Once upon a time, Arjan van de Ven arjanv@redhat.com said:
On Sat, 2004-04-10 at 21:57, Jurgen Botz wrote:
In the last few kernel RPMS there doesn't seem to be an sg module. It was there in ./2.6.4-1.300, it's missing in 305. What happened to it?
it's deprecated
How do I control my tape library robot? AFAIK that is only controlled via sg (the other SCSI modules ignore tape robots).
On Mon, Apr 12, 2004 at 03:40:32PM -0500, Chris Adams wrote:
module. It was there in ./2.6.4-1.300, it's missing in 305. What happened to it?
it's deprecated
How do I control my tape library robot? AFAIK that is only controlled via sg (the other SCSI modules ignore tape robots).
Use 2.4 or rebuild the kernel. Ditto with ide-scsi, both are needed for real world hardware. Oh and file a bug, and if it closed re-open it until you get a working answer.
On 2004.04.12 16:49, Alan Cox wrote:
On Mon, Apr 12, 2004 at 03:40:32PM -0500, Chris Adams wrote:
module. It was there in ./2.6.4-1.300, it's missing in 305. What happened to it?
it's deprecated
How do I control my tape library robot? AFAIK that is only controlled via sg (the other SCSI modules ignore tape robots).
Use 2.4 or rebuild the kernel. Ditto with ide-scsi, both are needed for real world hardware. Oh and file a bug, and if it closed re-open it until you get a working answer.
FC2T3 is still missing the ide-scsi module! And filing a bug doesn't work :-( I've had one in bugzilla since February 28, and no reaction whatsoever from the Red Hat kernel packagers...
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117095
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it? Put it in unsupported for all I care, but compile it, _please_.
Willem Riede.
On Sun, May 02, 2004 at 05:32:43PM -0400, Willem Riede wrote:
FC2T3 is still missing the ide-scsi module! And filing a bug doesn't work :-( I've had one in bugzilla since February 28, and no reaction whatsoever from the Red Hat kernel packagers...
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117095
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it?
And IDE multi-changers
Put it in unsupported for all I care, but compile it, _please_.
Unfortunately Arjan and Dave are out of range of my torture kit so all I can do is bitch at them as well (and if need be release rival kernel rpms which would be a pita to have to do)
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
About what's being discussed, I just ponder that failing to work correctly with common devices (usb-storage, etc) is a fair good way of killing a distribution. Ok, that's a will to make everything work with certain device drivers (like make scsi disks work with sr_mod (/dev/sr*) and (/dev/scd*)) but it is also very important to have in mind that, until it is done and working fine, the "deprecated" devices and modules and functions and whatsoever must still work in good shape.
The mainstream kernel 2.6.6-rc3-bk3 works with /dev/sg*. There are bugs, but the user is still able to deal with them and get job done. The fact that mainstream and FC2T3 differs in matters of functionality raises the fear that such incompatibilities grow bigger in the future and portability and interoperability among distributions (FC, SuSe, Conectiva, Slackware, etc...) become a subject of concern. That would be bloody sucking.
Regards,
Casimiro Alan Cox Public key for 0x68FA52A33F440F2D:
| On Sun, May 02, 2004 at 05:32:43PM -0400, Willem Riede wrote: | |> FC2T3 is still missing the ide-scsi module! And filing a bug |> doesn't work :-( I've had one in bugzilla since February 28, and |> no reaction whatsoever from the Red Hat kernel packagers... |> |> https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117095 |> |> Why not compile ide-scsi? Why inconvenience owners of tape drives |> that need it? | | | And IDE multi-changers | |> Put it in unsupported for all I care, but compile it, _please_. | | | Unfortunately Arjan and Dave are out of range of my torture kit so | all I can do is bitch at them as well (and if need be release rival | kernel rpms which would be a pita to have to do) | | Public key for 0x68FA52A33F440F2D - -----BEGIN PGP PUBLIC KEY BLOCK----- Version: GnuPG v1.2.4 (GNU/Linux)
mQGiBEAQK2cRBAC5dUcZbPeatM8tsmhIdfSOq2OMwXNM4eU4m7gARXml0iyaV3hs vNN9CP8JMdkuB2CohAasxf8zZ28OmSKG9QuUj5PNbRPk/HSj/yKzPB6yedHBNssq ajjJN19hoOB/YNCrXPecnUjJnTvQfgbC0CTNzM3zXfn8KNDpSa5kgZ6oZwCg3ZNx eKpHbKAwkkasK1nLW3r5M80D/iY0eB1BqIsL2vCFufb8W2wIX/NEBxLa5vf1uxI9 aFXTmGeUuRHvIQHlAiJDdXi/MP1tf/pQdM8nDIxd5BegfkaTDkSJf4m+K13wsEoM sDHx1uiQduQytSUjLcqVN7CueNWNrC7azaBHzFQgbpM0xL4rY/h4DvQg+X3h4qVg etToA/47CkSeltJALUom+17PEa3DP0fkIuSQu6H6YKbkNkfUFup2N2kSAThqYf5d aXjvehsN8GsgQAjOWyk3W+FbD+s/DQkLydiRL9STLXjY9NalkfVIhrn2a4djVYjW pakHC+x0RiV23nlRWzQ73QljyKDlSWsVB3KdijuOeo3Yb5XjhrRDQ2FzaW1pcm8g ZGUgQWxtZWlkYSBCYXJyZXRvIChQZXNzb2FsKSA8Y2FzaW1pcm9fYmFycmV0b0B1 b2wuY29tLmJyPohfBBMRAgAfBQJAECtnBQkAnjQABAsHAwIDFQIDAxYCAQIeAQIX gAAKCRBo+lKjP0QPLUMwAJ0fNhBH5wWGVappX8wdB8tuwkVSkACgqL/CG+84UgZF ATFBG7d2wkjVSNe5Ag0EQBArghAIAOwWVTsFepybl0mCSHoOF0vlcB0IlCywvBo3 nYLG6zUo8IUQpXg5R8YxBfXvsSpSK83QQElZmyB25gQm1BvEpdUMGs+VmfBddqfx SH69k0W/rWuwQ90Auk3mQDk38CB/rT/Hem3vZkiGoT8vPxWuCO5UJm+Y36Gg6tJr JdELhtAAC8aDpFx+T5Ye0lTh00F2dca8cR+gf7S/lZseWutvZGvziqTEvC8cTzmb af2x23o/RwAsy5wz1ljxh6iF59YeJsclNsQ0tBiRnXphqzcOPBorVUIF4EHS9sb+ tt7FhkhkSSDm2A9/veoIZmy81P2IniwAta1bEsz0EQsUSExhyiMAAwUIAIRADBcW DskFNLYrY4KRNmL192pOgRH6BQEw8n+mrn2uuRGCR6ZnoDm3v5/nbRLMBwHZQxLP EciYkPG9gTx/I74mQcgitqCCYiI2UzTaz/BbAMCIqrWyzGYav9sjaJtd0ko2eOZx TU53pw50dwV9bW+xnh9a1x8x1Qoj1bnOyhESOJ4THGFhnjeMON+DhU2N+eyax9Md jRUtxOJB6z7RVjELn6islKpEzH/1wKrqOFkL7BEr+3m2mjm8sBBs1gC3u+sNbgGB Ebv5sGBsMX4+LbBIh4aw+fanzHZhVSAjSgVl7RH7bgF6YqAKAfKB1EpRwTW3FTqH LMeUz8mQtbt8IpCITAQYEQIADAUCQBArggUJAJ40AAAKCRBo+lKjP0QPLZL/AJwJ OSAyi5Zi8OmeL7iuNeHAeY2AUwCgn9LHEbusFl2KXWHnlFHya8WTyGE= =aaqn - -----END PGP PUBLIC KEY BLOCK-----
On Mon, 2004-05-03 at 01:22, Casimiro de Almeida Barreto wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
About what's being discussed, I just ponder that failing to work correctly with common devices (usb-storage, etc) is a fair good way of killing a distribution.
the usb-storage bug is fixed in kernel 349 (see http://people.redhat.com). Actually it's worked around. cdrecord has a bug where it doesn't check a return value and interprets an error value as 255 sectors-io-size. Which is/was invalid for usb storage. SG just silently ignores the fact that that is an invalid value for usb storage and continues, at the risk of breaking the burn. As I said fixed in 349 by making usb-storage also grok 255 sector sized IO's.
Arjan van de Ven wrote:
the usb-storage bug is fixed in kernel 349 (see http://people.redhat.com). Actually it's worked around. cdrecord has a bug where it doesn't check a return value and interprets an error value as 255 sectors-io-size. Which is/was invalid for usb storage. SG just silently ignores the fact that that is an invalid value for usb storage and continues, at the risk of breaking the burn. As I said fixed in 349 by making usb-storage also grok 255 sector sized IO's.
I don't know about cdrecord, but grip and cdparanoia are still broken with 349 and my USB CD-RW drive. I still get:
Error trying to open /dev/sg0 exclusively (No such device or address). retrying in 1 second
...and failure to rip any tracks.
Also, 349 seems to ignore "selinux=0" argument?
:j
On Sun, 2004-05-02 at 23:26, Alan Cox wrote:
On Sun, May 02, 2004 at 05:32:43PM -0400, Willem Riede wrote:
FC2T3 is still missing the ide-scsi module! And filing a bug doesn't work :-( I've had one in bugzilla since February 28, and no reaction whatsoever from the Red Hat kernel packagers...
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117095
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it?
And IDE multi-changers
Put it in unsupported for all I care, but compile it, _please_.
Unfortunately Arjan and Dave are out of range of my torture kit so all I can do is bitch at them as well (and if need be release rival kernel rpms which would be a pita to have to do)
Odd. I thought we had come to a compromise on this which went along the lines of..
if device is an IDE CD drive print "use /dev/hd? directly" abort else passthrough to usual ide-scsi foo.
Arjan ?
Dave
On Mon, May 03, 2004 at 01:08:20AM +0100, Dave Jones wrote:
Odd. I thought we had come to a compromise on this which went along the lines of..
if device is an IDE CD drive print "use /dev/hd? directly" abort else passthrough to usual ide-scsi foo.
Multichangers are special, although even the nakamichi 5x IDE is pretty rare and its users know about ide-scsi hand setup 8)
On 2004.05.02 20:08, Dave Jones wrote:
On Sun, 2004-05-02 at 23:26, Alan Cox wrote:
On Sun, May 02, 2004 at 05:32:43PM -0400, Willem Riede wrote:
FC2T3 is still missing the ide-scsi module! And filing a bug doesn't work :-( I've had one in bugzilla since February 28, and no reaction whatsoever from the Red Hat kernel packagers...
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117095
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it?
And IDE multi-changers
Put it in unsupported for all I care, but compile it, _please_.
Unfortunately Arjan and Dave are out of range of my torture kit so all I can do is bitch at them as well (and if need be release rival kernel rpms which would be a pita to have to do)
Odd. I thought we had come to a compromise on this which went along the lines of..
if device is an IDE CD drive print "use /dev/hd? directly" abort else passthrough to usual ide-scsi foo.
That is consistent with what I proposed both on linux-kernel/scsi and in my bugzilla entry. But as ide-scsi is not compiled, that logic is moot...
[root@fallguy root]# grep BLK_DEV_IDESCSI /usr/src/l*/configs/* /usr/src/linux-2.6.5-1.327/configs/kernel-2.6.5-i586.config:# CONFIG_BLK_DEV_IDESCSI is not set /usr/src/linux-2.6.5-1.327/configs/kernel-2.6.5-i586-smp.config:# CONFIG_BLK_DEV_IDESCSI is not set /usr/src/linux-2.6.5-1.327/configs/kernel-2.6.5-i686.config:# CONFIG_BLK_DEV_IDESCSI is not set /usr/src/linux-2.6.5-1.327/configs/kernel-2.6.5-i686-smp.config:# CONFIG_BLK_DEV_IDESCSI is not set
Regards, Willem Riede.
On 2004.05.02 20:08, Dave Jones wrote:
On Sun, 2004-05-02 at 23:26, Alan Cox wrote:
On Sun, May 02, 2004 at 05:32:43PM -0400, Willem Riede wrote:
FC2T3 is still missing the ide-scsi module! And filing a bug doesn't work :-( I've had one in bugzilla since February 28, and no reaction whatsoever from the Red Hat kernel packagers...
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117095
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it?
And IDE multi-changers
Put it in unsupported for all I care, but compile it, _please_.
Unfortunately Arjan and Dave are out of range of my torture kit so all I can do is bitch at them as well (and if need be release rival kernel rpms which would be a pita to have to do)
Odd. I thought we had come to a compromise on this which went along the lines of..
if device is an IDE CD drive print "use /dev/hd? directly" abort else passthrough to usual ide-scsi foo.
Arjan ?
Dave
[root@fallguy root]# grep BLK_DEV_IDESCSI /usr/src/l*/configs/* /usr/src/linux-2.6.5-1.351/configs/kernel-2.6.5-i586.config:# CONFIG_BLK_DEV_IDESCSI is not set /usr/src/linux-2.6.5-1.351/configs/kernel-2.6.5-i586-smp.config:# CONFIG_BLK_DEV_IDESCSI is not set /usr/src/linux-2.6.5-1.351/configs/kernel-2.6.5-i686.config:# CONFIG_BLK_DEV_IDESCSI is not set /usr/src/linux-2.6.5-1.351/configs/kernel-2.6.5-i686-smp.config:# CONFIG_BLK_DEV_IDESCSI is not set
Ide-scsi is both stable and needed.
Sigh. Willem Riede.
On Monday 03 May 2004 00:32, Willem Riede wrote: ...
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it? Put it in unsupported for all I care, but compile it, _please_.
I have problems burning CDs, count one more please from me.
On Mon, 2004-05-03 at 14:29, Doncho N. Gunchev wrote:
On Monday 03 May 2004 00:32, Willem Riede wrote: ...
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it? Put it in unsupported for all I care, but compile it, _please_.
I have problems burning CDs, count one more please from me.
I'm sorry but without details about what is broken that is really not helpful.
On Monday 03 May 2004 15:45, Arjan van de Ven wrote:
On Mon, 2004-05-03 at 14:29, Doncho N. Gunchev wrote:
On Monday 03 May 2004 00:32, Willem Riede wrote: ...
Why not compile ide-scsi? Why inconvenience owners of tape drives that need it? Put it in unsupported for all I care, but compile it, _please_.
I have problems burning CDs, count one more please from me.I'm sorry but without details about what is broken that is really not helpful.
True, one of the CD-Writers I have access to - 'HL-DT-ST - CD-RW GCE-8525B' (LG) worked just fine when my bittorrent finished downloading FC2t3. With all previous 2.6 kernels starting from FC1 I had no luck. I had one minute delay (no activity at all) with them and cdrecord was failing to write the 'lead-in'. I'll test is more tonight and if there are problems I'll report them. I still wonder about ide-scsi. BTW: my 128MB flash works too, kudzu just did not add entry in /etc/fstab as it used to do with FC1... and thanks for the response.
On Mon, Apr 12, 2004 at 03:40:32PM -0500, Chris Adams wrote:
How do I control my tape library robot? AFAIK that is only controlled via sg (the other SCSI modules ignore tape robots).
Updating your firmware on Seagate drives with their "Seatools" software is another thing you need sg access for.
devel@lists.stg.fedoraproject.org
-
Alan Cox -
Arjan van de Ven -
Casimiro de Almeida Barreto -
Chris Adams -
Dave Jones -
Doncho N. Gunchev -
Jos Vos -
Jurgen Botz -
Willem Riede