GNUTLS is supported as a crypto provider since
commit cc4c7c7382
Author: Daniel P. Berrangé <berrange@redhat.com>
Date: Wed Jun 30 17:20:02 2021 +0100
crypto: introduce build system for gnutls crypto backend
So enable the LUKS tests in this config.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
If given a malformed LUKS header, it is possible that the algorithm
names end up being an empty string. This leads to confusing error
messages unless quoting is used to highlight where the empty string
is subsituted in the error message.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The unit test suite is shortly going to want to convert header
endianness separately from the main I/O functions.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
This will allow unit testing code to use the structs.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Both the master key and key slot passphrases are run through the PBKDF2
algorithm. The iterations count is expected to be generally very large
(many 10's or 100's of 1000s). It is hard to define a low level cutoff,
but we can certainly say that iterations count should be non-zero. A
zero count likely indicates an initialization mistake so reject it.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The LUKS header data on disk is a fixed size, however, there's expected
to be a gap between the end of the header and the first key slot to get
alignment with the 2nd sector on 4k drives. This wasn't originally part
of the LUKS spec, but was always part of the reference implementation,
so it is worth validating this.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
We already validate that LUKS keyslots don't overlap with the
header, or with each other. This closes the remaining hole in
validation of LUKS file regions.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
We already check that key material doesn't overlap between key slots,
and that it doesn't overlap with the payload. We didn't check for
overlap with the LUKS header.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Although the LUKS stripes are encoded in the keyslot header and so
potentially configurable, in pratice the cryptsetup impl mandates
this has the fixed value 4000. To avoid incompatibility apply the
same enforcement in QEMU too. This also caps the memory usage for
key material when QEMU tries to open a LUKS volume.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The LUKS spec requires that header strings are NUL-terminated, and our
code relies on that. Protect against maliciously crafted headers by
adding validation.
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Using FILE * APIs for writing the PSK file results in translation from
UNIX to DOS line endings on Windows. When the crypto PSK code later
loads the credentials the stray \r will result in failure to load the
PSK credentials into GNUTLS.
Rather than switching the FILE* APIs to open in binary format, just
switch to the more concise g_file_set_contents API.
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
If setting credentials fails, the handshake will later fail to complete
with an obscure error message which is hard to diagnose.
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Currently we check status of each submodule, before actually checking
if we're in a git repo. These status commands will all fail, but we
are hiding their output so we don't see it currently.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Upon failure, a libseccomp API returns actual errno value very
rarely. Fortunately, after its commit 34bf78ab (contained in
2.5.0 release), the SCMP_FLTATR_API_SYSRAWRC attribute can be set
which makes subsequent APIs return true errno on failure.
This is especially critical when seccomp_load() fails, because
generic -ECANCELED says nothing.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Random failure was observed when running qtests on Windows due to
"Broken pipe" detected by qmp_fd_receive(). What happened is that
the qtest executable sends testing data over a socket to the QEMU
under test but no response is received. The errno of the recv()
call from the qtest executable indicates ETIMEOUT, due to the qmp
chardev's tcp_chr_read() is never called to receive testing data
hence no response is sent to the other side.
tcp_chr_read() is registered as the callback of the socket watch
GSource. The reason of the callback not being called by glib, is
that the source check fails to indicate the source is ready. There
are two socket watch sources created to monitor the same socket
event object from the char-socket backend in update_ioc_handlers().
During the source check phase, qio_channel_socket_source_check()
calls WSAEnumNetworkEvents() to discover occurrences of network
events for the indicated socket, clear internal network event records,
and reset the event object. Testing shows that if we don't reset the
event object by not passing the event handle to WSAEnumNetworkEvents()
the symptom goes away and qtest runs very stably.
It seems we don't need to call WSAEnumNetworkEvents() at all, as we
don't parse the result of WSANETWORKEVENTS returned from this API.
We use select() to poll the socket status. Fix this instability by
dropping the WSAEnumNetworkEvents() call.
Some side notes:
During the testing, I removed the following codes in update_ioc_handlers():
remove_hup_source(s);
s->hup_source = qio_channel_create_watch(s->ioc, G_IO_HUP);
g_source_set_callback(s->hup_source, (GSourceFunc)tcp_chr_hup,
chr, NULL);
g_source_attach(s->hup_source, chr->gcontext);
and such change also makes the symptom go away.
And if I moved the above codes to the beginning, before the call to
io_add_watch_poll(), the symptom also goes away.
It seems two sources watching on the same socket event object is
the key that leads to the instability. The order of adding a source
watch seems to also play a role but I can't explain why.
Hopefully a Windows and glib expert could explain this behavior.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
There is no need to do a type cast on ssource->socket as it is
already declared as a SOCKET.
Suggested-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
In the win32 version qio_channel_create_socket_watch() body there is
no need to do a '#ifdef WIN32'.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Replace the existing logic to get the directory for temporary files
with g_get_tmp_dir(), which works for win32 too.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
When the user creates a LUKS-encrypted qcow2 image using the qemu-img
program, the passphrase is hashed using PBKDF2 with a dynamic
number of iterations. The number of iterations is determined by
measuring thread cpu time usage, such that it takes approximately
2 seconds to compute the hash.
Because Darwin doesn't implement getrusage(RUSAGE_THREAD), we get an
error message:
> qemu-img: test.qcow2: Unable to calculate thread CPU usage on this platform
for this command:
> qemu-img create --object secret,id=key,data=1234 -f qcow2 -o 'encrypt.format=luks,encrypt.key-secret=key' test.qcow2 100M
This patch implements qcrypto_pbkdf2_get_thread_cpu() for Darwin so that
the above command works.
Signed-off-by: Jungmin Park <pjm0616@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
* Highlight of this PR is Linus Heckemann's GHashTable patch which
brings massive general performance improvements of 9p server
somewhere between factor 6 .. 12.
* Bin Meng's g_mkdir patch is a preparatory patch for upcoming
Windows host support of 9p server.
* The rest of the patches in this PR are 9p test code restructuring
and refactoring changes to improve readability and to ease
maintenance of 9p test code on the long-term.
-----BEGIN PGP SIGNATURE-----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=MxSG
-----END PGP SIGNATURE-----
Merge tag 'pull-9p-20221024' of https://github.com/cschoenebeck/qemu into staging
9pfs: performance, Windows host prep, tests restructure
* Highlight of this PR is Linus Heckemann's GHashTable patch which
brings massive general performance improvements of 9p server
somewhere between factor 6 .. 12.
* Bin Meng's g_mkdir patch is a preparatory patch for upcoming
Windows host support of 9p server.
* The rest of the patches in this PR are 9p test code restructuring
and refactoring changes to improve readability and to ease
maintenance of 9p test code on the long-term.
# -----BEGIN PGP SIGNATURE-----
#
# iQJLBAABCgA1FiEEltjREM96+AhPiFkBNMK1h2Wkc5UFAmNWbs8XHHFlbXVfb3Nz
# QGNydWRlYnl0ZS5jb20ACgkQNMK1h2Wkc5V4cw/8CqoSJqoJixlP8kAGDYWq3CgF
# SKd09rIzLSWyyufAoZr1TqLwRrvEQRlZJSpL4fGvRpQLv0IQCu4x59ohHRob25Tm
# Fe7IxYBNuBwLW4yu+Y7FaujeGoYAi9Qw5q4ijq3/aSSiIeuXySKB2JmW71CQ+Tbe
# uwivsnMtWzQ7qsNwrtXYbxDs7UGkdsiW2sEQUS26GMApAXZoB+38hwtTW2Y9MOrC
# 58JuZza/fUVPzo0V1D0ggRawb5O2VTF5fz8aGFG4FvoyIW6DDZFSfnyre9QxivOl
# 5McWwSQ/D04vdEK9ornGPYr9YRGuP8g07p1EW9OfKeie4I41e9pS3UminK5lVCgo
# SfBHzz96efM5XR+Wnl4yVKowivmTqjwUU8lDqW2eB/7YBRuYUzrpxYe//UPv4q1J
# zaQV3pgwFAVkVJCnkcLCa1JQbH581bXSsuRlDdYqoRYfyzXoxbywNjvn9BXE0PrG
# WRecS//GyN3GVZYxMwb3H052110pYsYIg2YZ2H4QiqCwpEHHvy+L/ZXm19vbDm7B
# GYJQPUK8/y0NGwZsUYcUSx1TWlU9ZPwrbqZfv7e7+B6FL4VNjdaqb8PvS9admWSq
# LOSzrVVIus+nb7tP99d1Fb6oRyCy3x8E48gTr5UtTJHC4SAw/OBJmem6GOc/D490
# H7Dq8Y27qsQ6fT7iPm8=
# =MxSG
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 24 Oct 2022 06:54:07 EDT
# gpg: using RSA key 96D8D110CF7AF8084F88590134C2B58765A47395
# gpg: issuer "qemu_oss@crudebyte.com"
# gpg: Good signature from "Christian Schoenebeck <qemu_oss@crudebyte.com>" [unknown]
# gpg: WARNING: The key's User ID is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: ECAB 1A45 4014 1413 BA38 4926 30DB 47C3 A012 D5F4
# Subkey fingerprint: 96D8 D110 CF7A F808 4F88 5901 34C2 B587 65A4 7395
* tag 'pull-9p-20221024' of https://github.com/cschoenebeck/qemu: (23 commits)
tests/9p: remove unnecessary g_strdup() calls
tests/9p: merge v9fs_tunlinkat() and do_unlinkat()
tests/9p: merge v9fs_tlink() and do_hardlink()
tests/9p: merge v9fs_tsymlink() and do_symlink()
tests/9p: merge v9fs_tlcreate() and do_lcreate()
tests/9p: merge v9fs_tmkdir() and do_mkdir()
tests/9p: convert v9fs_tflush() to declarative arguments
tests/9p: simplify callers of twrite()
tests/9p: convert v9fs_twrite() to declarative arguments
tests/9p: simplify callers of tlopen()
tests/9p: convert v9fs_tlopen() to declarative arguments
tests/9p: simplify callers of treaddir()
tests/9p: convert v9fs_treaddir() to declarative arguments
tests/9p: simplify callers of tgetattr()
tests/9p: convert v9fs_tgetattr() to declarative arguments
tests/9p: simplify callers of tattach()
tests/9p: merge v9fs_tattach(), do_attach(), do_attach_rqid()
tests/9p: merge v9fs_tversion() and do_version()
tests/9p: simplify callers of twalk()
tests/9p: merge *walk*() functions
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
This is a leftover from before the recent function merge and
refactoring patches:
As these functions do not return control to the caller in
between, it is not necessary to duplicate strings passed to them.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <0f80141cde3904ed0591354059da49d1d60bcdbc.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify those 2 functions into a single function
v9fs_tunlinkat() by using a declarative function arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <1dea593edd464908d92501933c068388c01f1744.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify those 2 functions into a single function
v9fs_tlink() by using a declarative function arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <cb4d42203e1e4e6027df4924bbe4bdbc002f668b.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify those 2 functions into a single function
v9fs_tsymlink() by using a declarative function arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <563f3ad04fe596ce0ae1e2654d1d08237f18c830.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify those 2 functions into a single function
v9fs_tlcreate() by using a declarative function arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <4c01b2caa5f5b54a2020fc92701deadd2abf0571.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify those 2 functions into a single function
v9fs_tmkdir() by using a declarative function arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <b87b2c972921df980440ff5b2d3e6bb8163d6551.1664917004.git.qemu_oss@crudebyte.com>
Use declarative function arguments for function v9fs_tflush().
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <91b7b154298c500d100b05137146c2905c3acdec.1664917004.git.qemu_oss@crudebyte.com>
Now as twrite() is using a declarative approach, simplify the
code of callers of this function.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <7f280ec6a1f9d8afed46567a796562c4dc28afa9.1664917004.git.qemu_oss@crudebyte.com>
Use declarative function arguments for function v9fs_twrite().
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <be0326e2d9ab66f68c06b1766ddf103849d570b4.1664917004.git.qemu_oss@crudebyte.com>
Now as tlopen() is using a declarative approach, simplify the
code of callers of this function.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <f74b6153e079fc7a340e5cb575ee32e0fe1e0ae6.1664917004.git.qemu_oss@crudebyte.com>
Use declarative function arguments for function v9fs_tlopen().
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <765ab515353c56f88f0a163631f626a44e9565d6.1664917004.git.qemu_oss@crudebyte.com>
Now as treaddir() is using a declarative approach, simplify the
code of callers of this function.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <7cec6f2c7011a481806c34908893b7282702a7a6.1664917004.git.qemu_oss@crudebyte.com>
Use declarative function arguments for function v9fs_treaddir().
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <a66aae4ceb19ec12d245b8c7f33a639584c8e272.1664917004.git.qemu_oss@crudebyte.com>
Now as tgetattr() is using a declarative approach, simplify the
code of callers of this function.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <60c6a083f320b86f3172951445df7bbc895932e2.1664917004.git.qemu_oss@crudebyte.com>
Use declarative function arguments for function v9fs_tgetattr().
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <d340a91be96fbfecfb8dacdd7558223b3c0d0e2c.1664917004.git.qemu_oss@crudebyte.com>
Now as tattach() is using a declarative approach, simplify the
code of callers of this function.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <9b50e5b89a0072e84a9191d18c19a53546a28bba.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify those 3 functions into a single function
v9fs_tattach() by using a declarative function arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <a6756b30bf2a1b25729c5bbabd1c9534a8f20d6f.1664917004.git.qemu_oss@crudebyte.com>
As with previous patches, unify functions v9fs_tversion() and do_version()
into a single function v9fs_tversion() by using a declarative function
arguments approach.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <2d253491aaffd267ec295f056dda47456692cd0c.1664917004.git.qemu_oss@crudebyte.com>
Now as twalk() is using a declarative approach, simplify the
code of callers of this function.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <8b9d3c656ad43b6c953d6bdacd8d9f4c8e599b2a.1664917004.git.qemu_oss@crudebyte.com>
Introduce declarative function calls.
There are currently 4 different functions for sending a 9p 'Twalk'
request: v9fs_twalk(), do_walk(), do_walk_rqids() and
do_walk_expect_error(). They are all doing the same thing, just in a
slightly different way and with slightly different function arguments.
Merge those 4 functions into a single function by using a struct for
function call arguments and use designated initializers when calling
this function to turn usage into a declarative approach, which is
better readable and easier to maintain.
Also move private functions genfid(), split() and split_free() from
virtio-9p-test.c to virtio-9p-client.c.
Based-on: <E1odrya-0004Fv-97@lizzy.crudebyte.com>
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <607969dbfbc63c1be008df9131133711b046e979.1664917004.git.qemu_oss@crudebyte.com>
The previous implementation would iterate over the fid table for
lookup operations, resulting in an operation with O(n) complexity on
the number of open files and poor cache locality -- for every open,
stat, read, write, etc operation.
This change uses a hashtable for this instead, significantly improving
the performance of the 9p filesystem. The runtime of NixOS's simple
installer test, which copies ~122k files totalling ~1.8GiB from 9p,
decreased by a factor of about 10.
Signed-off-by: Linus Heckemann <git@sphalerite.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Greg Kurz <groug@kaod.org>
[CS: - Retain BUG_ON(f->clunked) in get_fid().
- Add TODO comment in clunk_fid(). ]
Message-Id: <20221004104121.713689-1-git@sphalerite.org>
[CS: - Drop unnecessary goto and out: label. ]
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
This patch is pure refactoring, it does not change behaviour.
virtio-9p-test.c grew to 1657 lines. Let's split this file up between
actual 9p test cases vs. 9p test client, to make it easier to
concentrate on the actual 9p tests.
Move the 9p test client code to a new unit virtio-9p-client.c, which
are basically all functions and types prefixed with v9fs_* already.
Note that some client wrapper functions (do_*) are preserved in
virtio-9p-test.c, simply because these wrapper functions are going to
be wiped with subsequent patches anyway.
As the global QGuestAllocator variable is moved to virtio-9p-client.c,
add a new function v9fs_set_allocator() to be used by virtio-9p-test.c
instead of fiddling with a global variable across units and libraries.
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Message-Id: <E1odrya-0004Fv-97@lizzy.crudebyte.com>
Use g_mkdir() to create a directory on all platforms.
Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <20220927110632.1973965-27-bmeng.cn@gmail.com>
Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Rather than poking directly into RAM, add the bootinfo block as a proper
ROM, so that it's restored when rebooting the system. This way, if the
guest corrupts any of the bootinfo items, but then tries to reboot,
it'll still be restored back to normal as expected.
Then, since the RNG seed needs to be fresh on each boot, regenerate the
RNG seed in the ROM when reseting the CPU.
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Message-Id: <20221023191340.36238-1-Jason@zx2c4.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
The only issue with FMA instructions is that there are _a lot_ of them (30
opcodes, each of which comes in up to 4 versions depending on VEX.W and
VEX.L; a total of 96 possibilities). However, they can be implement with
only 6 helpers, two for scalar operations and four for packed operations.
(Scalar versions do not do any merging; they only affect the bottom 32
or 64 bits of the output operand. Therefore, there is no separate XMM
and YMM of the scalar helpers).
First, we can reduce the number of helpers to one third by passing four
operands (one output and three inputs); the reordering of which operands
go to the multiply and which go to the add is done in emit.c.
Second, the different instructions also dispatch to the same softfloat
function, so the flags for float32_muladd and float64_muladd are passed
in the helper as int arguments, with a little extra complication to
handle FMADDSUB and FMSUBADD.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Following a change on the kernel side (see link), pass BI_RNG_SEED
instead of BI_VIRT_RNG_SEED. This should have no impact on
compatibility, as there will simply be no effect if it's an old kernel,
which is how things have always been. We then use this as an opportunity
to add this to q800, since now we can, which is a nice improvement.
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Laurent Vivier <laurent@vivier.eu>
Link: https://lore.kernel.org/lkml/20220923170340.4099226-3-Jason@zx2c4.com/
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Message-Id: <20220926113900.1256630-1-Jason@zx2c4.com>
[lv: s/^I/ /g]
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
F16C only consists of two instructions, which are a bit peculiar
nevertheless.
First, they access only the low half of an YMM or XMM register for the
packed-half operand; the exact size still depends on the VEX.L flag.
This is similar to the existing avx_movx flag, but not exactly because
avx_movx is hardcoded to affect operand 2. To this end I added a "ph"
format name; it's possible to reuse this approach for the VPMOVSX and
VPMOVZX instructions, though that would also require adding two more
formats for the low-quarter and low-eighth of an operand.
Second, VCVTPS2PH is somewhat weird because it *stores* the result of
the instruction into memory rather than loading it.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Daniel P. Berrangé