Hi Jan and Cyril,
Thanks for your question and suggestion;)
What I mean is that:
1. errno will be set to negative when someone function run failed, but it will
not be set back to 0 when someone function return 0.
2. Usually we use *errno* after someone function(Eg: read()) return negative.
But if we want to use errno after read() return 0, we should set errno to 0
before call read(), or if read() return >=0, the errno maybe keep using last
failed number which setted by someone earlier function than read().
Acccording to above, I checked all code in testcases/kernel/io/direct_io/, make
sure all errno be setted to 0 before use errno in *if return success* condition.
If you feel this patch is OK, I will send another email with clear commit
message which you suggested.
Thank you,
Zorro Lang
----- 原始邮件 -----
> 发件人: "Jan Stancek" <jstan...@redhat.com>
> 收件人: "Cyril Hrubis" <chru...@suse.cz>
> 抄送: "Zorro Lang" <zl...@redhat.com>, ltp-list@lists.sourceforge.net
> 发送时间: 星期一, 2014年 12 月 08日 下午 8:03:13
> 主题: Re: [LTP] [PATCH] diotest: set errno with zero before test with zero
>
>
>
>
>
> ----- Original Message -----
> > From: "Cyril Hrubis" <chru...@suse.cz>
> > To: "Zorro Lang" <zl...@redhat.com>
> > Cc: ltp-list@lists.sourceforge.net
> > Sent: Monday, 8 December, 2014 12:53:50 PM
> > Subject: Re: [LTP] [PATCH] diotest: set errno with zero before test with
> > zero
> >
> > Hi!
> > > ret = read(fd, buf, count);
> > > if (ret >= 0 || errno != errnum) {
> > > tst_resm(TFAIL, "read allows %s. returns %d: %s", msg, ret,
> > > strerror(errno));
> > > l_fail = TRUE;
> >
> > As far as I can see this code tests that read() has returned negative
> > value and in this case errno must be set.
> >
> > I.e.
> >
> > if ret >= 0 -> read hasn't failed -> FAILURE
> >
> > The errno != errnum is not evaluated unless the first condition has
> > failed (that means ret < 0) and in this case errno must be set.
> >
> > The condition is equivalent to !(ret < 0 && errno == errnum)
> >
> > Or am I missing something?
>
> I think he refers to output of tst_resm being possibly misleading
> in case ret >= 0, that call is using strerror(errno).
>
> Regards,
> Jan
>
> >
> > --
> > Cyril Hrubis
> > chru...@suse.cz
> >
> > ------------------------------------------------------------------------------
> > Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> > from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> > with Interactivity, Sharing, Native Excel Exports, App Integration & more
> > Get technology previously reserved for billion-dollar corporations, FREE
> > http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
> > _______________________________________________
> > Ltp-list mailing list
> > Ltp-list@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/ltp-list
> >
>
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list
