* Re: [PATCH] mpt2sas: New feature - Fast Load Support @ 2012-01-10 8:48 André-Sebastian Liebe 2012-01-10 10:53 ` Desai, Kashyap 0 siblings, 1 reply; 7+ messages in thread From: André-Sebastian Liebe @ 2012-01-10 8:48 UTC (permalink / raw) To: linux-scsi -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi guys, I've run into problems with the all new shiny 3.2 kernel. My system broke badly as it's depending on some devices which are attached to my LSI SAS 2008 hba. The mount scripts of my gentoo/openrc setup get triggered right after udev has settled. But with the new fast load support udev settles before my drives appear to the system, so mount fails for all those drives and of cause all subsequent services depending on them. Is there any switch to revert back to the old behavior of blocking until all ports scans are done? Thanks in advance André-Sebastian Liebe - -- gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.18 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBAgAGBQJPC/tXAAoJEIvQ0TmZSz9R+ywP/Ays9LD9+UgfrmU/U5mc2HxA n9QcZUK6CNGdw6QA8XsLKXPIOZYsCBiiLroNORzFfwRW7xsue1YMD6tR121TjmYL QA9C9Clf9WYJT8RTEkmECHkejvxAkBqu0FhMJJC3377L2/qAN2NSkz6VInzqAdqE My4948xRx0z7ysjiFqRwOeZgss70IowshwDO5mAbuAawOQZa3UZo72HZ8gVxbf4H xKhPLy6QraEJX3SMtTmJDvp46xJ5UN4yXosdOD7pkYUeCnTX7AhaO+bIuEPgGB9Q UaEawMoMq8cPVEXl4L5gvPkIKCWdnmCLTtDcOeqTxuHoyOx/qn8xvh2D4K2shWgf jaxPN9VGa6VBIxmZWjX4emtI2QqhjnaW8o7A+hz1K8rJO1SgCdGJxdB+gjLhXcLe ZYKLOJ5OFIX8JQk6IjUTLkRCnZpEG/WORU+fSU93dCglcRxnevPUvIs3rpTzstGd /DKN5JPXMIPvX3aJ5sWoXygpnq9oOfPAMDXB4jibkYbbnF4aqWEp7m+l7g84dH0N 2O0SFQm7VTb15Pp3ZIKGQhxfVxHwsd+fnoM1rpqkRRrF4YmovuFgsHr7qjE4astq /c9TUJzn/93fan/f0FrW9KlBE26zqwCp6HIbNe/HLSZgdMXGVWynwa3w86d17eyg givRtuynhgPB/KDyoavm =5BNV -----END PGP SIGNATURE----- -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: [PATCH] mpt2sas: New feature - Fast Load Support 2012-01-10 8:48 [PATCH] mpt2sas: New feature - Fast Load Support André-Sebastian Liebe @ 2012-01-10 10:53 ` Desai, Kashyap 2012-01-11 8:46 ` [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive André-Sebastian Liebe 0 siblings, 1 reply; 7+ messages in thread From: Desai, Kashyap @ 2012-01-10 10:53 UTC (permalink / raw) To: André-Sebastian Liebe, linux-scsi@vger.kernel.org There is a some scsi_mod parameter called "scan". If you pass those parameter at boot time you can change the behavior. (see below description) scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they are discovered. Async scans them in kernel threads, allowing boot to proceed. none ignores them, expecting user space to do the scan. ~ Kashyap > -----Original Message----- > From: linux-scsi-owner@vger.kernel.org [mailto:linux-scsi- > owner@vger.kernel.org] On Behalf Of André-Sebastian Liebe > Sent: Tuesday, January 10, 2012 2:18 PM > To: linux-scsi@vger.kernel.org > Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi guys, > > I've run into problems with the all new shiny 3.2 kernel. My system > broke badly as it's depending on some devices which are attached to my > LSI SAS 2008 hba. > The mount scripts of my gentoo/openrc setup get triggered right after > udev has settled. But with the new fast load support udev settles > before my drives appear to the system, so mount fails for all those > drives and of cause all subsequent services depending on them. > Is there any switch to revert back to the old behavior of blocking until > all ports scans are done? > > Thanks in advance > André-Sebastian Liebe > > - -- > gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe > <andre@lianse.eu> > Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2.0.18 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > iQIcBAEBAgAGBQJPC/tXAAoJEIvQ0TmZSz9R+ywP/Ays9LD9+UgfrmU/U5mc2HxA > n9QcZUK6CNGdw6QA8XsLKXPIOZYsCBiiLroNORzFfwRW7xsue1YMD6tR121TjmYL > QA9C9Clf9WYJT8RTEkmECHkejvxAkBqu0FhMJJC3377L2/qAN2NSkz6VInzqAdqE > My4948xRx0z7ysjiFqRwOeZgss70IowshwDO5mAbuAawOQZa3UZo72HZ8gVxbf4H > xKhPLy6QraEJX3SMtTmJDvp46xJ5UN4yXosdOD7pkYUeCnTX7AhaO+bIuEPgGB9Q > UaEawMoMq8cPVEXl4L5gvPkIKCWdnmCLTtDcOeqTxuHoyOx/qn8xvh2D4K2shWgf > jaxPN9VGa6VBIxmZWjX4emtI2QqhjnaW8o7A+hz1K8rJO1SgCdGJxdB+gjLhXcLe > ZYKLOJ5OFIX8JQk6IjUTLkRCnZpEG/WORU+fSU93dCglcRxnevPUvIs3rpTzstGd > /DKN5JPXMIPvX3aJ5sWoXygpnq9oOfPAMDXB4jibkYbbnF4aqWEp7m+l7g84dH0N > 2O0SFQm7VTb15Pp3ZIKGQhxfVxHwsd+fnoM1rpqkRRrF4YmovuFgsHr7qjE4astq > /c9TUJzn/93fan/f0FrW9KlBE26zqwCp6HIbNe/HLSZgdMXGVWynwa3w86d17eyg > givRtuynhgPB/KDyoavm > =5BNV > -----END PGP SIGNATURE----- > -- > To unsubscribe from this list: send the line "unsubscribe linux-scsi" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive 2012-01-10 10:53 ` Desai, Kashyap @ 2012-01-11 8:46 ` André-Sebastian Liebe 2012-01-12 6:30 ` Desai, Kashyap 0 siblings, 1 reply; 7+ messages in thread From: André-Sebastian Liebe @ 2012-01-11 8:46 UTC (permalink / raw) To: Desai, Kashyap; +Cc: linux-scsi [-- Attachment #1: Type: text/plain, Size: 2163 bytes --] Tanks for your suggestion Kashyap! I've tried it out but ran directly into an other problem. With scsi_mod.scan=sync the hba reported all 4 drives correctly but dropped one of it instantaneously. So I'm either stuck with kernel 3.1 and all booting up correctly or using 3.2 and manually mounting the drives behind the hba and starting the services depending on them thereafter. I've attached parts of the kernel log with scsi_mod.sync parameter set to sync/async -andre On 10.01.2012 11:53, Desai, Kashyap wrote: > There is a some scsi_mod parameter called "scan". > If you pass those parameter at boot time you can change the behavior. (see below description) > > scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they are > discovered. Async scans them in kernel threads, allowing boot to proceed. > none ignores them, expecting user space to do the scan. > > ~ Kashyap > >> -----Original Message----- >> From: linux-scsi-owner@vger.kernel.org [mailto:linux-scsi- >> owner@vger.kernel.org] On Behalf Of André-Sebastian Liebe >> Sent: Tuesday, January 10, 2012 2:18 PM >> To: linux-scsi@vger.kernel.org >> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support >> > Hi guys, > > I've run into problems with the all new shiny 3.2 kernel. My system > broke badly as it's depending on some devices which are attached to my > LSI SAS 2008 hba. > The mount scripts of my gentoo/openrc setup get triggered right after > udev has settled. But with the new fast load support udev settles > before my drives appear to the system, so mount fails for all those > drives and of cause all subsequent services depending on them. > Is there any switch to revert back to the old behavior of blocking until > all ports scans are done? > > Thanks in advance > André-Sebastian Liebe > >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html -- gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 [-- Attachment #2: 3.2.0-scsi_mod=async.txt --] [-- Type: text/plain, Size: 4740 bytes --] [ 20.860030] mpt2sas0: diag reset: SUCCESS [ 20.984933] mpt2sas0: Allocated physical memory: size(4005 kB) [ 20.984936] mpt2sas0: Current Controller Queue Depth(1754), Max Controller Queue Depth(2015) [ 20.984938] mpt2sas0: Scatter Gather Elements per IO(128) [ 21.177151] mpt2sas0: LSISAS2008: FWVersion(07.00.00.00), ChipRevision(0x03), BiosVersion(07.11.00.00) [ 21.177154] mpt2sas0: Protocol=(Initiator,Target), Capabilities=(Raid,TLR,EEDP,Snapshot Buffer,Diag Trace Buffer,Task Set Full,NCQ) [ 21.177298] mpt2sas0: sending port enable !! [ 22.772658] mpt2sas0: host_add: handle(0x0001), sas_addr(0x500605b002c89840), phys(8) [ 28.896722] mpt2sas0: port enable: SUCCESS [ 28.907610] scsi 10:0:0:0: Direct-Access ATA WDC WD20EARS-00S 0A80 PQ: 0 ANSI: 5 [ 28.907619] scsi 10:0:0:0: SATA: handle(0x0009), sas_addr(0x4433221101000000), phy(1), device_name(0x4ee250010b165a17) [ 28.907624] scsi 10:0:0:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(1) [ 28.907708] scsi 10:0:0:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 28.907713] scsi 10:0:0:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 28.917685] scsi 10:0:1:0: Direct-Access ATA WDC WD30EZRX-00A 0A80 PQ: 0 ANSI: 5 [ 28.917693] scsi 10:0:1:0: SATA: handle(0x000a), sas_addr(0x4433221103000000), phy(3), device_name(0x4ee250016daeb0cb) [ 28.917697] scsi 10:0:1:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(3) [ 28.917779] scsi 10:0:1:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 28.917784] scsi 10:0:1:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 28.928013] scsi 10:0:2:0: Direct-Access ATA WDC WD30EZRX-00M 0A80 PQ: 0 ANSI: 5 [ 28.928021] scsi 10:0:2:0: SATA: handle(0x000b), sas_addr(0x4433221107000000), phy(7), device_name(0x4ee25001277eb05a) [ 28.928025] scsi 10:0:2:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(7) [ 28.928107] scsi 10:0:2:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 28.928112] scsi 10:0:2:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 28.936672] scsi 10:0:3:0: Direct-Access ATA WDC WD20EARS-00S 0A80 PQ: 0 ANSI: 5 [ 28.936680] scsi 10:0:3:0: SATA: handle(0x000c), sas_addr(0x4433221105000000), phy(5), device_name(0x4ee250010a965a17) [ 28.936684] scsi 10:0:3:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(5) [ 28.936777] scsi 10:0:3:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 28.936782] scsi 10:0:3:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 28.937033] sd 10:0:0:0: Attached scsi generic sg2 type 0 [ 28.937234] sd 10:0:1:0: Attached scsi generic sg3 type 0 [ 28.937400] sd 10:0:2:0: Attached scsi generic sg4 type 0 [ 28.937560] sd 10:0:3:0: Attached scsi generic sg5 type 0 [ 28.941283] sd 10:0:3:0: [sde] 3907029168 512-byte logical blocks: (2.00 TB/1.81 TiB) [ 28.942043] sd 10:0:1:0: [sdc] physical block alignment offset: 4096 [ 28.942050] sd 10:0:1:0: [sdc] 5860533168 512-byte logical blocks: (3.00 TB/2.72 TiB) [ 28.942053] sd 10:0:1:0: [sdc] 4096-byte physical blocks [ 28.942228] sd 10:0:0:0: [sdb] 3907029168 512-byte logical blocks: (2.00 TB/1.81 TiB) [ 28.942316] sd 10:0:2:0: [sdd] physical block alignment offset: 4096 [ 28.942321] sd 10:0:2:0: [sdd] 5860533168 512-byte logical blocks: (3.00 TB/2.72 TiB) [ 28.942324] sd 10:0:2:0: [sdd] 4096-byte physical blocks [ 28.962776] sd 10:0:3:0: [sde] Write Protect is off [ 28.962780] sd 10:0:3:0: [sde] Mode Sense: 7f 00 00 08 [ 28.968120] sd 10:0:1:0: [sdc] Write Protect is off [ 28.968124] sd 10:0:1:0: [sdc] Mode Sense: 7f 00 00 08 [ 28.968892] sd 10:0:2:0: [sdd] Write Protect is off [ 28.968896] sd 10:0:2:0: [sdd] Mode Sense: 7f 00 00 08 [ 28.970328] sd 10:0:3:0: [sde] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 28.971181] sd 10:0:0:0: [sdb] Write Protect is off [ 28.971185] sd 10:0:0:0: [sdb] Mode Sense: 7f 00 00 08 [ 28.977990] sd 10:0:1:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 28.979059] sd 10:0:2:0: [sdd] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 28.981742] sd 10:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 29.010029] sde: sde1 [ 29.053382] sdd: sdd1 [ 29.058647] sdc: sdc1 [ 29.060751] sd 10:0:3:0: [sde] Attached SCSI disk [ 29.075114] sdb: sdb1 [ 29.115190] sd 10:0:2:0: [sdd] Attached SCSI disk [ 29.120141] sd 10:0:1:0: [sdc] Attached SCSI disk [ 29.143900] sd 10:0:0:0: [sdb] Attached SCSI disk [-- Attachment #3: 3.2.0-scsi_mod=sync.txt --] [-- Type: text/plain, Size: 6144 bytes --] [ 21.156696] mpt2sas0: diag reset: SUCCESS [ 21.281678] mpt2sas0: Allocated physical memory: size(4005 kB) [ 21.281681] mpt2sas0: Current Controller Queue Depth(1754), Max Controller Queue Depth(2015) [ 21.281683] mpt2sas0: Scatter Gather Elements per IO(128) [ 21.473817] mpt2sas0: LSISAS2008: FWVersion(07.00.00.00), ChipRevision(0x03), BiosVersion(07.11.00.00) [ 21.473821] mpt2sas0: Protocol=(Initiator,Target), Capabilities=(Raid,TLR,EEDP,Snapshot Buffer,Diag Trace Buffer,Task Set Full,NCQ) [ 21.473895] mpt2sas0: sending port enable !! [ 23.067477] mpt2sas0: host_add: handle(0x0001), sas_addr(0x500605b002c89840), phys(8) [ 29.193384] mpt2sas0: port enable: SUCCESS [ 29.204365] scsi 10:0:0:0: Direct-Access ATA WDC WD20EARS-00S 0A80 PQ: 0 ANSI: 5 [ 29.204373] scsi 10:0:0:0: SATA: handle(0x0009), sas_addr(0x4433221101000000), phy(1), device_name(0x4ee250010b165a17) [ 29.204377] scsi 10:0:0:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(1) [ 29.204464] scsi 10:0:0:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 29.204469] scsi 10:0:0:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 29.204641] sd 10:0:0:0: Attached scsi generic sg2 type 0 [ 29.209862] sd 10:0:0:0: [sdb] 3907029168 512-byte logical blocks: (2.00 TB/1.81 TiB) [ 29.214985] scsi 10:0:1:0: Direct-Access ATA WDC WD30EZRX-00M 0A80 PQ: 0 ANSI: 5 [ 29.214992] scsi 10:0:1:0: SATA: handle(0x000a), sas_addr(0x4433221107000000), phy(7), device_name(0x4ee25001277eb05a) [ 29.214996] scsi 10:0:1:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(7) [ 29.215082] scsi 10:0:1:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 29.215087] scsi 10:0:1:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 29.215272] sd 10:0:1:0: Attached scsi generic sg3 type 0 [ 29.220138] sd 10:0:1:0: [sdc] physical block alignment offset: 4096 [ 29.220145] sd 10:0:1:0: [sdc] 5860533168 512-byte logical blocks: (3.00 TB/2.72 TiB) [ 29.220149] sd 10:0:1:0: [sdc] 4096-byte physical blocks [ 29.225261] scsi 10:0:2:0: Direct-Access ATA WDC WD30EZRX-00A 0A80 PQ: 0 ANSI: 5 [ 29.225268] scsi 10:0:2:0: SATA: handle(0x000b), sas_addr(0x4433221103000000), phy(3), device_name(0x4ee250016daeb0cb) [ 29.225272] scsi 10:0:2:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(3) [ 29.225357] scsi 10:0:2:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 29.225362] scsi 10:0:2:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 29.225516] sd 10:0:2:0: Attached scsi generic sg4 type 0 [ 29.230289] sd 10:0:2:0: [sdd] physical block alignment offset: 4096 [ 29.230296] sd 10:0:2:0: [sdd] 5860533168 512-byte logical blocks: (3.00 TB/2.72 TiB) [ 29.230299] sd 10:0:2:0: [sdd] 4096-byte physical blocks [ 29.234222] scsi 10:0:3:0: Direct-Access ATA WDC WD20EARS-00S 0A80 PQ: 0 ANSI: 5 [ 29.234228] scsi 10:0:3:0: SATA: handle(0x000c), sas_addr(0x4433221105000000), phy(5), device_name(0x4ee250010a965a17) [ 29.234232] scsi 10:0:3:0: SATA: enclosure_logical_id(0x500605b002c89840), slot(5) [ 29.234318] scsi 10:0:3:0: atapi(n), ncq(y), asyn_notify(n), smart(y), fua(y), sw_preserve(y) [ 29.234323] scsi 10:0:3:0: qdepth(32), tagged(1), simple(0), ordered(0), scsi_level(6), cmd_que(1) [ 29.234474] sd 10:0:3:0: Attached scsi generic sg5 type 0 [ 29.234639] port-10:0: remove: sas_addr(0x4433221101000000) [ 29.234649] port-10:0: remove: sas_addr(0x4433221101000000), phy(1) [ 29.238313] sd 10:0:3:0: [sde] 3907029168 512-byte logical blocks: (2.00 TB/1.81 TiB) [ 29.239286] sd 10:0:0:0: [sdb] Write Protect is off [ 29.239290] sd 10:0:0:0: [sdb] Mode Sense: 7f 00 00 08 [ 29.239312] sd 10:0:0:0: [sdb] Asking for cache data failed [ 29.239316] sd 10:0:0:0: [sdb] Assuming drive cache: write through [ 29.239485] sd 10:0:0:0: [sdb] READ CAPACITY(16) failed [ 29.239488] sd 10:0:0:0: [sdb] Result: hostbyte=0x01 driverbyte=0x00 [ 29.239491] sd 10:0:0:0: [sdb] Sense not available. [ 29.239516] sd 10:0:0:0: [sdb] READ CAPACITY failed [ 29.239518] sd 10:0:0:0: [sdb] Result: hostbyte=0x01 driverbyte=0x00 [ 29.239521] sd 10:0:0:0: [sdb] Sense not available. [ 29.239556] sd 10:0:0:0: [sdb] Asking for cache data failed [ 29.239558] sd 10:0:0:0: [sdb] Assuming drive cache: write through [ 29.239561] sdb: detected capacity change from 2000398934016 to 0 [ 29.239666] sd 10:0:0:0: [sdb] READ CAPACITY(16) failed [ 29.239669] sd 10:0:0:0: [sdb] Result: hostbyte=0x01 driverbyte=0x00 [ 29.239672] sd 10:0:0:0: [sdb] Sense not available. [ 29.239708] sd 10:0:0:0: [sdb] READ CAPACITY failed [ 29.239710] sd 10:0:0:0: [sdb] Result: hostbyte=0x01 driverbyte=0x00 [ 29.239713] sd 10:0:0:0: [sdb] Sense not available. [ 29.239760] sd 10:0:0:0: [sdb] Asking for cache data failed [ 29.239762] sd 10:0:0:0: [sdb] Assuming drive cache: write through [ 29.239765] sd 10:0:0:0: [sdb] Attached SCSI disk [ 29.246471] sd 10:0:1:0: [sdc] Write Protect is off [ 29.246475] sd 10:0:1:0: [sdc] Mode Sense: 7f 00 00 08 [ 29.256672] sd 10:0:1:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 29.257079] sd 10:0:2:0: [sdd] Write Protect is off [ 29.257083] sd 10:0:2:0: [sdd] Mode Sense: 7f 00 00 08 [ 29.259340] sd 10:0:3:0: [sde] Write Protect is off [ 29.259343] sd 10:0:3:0: [sde] Mode Sense: 7f 00 00 08 [ 29.266542] sd 10:0:2:0: [sdd] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 29.267228] sd 10:0:3:0: [sde] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 29.315743] sde: sde1 [ 29.332165] sdc: sdc1 [ 29.349526] sdd: sdd1 [ 29.351274] sd 10:0:3:0: [sde] Attached SCSI disk [ 29.373777] sd 10:0:1:0: [sdc] Attached SCSI disk [ 29.390661] sd 10:0:2:0: [sdd] Attached SCSI disk [ 29.391060] mpt2sas0: removing handle(0x0009), sas_addr(0x4433221101000000) [ 29.391064] mpt2sas0: failure at drivers/scsi/mpt2sas/mpt2sas_transport.c:705/mpt2sas_transport_port_add()! ^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive 2012-01-11 8:46 ` [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive André-Sebastian Liebe @ 2012-01-12 6:30 ` Desai, Kashyap 2012-01-16 8:06 ` André-Sebastian Liebe 0 siblings, 1 reply; 7+ messages in thread From: Desai, Kashyap @ 2012-01-12 6:30 UTC (permalink / raw) To: André-Sebastian Liebe Cc: linux-scsi@vger.kernel.org, Nandigama, Nagalakshmi, Thakkar, Vishal Andre, Can you enable mpt2sas driver log level to 0x38c and re-send the driver logs in case of "scsi_mod.scan=sync" You need to pass boot time parameter mpt2sas.logging_level=0x38c ` Kashyap > -----Original Message----- > From: André-Sebastian Liebe [mailto:andre@lianse.eu] > Sent: Wednesday, January 11, 2012 2:17 PM > To: Desai, Kashyap > Cc: linux-scsi@vger.kernel.org > Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support - > scsi_mod.scan=sync drops drive > > Tanks for your suggestion Kashyap! > > I've tried it out but ran directly into an other problem. With > scsi_mod.scan=sync the hba reported all 4 drives correctly but dropped > one of it instantaneously. So I'm either stuck with kernel 3.1 and all > booting up correctly or using 3.2 and manually mounting the drives > behind the hba and starting the services depending on them thereafter. > > I've attached parts of the kernel log with scsi_mod.sync parameter set > to sync/async > > -andre > > > On 10.01.2012 11:53, Desai, Kashyap wrote: > > There is a some scsi_mod parameter called "scan". > > If you pass those parameter at boot time you can change the behavior. > > (see below description) > > > > scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they > are > > discovered. Async scans them in kernel threads, allowing > boot to proceed. > > none ignores them, expecting user space to do the scan. > > > > ~ Kashyap > > > >> -----Original Message----- > >> From: linux-scsi-owner@vger.kernel.org [mailto:linux-scsi- > >> owner@vger.kernel.org] On Behalf Of André-Sebastian Liebe > >> Sent: Tuesday, January 10, 2012 2:18 PM > >> To: linux-scsi@vger.kernel.org > >> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support > >> > > Hi guys, > > > > I've run into problems with the all new shiny 3.2 kernel. My system > > broke badly as it's depending on some devices which are attached to my > > LSI SAS 2008 hba. > > The mount scripts of my gentoo/openrc setup get triggered right after > > udev has settled. But with the new fast load support udev settles > > before my drives appear to the system, so mount fails for all those > > drives and of cause all subsequent services depending on them. > > Is there any switch to revert back to the old behavior of blocking > > until all ports scans are done? > > > > Thanks in advance > > André-Sebastian Liebe > > > >> -- > >> To unsubscribe from this list: send the line "unsubscribe linux-scsi" > >> in the body of a message to majordomo@vger.kernel.org More majordomo > >> info at http://vger.kernel.org/majordomo-info.html > > -- > gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe > <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 > 89BF AA1E 1F66 ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive 2012-01-12 6:30 ` Desai, Kashyap @ 2012-01-16 8:06 ` André-Sebastian Liebe 2012-01-17 9:26 ` Nandigama, Nagalakshmi 0 siblings, 1 reply; 7+ messages in thread From: André-Sebastian Liebe @ 2012-01-16 8:06 UTC (permalink / raw) To: Desai, Kashyap Cc: linux-scsi@vger.kernel.org, Nandigama, Nagalakshmi, Thakkar, Vishal [-- Attachment #1: Type: text/plain, Size: 3134 bytes --] Hi guys, sorry for the late reply, I haven't had the chance to reboot the system earlier. But here are the logs for both scan types. -andre On 12.01.2012 07:30, Desai, Kashyap wrote: > Andre, > > Can you enable mpt2sas driver log level to 0x38c and re-send the driver logs in case of > "scsi_mod.scan=sync" > > You need to pass boot time parameter mpt2sas.logging_level=0x38c > > ` Kashyap > >> -----Original Message----- >> From: André-Sebastian Liebe [mailto:andre@lianse.eu] >> Sent: Wednesday, January 11, 2012 2:17 PM >> To: Desai, Kashyap >> Cc: linux-scsi@vger.kernel.org >> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support - >> scsi_mod.scan=sync drops drive >> >> Tanks for your suggestion Kashyap! >> >> I've tried it out but ran directly into an other problem. With >> scsi_mod.scan=sync the hba reported all 4 drives correctly but dropped >> one of it instantaneously. So I'm either stuck with kernel 3.1 and all >> booting up correctly or using 3.2 and manually mounting the drives >> behind the hba and starting the services depending on them thereafter. >> >> I've attached parts of the kernel log with scsi_mod.sync parameter set >> to sync/async >> >> -andre >> >> >> On 10.01.2012 11:53, Desai, Kashyap wrote: >>> There is a some scsi_mod parameter called "scan". >>> If you pass those parameter at boot time you can change the behavior. >>> (see below description) >>> >>> scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they >> are >>> discovered. Async scans them in kernel threads, allowing >> boot to proceed. >>> none ignores them, expecting user space to do the scan. >>> >>> ~ Kashyap >>> >>>> -----Original Message----- >>>> From: linux-scsi-owner@vger.kernel.org [mailto:linux-scsi- >>>> owner@vger.kernel.org] On Behalf Of André-Sebastian Liebe >>>> Sent: Tuesday, January 10, 2012 2:18 PM >>>> To: linux-scsi@vger.kernel.org >>>> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support >>>> >>> Hi guys, >>> >>> I've run into problems with the all new shiny 3.2 kernel. My system >>> broke badly as it's depending on some devices which are attached to my >>> LSI SAS 2008 hba. >>> The mount scripts of my gentoo/openrc setup get triggered right after >>> udev has settled. But with the new fast load support udev settles >>> before my drives appear to the system, so mount fails for all those >>> drives and of cause all subsequent services depending on them. >>> Is there any switch to revert back to the old behavior of blocking >>> until all ports scans are done? >>> >>> Thanks in advance >>> André-Sebastian Liebe >>> >>>> -- >>>> To unsubscribe from this list: send the line "unsubscribe linux-scsi" >>>> in the body of a message to majordomo@vger.kernel.org More majordomo >>>> info at http://vger.kernel.org/majordomo-info.html >> >> -- >> gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe >> <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 >> 89BF AA1E 1F66 -- gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 [-- Attachment #2: mpt2saslog_async.txt.tar.gz --] [-- Type: application/x-gzip, Size: 21764 bytes --] [-- Attachment #3: mpt2saslog_sync.txt.tar.gz --] [-- Type: application/x-gzip, Size: 21085 bytes --] ^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive 2012-01-16 8:06 ` André-Sebastian Liebe @ 2012-01-17 9:26 ` Nandigama, Nagalakshmi 2012-01-17 12:52 ` André-Sebastian Liebe 0 siblings, 1 reply; 7+ messages in thread From: Nandigama, Nagalakshmi @ 2012-01-17 9:26 UTC (permalink / raw) To: André-Sebastian Liebe, Desai, Kashyap Cc: linux-scsi@vger.kernel.org, Thakkar, Vishal [-- Attachment #1: Type: text/plain, Size: 3653 bytes --] Andre, Request you to apply the attached patch to 3.2 kernel and test. Regards, Nagalakshmi -----Original Message----- From: André-Sebastian Liebe [mailto:andre@lianse.eu] Sent: Monday, January 16, 2012 1:36 PM To: Desai, Kashyap Cc: linux-scsi@vger.kernel.org; Nandigama, Nagalakshmi; Thakkar, Vishal Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive Hi guys, sorry for the late reply, I haven't had the chance to reboot the system earlier. But here are the logs for both scan types. -andre On 12.01.2012 07:30, Desai, Kashyap wrote: > Andre, > > Can you enable mpt2sas driver log level to 0x38c and re-send the > driver logs in case of "scsi_mod.scan=sync" > > You need to pass boot time parameter mpt2sas.logging_level=0x38c > > ` Kashyap > >> -----Original Message----- >> From: André-Sebastian Liebe [mailto:andre@lianse.eu] >> Sent: Wednesday, January 11, 2012 2:17 PM >> To: Desai, Kashyap >> Cc: linux-scsi@vger.kernel.org >> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support - >> scsi_mod.scan=sync drops drive >> >> Tanks for your suggestion Kashyap! >> >> I've tried it out but ran directly into an other problem. With >> scsi_mod.scan=sync the hba reported all 4 drives correctly but >> dropped one of it instantaneously. So I'm either stuck with kernel >> 3.1 and all booting up correctly or using 3.2 and manually mounting >> the drives behind the hba and starting the services depending on them thereafter. >> >> I've attached parts of the kernel log with scsi_mod.sync parameter >> set to sync/async >> >> -andre >> >> >> On 10.01.2012 11:53, Desai, Kashyap wrote: >>> There is a some scsi_mod parameter called "scan". >>> If you pass those parameter at boot time you can change the behavior. >>> (see below description) >>> >>> scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they >> are >>> discovered. Async scans them in kernel threads, allowing >> boot to proceed. >>> none ignores them, expecting user space to do the scan. >>> >>> ~ Kashyap >>> >>>> -----Original Message----- >>>> From: linux-scsi-owner@vger.kernel.org [mailto:linux-scsi- >>>> owner@vger.kernel.org] On Behalf Of André-Sebastian Liebe >>>> Sent: Tuesday, January 10, 2012 2:18 PM >>>> To: linux-scsi@vger.kernel.org >>>> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support >>>> >>> Hi guys, >>> >>> I've run into problems with the all new shiny 3.2 kernel. My system >>> broke badly as it's depending on some devices which are attached to >>> my LSI SAS 2008 hba. >>> The mount scripts of my gentoo/openrc setup get triggered right >>> after udev has settled. But with the new fast load support udev >>> settles before my drives appear to the system, so mount fails for >>> all those drives and of cause all subsequent services depending on them. >>> Is there any switch to revert back to the old behavior of blocking >>> until all ports scans are done? >>> >>> Thanks in advance >>> André-Sebastian Liebe >>> >>>> -- >>>> To unsubscribe from this list: send the line "unsubscribe linux-scsi" >>>> in the body of a message to majordomo@vger.kernel.org More >>>> majordomo info at http://vger.kernel.org/majordomo-info.html >> >> -- >> gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe >> <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE >> 6664 89BF AA1E 1F66 -- gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 [-- Attachment #2: scsi-misc-2[1].6.git-2cb6fc8c014b9b00c4487a79b8f6ed0da4121f45.patch --] [-- Type: application/octet-stream, Size: 3221 bytes --] From 2cb6fc8c014b9b00c4487a79b8f6ed0da4121f45 Mon Sep 17 00:00:00 2001 From: "nagalakshmi.nandigama@lsi.com" <nagalakshmi.nandigama@lsi.com> Date: Tue, 13 Dec 2011 09:29:15 +0530 Subject: [PATCH] [SCSI] mpt2sas: Removed redundant calling of _scsih_probe_devices() from _scsih_probe Removed redundant calling of _scsih_probe_devices() from _scsih_probe as it is getting called from _scsih_scan_finished. Also moved the function scsi_scan_host(shost) to get called after the volumes on warp drive are reported to the OS. Otherwise by the time the (ioc->hide_drives) flags is set, the volumes on warp drive are reported to the OS already. Also modified the initialization of reply queues only in case of driver load time in the function _base_make_ioc_operational(). Signed-off-by: Nagalakshmi Nandigama <nagalakshmi.nandigama@lsi.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com> --- drivers/scsi/mpt2sas/mpt2sas_base.c | 20 +++++++------------- drivers/scsi/mpt2sas/mpt2sas_scsih.c | 3 +-- 2 files changed, 8 insertions(+), 15 deletions(-) diff --git a/drivers/scsi/mpt2sas/mpt2sas_base.c b/drivers/scsi/mpt2sas/mpt2sas_base.c index 4819181..0b2c955 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_base.c +++ b/drivers/scsi/mpt2sas/mpt2sas_base.c @@ -4099,7 +4099,8 @@ _base_make_ioc_operational(struct MPT2SAS_ADAPTER *ioc, int sleep_flag) ioc->reply_free[i] = cpu_to_le32(reply_address); /* initialize reply queues */ - _base_assign_reply_queues(ioc); + if (ioc->is_driver_loading) + _base_assign_reply_queues(ioc); /* initialize Reply Post Free Queue */ reply_post_free = (long)ioc->reply_post_free; @@ -4147,24 +4148,17 @@ _base_make_ioc_operational(struct MPT2SAS_ADAPTER *ioc, int sleep_flag) if (ioc->is_driver_loading) { - - - - ioc->wait_for_discovery_to_complete = - _base_determine_wait_on_discovery(ioc); - return r; /* scan_start and scan_finished support */ - } - - - if (ioc->wait_for_discovery_to_complete && ioc->is_warpdrive) { - if (ioc->manu_pg10.OEMIdentifier == 0x80) { + if (ioc->is_warpdrive && ioc->manu_pg10.OEMIdentifier + == 0x80) { hide_flag = (u8) (ioc->manu_pg10.OEMSpecificFlags0 & MFG_PAGE10_HIDE_SSDS_MASK); if (hide_flag != MFG_PAGE10_HIDE_SSDS_MASK) ioc->mfg_pg10_hide_flag = hide_flag; } + ioc->wait_for_discovery_to_complete = + _base_determine_wait_on_discovery(ioc); + return r; /* scan_start and scan_finished support */ } - r = _base_send_port_enable(ioc, sleep_flag); if (r) return r; diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index 8cd4ebb..193e33e 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c +++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c @@ -8039,7 +8039,6 @@ _scsih_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto out_attach_fail; } - scsi_scan_host(shost); if (ioc->is_warpdrive) { if (ioc->mfg_pg10_hide_flag == MFG_PAGE10_EXPOSE_ALL_DISKS) ioc->hide_drives = 0; @@ -8053,8 +8052,8 @@ _scsih_probe(struct pci_dev *pdev, const struct pci_device_id *id) } } else ioc->hide_drives = 0; + scsi_scan_host(shost); - _scsih_probe_devices(ioc); return 0; out_attach_fail: -- 1.7.6.5 ^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive 2012-01-17 9:26 ` Nandigama, Nagalakshmi @ 2012-01-17 12:52 ` André-Sebastian Liebe 0 siblings, 0 replies; 7+ messages in thread From: André-Sebastian Liebe @ 2012-01-17 12:52 UTC (permalink / raw) To: Nandigama, Nagalakshmi Cc: Desai, Kashyap, linux-scsi@vger.kernel.org, Thakkar, Vishal [-- Attachment #1: Type: text/plain, Size: 3930 bytes --] Hey Kashyap, works for me (against 3.2.1). -andre On 17.01.2012 10:26, Nandigama, Nagalakshmi wrote: > Andre, > Request you to apply the attached patch to 3.2 kernel and test. > > Regards, > Nagalakshmi > > > -----Original Message----- > From: André-Sebastian Liebe [mailto:andre@lianse.eu] > Sent: Monday, January 16, 2012 1:36 PM > To: Desai, Kashyap > Cc: linux-scsi@vger.kernel.org; Nandigama, Nagalakshmi; Thakkar, Vishal > Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive > > Hi guys, > > sorry for the late reply, I haven't had the chance to reboot the system earlier. But here are the logs for both scan types. > > -andre > > On 12.01.2012 07:30, Desai, Kashyap wrote: >> Andre, >> >> Can you enable mpt2sas driver log level to 0x38c and re-send the >> driver logs in case of "scsi_mod.scan=sync" >> >> You need to pass boot time parameter mpt2sas.logging_level=0x38c >> >> ` Kashyap >> >>> -----Original Message----- >>> From: André-Sebastian Liebe [mailto:andre@lianse.eu] >>> Sent: Wednesday, January 11, 2012 2:17 PM >>> To: Desai, Kashyap >>> Cc: linux-scsi@vger.kernel.org >>> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support - >>> scsi_mod.scan=sync drops drive >>> >>> Tanks for your suggestion Kashyap! >>> >>> I've tried it out but ran directly into an other problem. With >>> scsi_mod.scan=sync the hba reported all 4 drives correctly but >>> dropped one of it instantaneously. So I'm either stuck with kernel >>> 3.1 and all booting up correctly or using 3.2 and manually mounting >>> the drives behind the hba and starting the services depending on them thereafter. >>> >>> I've attached parts of the kernel log with scsi_mod.sync parameter >>> set to sync/async >>> >>> -andre >>> >>> >>> On 10.01.2012 11:53, Desai, Kashyap wrote: >>>> There is a some scsi_mod parameter called "scan". >>>> If you pass those parameter at boot time you can change the behavior. >>>> (see below description) >>>> >>>> scsi_mod.scan= [SCSI] sync (default) scans SCSI busses as they >>> are >>>> discovered. Async scans them in kernel threads, allowing >>> boot to proceed. >>>> none ignores them, expecting user space to do the scan. >>>> >>>> ~ Kashyap >>>> >>>>> -----Original Message----- >>>>> From: linux-scsi-owner@vger.kernel.org [mailto:linux-scsi- >>>>> owner@vger.kernel.org] On Behalf Of André-Sebastian Liebe >>>>> Sent: Tuesday, January 10, 2012 2:18 PM >>>>> To: linux-scsi@vger.kernel.org >>>>> Subject: Re: [PATCH] mpt2sas: New feature - Fast Load Support >>>>> >>>> Hi guys, >>>> >>>> I've run into problems with the all new shiny 3.2 kernel. My system >>>> broke badly as it's depending on some devices which are attached to >>>> my LSI SAS 2008 hba. >>>> The mount scripts of my gentoo/openrc setup get triggered right >>>> after udev has settled. But with the new fast load support udev >>>> settles before my drives appear to the system, so mount fails for >>>> all those drives and of cause all subsequent services depending on them. >>>> Is there any switch to revert back to the old behavior of blocking >>>> until all ports scans are done? >>>> >>>> Thanks in advance >>>> André-Sebastian Liebe >>>> >>>>> -- >>>>> To unsubscribe from this list: send the line "unsubscribe linux-scsi" >>>>> in the body of a message to majordomo@vger.kernel.org More >>>>> majordomo info at http://vger.kernel.org/majordomo-info.html >>> >>> -- >>> gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe >>> <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE >>> 6664 89BF AA1E 1F66 > > > -- > gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 -- gpg key AA1E1F66 @ gpg-keyserver.de: André-Sebastian Liebe <andre@lianse.eu> Key fingerprint = 5436 5358 172C EB7D E414 C6DE 6664 89BF AA1E 1F66 [-- Attachment #2: mpt2sas_patched.txt.tar.gz --] [-- Type: application/x-gzip, Size: 21792 bytes --] ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2012-01-17 12:52 UTC | newest] Thread overview: 7+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2012-01-10 8:48 [PATCH] mpt2sas: New feature - Fast Load Support André-Sebastian Liebe 2012-01-10 10:53 ` Desai, Kashyap 2012-01-11 8:46 ` [PATCH] mpt2sas: New feature - Fast Load Support - scsi_mod.scan=sync drops drive André-Sebastian Liebe 2012-01-12 6:30 ` Desai, Kashyap 2012-01-16 8:06 ` André-Sebastian Liebe 2012-01-17 9:26 ` Nandigama, Nagalakshmi 2012-01-17 12:52 ` André-Sebastian Liebe
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox