| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20080216184108.GT13446@gollum.tnic>
Date: Sat, 16 Feb 2008 19:41:08 +0100
From: Borislav Petkov <petkovbb@...glemail.com>
To: Bartlomiej Zolnierkiewicz <bzolnier@...il.com>
Cc: Stefan Bader <stefan.bader@...onical.com>,
linux-kernel@...r.kernel.org, linux-ide@...r.kernel.org,
Jens Axboe <jens.axboe@...cle.com>
Subject: Re: Optiarc DVD RW AD-5200A audio playing
On Sat, Feb 16, 2008 at 07:23:58PM +0100, Bartlomiej Zolnierkiewicz wrote:
> On Saturday 16 February 2008, Borislav Petkov wrote:
> > On Sat, Feb 16, 2008 at 06:48:24PM +0100, Bartlomiej Zolnierkiewicz wrote:
> > > On Saturday 16 February 2008, Borislav Petkov wrote:
> > > > On Sat, Feb 16, 2008 at 04:24:01PM +0100, Bartlomiej Zolnierkiewicz wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > On Saturday 16 February 2008, Borislav Petkov wrote:
> > > > > > On Fri, Feb 15, 2008 at 02:53:27PM -0500, Stefan Bader wrote:
> > > > > > > Hello Borislav,
> > > > > > >
> > > > > > > I worked on a problem with an DVD driver (model=Optiarc DVD RW AD-5200A)
> > > > > > > which obviously has the same problem as some Matshita drives. The
> > > > > > > following patch was reported to enabled audio playing on this drive.
> > > > > > > Would this approach be suitable for upstream or are there other
> > > > > > > solutions to this problem?
> > > > > > >
> > > > > > > Regards,
> > > > > > > Stefan
> > > > > > >
> > > > > > > --- a/drivers/ide/ide-cd.c
> > > > > > > +++ b/drivers/ide/ide-cd.c
> > > > > > > @@ -2988,7 +2988,8 @@ int ide_cdrom_probe_capabilities (ide_drive_t *drive)
> > > > > > > if (strcmp(drive->id->model, "MATSHITADVD-ROM SR-8187") == 0 ||
> > > > > > > strcmp(drive->id->model, "MATSHITADVD-ROM SR-8186") == 0 ||
> > > > > > > strcmp(drive->id->model, "MATSHITADVD-ROM SR-8176") == 0 ||
> > > > > > > - strcmp(drive->id->model, "MATSHITADVD-ROM SR-8174") == 0)
> > > > > > > + strcmp(drive->id->model, "MATSHITADVD-ROM SR-8174") == 0 ||
> > > > > > > + strcmp(drive->id->model, "Optiarc DVD RW AD-5200A") == 0)
> > > > > > > CDROM_CONFIG_FLAGS(drive)->audio_play = 1;
> > > > > > >
> > > > > > > #if ! STANDARD_ATAPI
> > > > > >
> > > > > > Hi Stefan,
> > > > > >
> > > > > > just to make sure that the audioplay bit is not set in the capabilities page,
> > > > > > can you please try the following patch applied against 2.6.25-rc2 and send me
> > > > > > the output. Thanks!
> > > > > >
> > > > > > @Bart: by the way, this cdi->mask thingy is kinda unintuitive doing double
> > > > > > negation to check whether a feature is supported or not. Yeah, this comes from
> > > > > > "above," i.e. uniform cdrom layer. But still, shouldn't we use a cdi->caps_flags
> > > > > > or something similar instead, which mirrors the caps page bits setting?
> > > > >
> > > > > It seems so (at least having negative flags is very unintuitive) but they
> > > > > might be some reason for this ugliness, Jens?
> > > > >
> > > > > [ Please also remember that since cdrom layer is _uniform_ it may be not
> > > > > possible and/or desirable to have 1-1 mapping between caps page bits
> > > > > and the future cdi->caps_flags. ]
> > > > >
> > > > > > commit 435f0f4496a1b32af2d542f43b2370a890fe2f83
> > > > > > Author: Borislav Petkov <petkovbb@...il.com>
> > > > > > Date: Sat Feb 16 09:56:36 2008 +0100
> > > > > >
> > > > > > ide-cd: Enable audio play quirk for Optiarc DVD RW AD-5200A drive
> > > > > >
> > > > > > Reported-by: Stefan Bader <stefan.bader@...onical.com>
> > > > > > Signed-off-by: Borislav Petkov <petkovbb@...il.com>
> > > > > >
> > > > > > diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
> > > > > > index f77db6b..2c9d06e 100644
> > > > > > --- a/drivers/ide/ide-cd.c
> > > > > > +++ b/drivers/ide/ide-cd.c
> > > > > > @@ -1750,6 +1750,10 @@ int ide_cdrom_probe_capabilities (ide_drive_t *drive)
> > > > > > cdi->mask &= ~(CDC_DVD_RAM | CDC_RAM);
> > > > > > if (buf[8 + 3] & 0x10)
> > > > > > cdi->mask &= ~CDC_DVD_R;
> > > > > > + if (!(buf[8 + 4] & 0x01)) {
> > > > >
> > > > > Hmm, shouldn't there be '&& (cd->cd_flags & IDE_CD_FLAG_PLAY_AUDIO_OK)'
> > > > > to prevent false positives?
> > > >
> > > > I wanted to see whether the caps page reports the audioplay bit off...
> > >
> > > we still need IDE_CD_FLAG_PLAY_AUDIO_OK flag to be _set_ to enable the quirk
> > >
> > > > >
> > > > > > + printk(KERN_INFO "ide-cd: audio play not advertised in caps page,"
> > > > >
> > > > > Would be nice to also printk() the device name.
> > > >
> > > > ... but printing the device model is actually a good idea and this will rule out
> > > > false positives, so Stefan, please drop the previous patch and test the updated
> > > > one below. Thanks.
> > > >
> > > >
> > > > commit 6cc44b0ce5c9270b15d456eb9ffa91b855e4e0d0
> > > > Author: Borislav Petkov <petkovbb@...il.com>
> > > > Date: Sat Feb 16 09:56:36 2008 +0100
> > > >
> > > > ide-cd: Enable audio play quirk for Optiarc DVD RW AD-5200A drive
> > > >
> > > > Reported-by: Stefan Bader <stefan.bader@...onical.com>
> > > > Signed-off-by: Borislav Petkov <petkovbb@...il.com>
> > > >
> > > > diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
> > > > index f77db6b..4c9984f 100644
> > > > --- a/drivers/ide/ide-cd.c
> > > > +++ b/drivers/ide/ide-cd.c
> > > > @@ -1750,6 +1750,11 @@ int ide_cdrom_probe_capabilities (ide_drive_t *drive)
> > > > cdi->mask &= ~(CDC_DVD_RAM | CDC_RAM);
> > > > if (buf[8 + 3] & 0x10)
> > > > cdi->mask &= ~CDC_DVD_R;
> > > > + if (!(buf[8 + 4] & 0x01)) {
> > > > + printk(KERN_INFO "ide-cd: audio play not advertised in caps "
> > > > + "page for drive model [%s], enabling quirk.\n",
> > > > + drive->id->model);
> > >
> > > if IDE_CD_FLAG_PLAY_AUDIO_OK flag is not set the above message is _bogus_
> > > (because the quirk won't be enabled)
> > >
> > > [ how's about just deleting the whole printk() to preserve simplicity? ]
> > >
> > > > + }
> > > > if ((buf[8 + 4] & 0x01) || (cd->cd_flags & IDE_CD_FLAG_PLAY_AUDIO_OK))
> > > > cdi->mask &= ~CDC_PLAY_AUDIO;
> > > >
> > > > @@ -1929,6 +1934,7 @@ static const struct cd_list_entry ide_cd_quirks_list[] = {
> > > > { "MATSHITADVD-ROM SR-8186", NULL, IDE_CD_FLAG_PLAY_AUDIO_OK },
> > > > { "MATSHITADVD-ROM SR-8176", NULL, IDE_CD_FLAG_PLAY_AUDIO_OK },
> > > > { "MATSHITADVD-ROM SR-8174", NULL, IDE_CD_FLAG_PLAY_AUDIO_OK },
> > > > + { "Optiarc DVD RW AD-5200A", NULL, IDE_CD_FLAG_PLAY_AUDIO_OK },
> >
> > But the flag _is_ set because the drive is added to the quirk list (line above ^) and
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
> (fortunately) majority of ATAPI drives are not in the quirk list ;-)
>
> [ and they may have (buf[8 + 4] & 0x01) == 0 ]
Bart,
i think i probably didn't express myself clear enough before. The patch
serves the sole purpose to show exactly whether buf[8 + 4] & 0x01) == 0 and
is in no way the final solution. I just wanted to make sure the caps page
doesn't set the audioplay bit. The final solution would be to add the drive
to the quirks list only, i guess, in case you don't have a better idea.
Sorry for the misunderstanding :=).
>
> > ide_cdrom_setup() calls ide_cd_flags() to set the proper flags for cd->cd_flags and later on
> > ide_cdrom_probe_capabilities() works on the same cd->cd_flags for testing
> > IDE_CD_FLAG_PLAY_AUDIO_OK, or have i gone blind completely...?
> >
> > > > { NULL, NULL, 0 }
> > > > };
--
Regards/Gruß,
Boris.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists