On Wed, 30 Jun 2021 15:07:11 -0400
Eric DeVolder <firstname.lastname@example.org> wrote:
I believe I have corrected for all feedback on v4, but with
responses to certain feedback below.
In patch 1/6, Igor asks:
"you are adding empty template files here
but the later matching bios-tables-test is nowhere to be found
Was testcase lost somewhere along the way?
also it seems you add ERST only to pc/q35,
so why tests/data/acpi/microvm/ERST is here?"
I did miss setting up microvm. That has been corrected.
As for the question about lost test cases, if you are referring
to the new binary blobs for pc,q35, those were in patch
6/6. There is a qtest in patch 5/6. If I don't understand the
question, please indicate as such.
All I see in this series is
[PATCH v5 09/10] ACPI ERST: qtest for ERST
which is not related to bios-tables-test and blobs whatsoever.
Blobs are for use with bios-tables-test and I'm referring to
missing test case in bios-tables-test.c
In patch 3/6, Igor asks:
"Also spec (ERST) is rather (maybe intentionally) vague on specifics,
so it would be better that before a patch that implements hw part
were a doc patch describing concrete implementation. As model
you can use docs/specs/acpi_hest_ghes.rst or other docs/specs/acpi_* files.
I'd start posting/discussing that spec within these thread
to avoid spamming list until doc is settled up."
I'm thinking that this cover letter is the bulk of the spec? But as
you say, to avoid spamming the group, we can use this thread to make
suggested changes to this cover letter which I will then convert
into a spec, for v6.
In patch 3/6, in many places Igor mentions utilizing the hostmem
mapped directly in the guest in order to avoid need-less copying.
It is true that the ERST has an "NVRAM" mode that would allow for
all the simplifications Igor points out, however, Linux does not
support this mode. This mode puts the burden of managing the NVRAM
space on the OS. So this implementation, like BIOS, is the non-NVRAM
see per patch comments where copying is not necessary regardless of
the implemented mode.
I did go ahead and separate the registers from the exchange buffer,
which would facilitate the support of NVRAM mode.
/* NVRAM ERST Error Log Address Range is not supported yet */
static void pr_unimpl_nvram(void)
pr_warn("NVRAM ERST Log Address Range not implemented yet.\n");
static int __erst_write_to_nvram(const struct cper_record_header *record)
/* do not print message, because printk is not safe for NMI */
static int __erst_read_to_erange_from_nvram(u64 record_id, u64 *offset)
static int __erst_clear_from_nvram(u64 record_id)
it's inconvenient when you copy questions/parts of unfinished discussion
from previous revision with a little context.
Usually discussion should continue in the original thread and
once some sort of consensus is reached new series based on it
is posted. Above blob shouldn't be here. (You can look at how others
handle multiple revisions)
The way you do it now, makes reviewer to repeat job done earlier
to point to the the same issues, so it wastes your and reviewer's time.
So please finish discussions in threads they started at and then post
This patchset introduces support for the ACPI Error Record
Serialization Table, ERST.
For background and implementation information, please see
docs/specs/acpi_erst.txt, which is patch 2/10.
Suggested-by: Konrad Wilk <email@example.com>
Signed-off-by: Eric DeVolder <firstname.lastname@example.org>
- Create docs/specs/acpi_erst.txt, per Igor
- Separate PCI BARs for registers and memory, per Igor
- Convert debugging to use trace infrastructure, per Igor
- Various other fixups, per Igor
- Converted to a PCI device, per Igor.
- Updated qtest.
- Rearranged patches, per Igor.
- Converted to using a TYPE_MEMORY_BACKEND_FILE object rather than
internal array with explicit file operations, per Igor.
- Changed the way the qdev and base address are handled, allowing
ERST to be disabled at run-time. Also aligns better with other
- Added qtest/smoke test per Paolo Bonzini
- Split patch into smaller chunks, per Igor Mammedov
- Did away with use of ACPI packed structures, per Igor Mammedov
- initial post
Eric DeVolder (10):
ACPI ERST: bios-tables-test.c steps 1 and 2
ACPI ERST: specification for ERST support
ACPI ERST: PCI device_id for ERST
ACPI ERST: header file for ERST
ACPI ERST: support for ACPI ERST feature
ACPI ERST: build the ACPI ERST table
ACPI ERST: trace support
ACPI ERST: create ACPI ERST table for pc/x86 machines.
ACPI ERST: qtest for ERST
ACPI ERST: step 6 of bios-tables-test.c
docs/specs/acpi_erst.txt | 152 +++++++
hw/acpi/erst.c | 918 +++++++++++++++++++++++++++++++++++++++++++
hw/acpi/meson.build | 1 +
hw/acpi/trace-events | 14 +
hw/i386/acpi-build.c | 9 +
hw/i386/acpi-microvm.c | 9 +
include/hw/acpi/erst.h | 84 ++++
include/hw/pci/pci.h | 1 +
tests/data/acpi/microvm/ERST | Bin 0 -> 976 bytes
tests/data/acpi/pc/ERST | Bin 0 -> 976 bytes
tests/data/acpi/q35/ERST | Bin 0 -> 976 bytes
tests/qtest/erst-test.c | 129 ++++++
tests/qtest/meson.build | 2 +
13 files changed, 1319 insertions(+)
create mode 100644 docs/specs/acpi_erst.txt
create mode 100644 hw/acpi/erst.c
create mode 100644 include/hw/acpi/erst.h
create mode 100644 tests/data/acpi/microvm/ERST
create mode 100644 tests/data/acpi/pc/ERST
create mode 100644 tests/data/acpi/q35/ERST
create mode 100644 tests/qtest/erst-test.c