diff mbox series

[PULL,50/50] target/riscv: Fix vcompress with rvv_ta_all_1s

Message ID	20241031035319.731906-51-alistair.francis@wdc.com
State	New
Headers	show Return-Path: <qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org> From: Alistair Francis <alistair23@gmail.com> To: qemu-devel@nongnu.org Cc: alistair23@gmail.com, Anton Blanchard <antonb@tenstorrent.com>, Daniel Henrique Barboza <dbarboza@ventanamicro.com>, Alistair Francis <alistair.francis@wdc.com> Subject: [PULL 50/50] target/riscv: Fix vcompress with rvv_ta_all_1s Date: Thu, 31 Oct 2024 13:53:18 +1000 Message-ID: <20241031035319.731906-51-alistair.francis@wdc.com> In-Reply-To: <20241031035319.731906-1-alistair.francis@wdc.com> References: <20241031035319.731906-1-alistair.francis@wdc.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Received-SPF: pass client-ip=2607:f8b0:4864:20::631; envelope-from=alistair23@gmail.com; helo=mail-pl1-x631.google.com X-Spam_score_int: -17 X-Spam_score: -1.8 X-Spam_bar: - X-Spam_report: (-1.8 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action Precedence: list Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org
Series	[PULL,01/50] target/riscv/csr.c: Fix an access to VXSAT \| expand [PULL,01/50] target/riscv/csr.c: Fix an access to VXSAT [PULL,02/50] target/riscv: Add fw_dynamic_info32 for booting RV32 OpenSBI [PULL,03/50] target/riscv: Adjust PMP size for no-MMU RV64 QEMU running RV32 [PULL,04/50] target/riscv: Correct SXL return value for RV32 in RV64 QEMU [PULL,05/50] target/riscv: Detect sxl to set bit width for RV32 in RV64 [PULL,06/50] target/riscv: Correct mcause/scause bit width for RV32 in RV64 QEMU [PULL,07/50] target/riscv: Enable RV32 CPU support in RV64 QEMU [PULL,08/50] target/riscv: Add max32 CPU for RV64 QEMU [PULL,09/50] tests/avocado: Boot Linux for RV32 cpu on RV64 QEMU [PULL,10/50] hw/intc: Make zeroth priority register read-only [PULL,11/50] hw/intc: Don't clear pending bits on IRQ lowering [PULL,12/50] target/riscv: Set vtype.vill on CPU reset [PULL,13/50] hw/intc/riscv_aplic: Check and update pending when write sourcecfg [PULL,14/50] hw/char: riscv_htif: Use blocking qemu_chr_fe_write_all [PULL,15/50] hw/char: sifive_uart: Print uart characters async [PULL,16/50] target/riscv: expose *envcfg csr and priv to qemu-user as well [PULL,17/50] target/riscv: Add zicfilp extension [PULL,18/50] target/riscv: Introduce elp state and enabling controls for zicfilp [PULL,19/50] target/riscv: save and restore elp state on priv transitions [PULL,20/50] target/riscv: additional code information for sw check [PULL,21/50] target/riscv: tracking indirect branches (fcfi) for zicfilp [PULL,22/50] target/riscv: zicfilp `lpad` impl and branch tracking [PULL,23/50] disas/riscv: enable `lpad` disassembly [PULL,24/50] target/riscv: Expose zicfilp extension as a cpu property [PULL,25/50] target/riscv: Add zicfiss extension [PULL,26/50] target/riscv: introduce ssp and enabling controls for zicfiss [PULL,27/50] target/riscv: tb flag for shadow stack instructions [PULL,28/50] target/riscv: mmu changes for zicfiss shadow stack protection [PULL,29/50] target/riscv: AMO operations always raise store/AMO fault [PULL,30/50] target/riscv: update `decode_save_opc` to store extra word2 [PULL,31/50] target/riscv: implement zicfiss instructions [PULL,32/50] target/riscv: compressed encodings for sspush and sspopchk [PULL,33/50] disas/riscv: enable disassembly for zicfiss instructions [PULL,34/50] disas/riscv: enable disassembly for compressed sspush/sspopchk [PULL,35/50] target/riscv: Expose zicfiss extension as a cpu property [PULL,36/50] exec/memtxattr: add process identifier to the transaction attributes [PULL,37/50] hw/riscv: add riscv-iommu-bits.h [PULL,38/50] hw/riscv: add RISC-V IOMMU base emulation [PULL,39/50] pci-ids.rst: add Red Hat pci-id for RISC-V IOMMU device [PULL,40/50] hw/riscv: add riscv-iommu-pci reference device [PULL,41/50] hw/riscv/virt.c: support for RISC-V IOMMU PCIDevice hotplug [PULL,42/50] test/qtest: add riscv-iommu-pci tests [PULL,43/50] hw/riscv/riscv-iommu: add Address Translation Cache (IOATC) [PULL,44/50] hw/riscv/riscv-iommu: add ATS support [PULL,45/50] hw/riscv/riscv-iommu: add DBG support [PULL,46/50] qtest/riscv-iommu-test: add init queues test [PULL,47/50] docs/specs: add riscv-iommu [PULL,48/50] target/riscv/kvm: set 'aia_mode' to default in error path [PULL,49/50] target/riscv/kvm: clarify how 'riscv-aia' default works [PULL,50/50] target/riscv: Fix vcompress with rvv_ta_all_1s

Commit Message

Alistair Francis Oct. 31, 2024, 3:53 a.m. UTC

From: Anton Blanchard <antonb@tenstorrent.com>

vcompress packs vl or less fields into vd, so the tail starts after the
last packed field. This could be more clearly expressed in the ISA,
but for now this thread helps to explain it:

https://github.com/riscv/riscv-v-spec/issues/796

Signed-off-by: Anton Blanchard <antonb@tenstorrent.com>
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20241030043538.939712-1-antonb@tenstorrent.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
---
 target/riscv/vector_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff mbox series

Patch

diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index 072bd444b1..ccb32e6122 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -5132,7 +5132,7 @@  void HELPER(NAME)(void *vd, void *v0, void *vs1, void *vs2,               \
     }                                                                     \
     env->vstart = 0;                                                      \
     /* set tail elements to 1s */                                         \
-    vext_set_elems_1s(vd, vta, vl * esz, total_elems * esz);              \
+    vext_set_elems_1s(vd, vta, num * esz, total_elems * esz);             \
 }
 
 /* Compress into vd elements of vs2 where vs1 is enabled */