Date: Wed, 29 Sep 1999 21:42:27 +0200 (MET DST) From: Gerard Roudier <groudier@club-internet.fr> To: Kevin Street <street@iname.com> Cc: "Kenneth D. Merry" <ken@kdm.org>, scsi@FreeBSD.ORG Subject: Re: sym driver 0.3.0 Message-ID: <Pine.LNX.3.95.990929211758.385A-100000@localhost> In-Reply-To: <14321.29414.266150.585956@mired.eh.local>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 28 Sep 1999, Kevin Street wrote: > Is it cam or the ncr driver which reduces the tags to the correct value? > It would definitely be useful if the sym driver could dynamically > reduce the tags for devices that can't handle 64 openings. This had been a misunderstanding from me that is fixed in SYM.0.4.0. I have uploaded the patch yesterday. The problem appears to me when rereading the source code. There were no risk for your data. =20 > The only other parameters that are different are the Read/Write Retry > Counts. They start at 34/28 on the ncr and are increased to 75 when > the dump runs. The sym driver has them set to 75 at boot, so these > end up being the same. Basically, given slow devices and comparable configurations, no significant differences in performances should be observed between the sym and the ncr driver. By the way, I am much more interested in robustness and reliability testings that ncr .vs. sym benchmarks. I have no doubts about the performances of the sym driver that are guaranteed by design and are known for the linux sym53c8xx driver that is based on the same design regarding performances issues.=20 About benchmarks, comparisons with other HAs from other vendors will be welcome, especially using high end hard disks and Adaptec equivalent adapters. The README.sym file contains a summary of the sym driver features. Performances is probably the one people on this list are most interested in, but some others as the support the all NVRAM formats is a major feature for end-users and extremely user-friendly. You may also consider that the sym driver supports _full_ SPI2 minus linked commands (that IIRC are obsoleted by SPI3), which is pretty rare on the planet (some SPI2 features are still to be tested). > Thanks for the help. Thanks to you, too. G=E9rard. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.3.95.990929211758.385A-100000>