github-mirror/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-05-30 18:13:41 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity Graph

PCI/AER: Print TLP Log for errors introduced since PCIe r1.1

Browse Source 
When reporting an error, the AER driver prints the TLP Header / Prefix Log
only for errors enumerated in the AER_LOG_TLP_MASKS macro.

The macro was never amended since its introduction in 2006 with commit
6c2b374d74 ("PCI-Express AER implemetation: AER core and aerdriver").
At the time, PCIe r1.1 was the latest spec revision.

Amend the macro with errors defined since then to avoid omitting the TLP
Header / Prefix Log for newer errors.

The order of the errors in AER_LOG_TLP_MASKS follows PCIe r1.1 sec 6.2.7
rather than 7.10.2, because only the former documents for which errors a
TLP Header / Prefix is logged.  Retain this order.  The section number is
still 6.2.7 in today's PCIe r7.0.

For Completion Timeouts, the TLP Header / Prefix is only logged if the
Completion Timeout Prefix / Header Log Capable bit is set in the AER
Capabilities and Control register.  Introduce a tlp_header_logged() helper
to check whether the TLP Header / Prefix Log is populated and use it in
the two places which currently match against AER_LOG_TLP_MASKS directly.

For Uncorrectable Internal Errors, logging of the TLP Header / Prefix is
optional per PCIe r7.0 sec 6.2.7.  If needed, drivers could indicate
through a flag whether devices are capable and tlp_header_logged() could
then check that flag.

pcitools introduced macros for newer errors with commit 144b0911cc0b
("ls-ecaps: extend decode support for more fields for AER CE and UE
status"):
  https://git.kernel.org/pub/scm/utils/pciutils/pciutils.git/commit/?id=144b0911cc0b

Unfortunately some of those macros are overly long:
  PCI_ERR_UNC_POISONED_TLP_EGRESS
  PCI_ERR_UNC_DMWR_REQ_EGRESS_BLOCKED
  PCI_ERR_UNC_IDE_CHECK
  PCI_ERR_UNC_MISR_IDE_TLP
  PCI_ERR_UNC_PCRC_CHECK
  PCI_ERR_UNC_TLP_XLAT_EGRESS_BLOCKED

This seems unsuitable for <linux/pci_regs.h>, so shorten to:
  PCI_ERR_UNC_POISON_BLK
  PCI_ERR_UNC_DMWR_BLK
  PCI_ERR_UNC_IDE_CHECK
  PCI_ERR_UNC_MISR_IDE
  PCI_ERR_UNC_PCRC_CHECK
  PCI_ERR_UNC_XLAT_BLK

Note that some of the existing macros in <linux/pci_regs.h> do not match
exactly with pcitools (e.g. PCI_ERR_UNC_SDES versus PCI_ERR_UNC_SURPDN),
so it does not seem mandatory for them to be identical.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/5f707caf1260bd8f15012bb032f7da9a9b898aba.1756712066.git.lukas@wunner.de
This commit is contained in:
Lukas Wunner 2025-09-01 09:44:52 +02:00 committed by Bjorn Helgaas
parent 6633875250
commit c8ab5e888b
2 changed files with 34 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
drivers/pci/pcie/aer.c
View File

@ -96,11 +96,21 @@ struct aer_info {
};
#define AER_LOG_TLP_MASKS (PCI_ERR_UNC_POISON_TLP| \
PCI_ERR_UNC_POISON_BLK | \
PCI_ERR_UNC_ECRC| \
PCI_ERR_UNC_UNSUP| \
PCI_ERR_UNC_COMP_ABORT| \
PCI_ERR_UNC_UNX_COMP| \
PCI_ERR_UNC_MALF_TLP)
PCI_ERR_UNC_ACSV | \
PCI_ERR_UNC_MCBTLP | \
PCI_ERR_UNC_ATOMEG | \
PCI_ERR_UNC_DMWR_BLK | \
PCI_ERR_UNC_XLAT_BLK | \
PCI_ERR_UNC_TLPPRE | \
PCI_ERR_UNC_MALF_TLP | \
PCI_ERR_UNC_IDE_CHECK | \
PCI_ERR_UNC_MISR_IDE | \
PCI_ERR_UNC_PCRC_CHECK)
#define SYSTEM_ERROR_INTR_ON_MESG_MASK (PCI_EXP_RTCTL_SECEE| \
PCI_EXP_RTCTL_SENFEE| \
@ -796,6 +806,20 @@ static int aer_ratelimit(struct pci_dev *dev, unsigned int severity)
}
}
static bool tlp_header_logged(u32 status, u32 capctl)
{
/* Errors for which a header is always logged (PCIe r7.0 sec 6.2.7) */
if (status & AER_LOG_TLP_MASKS)
return true;
/* Completion Timeout header is only logged on capable devices */
if (status & PCI_ERR_UNC_COMP_TIME &&
capctl & PCI_ERR_CAP_COMP_TIME_LOG)
return true;
return false;
}
static void __aer_print_error(struct pci_dev *dev, struct aer_err_info *info)
{
const char **strings;
@ -910,7 +934,7 @@ void pci_print_aer(struct pci_dev *dev, int aer_severity,
status = aer->uncor_status;
mask = aer->uncor_mask;
info.level = KERN_ERR;
tlp_header_valid = status & AER_LOG_TLP_MASKS;
tlp_header_valid = tlp_header_logged(status, aer->cap_control);
}
info.status = status;
@ -1401,7 +1425,7 @@ int aer_get_device_error_info(struct aer_err_info *info, int i)
pci_read_config_dword(dev, aer + PCI_ERR_CAP, &aercc);
info->first_error = PCI_ERR_CAP_FEP(aercc);
if (info->status & AER_LOG_TLP_MASKS) {
if (tlp_header_logged(info->status, aercc)) {
info->tlp_header_valid = 1;
pcie_read_tlp_log(dev, aer + PCI_ERR_HEADER_LOG,
aer + PCI_ERR_PREFIX_LOG,

7
include/uapi/linux/pci_regs.h
View File

@ -776,6 +776,12 @@
#define PCI_ERR_UNC_MCBTLP 0x00800000 /* MC blocked TLP */
#define PCI_ERR_UNC_ATOMEG 0x01000000 /* Atomic egress blocked */
#define PCI_ERR_UNC_TLPPRE 0x02000000 /* TLP prefix blocked */
#define PCI_ERR_UNC_POISON_BLK 0x04000000 /* Poisoned TLP Egress Blocked */
#define PCI_ERR_UNC_DMWR_BLK 0x08000000 /* DMWr Request Egress Blocked */
#define PCI_ERR_UNC_IDE_CHECK 0x10000000 /* IDE Check Failed */
#define PCI_ERR_UNC_MISR_IDE 0x20000000 /* Misrouted IDE TLP */
#define PCI_ERR_UNC_PCRC_CHECK 0x40000000 /* PCRC Check Failed */
#define PCI_ERR_UNC_XLAT_BLK 0x80000000 /* TLP Translation Egress Blocked */
#define PCI_ERR_UNCOR_MASK 0x08 /* Uncorrectable Error Mask */
/* Same bits as above */
#define PCI_ERR_UNCOR_SEVER 0x0c /* Uncorrectable Error Severity */
@ -798,6 +804,7 @@
#define PCI_ERR_CAP_ECRC_CHKC 0x00000080 /* ECRC Check Capable */
#define PCI_ERR_CAP_ECRC_CHKE 0x00000100 /* ECRC Check Enable */
#define PCI_ERR_CAP_PREFIX_LOG_PRESENT 0x00000800 /* TLP Prefix Log Present */
#define PCI_ERR_CAP_COMP_TIME_LOG 0x00001000 /* Completion Timeout Prefix/Header Log Capable */
#define PCI_ERR_CAP_TLP_LOG_FLIT 0x00040000 /* TLP was logged in Flit Mode */
#define PCI_ERR_CAP_TLP_LOG_SIZE 0x00f80000 /* Logged TLP Size (only in Flit mode) */
#define PCI_ERR_HEADER_LOG 0x1c /* Header Log Register (16 bytes) */