On 7/1/2025 8:18 PM, Alison Schofield wrote: > On Fri, Jun 27, 2025 at 12:22:39PM +0000, Shiju Jose wrote: >>> -----Original Message----- >>> From: Terry Bowman <terry.bowman@xxxxxxx> >>> Sent: 26 June 2025 23:43 >>> To: dave@xxxxxxxxxxxx; Jonathan Cameron <jonathan.cameron@xxxxxxxxxx>; >>> dave.jiang@xxxxxxxxx; alison.schofield@xxxxxxxxx; dan.j.williams@xxxxxxxxx; >>> bhelgaas@xxxxxxxxxx; Shiju Jose <shiju.jose@xxxxxxxxxx>; >>> ming.li@xxxxxxxxxxxx; Smita.KoralahalliChannabasappa@xxxxxxx; >>> rrichter@xxxxxxx; dan.carpenter@xxxxxxxxxx; >>> PradeepVineshReddy.Kodamati@xxxxxxx; lukas@xxxxxxxxx; >>> Benjamin.Cheatham@xxxxxxx; >>> sathyanarayanan.kuppuswamy@xxxxxxxxxxxxxxx; terry.bowman@xxxxxxx; >>> linux-cxl@xxxxxxxxxxxxxxx >>> Cc: linux-kernel@xxxxxxxxxxxxxxx; linux-pci@xxxxxxxxxxxxxxx >>> Subject: [PATCH v10 12/17] cxl/pci: Unify CXL trace logging for CXL Endpoints >>> and CXL Ports >>> > big snip - > >>> -); >>> - >>> TRACE_EVENT(cxl_aer_uncorrectable_error, >>> - TP_PROTO(const struct cxl_memdev *cxlmd, u32 status, u32 fe, u32 >>> *hl), >>> - TP_ARGS(cxlmd, status, fe, hl), >>> + TP_PROTO(struct device *dev, u64 serial, u32 status, u32 fe, >>> + u32 *hl), >>> + TP_ARGS(dev, serial, status, fe, hl), >>> TP_STRUCT__entry( >>> - __string(memdev, dev_name(&cxlmd->dev)) >>> - __string(host, dev_name(cxlmd->dev.parent)) >>> + __string(name, dev_name(dev)) >>> + __string(parent, dev_name(dev->parent)) >> Hi Terry, >> >> Thanks for considering the feedback given in v9 regarding the compatibility issue >> with the rasdaemon. >> https://lore.kernel.org/all/959acc682e6e4b52ac0283b37ee21026@xxxxxxxxxx/ >> >> Probably some confusion w.r.t the feedback. >> Unfortunately TP_printk(...) is not an ABI that we need to keep stable, >> it's this structure, TP_STRUCT__entry(..) , that matters to the rasdaemon. >> > I'm not so sure you should be letting him off the hook for TP_printk ;) > It seems TP_printk should be kept aligned w TP_STRUCT_entry(). As a > user who often looks at TP_printk output, I'd say keep them all in > sync, and consider them ABI - ie. add to but don't modify. > > I agree. I will keep TP_printk and TP_STRUCT aligned by using 'memdev' and 'host'. The only change here will be TP_PROTO and will be: TP_PROTO(struct device *dev, u64 serial, u32 status), Let me know if that's not ok. -Terry >>> __field(u64, serial) >>> __field(u32, status) >>> __field(u32, first_error) >>> __array(u32, header_log, CXL_HEADERLOG_SIZE_U32) >>> ), >>> TP_fast_assign( >>> - __assign_str(memdev); >>> - __assign_str(host); >>> - __entry->serial = cxlmd->cxlds->serial; >>> + __assign_str(name); >>> + __assign_str(parent); >>> + __entry->serial = serial; >>> __entry->status = status; >>> __entry->first_error = fe; >>> /* >>> @@ -99,8 +72,8 @@ TRACE_EVENT(cxl_aer_uncorrectable_error, >>> */ >>> memcpy(__entry->header_log, hl, CXL_HEADERLOG_SIZE); >>> ), >>> - TP_printk("memdev=%s host=%s serial=%lld: status: '%s' first_error: >>> '%s'", >>> - __get_str(memdev), __get_str(host), __entry->serial, >>> + TP_printk("memdev=%s host=%s serial=%lld status='%s' >>> first_error='%s'", >>> + __get_str(name), __get_str(parent), __entry->serial, >>> show_uc_errs(__entry->status), >>> show_uc_errs(__entry->first_error) >>> ) > snip >
