Date: Fri, 2 Nov 2007 13:15:51 +0300 From: "Artem Kuchin" <matrix@itlegion.ru> To: "Pawel Jakub Dawidek" <pjd@FreeBSD.org>, "Dag-Erling Sm??rgrav" <des@des.no> Cc: freebsd-current@freebsd.org Subject: Re: BIO_FLUSH on twe driver. Why is it not there? Message-ID: <038501c81d3b$6d6460b0$0c00a8c0@Artem> References: <0fbb01c81be1$37e698f0$0c00a8c0@Artem> <863avptzir.fsf@ds4.des.no> <20071102095506.GC24455@garage.freebsd.pl>
next in thread | previous in thread | raw e-mail | index | archive | help
>> > However, twe is working via scsi subsystem and the authour of gjournal >> > said somewhere that he has had implemeneted BIO_FLISH for scsi and he >> > specifically mentioned that he has tested twe and twa and they both >> > support BIO_FLUSH. >> >> twa(4) uses CAM, but twe(4) doesn't (and never has). > >That was my mistake. Someone told me (I think it was jhb@), that twe(4) >turns off write cache automatically when there is no battery, so >BIO_FLUSH isn't really needed and if there is a battery, it also isn't >really needed. It would be nice to hide the warning in this case >somehow, but for now you should be safe by simply ignoring the warning. Um.. If battry is not installed write can be enabled easily. Here is my setting: Unit Write Cache (Controller ID 0) Unit 2 [RAID 1] Enabled Is your phrase "it also isn't really needed' related to this case too? 'Cause you said if BIO_FLUSH is not there and write cache is bigger than gjorunal's cache there could be trouble. Cache on my controller is 64MB, what;s the size of gjournal's cache? -- Artem
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?038501c81d3b$6d6460b0$0c00a8c0>