From mboxrd@z Thu Jan 1 00:00:00 1970 From: Khiem Nguyen Date: Fri, 12 Sep 2014 00:39:59 +0000 Subject: Re: [PATCH 0/5] arm: shmobile: Lager Add USB and SATA support Message-Id: <541240DF.6020604@renesas.com> List-Id: References: <1387370660-20481-1-git-send-email-valentine.barshak@cogentembedded.com> In-Reply-To: <1387370660-20481-1-git-send-email-valentine.barshak@cogentembedded.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: linux-sh@vger.kernel.org Hi Hiep-san, Thanks for your update. On 9/11/2014 8:26 PM, カオ　ミン　ヒェップ wrote: > Hi Khiem-san > >> It seems SATA performance (write speed) is degraded after CPUFreq is enabled >> (even performance governor is selected). >> >> Base on your test result so far: >> T1 Upstream-v3.16-rc5: 66 MB/s >> T2 v3.17-rc2 (CPUFreq disabled): ~ 66 MB/s >> T3 v3.17-rc2 (CPUFreq enable / on-demand governor): 42.61MB/s >> T4 v3.17-rc2 (CPUFreq disabled / performance governor ): 58.3MB/s >> >> The diff between T2 and T4 is ~8 MB/s (or 8% of 66 MB/s). >> I guess we should try to find the cause of this regression. >> (I wonder similar phenomenon occurs in backport version ...) [snip] > Have some miss in my way of the report yesterday. > As yesterday results, T2 v3.17-rc2 (CPUFreq disabled) should be (367MB/(5.3s+1.0s)) ->58.2MB/s [snip] > And I have just re-confirmed T1 (Upstream-v3.16-rc5), > On upstream-v3.16-rc5 SATA's writing speed should be 55.6MB/s as below. > There is no uniformity in the speed calculation of SATA in our test team. > That 66MB/s only is result of copy to SATA-HDD without umount the device before finish. > Sorry for the inconvenience. [snip] It makes sense now. SATA performance does not degrade after CPUFreq is enabled (select performance governor). >> T1 Upstream-v3.16-rc5 (CPUFreq disabled by default): -> 55.6MB/s >> T2 v3.17-rc2 (CPUFreq disabled) : -> 58.2MB/s >> T3 v3.17-rc2 (CPUFreq enable / on-demand governor) : 42.61MB/s >> T4 v3.17-rc2 (CPUFreq disabled / performance governor ): 58.3MB/s And we should take note SATA performance when CPUfreq/on-demand governor is selected. (perhaps, CPUFreq parameters should be tuned) Thanks for your testing, Hiep-san. >>> >>> Thank you! >>> Cao Minh Hiep. -- Best regards, KHIEM Nguyen