sbp2: improved handling of device quirks
cover sbp2: improved handling of device quirks
Hi all,
the following series of patches reworks and adds to sbp2's handling of buggy SBP-2 hardware:
[PATCH 1/4] sbp2: consolidate workarounds
-----------------------------------------
Basically a refactoring, eases the integration of further workarounds. This patch was posted on linux1394-devel before.
[PATCH 2/4] sbp2: add read_capacity workaround for iPod
-------------------------------------------------------
There is a bug in newer iPod firmwares which causes IO errors as soon as an iPod is plugged in, making it inaccessible. Usb-storage, which has to deal with the same problem, already has the same workaround we are adding here. The issue has only recently become known to also affect FireWire iPods.
This patch of the series is actually the reason why I'm Cc'ing LKML and AKPM: It'd be nice to have this merged sooner than later since distribution kernels ship in configurations now which uncover this device bug (particularly when EFI partion support is enabled).
This patch was posted on linux1394-devel before but I updated it today to cover two more affected iPod models.
Requires PATCH 1/4.
[PATCH 3/4] sbp2: make TSB42AA9 workaround specific to Momobay CX-1
-------------------------------------------------------------------
Makes an existing blacklist entry more specific, although no actual harm was done when this workaround caught devices which did not really need it.
This patch was posted on linux1394-devel before.
Requires PATCH 1/4.
[PATCH 4/4] sbp2: add ability to override hardwired blacklist
-------------------------------------------------------------
Some of the workarounds which are triggered by sbp2's built-in blacklist may be degrading or even disruptive if unintentionally applied to devices which don't require the workaround. As a quick countermeasure which does not require users to recompile the driver, a new flag for sbp2's respective module load parameter is provided which disables the blacklist lookup.
Note: There has no such problem occured in the past, however as we are adding more workarounds, such setbacks could become more likely.
This is a new patch.
Requires PATCH 1/4, also needs PATCH 2/4 to apply without rejects.
Signature · 3 lines
Stefan Richter -=====-=-==- -=-- -===- http://arcgraph.de/sr/
PATCH 1/4 sbp2: consolidate workarounds
sbp2: consolidate workarounds
Grand unification of the three types of workarounds we have so far.
The "skip mode page 8" workaround is now limited to devices which pretend to be of TYPE_DISK instead of TYPE_RBC. This workaround is no longer enabled for Initio bridges since scsi_lib has an own fix.
Patch update in anticipation of more workarounds:
- Add module parameter "workarounds".
- Deprecate parameter "force_inquiry_hack".
- Compose the blacklist of a compound type for better readability and extensibility.
- Remove a now unused #define.
---
first posted on 2006-02-23, updated on 2006-03-19
Documentation/feature-removal-schedule.txt | 9 + drivers/ieee1394/sbp2.c | 158 +++++++++++++++++------------ drivers/ieee1394/sbp2.h | 16 +- 3 files changed, 109 insertions(+), 74 deletions(-)
drivers/ieee1394/sbp2.c +94 -64
@@ -42,6 +42,7 @@#include <linux/kernel.h> #include <linux/list.h> #include <linux/string.h> +#include <linux/stringify.h> #include <linux/slab.h> #include <linux/interrupt.h> #include <linux/fs.h>
@@ -117,7 +118,8 @@ MODULE_PARM_DESC(serialize_io, "Serializ*/ static int max_sectors =SBP2_MAX_SECTORS ;module_param (max_sectors, int,0444 ); -MODULE_PARM_DESC (max_sectors,"Change max sectors per I/O supported (default = 255)" ); +MODULE_PARM_DESC(max_sectors, "Change max sectors per I/O supported (default = " + __stringify(SBP2_MAX_SECTORS) ")");/* * Exclusive login to sbp2 device? In most cases, the sbp2 driver should
@@ -135,18 +137,33 @@ module_param(exclusive_login, int, 0644)MODULE_PARM_DESC (exclusive_login,"Exclusive login to sbp2 device (default = 1)" );/* -* SCSI inquiry hack for really badly behaved sbp2 devices. Turn this on -* if your sbp2 device is not properly handling the SCSI inquiry command. -* This hack makes the inquiry look more like a typical MS Windows inquiry -* by enforcing 36 byte inquiry and avoiding access to mode_sense page 8. +* If any of the following workarounds is required for your device to work, +* please submit the kernel messages logged by sbp2 to the linux1394-devel +* mailing list. * -* If force_inquiry_hack=1 is required for your device to work, -* please submit the logged sbp2_firmware_revision value of this device to -* the linux1394-devel mailing list. -*/ +* - 128kB max transfer +* Limit transfer size. Necessary for some old bridges. +* +* - 36 byte inquiry +* When scsi_mod probes the device, let the inquiry command look like that +* from MS Windows. +* +* - skip mode page 8 +* Suppress sending of mode_sense for mode page 8 if the device pretends to +* support the SCSI Primary Block commands instead of Reduced Block Commands. +*/ +static int sbp2_default_workarounds; +module_param_named (workarounds, sbp2_default_workarounds, int,0644 ); +MODULE_PARM_DESC(workarounds, "Work around device bugs (default = 0" + ", 128kB max transfer = " __stringify(SBP2_WORKAROUND_128K_MAX_TRANS) + ", 36 byte inquiry = " __stringify(SBP2_WORKAROUND_INQUIRY_36) + ", skip mode page 8 = " __stringify(SBP2_WORKAROUND_MODE_SENSE_8) +", or a combination)" ); + +/* legacy parameter */ static int force_inquiry_hack;module_param (force_inquiry_hack, int,0644 ); -MODULE_PARM_DESC (force_inquiry_hack,"Force SCSI inquiry hack (default = 0)" ); +MODULE_PARM_DESC (force_inquiry_hack,"Deprecated, use 'workarounds'" );/* * Export information about protocols/devices supported by this driver.
@@ -266,14 +283,29 @@ static struct hpsb_protocol_driver sbp2_ };/* -* List of device firmwares that require the inquiry hack. -* Yields a few false positives but did not break other devices so far. -*/ -static u32 sbp2_broken_inquiry_list[] = { -0x00002800 ,/* Stefan Richter <[email protected]> */ -/* DViCO Momobay CX-1 */ -0x00000200 /* Andreas Plesch <[email protected]> */ -/* QPS Fire DVDBurner */ +* List of devices with known bugs. +* +* The firmware_revision field, masked with 0xffff00, is the best indicator +* for the type of bridge chip of a device. It yields a few false positives +* but this did not break correctly behaving devices so far. +*/ +static const struct { +u32 firmware_revision; +unsigned workarounds; +} sbp2_workarounds_table[] = { +/* TSB42AA9 */ { + .firmware_revision =0x002800 , + .workarounds =SBP2_WORKAROUND_INQUIRY_36 | +SBP2_WORKAROUND_MODE_SENSE_8 , + }, +/* Initio bridges, actually only needed for some older ones */ { + .firmware_revision =0x000200 , + .workarounds =SBP2_WORKAROUND_INQUIRY_36 , + }, +/* Symbios bridge */ { + .firmware_revision =0xa0b800 , + .workarounds =SBP2_WORKAROUND_128K_MAX_TRANS , + } };/**************************************
@@ -1450,7 +1482,8 @@ static void sbp2_parse_unit_directory(ststruct csr1212_dentry *dentry;u64 management_agent_addr;u32 command_set_spec_id, command_set, unit_characteristics, - firmware_revision, workarounds; + firmware_revision; +unsigned workarounds;int i;SBP2_DEBUG_ENTER ();
@@ -1506,12 +1539,8 @@ static void sbp2_parse_unit_directory(stcase SBP2_FIRMWARE_REVISION_KEY :/* Firmware revision */ firmware_revision = kv->value.immediate; -if (force_inquiry_hack) -SBP2_INFO ("sbp2_firmware_revision = %x" , - (unsigned int )firmware_revision); -else -SBP2_DEBUG ("sbp2_firmware_revision = %x" , - (unsigned int )firmware_revision); +SBP2_DEBUG ("sbp2_firmware_revision = %x" , + (unsigned int )firmware_revision);break ;default :
@@ -1519,42 +1548,37 @@ static void sbp2_parse_unit_directory(st } } -/* This is the start of our broken device checking. We try to hack -* around oddities and known defects. */ - workarounds =0x0 ; - -/* If the vendor id is 0xa0b8 (Symbios vendor id), then we have a -* bridge with 128KB max transfer size limitation. For sanity, we -* only voice this when the current max_sectors setting -* exceeds the 128k limit. By default, that is not the case. -* -* It would be really nice if we could detect this before the scsi -* host gets initialized. That way we can down-force the -* max_sectors to account for it. That is not currently -* possible. */ -if ((firmware_revision &0xffff00 ) == -SBP2_128KB_BROKEN_FIRMWARE && - (max_sectors *512 ) > (128 *1024 )) { -SBP2_WARN ("Node " NODE_BUS_FMT ": Bridge only supports 128KB max transfer size." , -NODE_BUS_ARGS (ud->ne->host, ud->ne->nodeid)); -SBP2_WARN ("WARNING: Current max_sectors setting is larger than 128KB (%d sectors)!" , - max_sectors); - workarounds |=SBP2_BREAKAGE_128K_MAX_TRANSFER ; - } - -/* Check for a blacklisted set of devices that require us to force -* a 36 byte host inquiry. This can be overriden as a module param -* (to force all hosts). */ -for (i =0 ; i <ARRAY_SIZE (sbp2_broken_inquiry_list); i++) { -if ((firmware_revision &0xffff00 ) == - sbp2_broken_inquiry_list[i]) { -SBP2_WARN ("Node " NODE_BUS_FMT ": Using 36byte inquiry workaround" , -NODE_BUS_ARGS (ud->ne->host, ud->ne->nodeid)); - workarounds |=SBP2_BREAKAGE_INQUIRY_HACK ; -break ;/* No need to continue. */ - } + workarounds = sbp2_default_workarounds; +if (force_inquiry_hack) { +SBP2_WARN ("force_inquiry_hack is deprecated. " +"Use parameter 'workarounds' instead." ); + workarounds |=SBP2_WORKAROUND_INQUIRY_36 ; } +for (i =0 ; i <ARRAY_SIZE (sbp2_workarounds_table); i++) { +if (sbp2_workarounds_table[i].firmware_revision != + (firmware_revision &0xffff00 )) +continue ; + workarounds |= sbp2_workarounds_table[i].workarounds; +break ; + } + +if (workarounds) +SBP2_INFO ("Workarounds for node " NODE_BUS_FMT ": " +"0x%x (firmware_revision 0x%x)" , +NODE_BUS_ARGS (ud->ne->host, ud->ne->nodeid), + workarounds, firmware_revision); + +/* We would need one SCSI host template for each target to adjust +* max_sectors on the fly, therefore warn only. */ +if (workarounds &SBP2_WORKAROUND_128K_MAX_TRANS && + (max_sectors *512 ) > (128 *1024 )) +SBP2_WARN ("Node " NODE_BUS_FMT ": Bridge only supports 128KB " +"max transfer size. WARNING: Current max_sectors " +"setting is larger than 128KB (%d sectors)" , +NODE_BUS_ARGS (ud->ne->host, ud->ne->nodeid), + max_sectors); +/* If this is a logical unit directory entry, process the parent * to get the values. */ if (ud->flags &UNIT_DIRECTORY_LUN_DIRECTORY ) {
@@ -2447,19 +2471,23 @@ static int sbp2scsi_slave_alloc(struct s scsi_id->sdev = sdev; -if (force_inquiry_hack || - scsi_id->workarounds &SBP2_BREAKAGE_INQUIRY_HACK ) { +if (scsi_id->workarounds &SBP2_WORKAROUND_INQUIRY_36 ) sdev->inquiry_len =36 ; - sdev->skip_ms_page_8 =1 ; - }return 0 ; }static int sbp2scsi_slave_configure (struct scsi_device *sdev) { +struct scsi_id_instance_data *scsi_id = + (struct scsi_id_instance_data *)sdev->host->hostdata[0 ]; +blk_queue_dma_alignment (sdev->request_queue, (512 -1 )); sdev->use_10_for_rw =1 ; sdev->use_10_for_ms =1 ; + +if (sdev->type ==TYPE_DISK && + scsi_id->workarounds &SBP2_WORKAROUND_MODE_SENSE_8 ) + sdev->skip_ms_page_8 =1 ;return 0 ; }
@@ -2603,7 +2631,9 @@ static int sbp2_module_init(void) scsi_driver_template.cmd_per_lun =1 ; } -/* Set max sectors (module load option). Default is 255 sectors. */ +if (sbp2_default_workarounds &SBP2_WORKAROUND_128K_MAX_TRANS && + (max_sectors *512 ) > (128 *1024 )) + max_sectors =128 *1024 /512 ; scsi_driver_template.max_sectors = max_sectors;/* Register our high level driver with 1394 stack */
drivers/ieee1394/sbp2.h +6 -10
@@ -227,11 +227,6 @@ struct sbp2_status_block {#define SBP2_SW_VERSION_ENTRY 0x00010483/* -* Other misc defines -*/ -#define SBP2_128KB_BROKEN_FIRMWARE 0xa0b800 - -/* * SCSI specific stuff */
@@ -239,6 +234,11 @@ struct sbp2_status_block {#define SBP2_MAX_SECTORS 255/* Max sectors supported */ #define SBP2_MAX_CMDS 8/* This should be safe */ +/* Flags for detected oddities and brokeness */ +#define SBP2_WORKAROUND_128K_MAX_TRANS 0x1 +#define SBP2_WORKAROUND_INQUIRY_36 0x2 +#define SBP2_WORKAROUND_MODE_SENSE_8 0x4 +/* This is the two dma types we use for cmd_dma below */ enum cmd_dma_types {CMD_DMA_NONE ,
@@ -268,10 +268,6 @@ struct sbp2_command_info { }; -/* A list of flags for detected oddities and brokeness. */ -#define SBP2_BREAKAGE_128K_MAX_TRANSFER 0x1 -#define SBP2_BREAKAGE_INQUIRY_HACK 0x2 -struct sbp2scsi_host_info ;/*
@@ -345,7 +341,7 @@ struct scsi_id_instance_data {struct Scsi_Host *scsi_host;/* Device specific workarounds/brokeness */ -u32 workarounds; +unsigned workarounds; };/* Sbp2 host data structure (one per IEEE1394 host) */
Documentation/feature-removal-schedule.txt +9 -0
@@ -56,6 +56,15 @@ Who: Jody McIntyre <scjody@steamballoon. --------------------------- +What: sbp2: module parameter "force_inquiry_hack" +When: July 2006 +Why: Superceded by parameter "workarounds". Both parameters are meant to be + used ad-hoc and for single devices only, i.e. not in modprobe.conf, + therefore the impact of this feature replacement should be low. +Who: Stefan Richter <[email protected]> + +--------------------------- + What: Video4Linux API 1 ioctls and video_decoder.h from Video devices. When: July 2006 Why: V4L1 AP1 was replaced by V4L2 API. during migration from 2.4 to 2.6
PATCH 2/4 sbp2: add read_capacity workaround for iPod
sbp2: add read_capacity workaround for iPod
Apple decided to copy some USB stupidity over to FireWire. The sector number returned by iPods from read_capacity is one too many. This may cause I/O errors, especially if the kernel is configured for EFI partition support. We use the same workaround as usb-storage but have to check for different model IDs.
http://marc.theaimsgroup.com/?t=114233262300001
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=187409
Acknowledgements: Diagnosis and therapy by Mathieu Chouquet-Stringer <[email protected]>, additional data about affected and unaffected Apple hardware from Vladimir Kotal, Sander De Graaf, Bryan Olmstead, Hugh Dixon.
---
first posted on 2006-03-19, updated on 2006-04-14 (more iPods added)
drivers/ieee1394/sbp2.c | 49 ++++++++++++++++++++++++++++++++++++++++++++---- drivers/ieee1394/sbp2.h | 1 2 files changed, 46 insertions(+), 4 deletions(-)
drivers/ieee1394/sbp2.h +1 -0
@@ -238,6 +238,7 @@ struct sbp2_status_block {#define SBP2_WORKAROUND_128K_MAX_TRANS 0x1#define SBP2_WORKAROUND_INQUIRY_36 0x2#define SBP2_WORKAROUND_MODE_SENSE_8 0x4 +#define SBP2_WORKAROUND_FIX_CAPACITY 0x8/* This is the two dma types we use for cmd_dma below */ enum cmd_dma_types {
drivers/ieee1394/sbp2.c +45 -4
93 lines not on this page · raw message
PATCH 3/4 sbp2: make TSB42AA9 workaround specific to Momobay CX-1
sbp2: make TSB42AA9 workaround specific to Momobay CX-1
The workarounds are not required for DViCO Momobay FX-3A and AFAIR not for Momobay CX-2. These contain an TSB42AA9A but feature the same firmware_revision value as the older DViCO Momobay CX-1.
---
first posted on 2006-03-19
drivers/ieee1394/sbp2.c +2 -1
@@ -300,8 +300,9 @@ static const struct {u32 model_id;unsigned workarounds; } sbp2_workarounds_table[] = { -/* TSB42AA9 */ { +/* DViCO Momobay CX-1 with TSB42AA9 bridge */ { .firmware_revision = 0x002800, + .model_id = 0x001010, .workarounds =SBP2_WORKAROUND_INQUIRY_36 | SBP2_WORKAROUND_MODE_SENSE_8, },
PATCH 4/4 sbp2: add ability to override hardwired blacklist
sbp2: add ability to override hardwired blacklist
In case the blacklist with workarounds for device bugs yields a false positive, the module load parameter can now also be used as an override instead of an addition to the blacklist.
---
drivers/ieee1394/sbp2.c | 29 ++++++++++++++++++----------- drivers/ieee1394/sbp2.h | 1 + 2 files changed, 19 insertions(+), 11 deletions(-)
drivers/ieee1394/sbp2.h +1 -0
@@ -239,6 +239,7 @@ struct sbp2_status_block {#define SBP2_WORKAROUND_INQUIRY_36 0x2#define SBP2_WORKAROUND_MODE_SENSE_8 0x4#define SBP2_WORKAROUND_FIX_CAPACITY 0x8 +#define SBP2_WORKAROUND_OVERRIDE 0x100/* This is the two dma types we use for cmd_dma below */ enum cmd_dma_types {
drivers/ieee1394/sbp2.c +18 -11
@@ -156,6 +156,11 @@ MODULE_PARM_DESC(exclusive_login, "Exclu* Tell sd_mod to correct the last sector number reported by read_capacity. * Avoids access beyond actual disk limits on devices with an off-by-one bug. * Don't use this with devices which don't have this bug. +* +* - override internal blacklist +* Instead of adding to the built-in blacklist, use only the workarounds +* specified in the module load parameter. +* Useful if a blacklist entry interfered with a non-broken device. */ static int sbp2_default_workarounds;module_param_named (workarounds, sbp2_default_workarounds, int,0644 );
@@ -164,6 +169,7 @@ MODULE_PARM_DESC(workarounds, "Work arou ", 36 byte inquiry = " __stringify(SBP2_WORKAROUND_INQUIRY_36) ", skip mode page 8 = " __stringify(SBP2_WORKAROUND_MODE_SENSE_8) ", fix capacity = " __stringify(SBP2_WORKAROUND_FIX_CAPACITY) + ", override internal blacklist = " __stringify(SBP2_WORKAROUND_OVERRIDE) ", or a combination)");/* legacy parameter */
@@ -1588,17 +1594,18 @@ static void sbp2_parse_unit_directory(st workarounds |=SBP2_WORKAROUND_INQUIRY_36 ; } -for (i =0 ; i <ARRAY_SIZE (sbp2_workarounds_table); i++) { -if (sbp2_workarounds_table[i].firmware_revision && - sbp2_workarounds_table[i].firmware_revision != - (firmware_revision &0xffff00 )) -continue ; -if (sbp2_workarounds_table[i].model_id && - sbp2_workarounds_table[i].model_id != ud->model_id) -continue ; - workarounds |= sbp2_workarounds_table[i].workarounds; -break ; - } +if (!(workarounds &SBP2_WORKAROUND_OVERRIDE )) +for (i =0 ; i <ARRAY_SIZE (sbp2_workarounds_table); i++) { +if (sbp2_workarounds_table[i].firmware_revision && + sbp2_workarounds_table[i].firmware_revision != + (firmware_revision &0xffff00 )) +continue ; +if (sbp2_workarounds_table[i].model_id && + sbp2_workarounds_table[i].model_id != ud->model_id) +continue ; + workarounds |= sbp2_workarounds_table[i].workarounds; +break ; + }if (workarounds) SBP2_INFO("Workarounds for node " NODE_BUS_FMT ": 0x%x "
Quoted 1 line · I wrote:
[PATCH 3/4] sbp2: make TSB42AA9 workaround specific to Momobay CX-1
...
Quoted 1 line
Requires PATCH 1/4.
Also requires patch 2/4 for a new struct member.
Signature · 3 lines
Stefan Richter -=====-=-==- -=-- -===- http://arcgraph.de/sr/
Quoted 4 lines · Stefan Richter
Hi all,
the following series of patches reworks and adds to sbp2's handling of buggy SBP-2 hardware:
All applied.
Thanks,
Jody
Quoted 66 lines
[PATCH 1/4] sbp2: consolidate workarounds
-----------------------------------------
Basically a refactoring, eases the integration of further workarounds. This patch was posted on linux1394-devel before.[PATCH 2/4] sbp2: add
read_capacityworkaround for iPod
-------------------------------------------------------
There is a bug in newer iPod firmwares which causes IO errors as soon as an iPod is plugged in, making it inaccessible. Usb-storage, which has to deal with the same problem, already has the same workaround we are adding here. The issue has only recently become known to also affect FireWire iPods.This patch of the series is actually the reason why I'm Cc'ing LKML and AKPM: It'd be nice to have this merged sooner than later since distribution kernels ship in configurations now which uncover this device bug (particularly when EFI partion support is enabled).
This patch was posted on linux1394-devel before but I updated it today to cover two more affected iPod models.
Requires PATCH 1/4.[PATCH 3/4] sbp2: make TSB42AA9 workaround specific to Momobay CX-1
-------------------------------------------------------------------
Makes an existing blacklist entry more specific, although no actual harm was done when this workaround caught devices which did not really need it.This patch was posted on linux1394-devel before.
Requires PATCH 1/4.[PATCH 4/4] sbp2: add ability to override hardwired blacklist
-------------------------------------------------------------
Some of the workarounds which are triggered by sbp2's built-in blacklist may be degrading or even disruptive if unintentionally applied to devices which don't require the workaround. As a quick countermeasure which does not require users to recompile the driver, a new flag for sbp2's respective module load parameter is provided which disables the blacklist lookup.Note: There has no such problem occured in the past, however as we are adding more workarounds, such setbacks could become more likely.
This is a new patch.
Requires PATCH 1/4, also needs PATCH 2/4 to apply without rejects.Signature · 14 lines
Stefan Richter -=====-=-==- -=-- -===- http://arcgraph.de/sr/ ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642 _______________________________________________ mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux1394-devel