable the
automatic-level logic.
But the PC compat code was not updated and the existing "[x]level"
compat properties broke compatibility for people using features that
triggered the auto-level code. To keep previous behavior, we should set
"min-[x]level" instead of "[x]level" on compat_props.
This was not a problem for most cases, because old machine-types don't
have full-cpuid-auto-level enabled. The only common use case it broke
was the CPUID[7] auto-level code, that was already enabled since the
first CPUID[7] feature was introduced (in QEMU 1.4.0).
This causes the regression reported at:
https://bugzilla.redhat.com/show_bug.cgi?id=1454641
Change the PC compat code to use "min-[x]level" instead of "[x]level" on
compat_props, and add new test cases to ensure we don't break this
again.
Reported-by: "Guo, Zhiyi" <zhguo@redhat.com>
Fixes: c39c0edf9b ("target-i386: Automatically set level/xlevel/xlevel2 when needed")
Cc: qemu-stable@nongnu.org
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit 1f43571604)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Currently objects specified on the command-line are only partially
cleaned up when 'object_del' is issued in either HMP or QMP: the
object itself is fully finalized, but the QemuOpts are not removed.
This results in the following behavior:
x86_64-softmmu/qemu-system-x86_64 -monitor stdio \
-object memory-backend-ram,id=ram1,size=256M
QEMU 2.7.91 monitor - type 'help' for more information
(qemu) object_del ram1
(qemu) object_del ram1
object 'ram1' not found
(qemu) object_add memory-backend-ram,id=ram1,size=256M
Duplicate ID 'ram1' for object
Try "help object_add" for more information
which can be an issue for use-cases like memory hotplug.
This happens on the HMP side because hmp_object_add() attempts to
create a temporary QemuOpts entry with ID 'ram1', which ends up
conflicting with the command-line-created entry, since it was never
cleaned up during the previous hmp_object_del() call.
We address this by adding a check in user_creatable_del(), which
is called by both qmp_object_del() and hmp_object_del() to handle
the actual object cleanup, to determine whether an option group entry
matching the object's ID is present and removing it if it is.
Note that qmp_object_add() never attempts to create a temporary
QemuOpts entry, so it does not encounter the duplicate ID error,
which is why this isn't generally visible in libvirt.
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: Eric Blake <eblake@redhat.com>
Cc: Daniel Berrange <berrange@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1496531612-22166-3-git-send-email-mdroth@linux.vnet.ibm.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
(cherry picked from commit c645d5acee)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
check-qom-proplist originally added tests for verifying that
object-creation helpers object_new_with_{props,propv} behaved in
similar fashion to the "traditional" method involving setting each
individual property separately after object creation rather than
via a single call.
Another similar "helper" for creating Objects exists in the form of
objects specified via -object command-line parameters. By that
rationale, we extend check-qom-proplist to include similar checks
for command-line-created objects by employing the same
qemu_opts_parse()-based parsing the vl.c employs.
This parser has a side-effect of parsing the object's options into
a QemuOpt structure and registering this in the global QemuOptsList
using the Object's ID. This can conflict with future Object instances
that attempt to use the same ID if we don't ensure this is cleaned
up as part of Object finalization, so we include a FIXME stub to test
for this case, which will then be resolved in a subsequent patch.
Suggested-by: Daniel Berrange <berrange@redhat.com>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: Eric Blake <eblake@redhat.com>
Cc: Daniel Berrange <berrange@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <1496531612-22166-2-git-send-email-mdroth@linux.vnet.ibm.com>
[Comment formatting tidied up]
Signed-off-by: Markus Armbruster <armbru@redhat.com>
(cherry picked from commit a1af255f06)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Virtio serial device controls the lifetime of virtio-serial-bus and
virtio-serial-bus links back to the device via its hotplug-handler
property. This extra ref-count prevents the device from getting
finalized, leaving the VirtIODevice memory listener registered and
leading to use-after-free later on.
This patch addresses the same issue as Fam Zheng's
"virtio-scsi: Unset hotplug handler when unrealize"
only for a different virtio device.
Cc: qemu-stable@nongnu.org
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
(cherry picked from commit f811f97040)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
This fixes an assertion failure that was triggered by qemu-iotests 129
on some CI host, while the same test case didn't seem to fail on other
hosts.
Essentially the problem is that the blk_unref(s->target) in
mirror_exit() doesn't necessarily mean that the BlockBackend goes away
immediately. It is possible that the job completion was triggered nested
in mirror_drain(), which looks like this:
BlockBackend *target = s->target;
blk_ref(target);
blk_drain(target);
blk_unref(target);
In this case, the write permissions for s->target are retained until
after blk_drain(), which makes removing mirror_top_bs fail for the
active commit case (can't have a writable backing file in the chain
without the filter driver).
Explicitly dropping the permissions first means that the additional
reference doesn't hurt and the job can complete successfully even if
called from the nested blk_drain().
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 63c8ef2890)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
We document that *file is valid if the return is not an error and
includes BDRV_BLOCK_OFFSET_VALID, but forgot to obey this contract
when a driver (such as blkdebug) lacks a callback. Messed up in
commit 67a0fd2 (v2.6), when we added the file parameter.
Enhance qemu-iotest 177 to cover this, using a sequence that would
print garbage or even SEGV, because it was dererefencing through
uninitialized memory. [The resulting test output shows that we
have less-than-ideal block status from the blkdebug driver, but
that's a separate fix coming up soon.]
Setting *file on all paths that return BDRV_BLOCK_OFFSET_VALID is
enough to fix the crash, but we can go one step further: always
setting *file, even on error, means that a broken caller that
blindly dereferences file without checking for error is now more
likely to get a reliable SEGV instead of randomly acting on garbage,
making it easier to diagnose such buggy callers. Adding an
assertion that file is set where expected doesn't hurt either.
CC: qemu-stable@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 81c219ac6c)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Since we are already in coroutine context during the body of
bdrv_co_get_block_status(), we can shave off a few layers of
wrappers when recursing to query the protocol when a format driver
returned BDRV_BLOCK_RAW.
Note that we are already using the correct recursion later on in
the same function, when probing whether the protocol layer is sparse
in order to find out if we can add BDRV_BLOCK_ZERO to an existing
BDRV_BLOCK_DATA|BDRV_BLOCK_OFFSET_VALID.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170504173745.27414-1-eblake@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit ee29d6adef)
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Use blkdebug's new geometry constraints to emulate setups that
have needed past regression fixes: write zeroes asserting
when running through a loopback block device with max-transfer
smaller than cluster size, and discard rounding away portions
of requests not aligned to preferred boundaries. Also, add
coverage that the block layer is honoring max transfer limits.
For now, a single iotest performs all actions, with the idea
that we can add future blkdebug constraint test cases in the
same file; but it can be split into multiple iotests if we find
reason to run one portion of the test in more setups than what
are possible in the other.
For reference, the final portion of the test (checking whether
discard passes as much as possible to the lowest layers of the
stack) works as follows:
qemu-io: discard 30M at 80000001, passed to blkdebug
blkdebug: discard 511 bytes at 80000001, -ENOTSUP (smaller than
blkdebug's 512 align)
blkdebug: discard 14371328 bytes at 80000512, passed to qcow2
qcow2: discard 739840 bytes at 80000512, -ENOTSUP (smaller than
qcow2's 1M align)
qcow2: discard 13M bytes at 77M, succeeds
blkdebug: discard 15M bytes at 90M, passed to qcow2
qcow2: discard 15M bytes at 90M, succeeds
blkdebug: discard 1356800 bytes at 105M, passed to qcow2
qcow2: discard 1M at 105M, succeeds
qcow2: discard 308224 bytes at 106M, -ENOTSUP (smaller than qcow2's
1M align)
blkdebug: discard 1 byte at 111457280, -ENOTSUP (smaller than
blkdebug's 512 align)
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170429191419.30051-10-eblake@redhat.com
[mreitz: For cooperation with image locking, add -r to the qemu-io
invocation which verifies the image content]
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 40812d9373)
Conflicts:
tests/qemu-iotests/group
* dropped context dependency on other test groups
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Make it easier to simulate various unusual hardware setups (for
example, recent commits 3482b9b and b8d0a98 affect the Dell
Equallogic iSCSI with its 15M preferred and maximum unmap and
write zero sizing, or b2f95fe deals with the Linux loopback
block device having a max_transfer of 64k), by allowing blkdebug
to wrap any other device with further restrictions on various
alignments.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170429191419.30051-9-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 430b26a82d)
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Rather than store into a local variable, then copy to the struct
if the value is valid, then reporting errors otherwise, it is
simpler to just store into the struct and report errors if the
value is invalid. This however requires that the struct store
a 64-bit number, rather than a narrower type. Likewise, setting
a sane errno value in ret prior to the sequence of parsing and
jumping to out: on error makes it easier for the next patch to
add a chain of similar checks.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170429191419.30051-8-eblake@redhat.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 3dc834f879)
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
In order to test the effects of artificial geometry constraints
on operations like write zero or discard, we first need blkdebug
to manage these actions. It also allows us to inject errors on
those operations, just like we can for read/write/flush.
We can also test the contract promised by the block layer; namely,
if a device has specified limits on alignment or maximum size,
then those limits must be obeyed (for now, the blkdebug driver
merely inherits limits from whatever it is wrapping, but the next
patch will further enhance it to allow specific limit overrides).
This patch intentionally refuses to service requests smaller than
the requested alignments; this is because an upcoming patch adds
a qemu-iotest to prove that the block layer is correctly handling
fragmentation, but the test only works if there is a way to tell
the difference at artificial alignment boundaries when blkdebug is
using a larger-than-default alignment. If we let the blkdebug
layer always defer to the underlying layer, which potentially has
a smaller granularity, the iotest will be thwarted.
Tested by setting up an NBD server with export 'foo', then invoking:
$ ./qemu-io
qemu-io> open -o driver=blkdebug blkdebug::nbd://localhost:10809/foo
qemu-io> d 0 15M
qemu-io> w -z 0 15M
Pre-patch, the server never sees the discard (it was silently
eaten by the block layer); post-patch it is passed across the
wire. Likewise, pre-patch the write is always passed with
NBD_WRITE (with 15M of zeroes on the wire), while post-patch
it can utilize NBD_WRITE_ZEROES (for less traffic).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170429191419.30051-7-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit 63188c2450)
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Rather than repeat the logic at each caller of checking if a Rule
exists that warrants an error injection, fold that logic into
inject_error(); and rename it to rule_check() for legibility.
This will help the next patch, which adds two more callers that
need to check rules for the potential of injecting errors.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170429191419.30051-6-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit d157ed5f72)
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Commits 04ed95f4 and 1a62d0ac updated the block layer to auto-fragment
any I/O to fit within device boundaries. Additionally, when using a
minimum alignment of 4k, we want to ensure the block layer does proper
read-modify-write rather than requesting I/O on a slice of a sector.
Let's enforce that the contract is obeyed when using blkdebug. For
now, blkdebug only allows alignment overrides, and just inherits other
limits from whatever device it is wrapping, but a future patch will
further enhance things.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170429191419.30051-5-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
(cherry picked from commit e0ef439588)
* prereq for 81c219a
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The tx_bh or tx_timer will free in virtio_net_del_queue() function, when
removing virtio-net queues if the guest doesn't support multiqueue. But
it might be still referenced by virtio_net_set_status(), which needs to
be set NULL. And also the tx_waiting needs to be set zero to prevent
virtio_net_set_status() accessing tx_bh or tx_timer.
Cc: qemu-stable@nongnu.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit f989c30cf8)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Prior to the virtio-ccw-2.7 machine (and commit 2a79eb1a), our virtio
devices residing under the virtual-css bus do not have qdev_path based
migration stream identifiers (because their qdev_path is NULL). The ids
are instead generated when the device is registered as a composition of
the so called idstr, which takes the vmsd name as its value, and an
instance_id, which is which is calculated as a maximal instance_id
registered with the same idstr plus one, or zero (if none was registered
previously).
That means, under certain circumstances, one device might try, and even
succeed, to load the state of a different device. This can lead to
trouble.
Let us fail the migration if the above problem is detected during load.
How to reproduce the problem:
1) start qemu-system-s390x making sure you have the following devices
defined on your command line:
-device virtio-rng-ccw,id=rng1,devno=fe.0.0001
-device virtio-rng-ccw,id=rng2,devno=fe.0.0002
2) detach the devices and reattach in reverse order using the monitor:
(qemu) device_del rng1
(qemu) device_del rng2
(qemu) device_add virtio-rng-ccw,id=rng2,devno=fe.0.0002
(qemu) device_add virtio-rng-ccw,id=rng1,devno=fe.0.0001
3) save the state of the vm into a temporary file and quit QEMU:
(qemu) migrate "exec:gzip -c > /tmp/tmp_vmstate.gz"
(qemu) q
4) use your command line from step 1 with
-incoming "exec:gzip -c -d /tmp/tmp_vmstate.gz"
appended to reproduce the problem (while trying to to load the saved vm)
CC: qemu-stable@nongnu.org
Signed-off-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Message-Id: <20170518111405.56947-1-pasic@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
(cherry picked from commit 8ed179c937)
* removed context dep on d8d98db5
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
This commit fixes a bug which causes the guest to hang. The bug was
observed upon a "receive overrun" (bit #6 of the ICR register)
interrupt which could be triggered post migration in a heavy traffic
environment. Even though the "receive overrun" bit (#6) is masked out
by the IMS register (refer to the log below) the driver still receives
an interrupt as the "receive overrun" bit (#6) causes the "Other" -
bit #24 of the ICR register - bit to be set as documented below. The
driver handles the interrupt and clears the "Other" bit (#24) but
doesn't clear the "receive overrun" bit (#6) which leads to an
infinite loop. Apparently the Windows driver expects that the "receive
overrun" bit and other ones - documented below - to be cleared when
the "Other" bit (#24) is cleared.
So to sum that up:
1. Bit #6 of the ICR register is set by heavy traffic
2. As a results of setting bit #6, bit #24 is set
3. The driver receives an interrupt for bit 24 (it doesn't receieve an
interrupt for bit #6 as it is masked out by IMS)
4. The driver handles and clears the interrupt of bit #24
5. Bit #6 is still set.
6. 2 happens all over again
The Interrupt Cause Read - ICR register:
The ICR has the "Other" bit - bit #24 - that is set when one or more
of the following ICR register's bits are set:
LSC - bit #2, RXO - bit #6, MDAC - bit #9, SRPD - bit #16, ACK - bit
#17, MNG - bit #18
This bug can occur with any of these bits depending on the driver's
behaviour and the way it configures the device. However, trying to
reproduce it with any bit other than RX0 is challenging and came to
failure as the drivers don't implement most of these bits, trying to
reproduce it with LSC (Link Status Change - bit #2) bit didn't succeed
too as it seems that Windows handles this bit differently.
Log sample of the storm:
27563@1494850819.411877:e1000e_irq_pending_interrupts ICR PENDING: 0x1000000 (ICR: 0x815000c2, IMS: 0x1a00004)
27563@1494850819.411900:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
27563@1494850819.411915:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
27563@1494850819.412380:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
27563@1494850819.412395:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
27563@1494850819.412436:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
27563@1494850819.412441:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
27563@1494850819.412998:e1000e_irq_pending_interrupts ICR PENDING: 0x1000000 (ICR: 0x815000c2, IMS: 0x1a00004)
* This bug behaviour wasn't observed with the Linux driver.
This commit solves:
https://bugzilla.redhat.com/show_bug.cgi?id=1447935https://bugzilla.redhat.com/show_bug.cgi?id=1449490
Cc: qemu-stable@nongnu.org
Signed-off-by: Sameeh Jubran <sjubran@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit 82342e91b6)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
This matches the qbus_set_hotplug_handler in realize, and it releases
the final reference to the embedded VirtIODevice so that it is
properly finalized.
A use-after-free is fixed with this patch, indirectly:
virtio_device_instance_finalize wasn't called at hot-unplug, and the
vdev->listener would be a dangling pointer in the global and the per
address space listener list. See also RHBZ 1449031.
Cc: qemu-stable@nongnu.org
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <20170518102808.30046-1-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 2cbe2de545)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
According to section 2.1.2 of the virtio-1 specification:
"The device SHOULD set DEVICE_NEEDS_RESET when it enters an error state that
a reset is needed. If DRIVER_OK is set, after it sets DEVICE_NEEDS_RESET,
the device MUST send a device configuration change notification to the
driver."
Commit "f5ed36635d8f virtio: stop virtqueue processing if device is broken"
introduced a virtio_error() call that just does that:
- internally mark the device as broken
- set the DEVICE_NEEDS_RESET bit in the status
- send a configuration change notification
Unfortunately, virtio_notify_vector(), called by virtio_notify_config(),
returns right away when the device is marked as broken and the notification
isn't sent in this case.
The spec doesn't say whether a broken device can send notifications
in other situations or not. But since the driver isn't supposed to do
anything but to reset the device, it makes sense to keep the check in
virtio_notify_config().
Marking the device as broken AFTER the configuration change notification was
sent is enough to fix the issue.
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit 66453cff9e)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The code that tries to reopen a BlockDriverState in stream_start()
when the creation of a new block job fails crashes because it attempts
to dereference a pointer that is known to be NULL.
This is a regression introduced in a170a91fd3,
likely because the code was copied from stream_complete().
Cc: qemu-stable@nongnu.org
Reported-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Alberto Garcia <berto@igalia.com>
Tested-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 525989a50a)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The curl driver has a ugly hack where, if it cannot find an empty CURLState,
it just uses aio_poll to wait for one to be empty. This is probably
buggy when used together with dataplane, and the simplest way to fix it
is to use coroutines instead.
A more immediate effect of the bug however is that it can cause a
recursive call to curl_readv_bh_cb and recursively taking the
BDRVCURLState mutex. This causes a deadlock.
The fix is to unlock the mutex around aio_poll, but for cleanliness we
should also take the mutex around all calls to curl_init_state, even if
reaching the unlock/lock pair is impossible. The same is true for
curl_clean_state.
Reported-by: Kun Wei <kuwei@redhat.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20170515100059.15795-4-pbonzini@redhat.com
Cc: qemu-stable@nongnu.org
Cc: Jeff Cody <jcody@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jeff Cody <jcody@redhat.com>
(cherry picked from commit 456af34629)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Return value of read/write simcalls is not calculated correctly in case
of operations crossing page boundary and in case of short reads/writes.
Read and write simcalls should return the size of data actually
read/written or -1 in case of error.
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
(cherry picked from commit 347ec03093)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Read and write simcalls map physical memory to access I/O buffers, but
'read' simcall need to map it for writing and 'write' simcall need to
map it for reading, i.e. the opposite of what they do now. Fix that.
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
(cherry picked from commit 30c2afd151)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
On error path (like i/o error in one of the coroutines), it's required to
- wait for coroutines completion before cleaning the common structures
- reenter dependent coroutines so they ever finish
Introduced in 2d9187bc65.
Cc: qemu-stable@nongnu.org
Signed-off-by: Anton Nefedov <anton.nefedov@virtuozzo.com>
Reviewed-by: Peter Lieven <pl@kamp.de>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit b91127edd0)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The main loop uses aio_disable_external()/aio_enable_external() to
temporarily disable processing of external AioContext clients like
device emulation.
This allows monitor commands to quiesce I/O and prevent the guest from
submitting new requests while a monitor command is in progress.
The aio_enable_external() API is currently broken when an IOThread is in
aio_poll() waiting for fd activity when the main loop re-enables
external clients. Incrementing ctx->external_disable_cnt does not wake
the IOThread from ppoll(2) so fd processing remains suspended and leads
to unresponsive emulated devices.
This patch adds an aio_notify() call to aio_enable_external() so the
IOThread is kicked out of ppoll(2) and will re-arm the file descriptors.
The bug can be reproduced as follows:
$ qemu -M accel=kvm -m 1024 \
-object iothread,id=iothread0 \
-device virtio-scsi-pci,iothread=iothread0,id=virtio-scsi-pci0 \
-drive if=none,id=drive0,aio=native,cache=none,format=raw,file=test.img \
-device scsi-hd,id=scsi-hd0,drive=drive0 \
-qmp tcp::5555,server,nowait
$ scripts/qmp/qmp-shell localhost:5555
(qemu) blockdev-snapshot-sync device=drive0 snapshot-file=sn1.qcow2
mode=absolute-paths format=qcow2
After blockdev-snapshot-sync completes the SCSI disk will be
unresponsive. This leads to request timeouts inside the guest.
Reported-by: Qianqian Zhu <qizhu@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 20170508180705.20609-1-stefanha@redhat.com
Suggested-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit 321d1dba8b)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Qemu2.7~2.9 and vhost user for dpdk 17.02 release work together
to cause failures of new connection when negotiating to set MQ.
(one queue pair works well).
Because there exist some bugs in qemu code when introducing
VHOST_USER_PROTOCOL_F_REPLY_ACK to qemu. When vhost_user_set_mem_table
is invoked to deal with the vhost message VHOST_USER_SET_MEM_TABLE
for the second time, qemu indeed doesn't send the messge (The message
needs to be sent only once)but still will be waiting for dpdk's reply
ack, then, qemu is always freezing, while DPDK is always waiting for
next vhost message from qemu.
The patch aims to fix the bug, MQ can work well.
The same bug is found in function vhost_user_net_set_mtu, it is fixed
at the same time.
DPDK related patch is as following:
http://www.dpdk.org/dev/patchwork/patch/23955/
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
Cc: qemu-stable@nongnu.org
Fixes: ca525ce561 ("vhost-user: Introduce a new protocol feature REPLY_ACK.")
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Jens Freimann <jfreiman@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
(cherry picked from commit 60cd11024f)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Opening the backing image for the second time is bad, especially here
when it is also in use as the active image as the source. The
drive-backup job itself doesn't read from target->backing for COW,
instead it gets data from the write notifier, so it's not a big problem.
However, exporting the target to NBD etc. won't work, because of the
likely stale metadata cache.
Use BDRV_O_NO_BACKING in this case and manually set up the backing
BdrvChild.
Cc: qemu-stable@nongnu.org
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit fc0932fdcf)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
We now have macros in place to make it less verbose to add a scalar
to QDict and QList, so use them.
Patch created mechanically via:
spatch --sp-file scripts/coccinelle/qobject.cocci \
--macro-file scripts/cocci-macro-file.h --dir . --in-place
then touched up manually to fix a couple of '?:' back to original
spacing, as well as avoiding a long line in monitor.c.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-7-eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
(cherry picked from commit 46f5ac205a)
* prereq for fc0932f
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Rather than making lots of callers wrap a scalar in a QInt, QString,
or QBool, provide helper macros that do the wrapping automatically.
Update the Coccinelle script to make mass conversions easy, although
the conversion itself will be done as a separate patches to ease
review and backport efforts.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-6-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
(cherry picked from commit a92c21591b)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
We have macros in place to make it less verbose to add a subtype
of QObject to both QDict and QList. While we have made cleanups
like this in the past (see commit fcfcd8ffc, for example), having
it be automated by Coccinelle makes it easier to maintain.
Patch created mechanically via:
spatch --sp-file scripts/coccinelle/qobject.cocci \
--macro-file scripts/cocci-macro-file.h --dir . --in-place
then I verified that no manual touchups were required.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-5-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
(cherry picked from commit de6e7951fe)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
We have macros in place to make it less verbose to add a subtype
of QObject to both QDict and QList. While we have made cleanups
like this in the past (see commit fcfcd8ffc, for example), having
it be automated by Coccinelle makes it easier to maintain.
The script is separate from the cleanups, for ease of review and
backporting. A later patch will then add further possible cleanups.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20170427215821.19397-4-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
(cherry picked from commit a2f3453ebc)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
When trying to remove a file from a directory, both created in non-mapped
mode, the file remains and EBADF is returned to the guest.
This is a regression introduced by commit "df4938a6651b 9pfs: local:
unlinkat: don't follow symlinks" when fixing CVE-2016-9602. It changed the
way we unlink the metadata file from
ret = remove("$dir/.virtfs_metadata/$name");
if (ret < 0 && errno != ENOENT) {
/* Error out */
}
/* Ignore absence of metadata */
to
fd = openat("$dir/.virtfs_metadata")
unlinkat(fd, "$name")
if (ret < 0 && errno != ENOENT) {
/* Error out */
}
/* Ignore absence of metadata */
If $dir was created in non-mapped mode, openat() fails with ENOENT and
we pass -1 to unlinkat(), which fails in turn with EBADF.
We just need to check the return of openat() and ignore ENOENT, in order
to restore the behaviour we had with remove().
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
[groug: rewrote the comments as suggested by Eric]
(cherry picked from commit 6a87e7929f)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Commit f0c9d64a exposed the issue that with a xenfv machine using
pci passthrough, acpi pci hotplug code was being executed by mistake.
Guard calls to acpi_pcihp_device_plug_cb (and corresponding
acpi_pcihp_device_unplug_cb) with a check for xen_enabled(). Without
this check I am seeing an error that the bus doesn't have the
acpi-pcihp-bsel property set.
Signed-off-by: Bruce Rogers <brogers@suse.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 153eba4726)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The block layer takes care of removing the bs->file child if the block
driver's bdrv_open()/bdrv_file_open() implementation fails. The block
driver therefore does not need to do so, and indeed should not unless it
sets bs->file to NULL afterwards -- because if this is not done, the
bdrv_unref_child() in bdrv_open_inherit() will dereference the freed
memory block at bs->file afterwards, which is not good.
We can now decide whether to add a "bs->file = NULL;" after each of the
offending bdrv_unref_child() invocations, or just drop them altogether.
The latter is simpler, so let's do that.
Cc: qemu-stable <qemu-stable@nongnu.org>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit de234897b6)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
If a pci device is not reset by VM (by writing into config space)
and unplugged by VM, after that when VM reboots, qemu may assert:
pcibus_reset: Assertion `bus->irq_count[i] == 0' failed
Cc: qemu-stable@nongnu.org
Signed-off-by: herongguang <herongguang.he@huawei.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 3936161f1f)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Historically the migration data channel has only needed to be
unidirectional. Thus the 'exec:' protocol was requesting an
I/O channel with O_RDONLY on incoming side, and O_WRONLY on
the outgoing side.
This is fine for classic migration, but if you then try to run
TLS over it, this fails because the TLS handshake requires a
bi-directional channel.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
(cherry picked from commit 062d81f0e9)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Reproducer:
$ ./qemu-img info ''
qemu-img: ./block.c:1008: bdrv_open_driver: Assertion
`!drv->bdrv_needs_filename || bs->filename[0]' failed.
[1] 26105 abort (core dumped) ./qemu-img info ''
This patch fixes this to be:
$ ./qemu-img info ''
qemu-img: Could not open '': The 'file' block driver requires a file
name
Cc: qemu-stable <qemu-stable@nongnu.org>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 4a0082401a)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
It does not make much sense to use a backing image for the target when
you concatenate multiple images (because then there is no correspondence
between the source images' backing files and the target's); but it was
still possible to give one by using -o backing_file=X instead of -B X.
Fix this by moving the check.
(Also, change the error message because -B is not the only way to
specify the backing file, evidently.)
Cc: qemu-stable <qemu-stable@nongnu.org>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
* applied patch from v1 of series as suggested by author
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
After storing the creation options for the new image into @opts, we
fetch some things for our own information, like the backing file name,
or whether to use encryption or preallocation.
With the -n parameter, there will not be any creation options; this is
not too bad because this just means that querying a NULL @opts will
always return the default value.
However, we also use @opts for the --object options. Therefore, @opts is
not necessarily NULL if -n was specified; instead, it may contain those
options. In practice, this probably does not cause any problems because
there most likely is no object that supports any of the parameters we
query here, but this is neither something we should rely on nor does
this variable reuse make the code very nice to read.
Therefore, just use an own variable for the --object options.
Cc: qemu-stable <qemu-stable@nongnu.org>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
* applied patch from v1 of series as suggested by author
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Otherwise the qemu-img process will exit with EXIT_SUCCESS instead of
EXIT_FAILURE.
Cc: qemu-stable <qemu-stable@nongnu.org>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
* applied directly to stable, upstream code has issue fixed via a
refactoring introduced by 9fd77f9, which isn't targetted for stable
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
The QGA schema states:
@can-offline: Whether offlining the VCPU is possible. This member
is always filled in by the guest agent when the structure
is returned, and always ignored on input (hence it can be
omitted then).
Currently 'can-offline' is missing entirely from the reply. This causes
errors in libvirt which is expecting the reply to be compliant with the
schema docs.
BZ#1438735: https://bugzilla.redhat.com/show_bug.cgi?id=1438735
Signed-off-by: Sameeh Jubran <sameeh@daynix.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
(cherry picked from commit 54858553de)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
2017-07-31 15:25:56 -05:00
11081 changed files with 710577 additions and 2280270 deletions
/* translator may have removed instructions, update final count */
g_assert(num_insns<=ptb->n);
ptb->n=num_insns;
/* collect instrumentation requests */
qemu_plugin_tb_trans_cb(cpu,ptb);
/* inject the instrumentation at the appropriate places */
plugin_gen_inject(ptb);
}
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
