From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chen Gang Subject: Re: =?UTF-8?B?5Zue5aSN77yaIFtQQVRDSCBsaW51eC1uZXh0XSBuZXQvZGNjcC8=?= =?UTF-8?B?dGltZXIuYzogdXNlICd1NjQnIGluc3RlYWQgb2YgJ3M2NCcgdG8gYXZvaWQgY28=?= =?UTF-8?B?bXBpbGVyJ3Mgd2FybmluZw==?= Date: Thu, 22 May 2014 09:01:39 +0800 Message-ID: <537D4C73.40901@gmail.com> References: <1546102096.46113.1400718399706.JavaMail.root@bj-mail03.pku.edu.cn> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: gerrit@erg.abdn.ac.uk, Guan Xuetao , David Miller , dccp@vger.kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: =?UTF-8?B?566h6Zuq5rab?= Return-path: In-Reply-To: <1546102096.46113.1400718399706.JavaMail.root@bj-mail03.pku.edu.cn> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On 05/22/2014 08:26 AM, =E7=AE=A1=E9=9B=AA=E6=B6=9B wrote: >=20 > ----- Chen Gang =E5=86=99=E9=81=93=EF=BC=9A >> 'dccp_timestamp_seed' is initialized once by ktime_get_real() in >> dccp_timestamping_init(). It is always less than ktime_get_real() >> in dccp_timestamp(). >> >> Then, ktime_us_delta() in dccp_timestamp() will always return positi= ve >> number. So can use manual type cast to let compiler and do_div() kno= w >> about it to avoid warning. >> >> The related warning (with allmodconfig under unicore32): >> >> CC [M] net/dccp/timer.o >> net/dccp/timer.c: In function =E2=80=98dccp_timestamp=E2=80=99: >> net/dccp/timer.c:285: warning: comparison of distinct pointer type= s lacks a cast >> >> >> Signed-off-by: Chen Gang >> --- >> net/dccp/timer.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/net/dccp/timer.c b/net/dccp/timer.c >> index 16f0b22..1cd46a3 100644 >> --- a/net/dccp/timer.c >> +++ b/net/dccp/timer.c >> @@ -280,7 +280,7 @@ static ktime_t dccp_timestamp_seed; >> */ >> u32 dccp_timestamp(void) >> { >> - s64 delta =3D ktime_us_delta(ktime_get_real(), dccp_timestamp_seed= ); >> + u64 delta =3D (u64)ktime_us_delta(ktime_get_real(), dccp_timestamp= _seed); >=20 > Do you assume that delta should be very small? > Otherwise, return value will be different if data type is changed. >=20 'u64' is a very very large number. after calculation, if it is based on nano second (although I am not quite sure whether it is based on it). a hour, 3,600,000,000,000ns a day, 90,000,000,000,000ns a year, 50,000,000,000,000,000ns 10 years, 500,000,000,000,000,000ns 100 years, 5,000,000,000,000,000,000ns 4G * 4G =3D 16,000,000,000,000,000,000ns So we can assume it will never overflow for 'u64'. Thanks. --=20 Chen Gang Open, share, and attitude like air, water, and life which God blessed