From: Luis Machado Date: Mon, 20 Feb 2023 09:56:19 +0000 (+0000) Subject: sve: Fix signal frame z/v register restore X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=1bb099a2449347510e164e253aea14a186fef0b1;p=binutils-gdb.git sve: Fix signal frame z/v register restore While doing some SME work, I ran into the situation where the Z register contents restored from a signal frame are incorrect if the signal frame only contains fpsimd state and no sve state. This happens because we only restore the v register values in that case, and don't do anything for the z registers. Fix this by initializing the z registers to 0 and then copying over the overlapping part of the v registers to the z registers. While at it, refactor the code a bit to simplify it and make it smaller. Regression-tested on aarch64-linux Ubuntu 22.04/20.04. Reviewed-by: Thiago Jung Bauermann --- diff --git a/gdb/aarch64-linux-tdep.c b/gdb/aarch64-linux-tdep.c index b183a3c9a38..bdd5cb05c10 100644 --- a/gdb/aarch64-linux-tdep.c +++ b/gdb/aarch64-linux-tdep.c @@ -196,14 +196,13 @@ read_aarch64_ctx (CORE_ADDR ctx_addr, enum bfd_endian byte_order, /* Given CACHE, use the trad_frame* functions to restore the FPSIMD registers from a signal frame. - VREG_NUM is the number of the V register being restored, OFFSET is the - address containing the register value, BYTE_ORDER is the endianness and - HAS_SVE tells us if we have a valid SVE context or not. */ + FPSIMD_CONTEXT is the address of the signal frame context containing FPSIMD + data. */ static void -aarch64_linux_restore_vreg (struct trad_frame_cache *cache, int num_regs, - int vreg_num, CORE_ADDR offset, - enum bfd_endian byte_order, bool has_sve) +aarch64_linux_restore_vregs (struct gdbarch *gdbarch, + struct trad_frame_cache *cache, + CORE_ADDR fpsimd_context) { /* WARNING: SIMD state is laid out in memory in target-endian format. @@ -215,11 +214,22 @@ aarch64_linux_restore_vreg (struct trad_frame_cache *cache, int num_regs, 2 - If the target is little endian, then SIMD state is little endian, so no byteswap is needed. */ - if (byte_order == BFD_ENDIAN_BIG) + enum bfd_endian byte_order = gdbarch_byte_order (gdbarch); + int num_regs = gdbarch_num_regs (gdbarch); + aarch64_gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); + + for (int i = 0; i < 32; i++) { + CORE_ADDR offset = (fpsimd_context + AARCH64_FPSIMD_V0_OFFSET + + (i * AARCH64_FPSIMD_VREG_SIZE)); + gdb_byte buf[V_REGISTER_SIZE]; - if (target_read_memory (offset, buf, V_REGISTER_SIZE) != 0) + /* Read the contents of the V register. */ + if (target_read_memory (offset, buf, V_REGISTER_SIZE)) + error (_("Failed to read fpsimd register from signal context.")); + + if (byte_order == BFD_ENDIAN_BIG) { size_t size = V_REGISTER_SIZE/2; @@ -234,50 +244,66 @@ aarch64_linux_restore_vreg (struct trad_frame_cache *cache, int num_regs, store_unsigned_integer (buf + size , size, BFD_ENDIAN_LITTLE, u64); /* Now we can store the correct bytes for the V register. */ - trad_frame_set_reg_value_bytes (cache, AARCH64_V0_REGNUM + vreg_num, + trad_frame_set_reg_value_bytes (cache, AARCH64_V0_REGNUM + i, {buf, V_REGISTER_SIZE}); trad_frame_set_reg_value_bytes (cache, num_regs + AARCH64_Q0_REGNUM - + vreg_num, {buf, Q_REGISTER_SIZE}); + + i, {buf, Q_REGISTER_SIZE}); trad_frame_set_reg_value_bytes (cache, num_regs + AARCH64_D0_REGNUM - + vreg_num, {buf, D_REGISTER_SIZE}); + + i, {buf, D_REGISTER_SIZE}); trad_frame_set_reg_value_bytes (cache, num_regs + AARCH64_S0_REGNUM - + vreg_num, {buf, S_REGISTER_SIZE}); + + i, {buf, S_REGISTER_SIZE}); trad_frame_set_reg_value_bytes (cache, num_regs + AARCH64_H0_REGNUM - + vreg_num, {buf, H_REGISTER_SIZE}); + + i, {buf, H_REGISTER_SIZE}); trad_frame_set_reg_value_bytes (cache, num_regs + AARCH64_B0_REGNUM - + vreg_num, {buf, B_REGISTER_SIZE}); + + i, {buf, B_REGISTER_SIZE}); - if (has_sve) + if (tdep->has_sve ()) trad_frame_set_reg_value_bytes (cache, num_regs + AARCH64_SVE_V0_REGNUM - + vreg_num, {buf, V_REGISTER_SIZE}); + + i, {buf, V_REGISTER_SIZE}); } - return; - } + else + { + /* Little endian, just point at the address containing the register + value. */ + trad_frame_set_reg_addr (cache, AARCH64_V0_REGNUM + i, offset); + trad_frame_set_reg_addr (cache, num_regs + AARCH64_Q0_REGNUM + i, + offset); + trad_frame_set_reg_addr (cache, num_regs + AARCH64_D0_REGNUM + i, + offset); + trad_frame_set_reg_addr (cache, num_regs + AARCH64_S0_REGNUM + i, + offset); + trad_frame_set_reg_addr (cache, num_regs + AARCH64_H0_REGNUM + i, + offset); + trad_frame_set_reg_addr (cache, num_regs + AARCH64_B0_REGNUM + i, + offset); - /* Little endian, just point at the address containing the register - value. */ - trad_frame_set_reg_addr (cache, AARCH64_V0_REGNUM + vreg_num, offset); - trad_frame_set_reg_addr (cache, num_regs + AARCH64_Q0_REGNUM + vreg_num, - offset); - trad_frame_set_reg_addr (cache, num_regs + AARCH64_D0_REGNUM + vreg_num, - offset); - trad_frame_set_reg_addr (cache, num_regs + AARCH64_S0_REGNUM + vreg_num, - offset); - trad_frame_set_reg_addr (cache, num_regs + AARCH64_H0_REGNUM + vreg_num, - offset); - trad_frame_set_reg_addr (cache, num_regs + AARCH64_B0_REGNUM + vreg_num, - offset); - - if (has_sve) - trad_frame_set_reg_addr (cache, num_regs + AARCH64_SVE_V0_REGNUM - + vreg_num, offset); + if (tdep->has_sve ()) + trad_frame_set_reg_addr (cache, num_regs + AARCH64_SVE_V0_REGNUM + + i, offset); + } + if (tdep->has_sve ()) + { + /* If SVE is supported for this target, zero out the Z + registers then copy the first 16 bytes of each of the V + registers to the associated Z register. Otherwise the Z + registers will contain uninitialized data. */ + std::vector z_buffer (tdep->vq * 16); + + /* We have already handled the endianness swap above, so we don't need + to worry about it here. */ + memcpy (z_buffer.data (), buf, V_REGISTER_SIZE); + trad_frame_set_reg_value_bytes (cache, + AARCH64_SVE_Z0_REGNUM + i, + z_buffer); + } + } } /* Implement the "init" method of struct tramp_frame. */ @@ -432,16 +458,7 @@ aarch64_linux_sigframe_init (const struct tramp_frame *self, /* If there was no SVE section then set up the V registers. */ if (sve_regs == 0) - { - for (int i = 0; i < 32; i++) - { - CORE_ADDR offset = (fpsimd + AARCH64_FPSIMD_V0_OFFSET - + (i * AARCH64_FPSIMD_VREG_SIZE)); - - aarch64_linux_restore_vreg (this_cache, num_regs, i, offset, - byte_order, tdep->has_sve ()); - } - } + aarch64_linux_restore_vregs (gdbarch, this_cache, fpsimd); } trad_frame_set_id (this_cache, frame_id_build (sp, func));