From mboxrd@z Thu Jan 1 00:00:00 1970 Subject: Re: [edk2-devel] [PATCH v1 1/1] UefiCpuPkg: Remove PEI/DXE instances of CpuTimerLib. To: Zeng, Star ,devel@edk2.groups.io From: "Jason Lou" X-Originating-Location: CN (192.102.204.55) X-Originating-Platform: Windows Chrome 89 User-Agent: GROUPS.IO Web Poster MIME-Version: 1.0 Date: Thu, 25 Mar 2021 00:58:00 -0700 References: In-Reply-To: Message-ID: <20843.1616659080240333148@groups.io> Content-Type: multipart/alternative; boundary="1i0vf4CBg9MaUTZJsvgH" --1i0vf4CBg9MaUTZJsvgH Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hi Star, sorry for my late response. Yes, HOB search time is related to the sequence of HOB generation. Some data collected on the test platform: 1. The time range required to search for the specified HOB: 20ns ~ 2700ns Search for the 1st HOB(PHIT HOB): 20ns Search for the last HOB: about 2700ns 2. The time required to search for CpuCrystalFrequencyHob HOB: about 2000n= s. The actual delay of NanoSecondDelay function is longer than the expected d= elay. If any questions, please feel free let me know, thanks Jason Lou --1i0vf4CBg9MaUTZJsvgH Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hi Star, sorry for my late response.

Yes, HOB search time is rel= ated to the sequence of HOB generation.

Some data collected on the test platform:
1. The time range required to search for the specified HOB: 20ns ~ 27= 00ns
     Search for the 1st HOB(PHIT HOB): 20ns
     Search for the last HOB: about 2700ns
2. The time required to search for CpuCrystalFrequencyHob HOB: about = 2000ns.

The actual delay of NanoSecondDelay function is longer t= han the expected delay.

If any questions, please feel free let m= e know, thanks

Jason Lou

--1i0vf4CBg9MaUTZJsvgH--