For SME2, we need to expose the new ZT0 register in the gdbstub XML.
gdb documents that the requirements are:
> The ‘org.gnu.gdb.aarch64.sme2’ feature is optional. If present,
> then the ‘org.gnu.gdb.aarch64.sme’ feature must also be present.
> The ‘org.gnu.gdb.aarch64.sme2’ feature should contain the
> following:
>
> - ZT0 is a register of 512 bits (64 bytes). It is defined as a
> vector of bytes.
Implement this.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20251017153027.969016-2-peter.maydell@linaro.org
Update aarch64-core.xml to include field definitions for PSTATE, which
in gdb is modelled in the cpsr (current program status register)
pseudo-register, named after the actual cpsr register in armv7.
Defining the fields layout of the register allows easy inspection of for
example, the current exception level (EL):
For example. Before booting a Linux guest, EL=2, but after booting and
Ctrl-C'ing in gdb, we get EL=0:
(gdb) info registers $cpsr
cpsr 0x20402009 [ SP EL=2 BTYPE=0 PAN C ]
(gdb) cont
Continuing.
^C
Thread 2 received signal SIGINT, Interrupt.
0x0000ffffaaff286c in ?? ()
(gdb) info registers $cpsr
cpsr 0x20001000 [ EL=0 BTYPE=0 SSBS C ]
The aarch64-core.xml has been updated to match exactly the version
retrieved from upstream gdb, retrieved in 2025-05-19 from HEAD commit
9f4dc0b137c86f6ff2098cb1ab69442c69d6023d.
Link: https://sourceware.org/git/?p=binutils-gdb.git;a=blob;f=gdb/features/aarch64-core.xml;h=b8046510b9a085d30463d37b3ecc8d435f5fb7a4;hb=HEAD
Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org>
Message-Id: <20250519-gdbstub-aarch64-pstate-xml-v1-1-b4dbe87fe7c6@linaro.org>
[AJB: expanded upstream link]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-ID: <20250603110204.838117-18-alex.bennee@linaro.org>
Since GDB 13.1(GDB commit ea3352172), GDB LoongArch changed to use
fcc0-7 instead of fcc register. This commit partially reverts commit
2f149c759 (`target/loongarch: Update gdb_set_fpu() and gdb_get_fpu()`)
to match the behavior of GDB.
Note that it is a breaking change for GDB 13.0 or earlier, but it is
also required for GDB 13.1 or later to work.
Signed-off-by: Jiajie Chen <c@jia.je>
Acked-by: Song Gao <gaosong@loongson.cn>
Message-Id: <20230808054315.3391465-1-c@jia.je>
Signed-off-by: Song Gao <gaosong@loongson.cn>
Both TCG and KVM emulate ckc, cputm, last_break and prefix, and it's
quite useful to have them during debugging. Right now they are grouped
together with KVM-only pp, pfault_token, pfault_select and
pfault_compare in s390-virt.xml, and are not available when debugging
TCG-emulated code.
Move KVM-only registers into the new s390-virt-kvm.xml file. Advertise
s390-virt.xml always, and the new s390-virt-kvm.xml only for KVM.
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-Id: <20230314101813.174874-1-iii@linux.ibm.com>
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
So that we can avoid the "older gdb crashes" problem described in
commit 5787d17a42 and which caused us to disable reporting pauth
information via the gdbstub, newer gdb is going to implement support
for recognizing the pauth information via a new feature name:
org.gnu.gdb.aarch64.pauth_v2
Older gdb won't recognize this feature name, so we can re-enable the
pauth support under the new name without risking them crashing.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20230406150827.3322670-1-peter.maydell@linaro.org
Mirroring the upstream gdb xml files, the two stack boundary
registers are separated out.
Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Before this commit, there were contradictory descriptions about size of EFER
register.
Line 113 says the size is 8 bytes.
Line 129 says the size is 4 bytes.
As a result, when GDB is debugging an OS running on QEMU, the GDB cannot
read 'g' packets correctly. This 'g' packet transmits values of each
registers of machine emulated by QEMU to GDB. QEMU, the packet sender,
assign 4 bytes for EFER in 'g' packet based on the line 113.
GDB, the packet receiver, extract 8 bytes for EFER in 'g' packet based on
the line 129. Therefore, all registers located behind EFER in 'g' packet
has been shifted 4 bytes in GDB.
After this commit, GDB can read 'g' packets correctly.
Signed-off-by: TaiseiIto <taisei1212@outlook.jp>
Message-Id: <TY0PR0101MB4285F637209075C9F65FCDA6A4479@TY0PR0101MB4285.apcprd01.prod.exchangelabs.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The fixed register numbering in the various GDB feature files for
RISC-V only exists because these files were originally copied from the
GDB source tree.
However, the fixed numbering only exists in the GDB source tree so
that GDB, when it connects to a target that doesn't provide a target
description, will use a specific numbering scheme.
That numbering scheme is designed to be compatible with the first
versions of QEMU (for RISC-V), that didn't send a target description,
and relied on a fixed numbering scheme.
Because of the way that QEMU manages its target descriptions,
recording the number of registers in each feature, and just relying on
GDB's numbering starting from 0, then I propose that we remove all the
fixed numbering from the RISC-V feature xml files, and just rely on
the standard numbering scheme. Plenty of other targets manage their
xml files this way, e.g. ARM, AArch64, Loongarch, m68k, rx, and s390.
Signed-off-by: Andrew Burgess <aburgess@redhat.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Palmer Dabbelt <palmer@rivosinc.com>
Message-Id: <6069395f90e6fc24dac92197be815fedf42f5974.1661934573.git.aburgess@redhat.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
While testing some changes to GDB's handling for the RISC-V registers
fcsr, fflags, and frm, I spotted that QEMU includes these registers
twice in the target description it sends to GDB, once in the fpu
feature, and once in the csr feature.
Right now things basically work OK, QEMU maps these registers onto two
different register numbers, e.g. fcsr maps to both 68 and 73, and GDB
can use either of these to access the register.
However, GDB's target descriptions don't really work this way, each
register should appear just once in a target description, mapping the
register name onto the number GDB should use when accessing the
register on the target. Duplicate register names actually result in
duplicate registers on the GDB side, however, as the registers have
the same name, the user can only access one of these registers.
Currently GDB has a hack in place, specifically for RISC-V, to spot
the duplicate copies of these three registers, and hide them from the
user, ensuring the user only ever sees a single copy of each.
In this commit I propose fixing this issue on the QEMU side, and in
the process, simplify the fpu register handling a little.
I think we should, remove fflags, frm, and fcsr from the two (32-bit
and 64-bit) fpu feature xml files. These files will only contain the
32 core floating point register f0 to f31. The fflags, frm, and fcsr
registers will continue to be advertised in the csr feature as they
currently are.
With that change made, I will simplify riscv_gdb_get_fpu and
riscv_gdb_set_fpu, removing the extra handling for the 3 status
registers.
Signed-off-by: Andrew Burgess <aburgess@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <0fbf2a5b12e3210ff3867d5cf7022b3f3462c9c8.1661934573.git.aburgess@redhat.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Cortex-M CPUs with MVE should advertise this fact to gdb, using the
org.gnu.gdb.arm.m-profile-mve XML feature, which defines the VPR
register. Presence of this feature also tells gdb to create
pseudo-registers Q0..Q7, so we do not need to tell gdb about them
separately.
Note that unless you have a very recent GDB that includes this fix:
http://patches-tcwg.linaro.org/patch/58133/ gdb will mis-print the
individual fields of the VPR register as zero (but showing the whole
thing as hex, eg with "print /x $vpr" will give the correct value).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20211101160814.5103-1-peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Currently we send VFP XML which includes D0..D15 or D0..D31, plus
FPSID, FPSCR and FPEXC. The upstream GDB tolerates this, but its
definition of this XML feature does not include FPSID or FPEXC. In
particular, for M-profile cores there are no FPSID or FPEXC
registers, so advertising those is wrong.
Move FPSID and FPEXC into their own bit of XML which we only send for
A and R profile cores. This brings our definition of the XML
org.gnu.gdb.arm.vfp feature into line with GDB's own (at least for
non-Neon cores...) and means we don't claim to have FPSID and FPEXC
on M-profile.
(It seems unlikely to me that any gdbstub users really care about
being able to look at FPEXC and FPSID; but we've supplied them to gdb
for a decade and it's not hard to keep doing so.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210921162901.17508-5-peter.maydell@linaro.org
GDB's remote protocol requires M-profile cores to use the feature
name 'org.gnu.gdb.arm.m-profile' instead of the 'org.gnu.gdb.arm.core'
feature used for A- and R-profile cores. We weren't doing this, which
meant GDB treated our M-profile cores like A-profile ones. This mostly
doesn't matter, but for instance means that it doesn't correctly
handle backtraces where an M-profile exception frame is involved.
Ship a copy of GDB's arm-m-profile.xml and use it on the M-profile
cores. The integer registers have the same offsets as the
arm-core.xml, but register 25 is the M-profile XPSR rather than the
A-profile CPSR, so we need to update arm_cpu_gdb_read_register() and
arm_cpu_gdb_write_register() to handle XSPR reads and writes.
Fixes: https://bugs.launchpad.net/qemu/+bug/1877136
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20200507134755.13997-1-peter.maydell@linaro.org
Currently "cf-core.xml" is sent to GDB when using any m68k flavor. Thing is
it uses the "org.gnu.gdb.coldfire.core" feature name and gdb 8.3 then expects
a coldfire FPU instead of the default m68881 FPU.
This is not OK because the m68881 floats registers are 96 bits wide so it
crashes GDB with the following error message:
(gdb) target remote localhost:7960
Remote debugging using localhost:7960
warning: Register "fp0" has an unsupported size (96 bits)
warning: Register "fp1" has an unsupported size (96 bits)
...
Remote 'g' packet reply is too long (expected 148 bytes, got 180 bytes): \
00000000000[...]0000
With this patch: qemu-system-m68k -M none -cpu m68020 -s -S
(gdb) tar rem :1234
Remote debugging using :1234
warning: No executable has been specified and target does not support
determining executable automatically. Try using the "file" command.
0x00000000 in ?? ()
(gdb) p $fp0
$1 = nan(0xffffffffffffffff)
Signed-off-by: KONRAD Frederic <frederic.konrad@adacore.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <1588094279-17913-3-git-send-email-frederic.konrad@adacore.com>
Message-Id: <20200430190122.4592-10-alex.bennee@linaro.org>
This patch enables a debugger to read the current privilege level via a virtual
"priv" register. When compiled with CONFIG_USER_ONLY the register is still
visible but always reports the value zero.
Signed-off-by: Jonathan Behrens <jonathan@fintelia.io>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
The machine description we send is being (silently) thrown on the floor
by GDB and GDB silently uses the default machine description, because
the xml parse fails on <feature> nested within <feature>.
Changes to the xml in qemu source code have no effect.
In addition, the default machine description has fs_base, which fails to
be retrieved, which breaks the whole register window. Add it and the
other control registers.
Signed-off-by: Doug Gale <doug16k@gmail.com>
Message-Id: <20190124040457.2546-1-doug16k@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add an XML description for SSE registers (XMM+MXCSR) for both X86
and X86-64 architectures in the GDB stub:
- configure: Define gdb_xml_files for the X86 targets (32 and 64bit).
- gdb-xml/i386-32bit-sse.xml & gdb-xml/i386-64bit-sse.xml: The XML files
that contain a description of the XMM + MXCSR registers.
- gdb-xml/i386-32bit.xml & gdb-xml/i386-64bit.xml: wrappers that include
the XML file of the core registers and the other XML file of the SSE registers.
- target/i386/cpu.c: Modify the gdb_core_xml_file to the new XML wrapper,
modify the gdb_num_core_regs to fit the registers number defined in each
XML file.
Signed-off-by: Abdallah Bouassida <abdallah.bouassida@lauterbach.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch implements XML target description support for X86 and X86-64
architectures in the GDB stub, as the way with ARM and PowerPC:
- gdb-xml/32bit-core.xml & gdb-xml/64bit-core.xml: Adding the XML target
description files, these files are picked from GDB source code.
- configure: Define gdb_xml_files for X86 targets.
- target/i386/cpu.c: Define gdb_core_xml_file and gdb_arch_name to add
XML awareness for this architecture, modify the gdb_num_core_regs to
fit the registers number defined in each XML file.
Signed-off-by: Abdallah Bouassida <abdallah.bouassida@lauterbach.com>
Message-Id: <2b3c8119-1602-28c7-eab4-296593877103@lauterbach.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Let's support reading and writing of control registers for kvm and tcg.
We have to take care of flushing the tlb (tcg) and pushing the changed
registers into kvm.
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
