Before moving the existing DFP instructions to decodetree, drop the
nip update that shouldn't be done for these instructions.
Signed-off-by: Luis Pires <luis.pires@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20211029192417.400707-9-luis.pires@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Implement the following PowerISA v3.1 instruction:
dctfixqq: DFP Convert To Fixed Quadword Quad
Signed-off-by: Luis Pires <luis.pires@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20211029192417.400707-8-luis.pires@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Implement the following PowerISA v3.1 instruction:
dcffixqq: DFP Convert From Fixed Quadword Quad
Signed-off-by: Luis Pires <luis.pires@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20211029192417.400707-5-luis.pires@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Move REQUIRE_ALTIVEC to translate.c and rename it to REQUIRE_VECTOR.
Signed-off-by: Bruno Larsen <bruno.larsen@eldorado.org.br>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Signed-off-by: Fernando Valle <fernando.valle@eldorado.org.br>
Signed-off-by: Luis Pires <luis.pires@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <20211029192417.400707-3-luis.pires@eldorado.org.br>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Implement the following PowerISA v3.1 instruction:
cnttzdm: Count Trailing Zeros Doubleword Under Bit Mask
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Luis Pires <luis.pires@eldorado.org.br>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20211029202424.175401-9-matheus.ferst@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Implement the following PowerISA v3.1 instruction:
cntlzdm: Count Leading Zeros Doubleword Under Bit Mask
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Luis Pires <luis.pires@eldorado.org.br>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20211029202424.175401-8-matheus.ferst@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Fernando Eckhardt Valle <fernando.valle@eldorado.org.br>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20211029202424.175401-5-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Move load floating point instructions (lfs, lfsu, lfsx, lfsux, lfd, lfdu, lfdx, lfdux)
and store floating point instructions(stfs, stfsu, stfsx, stfsux, stfd, stfdu, stfdx,
stfdux) from legacy system to decodetree.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Fernando Eckhardt Valle <fernando.valle@eldorado.org.br>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20211029202424.175401-4-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Move resolve_PLS_D from fixedpoint-impl.c.inc to translate.c
because this way the function can be used not only by fixed
point instructions.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Fernando Eckhardt Valle <phervalle@gmail.com>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20211029202424.175401-3-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The do_ea_calc function will calculate the effective address(EA)
according to PowerIsa 3.1. With that, it was replaced part of
do_ldst() that calculates the EA by this new function.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Fernando Eckhardt Valle (pherde) <phervalle@gmail.com>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20211029202424.175401-2-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
In commit 8f0a4b6a9b, we started to require L=0 for ppc32 to match what
The Programming Environments Manual say:
"For 32-bit implementations, the L field must be cleared, otherwise
the instruction form is invalid."
The stricter behavior, however, broke AROS boot on sam460ex, which is a
regression from 6.0. This patch partially reverts the change, raising
the exception only for CPUs known to require L=0 (e500 and e500mc) and
logging a guest error for other cases.
Both behaviors are acceptable by the PowerISA, which allows "the system
illegal instruction error handler to be invoked or yield boundedly
undefined results."
Reported-by: BALATON Zoltan <balaton@eik.bme.hu>
Fixes: 8f0a4b6a9b ("target/ppc: Move cmp/cmpi/cmpl/cmpli to decodetree")
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20210720135507.2444635-1-matheus.ferst@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Additionally, REQUIRE_64BIT when L=1 to match what is specified in The
Programming Environments Manual:
"For 32-bit implementations, the L field must be cleared, otherwise the
instruction form is invalid."
Some CPUs are known to deviate from this specification by ignoring the
L bit [1]. The stricter behavior, however, can help users that test
software with qemu, making it more likely to detect bugs that would
otherwise be silent.
If deemed necessary, a future patch can adapt this behavior based on
the specific CPU model.
[1] The 601 manual is the only one I've found that explicitly states
that the L bit is ignored, but we also observe this behavior in a 7447A
v1.2.
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20210601193528.2533031-15-matheus.ferst@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
[dwg: Corrected whitespace error]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
These are all connected by macros in the legacy decoding.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20210601193528.2533031-9-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
These are all connected by macros in the legacy decoding.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20210601193528.2533031-7-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The illegal suffix behavior matches what was observed in a
POWER10 DD2.0 machine.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20210601193528.2533031-6-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
TARGET_WORDS_BIGENDIAN may not match the machine endianness if that's a
runtime-configurable parameter.
Fixes: bcb0b7b1a1
Fixes: afae37d98a
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/212
Signed-off-by: Giuseppe Musacchio <thatlemon@gmail.com>
Message-Id: <20210518133020.58927-1-thatlemon@gmail.com>
Tested-by: Paul A. Clarke <pc@us.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Make the implementation match the lxvwsx one.
The code is now shorter smaller and potentially faster as the
translation will use the host SIMD capabilities if available.
No functional change.
Signed-off-by: Giuseppe Musacchio <thatlemon@gmail.com>
Message-Id: <a463dea379da4cb3a22de49c678932f74fb15dd7.1604912739.git.thatlemon@gmail.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Implement the "Load VSX Vector Word & Splat Indexed" opcode, introduced
in Power ISA v3.0.
Buglink: https://bugs.launchpad.net/qemu/+bug/1793608
Signed-off-by: Giuseppe Musacchio <thatlemon@gmail.com>
Message-Id: <d7d533e18c2bc10d924ee3e09907ff2b41fddb3a.1604912739.git.thatlemon@gmail.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Here's my first pull request for qemu-5.2, which has quite a few
accumulated things. Highlights are:
* Preliminary support for POWER10 (Power ISA 3.1) instruction emulation
* Add documentation on the (very confusing) pseries NUMA configuration
* Fix some bugs handling edge cases with XICS, XIVE and kernel_irqchip
* Fix icount for a number of POWER registers
* Many cleanups to error handling in XIVE code
* Validate size of -prom-env data
-----BEGIN PGP SIGNATURE-----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=L9Wi
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/dgibson/tags/ppc-for-5.2-20200818' into staging
ppc patch queue 2020-08-18
Here's my first pull request for qemu-5.2, which has quite a few
accumulated things. Highlights are:
* Preliminary support for POWER10 (Power ISA 3.1) instruction emulation
* Add documentation on the (very confusing) pseries NUMA configuration
* Fix some bugs handling edge cases with XICS, XIVE and kernel_irqchip
* Fix icount for a number of POWER registers
* Many cleanups to error handling in XIVE code
* Validate size of -prom-env data
# gpg: Signature made Tue 18 Aug 2020 05:18:36 BST
# gpg: using RSA key 75F46586AE61A66CC44E87DC6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>" [full]
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>" [full]
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>" [full]
# gpg: aka "David Gibson (kernel.org) <dwg@kernel.org>" [unknown]
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-5.2-20200818: (40 commits)
spapr/xive: Use xive_source_esb_len()
nvram: Exit QEMU if NVRAM cannot contain all -prom-env data
spapr/xive: Simplify error handling of kvmppc_xive_cpu_synchronize_state()
ppc/xive: Simplify error handling in xive_tctx_realize()
spapr/xive: Simplify error handling in kvmppc_xive_connect()
ppc/xive: Fix error handling in vmstate_xive_tctx_*() callbacks
spapr/xive: Fix error handling in kvmppc_xive_post_load()
spapr/kvm: Fix error handling in kvmppc_xive_pre_save()
spapr/xive: Rework error handling of kvmppc_xive_set_source_config()
spapr/xive: Rework error handling in kvmppc_xive_get_queues()
spapr/xive: Rework error handling of kvmppc_xive_[gs]et_queue_config()
spapr/xive: Rework error handling of kvmppc_xive_cpu_[gs]et_state()
spapr/xive: Rework error handling of kvmppc_xive_mmap()
spapr/xive: Rework error handling of kvmppc_xive_source_reset()
spapr/xive: Rework error handling of kvmppc_xive_cpu_connect()
spapr: Simplify error handling in spapr_phb_realize()
spapr/xive: Convert KVM device fd checks to assert()
ppc/xive: Introduce dedicated kvm_irqchip_in_kernel() wrappers
ppc/xive: Rework setup of XiveSource::esb_mmio
target/ppc: Integrate icount to purr, vtb, and tbu40
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
With Makefiles that have automatically generated dependencies, you
generated includes are set as dependencies of the Makefile, so that they
are built before everything else and they are available when first
building the .c files.
Alternatively you can use a fine-grained dependency, e.g.
target/arm/translate.o: target/arm/decode-neon-shared.inc.c
With Meson you have only one choice and it is a third option, namely
"build at the beginning of the corresponding target"; the way you
express it is to list the includes in the sources of that target.
The problem is that Meson decides if something is a source vs. a
generated include by looking at the extension: '.c', '.cc', '.m', '.C'
are sources, while everything else is considered an include---including
'.inc.c'.
Use '.c.inc' to avoid this, as it is consistent with our other convention
of using '.rst.inc' for included reStructuredText files. The editorconfig
file is adjusted.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When emulating certain floating point instructions or vector instructions on
PowerPC machines, QEMU did not properly generate the SPE/Embedded Floating-
Point Unavailable interrupt. See the buglink further below for references to
the relevant NXP documentation.
This patch fixes the behavior of some evfs* instructions that were
incorrectly emitting the interrupt.
More importantly, this patch fixes the behavior of several efd* and ev*
instructions that were not generating the interrupt. Triggering the
interrupt for these instructions fixes lazy FPU/vector context switching on
some operating systems like Linux.
Without this patch, the result of some double-precision arithmetic could be
corrupted due to the lack of proper saving and restoring of the upper
32-bit part of the general-purpose registers.
Buglink: https://bugs.launchpad.net/qemu/+bug/1888918
Buglink: https://bugs.launchpad.net/qemu/+bug/1611394
Signed-off-by: Matthieu Bucchianeri <matthieu.bucchianeri@leostella.com>
Message-Id: <20200727175553.32276-1-matthieu.bucchianeri@leostella.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
vmulhsd: Vector Multiply High Signed Doubleword
vmulhud: Vector Multiply High Unsigned Doubleword
Signed-off-by: Lijun Pan <ljp@linux.ibm.com>
Message-Id: <20200724045845.89976-5-ljp@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
vmulhsw: Vector Multiply High Signed Word
vmulhuw: Vector Multiply High Unsigned Word
Signed-off-by: Lijun Pan <ljp@linux.ibm.com>
Message-Id: <20200724045845.89976-4-ljp@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Convert the original implementation of vmuluwm to the more generic
tcg_gen_gvec_mul.
Signed-off-by: Lijun Pan <ljp@linux.ibm.com>
Message-Id: <20200701234344.91843-5-ljp@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Fix double-call to tcg_temp_new_i64(), where a temp is allocated both at
declaration time and further down the implementation of gen_evmwsmiaa().
Note that gen_evmwsmia() and gen_evmwsmiaa() are still not implemented
correctly, as they invoke gen_evmwsmi() which may return early, but the
return is not propagated. This will be fixed in my patch for bug #1888918.
Signed-off-by: Matthieu Bucchianeri <matthieu.bucchianeri@leostella.com>
Message-Id: <20200727172114.31415-1-matthieu.bucchianeri@leostella.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
We can now unify the implementation of the 3 VSPLTI instructions.
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
"Deferred" was misspelled as "differed" in some comments, correct this
typo,
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-Id: <20200214155748.0896B745953@zero.eik.bme.hu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
In previous implementation, invocation of TCG shift function could request
shift of TCG variable by 64 bits when variable 'sh' is 0, which is not
supported in TCG (values can be shifted by 0 to 63 bits). This patch fixes
this by using two separate invocation of TCG shift functions, with maximum
shift amount of 32.
Name of variable 'shifted' is changed to 'carry' so variable naming
is similar to old helper implementation.
Variables 'avrA' and 'avrB' are replaced with variable 'avr'.
Fixes: 4e6d0920e7
Reported-by: "Paul A. Clark" <pc@us.ibm.com>
Reported-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Suggested-by: Aleksandar Markovic <aleksandar.markovic@rt-rk.com>
Signed-off-by: Stefan Brankovic <stefan.brankovic@rt-rk.com>
Message-Id: <1570196639-7025-2-git-send-email-stefan.brankovic@rt-rk.com>
Tested-by: Paul A. Clarke <pc@us.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
ISA 3.0B added a set of Floating-Point Status and Control Register (FPSCR)
instructions: mffsce, mffscdrn, mffscdrni, mffscrn, mffscrni, mffsl.
This patch adds support for 'mffsce' instruction.
'mffsce' is identical to 'mffs', except that it also clears the exception
enable bits in the FPSCR.
On CPUs without support for 'mffsce' (below ISA 3.0), the
instruction will execute identically to 'mffs'.
Signed-off-by: Paul A. Clarke <pc@us.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <1568817082-1384-1-git-send-email-pc@us.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
ISA 3.0B added a set of Floating-Point Status and Control Register (FPSCR)
instructions: mffsce, mffscdrn, mffscdrni, mffscrn, mffscrni, mffsl.
This patch adds support for 'mffscrn' and 'mffscrni' instructions.
'mffscrn' and 'mffscrni' are similar to 'mffsl', except they do not return
the status bits (FI, FR, FPRF) and they also set the rounding mode in the
FPSCR.
On CPUs without support for 'mffscrn'/'mffscrni' (below ISA 3.0), the
instructions will execute identically to 'mffs'.
Signed-off-by: Paul A. Clarke <pc@us.ibm.com>
Message-Id: <1568817081-1345-1-git-send-email-pc@us.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Since I found this two instructions implemented with tcg, I refactored
them so they are consistent with other similar implementations that
I introduced in this patch.
Also, a new dual macro GEN_VXFORM_TRANS_DUAL is added. This macro is
used if one instruction is realized with direct translation, and second
one with a helper.
Signed-off-by: Stefan Brankovic <stefan.brankovic@rt-rk.com>
Message-Id: <1566898663-25858-4-git-send-email-stefan.brankovic@rt-rk.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
A class of instructions of the form:
op Target,A,B
which operate like:
Target = Target * A + B
have a bit set which distinguishes them from instructions that operate as:
Target = Target * B + A
This bit is not being checked properly (using PPC_BIT macro), so all
instructions in this class are operating incorrectly as the second form
above. The bit was being checked as if it were part of a 64-bit
instruction opcode, rather than a proper 32-bit opcode. Fix by using the
macro (PPC_BIT32) which treats the opcode as a 32-bit quantity.
Fixes: c9f4e4d8b6 ("target/ppc: improve VSX_FMADD with new GEN_VSX_HELPER_VSX_MADD macro")
Signed-off-by: Paul A. Clarke <pc@us.ibm.com>
Message-Id: <1566401321-22419-1-git-send-email-pc@us.ibm.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Tested-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
ISA 3.0B added a set of Floating-Point Status and Control Register (FPSCR)
instructions: mffsce, mffscdrn, mffscdrni, mffscrn, mffscrni, mffsl.
This patch adds support for 'mffsl'.
'mffsl' is identical to 'mffs', except it only returns mode, status, and enable
bits from the FPSCR.
On CPUs without support for 'mffsl' (below ISA 3.0), the 'mffsl' instruction
will execute identically to 'mffs'.
Note: I renamed FPSCR_RN to FPSCR_RN0 so I could create an FPSCR_RN mask which
is both bits of the FPSCR rounding mode, as defined in the ISA.
I also fixed a typo in the definition of FPSCR_FR.
Signed-off-by: Paul A. Clarke <pc@us.ibm.com>
v4:
- nit: added some braces to resolve a checkpatch complaint.
v3:
- Changed tcg_gen_and_i64 to tcg_gen_andi_i64, eliminating the need for a
temporary, per review from Richard Henderson.
v2:
- I found that I copied too much of the 'mffs' implementation.
The 'Rc' condition code bits are not needed for 'mffsl'. Removed.
- I now free the (renamed) 'tmask' temporary.
- I now bail early for older ISA to the original 'mffs' implementation.
Message-Id: <1565982203-11048-1-git-send-email-pc@us.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Optimize Altivec instruction vclzw (Vector Count Leading Zeros Word).
This instruction counts the number of leading zeros of each word element
in source register and places result in the appropriate word element of
destination register.
Counting is to be performed in four iterations of for loop(one for each
word elemnt of source register vB). Every iteration consists of loading
appropriate word element from source register, counting leading zeros
with tcg_gen_clzi_i32, and saving the result in appropriate word element
of destination register.
Signed-off-by: Stefan Brankovic <stefan.brankovic@rt-rk.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <1563200574-11098-7-git-send-email-stefan.brankovic@rt-rk.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Optimize Altivec instruction vclzd (Vector Count Leading Zeros Doubleword).
This instruction counts the number of leading zeros of each doubleword element
in source register and places result in the appropriate doubleword element of
destination register.
Using tcg-s count leading zeros instruction two times(once for each
doubleword element of source register vB) and placing result in
appropriate doubleword element of destination register vD.
Signed-off-by: Stefan Brankovic <stefan.brankovic@rt-rk.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <1563200574-11098-6-git-send-email-stefan.brankovic@rt-rk.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Optimize altivec instruction vgbbd (Vector Gather Bits by Bytes by Doubleword)
All ith bits (i in range 1 to 8) of each byte of doubleword element in
source register are concatenated and placed into ith byte of appropriate
doubleword element in destination register.
Following solution is done for both doubleword elements of source register
in parallel, in order to reduce the number of instructions needed(that's why
arrays are used):
First, both doubleword elements of source register vB are placed in
appropriate element of array avr. Bits are gathered in 2x8 iterations(2 for
loops). In first iteration bit 1 of byte 1, bit 2 of byte 2,... bit 8 of
byte 8 are in their final spots so avr[i], i={0,1} can be and-ed with
tcg_mask. For every following iteration, both avr[i] and tcg_mask variables
have to be shifted right for 7 and 8 places, respectively, in order to get
bit 1 of byte 2, bit 2 of byte 3.. bit 7 of byte 8 in their final spots so
shifted avr values(saved in tmp) can be and-ed with new value of tcg_mask...
After first 8 iteration(first loop), all the first bits are in their final
places, all second bits but second bit from eight byte are in their places...
only 1 eight bit from eight byte is in it's place). In second loop we do all
operations symmetrically, in order to get other half of bits in their final
spots. Results for first and second doubleword elements are saved in
result[0] and result[1] respectively. In the end those results are saved in
appropriate doubleword element of destination register vD.
Signed-off-by: Stefan Brankovic <stefan.brankovic@rt-rk.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <1563200574-11098-5-git-send-email-stefan.brankovic@rt-rk.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>