qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
To: Jonathan Cameron <Jonathan.Cameron@Huawei.com>
Cc: Igor Mammedov <imammedo@redhat.com>,
	Shiju Jose <shiju.jose@huawei.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Ani Sinha <anisinha@redhat.com>,
	Dongjiu Geng <gengdongjiu1@gmail.com>,
	<linux-kernel@vger.kernel.org>, <qemu-arm@nongnu.org>,
	<qemu-devel@nongnu.org>
Subject: Re: [PATCH 09/15] acpi/ghes: make the GHES record generation more generic
Date: Thu, 26 Sep 2024 16:19:15 +0200	[thread overview]
Message-ID: <20240926161726.3945f835@foz.lan> (raw)
In-Reply-To: <20240926130056.00001e12@Huawei.com>

Em Thu, 26 Sep 2024 13:00:56 +0100
Jonathan Cameron <Jonathan.Cameron@Huawei.com> escreveu:

> On Wed, 25 Sep 2024 06:04:14 +0200
> Mauro Carvalho Chehab <mchehab+huawei@kernel.org> wrote:
> 
> > Split the code into separate functions to allow using the
> > common CPER filling code by different error sources.
> > 
> > The generic code was moved to ghes_record_cper_errors(),
> > and ghes_gen_err_data_uncorrectable_recoverable() now contains
> > only a logic to fill GEGB part of the record.
> > 
> > The remaining code to generate a memory error now belongs to
> > acpi_ghes_record_errors() function.
> > 
> > A further patch will give it a better name.  
> 
> That bit is fine, I'm less sure about
> ghes_gen_err_data_uncorrectable_recoverable()
> Maybe you refactor that later, but I'd suggest doing so in this
> patch to make it 
> ghes_gen_data() with the uncorrectable and recoverable bits
> passed in as parameters.

For now, no need. When using the error injection script, such
function is not used. The script can already play with what's
defined there.

Besides that, I tried to generalize it, but it is not trivial,
as passing some things as parameter is really hard. So, instead,
I opted to keep the code as-is. It can be changed later once we
add internal events that require a different setting than what
we have with ARMv8 memory errors via SEA.
> 
> Jonathan
> 
> > 
> > Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
> > ---
> >  hw/acpi/ghes.c         | 118 +++++++++++++++++++++++++----------------
> >  include/hw/acpi/ghes.h |   3 ++
> >  2 files changed, 74 insertions(+), 47 deletions(-)
> > 
> > diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c
> > index 340a0263faf8..307b5a41d539 100644
> > --- a/hw/acpi/ghes.c
> > +++ b/hw/acpi/ghes.c
> > @@ -181,51 +181,30 @@ static void acpi_ghes_build_append_mem_cper(GArray *table,
> >      build_append_int_noprefix(table, 0, 7);
> >  }
> >  
> > -static int acpi_ghes_record_mem_error(uint64_t error_block_address,
> > -                                      uint64_t error_physical_addr)
> > +static void
> > +ghes_gen_err_data_uncorrectable_recoverable(GArray *block,
> > +                                            const uint8_t *section_type,
> > +                                            int data_length)
> >  {  
> 
> That's an ugly name .  Suggestion below on instead passing parameters
> for the uncorrectable and recoverable parts and amking this
> ghes_gen_err_data()
> 
> > -    GArray *block;
> > -
> > -    /* Memory Error Section Type */
> > -    const uint8_t uefi_cper_mem_sec[] =
> > -          UUID_LE(0xA5BC1114, 0x6F64, 0x4EDE, 0xB8, 0x63, 0x3E, 0x83, \
> > -                  0xED, 0x7C, 0x83, 0xB1);
> > -
> >      /* invalid fru id: ACPI 4.0: 17.3.2.6.1 Generic Error Data,
> >       * Table 17-13 Generic Error Data Entry
> >       */
> >      QemuUUID fru_id = {};
> > -    uint32_t data_length;
> >  
> > -    block = g_array_new(false, true /* clear */, 1);
> > -
> > -    /* This is the length if adding a new generic error data entry*/
> > -    data_length = ACPI_GHES_DATA_LENGTH + ACPI_GHES_MEM_CPER_LENGTH;
> >      /*
> > -     * It should not run out of the preallocated memory if adding a new generic
> > -     * error data entry
> > +     * Calculate the size with this block. No need to check for
> > +     * too big CPER, as CPER size is checked at ghes_record_cper_errors()
> >       */
> > -    assert((data_length + ACPI_GHES_GESB_SIZE) <=
> > -            ACPI_GHES_MAX_RAW_DATA_LENGTH);
> > +    data_length += ACPI_GHES_GESB_SIZE;
> >  
> >      /* Build the new generic error status block header */
> >      acpi_ghes_generic_error_status(block, ACPI_GEBS_UNCORRECTABLE,
> >          0, 0, data_length, ACPI_CPER_SEV_RECOVERABLE);
> >  
> >      /* Build this new generic error data entry header */
> > -    acpi_ghes_generic_error_data(block, uefi_cper_mem_sec,
> > +    acpi_ghes_generic_error_data(block, section_type,
> >          ACPI_CPER_SEV_RECOVERABLE, 0, 0,
> >          ACPI_GHES_MEM_CPER_LENGTH, fru_id, 0);  
> Maybe should just pass in ACPI_CPER_SEV_RECOVERABLE 
> and ACPI_GEBS_UNCORRECTABLE in parameters.
> 
> Main advantage being that should allow reuse for other combinations
> and it gets rid of the nasty function name!
> 
> > -
> > -    /* Build the memory section CPER for above new generic error data entry */
> > -    acpi_ghes_build_append_mem_cper(block, error_physical_addr);
> > -
> > -    /* Write the generic error data entry into guest memory */
> > -    cpu_physical_memory_write(error_block_address, block->data, block->len);
> > -
> > -    g_array_free(block, true);
> > -
> > -    return 0;
> >  }
> >  
> >  /*
> > @@ -399,14 +378,19 @@ void acpi_ghes_add_fw_cfg(AcpiGhesState *ags, FWCfgState *s,
> >      ags->present = true;
> >  }
> >  
> > -int acpi_ghes_record_errors(uint16_t source_id, uint64_t physical_address)
> > +void ghes_record_cper_errors(const void *cper, size_t len,
> > +                             uint16_t source_id, Error **errp)
> >  {  
> 
> > -    return ret;
> > +    read_ack_register = cpu_to_le64(0);
> > +    /*
> > +        * Clear the Read Ack Register, OSPM will write it to 1 when
> > +        * it acknowledges this error.
> > +        */  
> 
> Indent issue.
> 
> > +    cpu_physical_memory_write(read_ack_register_addr,
> > +        &read_ack_register, sizeof(uint64_t));
> > +
> > +    /* Write the generic error data entry into guest memory */
> > +    cpu_physical_memory_write(error_block_addr, cper, len);
> > +
> > +    return;
> > +}  
> 



Thanks,
Mauro


  reply	other threads:[~2024-09-26 14:20 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-25  4:04 [PATCH 00/15] Prepare GHES driver to support error injection Mauro Carvalho Chehab
2024-09-25  4:04 ` [PATCH 01/15] acpi/ghes: get rid of ACPI_HEST_SRC_ID_RESERVED Mauro Carvalho Chehab
2024-09-25 14:02   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 02/15] acpi/ghes: simplify acpi_ghes_record_errors() code Mauro Carvalho Chehab
2024-09-25 14:09   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 03/15] acpi/ghes: simplify the per-arch caller to build HEST table Mauro Carvalho Chehab
2024-09-25 14:11   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 04/15] acpi/ghes: better handle source_id and notification Mauro Carvalho Chehab
2024-09-25 14:13   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 05/15] acpi/ghes: Fix acpi_ghes_record_errors() argument Mauro Carvalho Chehab
2024-09-25 14:15   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 06/15] acpi/ghes: Remove a duplicated out of bounds check Mauro Carvalho Chehab
2024-09-25 14:15   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 07/15] acpi/ghes: Change the type for source_id Mauro Carvalho Chehab
2024-09-25 14:16   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 08/15] acpi/ghes: Prepare to support multiple sources on ghes Mauro Carvalho Chehab
2024-09-25 14:23   ` Jonathan Cameron via
2024-10-01  5:29     ` Mauro Carvalho Chehab
2024-09-25  4:04 ` [PATCH 09/15] acpi/ghes: make the GHES record generation more generic Mauro Carvalho Chehab
2024-09-26 12:00   ` Jonathan Cameron via
2024-09-26 14:19     ` Mauro Carvalho Chehab [this message]
2024-09-25  4:04 ` [PATCH 10/15] acpi/ghes: move offset calculus to a separate function Mauro Carvalho Chehab
2024-09-26 12:03   ` Jonathan Cameron via
2024-10-01  5:38     ` Mauro Carvalho Chehab
2024-09-25  4:04 ` [PATCH 11/15] acpi/ghes: better name GHES memory error function Mauro Carvalho Chehab
2024-09-26 12:07   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 12/15] acpi/ghes: don't crash QEMU if ghes GED is not found Mauro Carvalho Chehab
2024-09-26 12:09   ` Jonathan Cameron via
2024-09-26 14:24     ` Mauro Carvalho Chehab
2024-09-25  4:04 ` [PATCH 13/15] acpi/ghes: rename etc/hardware_error file macros Mauro Carvalho Chehab
2024-09-26 12:11   ` Jonathan Cameron via
2024-09-25  4:04 ` [PATCH 14/15] better name the offset of the hardware error firmware Mauro Carvalho Chehab
2024-09-26 12:12   ` Jonathan Cameron via
2024-10-01  6:02     ` Mauro Carvalho Chehab
2024-09-25  4:04 ` [PATCH 15/15] docs: acpi_hest_ghes: fix documentation for CPER size Mauro Carvalho Chehab
2024-09-26 12:13   ` Jonathan Cameron via

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240926161726.3945f835@foz.lan \
    --to=mchehab+huawei@kernel.org \
    --cc=Jonathan.Cameron@Huawei.com \
    --cc=anisinha@redhat.com \
    --cc=gengdongjiu1@gmail.com \
    --cc=imammedo@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=shiju.jose@huawei.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).