|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [PATCH 5/5] Replace RegisterEventSource with TraceLoggingRegister
On 09/02/2026 15:03, Owen Smith wrote: >> Could you add some comments to explain the potential issue with >> RegisterEventSource? > > RegisterEventSource is marked deprecated - CodeQL detected this issue > https://learn.microsoft.com/en-us/windows-hardware/drivers/devtest/28735-banned-crimson-api-usage > I will add a comment to this effect. > >> I think it'd be great if you could provide similar patches for >> xencons_monitor and xenagent as well. > > Similar patches will be created for the other user mode components. > >>> --- >>> +TRACELOGGING_DEFINE_PROVIDER(g_xenbus_monitor, >> >> >> The provider handle name should use the same convention as other global >> variables (e.g. MonitorTraceLoggingProvider). >> >>> + MONITOR_NAME, >>> + //{E8ADEEB7-6DD1-447E-A49B-A6571CC74039} >>> + (0xe8adeeb7, 0x6dd1, 0x447e, 0xa4, 0x9b, >>> 0xa6, 0x57, 0x1c, 0xc7, 0x40, 0x39) >> >> Is it worth having a DEFINE_GUID for this somewhere? > > This is how the macro is documented - I'd assume the macro will internally > define the guid correctly, > using DEFINE_GUID will cause issues with the trace macro's expansion. > I see, thanks. Was the GUID generated from MONITOR_NAME using the code given in TraceLoggingProvider.h or was it randomly generated? Also, a codestyle nit: the parentheses of function calls are attached to the last argument: ..., MONITOR_NAME, //{E8ADEEB7-6DD1-447E-A49B-A6571CC74039} (0xe8adeeb7, 0x6dd1, 0x447e, 0xa4, 0x9b, 0xa6, 0x57, 0x1c, 0xc7, 0x40, 0x39)); (and in the TraceLoggingWrite calls elsewhere) >>> + case LOG_INFO: >>> + TraceLoggingWrite(g_xenbus_monitor, >>> + "Information", >>> + TraceLoggingLevel(WINEVENT_LEVEL_INFO), >>> + TraceLoggingString(Buffer, "Info") >> >> Buffer is a PTSTR, so we should have an appropriate #define for >> TraceLoggingString. > > >>> - Log("[%u]: %s [%s]", >>> + LogInfo("[%u]: %s [%s]", >>> SessionId, >>> Name, >>> WTSStateName(State)); >> >> Could you align the arguments here (and elsewhere)? > > Sure > > > > Owen > > -- | Vates XCP-ng & Xen Orchestra - Vates solutions web: https://vates.tech
|
![]() |
Lists.xenproject.org is hosted with RackSpace, monitoring our |