On Wed, 5 Nov 2025 21:44:49 +1000 Gavin Shan <[email protected]> wrote:
> In the situation where host and guest has 64KiB and 4KiB page sizes, > one problematic host page affects 16 guest pages. we need to send 16 > consective errors in this specific case. > > Extend acpi_ghes_memory_errors() to support multiple CPERs after the > hunk of code to generate the GHES error status is pulled out from > ghes_gen_err_data_uncorrectable_recoverable(). The status field of > generic error status block is also updated accordingly if multiple > error data entries are contained in the generic error status block. > > Signed-off-by: Gavin Shan <[email protected]> Hi Gavin, Mostly fine, but a few comments on the defines added and a question on what the multiple things are meant to mean? > diff --git a/hw/acpi/ghes.c b/hw/acpi/ghes.c > index a9c08e73c0..527b85c8d8 100644 > --- a/hw/acpi/ghes.c > +++ b/hw/acpi/ghes.c > @@ -57,8 +57,12 @@ > /* The memory section CPER size, UEFI 2.6: N.2.5 Memory Error Section */ > #define ACPI_GHES_MEM_CPER_LENGTH 80 > > -/* Masks for block_status flags */ > -#define ACPI_GEBS_UNCORRECTABLE 1 > +/* Bits for block_status flags */ > +#define ACPI_GEBS_UNCORRECTABLE 0 > +#define ACPI_GEBS_CORRECTABLE 1 > +#define ACPI_GEBS_MULTIPLE_UNCORRECTABLE 2 > +#define ACPI_GEBS_MULTIPLE_CORRECTABLE 3 So this maps to the bits in block status. I'm not actually sure what these multiple variants are meant to tell us. The multiple error blocks example referred to by the spec is a way to represent the same error applying to multiple places. So that's one error, many blocks. I have no idea if we set these bits in that case. Based on a quick look I don't think linux even takes any notice. THere are defines in actbl1.h but I'm not seeing any use made of them. > +#define ACPI_GEBS_ERROR_DATA_ENTRIES 4 This is bits 4-13 and the define isn't used. I'd drop it.
