-
Notifications
You must be signed in to change notification settings - Fork 221
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
CV32E40Pv2 Verification update #2369
CV32E40Pv2 Verification update #2369
Conversation
…_pulp_hwloop_count_range_test for large lpcount Signed-off-by: Vaibhav Jain <[email protected]>
…m update to use directed_streams in debug rom Signed-off-by: Vaibhav Jain <[email protected]>
…structions mainly wfi with pulp Signed-off-by: Vaibhav Jain <[email protected]>
Signed-off-by: dd-baoshan <[email protected]>
…s_0130 Add new test in regression yaml
Signed-off-by: dd-baoshan <[email protected]>
…s_0131 Redo updates for Increasing debug rom space
Signed-off-by: dd-baoshan <[email protected]>
Signed-off-by: dd-baoshan <[email protected]>
Signed-off-by: dd-baoshan <[email protected]>
…nously and ebreakm is enable in debug mode; Add plusarg to disable sampling Signed-off-by: dd-baoshan <[email protected]>
…s_0202 Cv32e40p/bsm update tb files 0202
Signed-off-by: Pascal Gouedo <[email protected]>
Set to Warning temporarily. Signed-off-by: Pascal Gouedo <[email protected]>
cv32e40p/dev_dd_pgo
… relaxation seems to have bad effect on 20230905 toolchain. Signed-off-by: Pascal Gouedo <[email protected]>
Restored old linker scripts
Signed-off-by: dd-baoshan <[email protected]>
Signed-off-by: dd-baoshan <[email protected]>
…s_0205 Cv32e40p/bsm update tb files 0205
Signed-off-by: Pascal Gouedo <[email protected]>
New linker scripts to resolve gp relaxation (gcc issue #92).
…without gp relaxation. Signed-off-by: Pascal Gouedo <[email protected]>
Signed-off-by: Pascal Gouedo <[email protected]>
Cv32e40p/dev dd pgo
Signed-off-by: dd-baoshan <[email protected]>
…s_0206 Minor fix
Signed-off-by: dd-baoshan <[email protected]>
…s_0206 Fix issue in handling exception trap intercept with debug trigger match
…be copied inside the assembly test
OHG sync from cv32e40p/dev
First round of coverage enhancements
added missing hpmcounterh regs
Signed-off-by: Pascal Gouedo <[email protected]>
Corrected duplicated definition of F bins.
Signed-off-by: dd-baoshan <[email protected]>
…instr and data obi interface Signed-off-by: dd-baoshan <[email protected]>
…s_WW06 Cv32e40p/bsm update tb files ww06
…could occur from cfg field missing inside test Object OR --cfg switch missing from cmd)
…nnot be overriden at all by the cmd switch option
cv_regress fixes & tests update
…ent (MPP hardwired to 11)
Fixes for CSR tests
…rrupt_covg_v2) Signed-off-by: dd-baoshan <[email protected]>
…s_WW06 Add directed tests to improve tb functional coverage holes (uvme_inte…
@@ -242,8 +242,23 @@ SECTIONS | |||
. = SEGMENT_START("ldata-segment", .); | |||
. = ALIGN(32 / 8); | |||
__bss_end = .; | |||
/* In toolchain, x3 is the gp global pointer register and gp relaxation occurs. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for this detailed comment. (Did I mention I hate the toolchain?)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Toolchain is just implementing what is defined in RISC-V ABI 😀.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM @XavierAubert. Fortunately for your reviewer many of the files are YAMLs and I did not review in detail. I have a few comments about license headers, but none of these should gate the merge of this PR.
Passes a local CI regression.
/* Script for -z combreloc */ | ||
/* Copyright (C) 2014-2020 Free Software Foundation, Inc. | ||
Copyright (C) 2019 ETH Zürich and University of Bologna | ||
Copyright (C) 2020 OpenHW Group |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a new file and so should have an updated license header. Ideally it would update the copyright holder and date and use the SPDX Header:
# Copyright 2024 Dolphin Design
# SPDX-License-Identifier: Apache-2.0 WITH SHL-2.1
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a copy of the original linker script.
I didn't change the header as it was already like that in v1.
@@ -0,0 +1,183 @@ | |||
/* | |||
* Copyright 2023 Dolphin Design |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
2024?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sorry as I was overlooked when using the template.
/////////////////////////////////////////////////////////////////////////////// | ||
// | ||
// Copyright 2023 OpenHW Group | ||
// Copyright 2023 Dolphin Design |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
2024
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sorry as I was overlooked when using the template.
// BELOW ARE USE FOR SPECIAL HACKS PURPOSE - START | ||
|
||
// 1 - To cover directives instr/data gnt assert-deassert when req is low | ||
`define TB_HACK_1_OBI_GNT(TYPE) initial begin : hack_obi_intf_gnt_signal_1_``TYPE \ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Maybe this could go in a file such as uvmt_cv32e40p_macros.sv
, similar to what is in place for the environment files.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will update this in next PR. Thanks.
Below is the pull request containing the last two weeks of work on CV32E40Pv2 verification. As usual, everything is organized in a more readable way, by topic.
Tests addition, regression lists updates
New tests to address hwloop, and hwloop in debug (New tests XavierAubert/core-v-verif#121 & Add new test in regression yaml XavierAubert/core-v-verif#139)
Added new test to cover directive holes on gnt signal for instr and data obi interface Cv32e40p/bsm update tb files ww06 XavierAubert/core-v-verif#161
Added new tests to improve functional coverage holes Add directed tests to improve tb functional coverage holes (uvme_inte… XavierAubert/core-v-verif#164
Added a bunch of new tests to improve decoder RTL coverage (illegal instructions for custom opcodes and F instructions, illegal, readonly and legal CSRs accesses, see PRs First round of coverage enhancements XavierAubert/core-v-verif#158 & Fixes for CSR tests XavierAubert/core-v-verif#163)
Also updated v1 test to improve our coverage 2399b9f
Functionnal Coverage updates
Functional coverage enhacements on interrupts done by @dd-baoshan
Various Updates and Fixes
PR Cv32e40p/bsm update tb files 0202 XavierAubert/core-v-verif#141 to address updates mentionned in CV32E40Pv2 Verification update #2357
Solving pseudo instruction "LA" not translating to correct instructions corev-gcc#92 with New linker scripts to resolve gp relaxation (gcc issue #92). XavierAubert/core-v-verif#147, cv32e40p/dev_dd_pgo XavierAubert/core-v-verif#144
Corrected two directed tests for hwloop & hwloop in debug mode Cv32e40p/dev dd pgo XavierAubert/core-v-verif#148
Fix issue in handling exception trap intercept with debug trigger match 22e410c & 4103a32
Solved a bug where logs were missing if a test was run in parallel (See Correction trial about missing log when parallel execution. XavierAubert/core-v-verif#154 & Correction to avoid failing execution when missing log file. XavierAubert/core-v-verif#155)
cv_regress script fix, changed behaviour of num field as it was confusing cv_regress fixes & tests update XavierAubert/core-v-verif#162