| Message ID | 1494883680-25551-9-git-send-email-tbaicar@codeaurora.org (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
On Mon, May 15, 2017 at 03:27:57PM -0600, Tyler Baicar wrote: > UEFI spec allows for non-standard section in Common Platform Error > Record. This is defined in section N.2.3 of UEFI version 2.5. > > Currently if the CPER section's type (UUID) does not match with > one of the section types that the kernel knows how to parse, the > section is skipped. Therefore, user is not able to see > such CPER data, for instance, error record of non-standard section. > > This change prints out the raw data in hex in the dmesg buffer so > that non-standard sections are reported to the user. Non-standard > section type errors should be reported to the user because these > can include errors which are vendor specific. The data length is > taken from Error Data length field of Generic Error Data Entry. > > The following is a sample output from dmesg: > [ 140.739180] {1}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 2 > [ 140.739182] {1}[Hardware Error]: It has been corrected by h/w and requires no further action > [ 140.739191] {1}[Hardware Error]: event severity: corrected > [ 140.739196] {1}[Hardware Error]: time: precise 2017-03-15 20:37:35 > [ 140.739197] {1}[Hardware Error]: Error 0, type: corrected > [ 140.739203] {1}[Hardware Error]: section type: unknown, d2e2621c-f936-468d-0d84-15a4ed015c8b > [ 140.739205] {1}[Hardware Error]: section length: 0x238 > [ 140.739210] {1}[Hardware Error]: 00000000: 4d415201 4d492031 453a4d45 435f4343 .RAM1 IMEM:ECC_C > [ 140.739214] {1}[Hardware Error]: 00000010: 53515f45 44525f42 00000000 00000000 E_QSB_RD........ > [ 140.739217] {1}[Hardware Error]: 00000020: 00000000 00000000 00000000 00000000 ................ > [ 140.739220] {1}[Hardware Error]: 00000030: 00000000 00000000 01010000 01010000 ................ > [ 140.739223] {1}[Hardware Error]: 00000040: 00000000 00000000 00000005 00000000 ................ > [ 140.739226] {1}[Hardware Error]: 00000050: 01010000 00000000 00000001 00dddd00 ................ Let me repeat myself from the last time: "Kill all those prefixes: " Hardware error from APEI Generic Hardware Error Source: 2 It has been corrected by h/w and requires no further action event severity: corrected time: precise 2017-03-15 20:37:35 Error 0, type: corrected section type: unknown, d2e2621c-f936-468d-0d84-15a4ed015c8b section length: 568 (0x238) 00000000: 4d415201 4d492031 453a4d45 435f4343 .RAM1 IMEM:ECC_C 00000010: 53515f45 44525f42 00000000 00000000 E_QSB_RD........ 00000020: 00000000 00000000 00000000 00000000 ................ 00000030: 00000000 00000000 01010000 01010000 ................ 00000040: 00000000 00000000 00000005 00000000 ................ 00000050: 01010000 00000000 00000001 00dddd00 ................ " to the important info only."
On 5/16/2017 8:29 AM, Borislav Petkov wrote: > On Mon, May 15, 2017 at 03:27:57PM -0600, Tyler Baicar wrote: >> UEFI spec allows for non-standard section in Common Platform Error >> Record. This is defined in section N.2.3 of UEFI version 2.5. >> >> Currently if the CPER section's type (UUID) does not match with >> one of the section types that the kernel knows how to parse, the >> section is skipped. Therefore, user is not able to see >> such CPER data, for instance, error record of non-standard section. >> >> This change prints out the raw data in hex in the dmesg buffer so >> that non-standard sections are reported to the user. Non-standard >> section type errors should be reported to the user because these >> can include errors which are vendor specific. The data length is >> taken from Error Data length field of Generic Error Data Entry. >> >> The following is a sample output from dmesg: >> [ 140.739180] {1}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 2 >> [ 140.739182] {1}[Hardware Error]: It has been corrected by h/w and requires no further action >> [ 140.739191] {1}[Hardware Error]: event severity: corrected >> [ 140.739196] {1}[Hardware Error]: time: precise 2017-03-15 20:37:35 >> [ 140.739197] {1}[Hardware Error]: Error 0, type: corrected >> [ 140.739203] {1}[Hardware Error]: section type: unknown, d2e2621c-f936-468d-0d84-15a4ed015c8b >> [ 140.739205] {1}[Hardware Error]: section length: 0x238 >> [ 140.739210] {1}[Hardware Error]: 00000000: 4d415201 4d492031 453a4d45 435f4343 .RAM1 IMEM:ECC_C >> [ 140.739214] {1}[Hardware Error]: 00000010: 53515f45 44525f42 00000000 00000000 E_QSB_RD........ >> [ 140.739217] {1}[Hardware Error]: 00000020: 00000000 00000000 00000000 00000000 ................ >> [ 140.739220] {1}[Hardware Error]: 00000030: 00000000 00000000 01010000 01010000 ................ >> [ 140.739223] {1}[Hardware Error]: 00000040: 00000000 00000000 00000005 00000000 ................ >> [ 140.739226] {1}[Hardware Error]: 00000050: 01010000 00000000 00000001 00dddd00 ................ > Let me repeat myself from the last time: > > "Kill all those prefixes: > > " Hardware error from APEI Generic Hardware Error Source: 2 > It has been corrected by h/w and requires no further action > event severity: corrected > time: precise 2017-03-15 20:37:35 > Error 0, type: corrected > section type: unknown, d2e2621c-f936-468d-0d84-15a4ed015c8b > section length: 568 (0x238) > 00000000: 4d415201 4d492031 453a4d45 435f4343 .RAM1 IMEM:ECC_C > 00000010: 53515f45 44525f42 00000000 00000000 E_QSB_RD........ > 00000020: 00000000 00000000 00000000 00000000 ................ > 00000030: 00000000 00000000 01010000 01010000 ................ > 00000040: 00000000 00000000 00000005 00000000 ................ > 00000050: 01010000 00000000 00000001 00dddd00 ................ > " > > to the important info only." Hello Boris, I meant to respond to this comment after I sent the v16 patch series, but you beat me to it :) These prefixes are common to all the GHES/CPER printing to the kernel logs. The first value here '{1}' is an increment based on the number of error records that have been printed during the current boot. This value can be very helpful when trying to parse the log which could have hundreds or thousands of these errors. Just yesterday I saw a log with ~250 records printed. It was a lot easier to see that incremented number and know how many were in the log than to actually parse the log to count that information. The '[Hardware Error]' print I could see doing away with, but it does actually have value when you're looking through the logs. It helps these errors stand out especially to people who aren't looking for them. These hardware errors shouldn't be happening, so it makes sense for them to stand out in the logs. And it also helps to find all these records in a log that could be littered with a lot of other prints. I find myself doing "dmesg | grep 'Hardware error'" all the time. Hopefully that is enough justification to keep these. If not, then I can add a separate patch in this series to remove them. Thanks, Tyler
On Tue, May 16, 2017 at 10:44:43AM -0600, Baicar, Tyler wrote: > I meant to respond to this comment after I sent the v16 patch series, but > you beat me to it :) > > These prefixes are common to all the GHES/CPER printing to the kernel logs. I don't mean that - I meant to remove them from this particular example in the commit message along with the timestamps. Because they are simply useless there. Of course they should stay in the actual printk calls.
On 5/16/2017 10:55 AM, Borislav Petkov wrote: > On Tue, May 16, 2017 at 10:44:43AM -0600, Baicar, Tyler wrote: >> I meant to respond to this comment after I sent the v16 patch series, but >> you beat me to it :) >> >> These prefixes are common to all the GHES/CPER printing to the kernel logs. > I don't mean that - I meant to remove them from this particular example > in the commit message along with the timestamps. Because they are simply > useless there. > > Of course they should stay in the actual printk calls. Ah, okay. I'll remove all that from the commit message.
diff --git a/drivers/firmware/efi/cper.c b/drivers/firmware/efi/cper.c index eac0854..d5a5855 100644 --- a/drivers/firmware/efi/cper.c +++ b/drivers/firmware/efi/cper.c @@ -585,8 +585,15 @@ static void cper_print_tstamp(const char *pfx, else goto err_section_too_small; #endif - } else - printk("%s""section type: unknown, %pUl\n", newpfx, sec_type); + } else { + const void *err = acpi_hest_get_payload(gdata); + + printk("%ssection type: unknown, %pUl\n", newpfx, sec_type); + printk("%ssection length: %#x\n", newpfx, + gdata->error_data_length); + print_hex_dump(newpfx, "", DUMP_PREFIX_OFFSET, 16, 4, err, + gdata->error_data_length, true); + } return;
UEFI spec allows for non-standard section in Common Platform Error Record. This is defined in section N.2.3 of UEFI version 2.5. Currently if the CPER section's type (UUID) does not match with one of the section types that the kernel knows how to parse, the section is skipped. Therefore, user is not able to see such CPER data, for instance, error record of non-standard section. This change prints out the raw data in hex in the dmesg buffer so that non-standard sections are reported to the user. Non-standard section type errors should be reported to the user because these can include errors which are vendor specific. The data length is taken from Error Data length field of Generic Error Data Entry. The following is a sample output from dmesg: [ 140.739180] {1}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 2 [ 140.739182] {1}[Hardware Error]: It has been corrected by h/w and requires no further action [ 140.739191] {1}[Hardware Error]: event severity: corrected [ 140.739196] {1}[Hardware Error]: time: precise 2017-03-15 20:37:35 [ 140.739197] {1}[Hardware Error]: Error 0, type: corrected [ 140.739203] {1}[Hardware Error]: section type: unknown, d2e2621c-f936-468d-0d84-15a4ed015c8b [ 140.739205] {1}[Hardware Error]: section length: 0x238 [ 140.739210] {1}[Hardware Error]: 00000000: 4d415201 4d492031 453a4d45 435f4343 .RAM1 IMEM:ECC_C [ 140.739214] {1}[Hardware Error]: 00000010: 53515f45 44525f42 00000000 00000000 E_QSB_RD........ [ 140.739217] {1}[Hardware Error]: 00000020: 00000000 00000000 00000000 00000000 ................ [ 140.739220] {1}[Hardware Error]: 00000030: 00000000 00000000 01010000 01010000 ................ [ 140.739223] {1}[Hardware Error]: 00000040: 00000000 00000000 00000005 00000000 ................ [ 140.739226] {1}[Hardware Error]: 00000050: 01010000 00000000 00000001 00dddd00 ................ ... The raw data from the error can then be decoded using vendor specific tools. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> --- drivers/firmware/efi/cper.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)