wendyshi
Tue Oct 28 12:49:36 CST 2003
------=_NextPart_0001_DC732527
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Phil is right. We are backporting ATA_PASS_THROUGH to XP and 2000. For XP,
the complete fix will be available in XP SP2 (but I am not sure when SP2
will be ready). An intermediate hotfix will be available in a KB article in
about a month or two. I can post a message here once it's ready. For 2000,
it's still in progress.
This posting is provided "AS IS" with no warranties, and confers no rights.
--------------------
NNTP-Posting-Date: Fri, 24 Oct 2003 13:09:23 -0500
Reply-To: "Phil Barila" <PBarila@Seagate.com>
From: "Phil Barila" <PBarila@Barila.com>
Newsgroups: microsoft.public.development.device.drivers
References: <eec5VPYmDHA.2312@TK2MSFTNGP12.phx.gbl>
<OKAk4KlmDHA.2772@TK2MSFTNGP10.phx.gbl>
<u2VFTSlmDHA.2592@TK2MSFTNGP10.phx.gbl>
Subject: Re: How to find port resources used by hard drives?
Date: Fri, 24 Oct 2003 12:09:26 -0600
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1158
X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Message-ID: <ssGdnfHUVpJO9wSiRVn-tw@4dv.net>
Lines: 58
NNTP-Posting-Host: 192.55.20.36
X-Trace:
sv3-TxTVAqvpY0jEPpbHk7ofWe/cOq+Lr2xSe6scVUHClITk5MW/cvcsg5F5bQwxNOr9OdZP3TSu
kbu45X4!lOOkKqIeMBYbbIut3nrdplUiTJjaN/NrY/YUxnijOYS3y0+ykWDpdvCvH/cHmSSkwGr+
Aw==
X-Complaints-To: abuse@4dv.net
X-DMCA-Complaints-To: abuse@4dv.net
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your
complaint properly
X-Postfilter: 1.1
Path:
cpmsftngxa06.phx.gbl!TK2MSFTNGP08.phx.gbl!newsfeed00.sul.t-online.de!t-onlin
e.de!news-lei1.dfn.de!news-fra1.dfn.de!npeer.de.kpn-eurorings.net!small1.nnt
p.aus1.giganews.com!border1.nntp.aus1.giganews.com!intern1.nntp.aus1.giganew
s.com!nntp.giganews.com!nntp.4dv.net!news.4dv.net.POSTED!not-for-mail
Xref: cpmsftngxa06.phx.gbl microsoft.public.development.device.drivers:32965
X-Tomcat-NG: microsoft.public.development.device.drivers
"Josh Dahlby" <n/a> wrote in message
news:u2VFTSlmDHA.2592@TK2MSFTNGP10.phx.gbl...
> I don't need to support RAID drives, so I suppose I should refine my
> statement to being any drive exposing the ATA task file registers. I need
> to support the standard IDE controller, serial ATA, and probably a few
> specific IDE add-in cards.
>
> I have worked out how to issue ATA commands to the drive without
interfering
> with the real drivers use of it. I've run millions of iterations of the
> command I need while dong things like defragging, so I'm pretty sure I've
> got it right. Well, for single processor systems anyway, which is all I
> need to support at this time. Anyway, I don't have a choice. Because
there
> is no way to issue 48 bit ATA commands on XP and 2000, I have to write
> directly to the registers. If Microsoft would backport
> IOCTL_ATA_PASS_THROUGH to XP and 2000, then I wouldn't even have to write
a
> driver, but that doesn't seem likely.
What *are* you doing? I know a way to write the device registers without
getting the ATA driver confused, but it's a hack that's not fit for
distribution outside of a lab. And if I ever find out what product you are
working on, I will ensure that nobody I care about buys it until I know you
aren't doing that anymore.
What 48 bit command are you trying to issue that isn't already covered by
the 48 bit support in 2000 SP3 & 4 and XP SP1? You do know that full 48 bit
support is available today, right? You do also know that a FLUSH CACHE is
the same as FLUSH CACHE EXT, except that you won't get a valid error output
if it fails to flush an LBA that is more than 28 bits? Are you playing with
the HPA?
If you still need ATA_PASS_THROUGH after all that, XP is done, and you can
ask the Support group for the hotfix for the cost of the phone call.
Windows 2000 is in process.
> Certainly if you have suggestions on how to issue a 48 bit ATA command to
> the drive without doing it myself, I'd be interested to find out how. I
> really don't want to write this driver as it adds quite a bit of
complexity
> to this application, but there doesn't seem to be any way around it.
Again, what 48 bit command do you need beyond READ & WRITE? Thsoe can be
handled by opening the device for R/W and issuing Read/WriteFile() with
offset and length that are integer multiples of the disk sector size. Since
you are talking about ATA drives, for the time being, that's precisely 512
bytes.
Phil
--
Philip D. Barila Windows DDK MVP
Seagate Technology, LLC
(720) 684-1842
As if I need to say it: Not speaking for Seagate.
E-mail address is pointed at a domain squatter. Use reply-to instead.
------=_NextPart_0001_DC732527
Content-Type: text/x-rtf
Content-Transfer-Encoding: 7bit
{\rtf1\ansi\ansicpg1252\deff0\deflang1033{\fonttbl{\f0\fnil\fprq2\fcharset0 MS Sans Serif;}}
\viewkind4\uc1\pard\f0\fs20 Phil is right. We are backporting ATA_PASS_THROUGH to XP and 2000. For XP, the complete fix will be available in XP SP2 (but I am not sure when SP2 will be ready). An intermediate hotfix will be available in a KB article in about a month or two. I can post a message here once it's ready. For 2000, it's still in progress.
\par
\par
\par This posting is provided "AS IS" with no warranties, and confers no rights.
\par
\par
\par
\par \pard\li720 --------------------
\par NNTP-Posting-Date: Fri, 24 Oct 2003 13:09:23 -0500
\par Reply-To: "Phil Barila" <PBarila@Seagate.com>
\par From: "Phil Barila" <PBarila@Barila.com>
\par Newsgroups: microsoft.public.development.device.drivers
\par References: <eec5VPYmDHA.2312@TK2MSFTNGP12.phx.gbl> <OKAk4KlmDHA.2772@TK2MSFTNGP10.phx.gbl> <u2VFTSlmDHA.2592@TK2MSFTNGP10.phx.gbl>
\par Subject: Re: How to find port resources used by hard drives?
\par Date: Fri, 24 Oct 2003 12:09:26 -0600
\par X-Priority: 3
\par X-MSMail-Priority: Normal
\par X-Newsreader: Microsoft Outlook Express 6.00.2800.1158
\par X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
\par Message-ID: <ssGdnfHUVpJO9wSiRVn-tw@4dv.net>
\par Lines: 58
\par NNTP-Posting-Host: 192.55.20.36
\par X-Trace: sv3-TxTVAqvpY0jEPpbHk7ofWe/cOq+Lr2xSe6scVUHClITk5MW/cvcsg5F5bQwxNOr9OdZP3TSukbu45X4!lOOkKqIeMBYbbIut3nrdplUiTJjaN/NrY/YUxnijOYS3y0+ykWDpdvCvH/cHmSSkwGr+Aw==
\par X-Complaints-To: abuse@4dv.net
\par X-DMCA-Complaints-To: abuse@4dv.net
\par X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
\par X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
\par X-Postfilter: 1.1
\par Path: cpmsftngxa06.phx.gbl!TK2MSFTNGP08.phx.gbl!newsfeed00.sul.t-online.de!t-online.de!news-lei1.dfn.de!news-fra1.dfn.de!npeer.de.kpn-eurorings.net!small1.nntp.aus1.giganews.com!border1.nntp.aus1.giganews.com!intern1.nntp.aus1.giganews.com!nntp.giganews.com!nntp.4dv.net!news.4dv.net.POSTED!not-for-mail
\par Xref: cpmsftngxa06.phx.gbl microsoft.public.development.device.drivers:32965
\par X-Tomcat-NG: microsoft.public.development.device.drivers
\par
\par "Josh Dahlby" <n/a> wrote in message
\par news:u2VFTSlmDHA.2592@TK2MSFTNGP10.phx.gbl...
\par > I don't need to support RAID drives, so I suppose I should refine my
\par > statement to being any drive exposing the ATA task file registers. I need
\par > to support the standard IDE controller, serial ATA, and probably a few
\par > specific IDE add-in cards.
\par >
\par > I have worked out how to issue ATA commands to the drive without
\par interfering
\par > with the real drivers use of it. I've run millions of iterations of the
\par > command I need while dong things like defragging, so I'm pretty sure I've
\par > got it right. Well, for single processor systems anyway, which is all I
\par > need to support at this time. Anyway, I don't have a choice. Because
\par there
\par > is no way to issue 48 bit ATA commands on XP and 2000, I have to write
\par > directly to the registers. If Microsoft would backport
\par > IOCTL_ATA_PASS_THROUGH to XP and 2000, then I wouldn't even have to write
\par a
\par > driver, but that doesn't seem likely.
\par
\par What *are* you doing? I know a way to write the device registers without
\par getting the ATA driver confused, but it's a hack that's not fit for
\par distribution outside of a lab. And if I ever find out what product you are
\par working on, I will ensure that nobody I care about buys it until I know you
\par aren't doing that anymore.
\par
\par What 48 bit command are you trying to issue that isn't already covered by
\par the 48 bit support in 2000 SP3 & 4 and XP SP1? You do know that full 48 bit
\par support is available today, right? You do also know that a FLUSH CACHE is
\par the same as FLUSH CACHE EXT, except that you won't get a valid error output
\par if it fails to flush an LBA that is more than 28 bits? Are you playing with
\par the HPA?
\par
\par If you still need ATA_PASS_THROUGH after all that, XP is done, and you can
\par ask the Support group for the hotfix for the cost of the phone call.
\par Windows 2000 is in process.
\par
\par > Certainly if you have suggestions on how to issue a 48 bit ATA command to
\par > the drive without doing it myself, I'd be interested to find out how. I
\par > really don't want to write this driver as it adds quite a bit of
\par complexity
\par > to this application, but there doesn't seem to be any way around it.
\par
\par Again, what 48 bit command do you need beyond READ & WRITE? Thsoe can be
\par handled by opening the device for R/W and issuing Read/WriteFile() with
\par offset and length that are integer multiples of the disk sector size. Since
\par you are talking about ATA drives, for the time being, that's precisely 512
\par bytes.
\par
\par Phil
\par --
\par Philip D. Barila Windows DDK MVP
\par Seagate Technology, LLC
\par (720) 684-1842
\par As if I need to say it: Not speaking for Seagate.
\par E-mail address is pointed at a domain squatter. Use reply-to instead.
\par
\par
\par \pard
\par
\par }
------=_NextPart_0001_DC732527--