From: Finn Thain <fthain@telegraphics.com.au>
To: "James E.J. Bottomley" <JBottomley@odin.com>,
"Martin K. Petersen" <martin.petersen@oracle.com>,
Michael Schmitz <schmitzmic@gmail.com>,
linux-m68k@vger.kernel.org, linux-scsi@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: [PATCH v4 18/78] ncr5380: Eliminate USLEEP_WAITLONG delay
Date: Sun, 03 Jan 2016 16:05:19 +1100 [thread overview]
Message-ID: <20160103050505.811790912@telegraphics.com.au> (raw)
In-Reply-To: 20160103050501.042035135@telegraphics.com.au
[-- Attachment #1: ncr5380-eliminate-USLEEP_WAITLONG --]
[-- Type: text/plain, Size: 4650 bytes --]
Linux 2.1.105 introduced the USLEEP_WAITLONG delay, apparently "needed for
Mustek scanners". It is intended to stall the issue queue for 5 seconds.
There are a number of problems with this.
1. Only g_NCR5380 enables the delay, which implies that the other five
drivers using the NCR5380.c core driver remain incompatible with
Mustek scanners.
2. The delay is not implemented by atari_NCR5380.c, which is problematic
for re-unifying the two core driver forks.
3. The delay is implemented using NCR5380_set_timer() which makes it
unreliable. A new command queued by the mid-layer cancels the delay.
4. The delay is applied indiscriminately in several situations in which
NCR5380_select() returns -1. These are-- reselection by the target,
failure of the target to assert BSY, and failure of the target to
assert REQ. It's clear from the comments that USLEEP_WAITLONG is not
relevant to the reselection case. And reportedly, these scanners do
not disconnect.
5. atari_NCR5380.c was forked before Linux 2.1.105, so it was spared some
of the damage done to NCR5380.c. In this case, the atari_NCR5380.c core
driver was more standard-compliant and may not have needed any
workaround like the USLEEP_WAITLONG kludge. The compliance issue was
addressed in the previous patch.
If these scanners still don't work, we need a better solution. Retrying
selection until EH aborts a command offers equivalent robustness. Bugs in
the existing driver prevent EH working correctly but this is addressed in
a subsequent patch. Remove USLEEP_WAITLONG.
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Tested-by: Ondrej Zary <linux@rainbow-software.org>
---
drivers/scsi/NCR5380.c | 19 +++++--------------
drivers/scsi/g_NCR5380.c | 1 -
2 files changed, 5 insertions(+), 15 deletions(-)
Index: linux/drivers/scsi/NCR5380.c
===================================================================
--- linux.orig/drivers/scsi/NCR5380.c 2016-01-03 16:03:19.000000000 +1100
+++ linux/drivers/scsi/NCR5380.c 2016-01-03 16:03:20.000000000 +1100
@@ -468,10 +468,6 @@ static void NCR5380_print_phase(struct S
#ifndef USLEEP_POLL
#define USLEEP_POLL msecs_to_jiffies(200)
#endif
-#ifndef USLEEP_WAITLONG
-/* RvC: (reasonable time to wait on select error) */
-#define USLEEP_WAITLONG USLEEP_SLEEP
-#endif
/*
* Function : int should_disconnect (unsigned char cmd)
@@ -619,8 +615,8 @@ static void prepare_info(struct Scsi_Hos
"can_queue %d, cmd_per_lun %d, "
"sg_tablesize %d, this_id %d, "
"flags { %s%s%s%s}, "
-#if defined(USLEEP_POLL) && defined(USLEEP_WAITLONG)
- "USLEEP_POLL %lu, USLEEP_WAITLONG %lu, "
+#if defined(USLEEP_POLL) && defined(USLEEP_SLEEP)
+ "USLEEP_POLL %lu, USLEEP_SLEEP %lu, "
#endif
"options { %s} ",
instance->hostt->name, instance->io_port, instance->n_io_port,
@@ -631,8 +627,8 @@ static void prepare_info(struct Scsi_Hos
hostdata->flags & FLAG_DTC3181E ? "DTC3181E " : "",
hostdata->flags & FLAG_NO_PSEUDO_DMA ? "NO_PSEUDO_DMA " : "",
hostdata->flags & FLAG_TOSHIBA_DELAY ? "TOSHIBA_DELAY " : "",
-#if defined(USLEEP_POLL) && defined(USLEEP_WAITLONG)
- USLEEP_POLL, USLEEP_WAITLONG,
+#if defined(USLEEP_POLL) && defined(USLEEP_SLEEP)
+ USLEEP_POLL, USLEEP_SLEEP,
#endif
#ifdef AUTOPROBE_IRQ
"AUTOPROBE_IRQ "
@@ -1030,15 +1026,10 @@ static void NCR5380_main(struct work_str
if (!NCR5380_select(instance, tmp)) {
/* OK or bad target */
} else {
- /* RvC: device failed, so we wait a long time
- this is needed for Mustek scanners, that
- do not respond to commands immediately
- after a scan */
- printk(KERN_DEBUG "scsi%d: device %d did not respond in time\n", instance->host_no, tmp->device->id);
LIST(tmp, hostdata->issue_queue);
tmp->host_scribble = (unsigned char *) hostdata->issue_queue;
hostdata->issue_queue = tmp;
- NCR5380_set_timer(hostdata, USLEEP_WAITLONG);
+ done = 0;
}
} /* if hostdata->selecting */
if (hostdata->connected
Index: linux/drivers/scsi/g_NCR5380.c
===================================================================
--- linux.orig/drivers/scsi/g_NCR5380.c 2016-01-03 16:03:06.000000000 +1100
+++ linux/drivers/scsi/g_NCR5380.c 2016-01-03 16:03:20.000000000 +1100
@@ -59,7 +59,6 @@
/* settings for DTC3181E card with only Mustek scanner attached */
#define USLEEP_POLL msecs_to_jiffies(10)
#define USLEEP_SLEEP msecs_to_jiffies(200)
-#define USLEEP_WAITLONG msecs_to_jiffies(5000)
#define AUTOPROBE_IRQ
next prev parent reply other threads:[~2016-01-03 5:18 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-03 5:05 [PATCH v4 00/78] More fixes, cleanup and modernization for NCR5380 drivers Finn Thain
2016-01-03 5:05 ` [PATCH v4 01/78] atari_scsi: Fix SCSI host ID setting Finn Thain
2016-01-03 5:05 ` [PATCH v4 02/78] ncr5380: Remove redundant static variable initializers Finn Thain
2016-01-03 5:05 ` [PATCH v4 03/78] ncr5380: Eliminate PDEBUG*, TDEBUG* and DTCDEBUG* macros Finn Thain
2016-01-03 5:05 ` [PATCH v4 04/78] ncr5380: Remove more pointless macros Finn Thain
2016-01-03 5:05 ` [PATCH v4 05/78] ncr5380: Remove NCR5380_local_declare and NCR5380_setup macros Finn Thain
2016-01-03 5:05 ` [PATCH v4 06/78] ncr5380: Remove NCR5380_instance_name macro Finn Thain
2016-01-03 5:05 ` [PATCH v4 07/78] ncr5380: Split NCR5380_init() into two functions Finn Thain
2016-01-03 5:05 ` [PATCH v4 08/78] ncr5380: Move NCR53C400-specific code Finn Thain
2016-01-03 5:05 ` [PATCH v4 09/78] atari_NCR5380: Reset bus on driver initialization if required Finn Thain
2016-01-03 5:05 ` [PATCH v4 10/78] atari_NCR5380: Remove RESET_BOOT, CONFIG_ATARI_SCSI_TOSHIBA_DELAY and CONFIG_ATARI_SCSI_RESET_BOOT Finn Thain
2016-01-03 5:05 ` [PATCH v4 11/78] ncr5380: Simplify bus reset handlers Finn Thain
2016-01-03 5:05 ` [PATCH v4 12/78] ncr5380: Remove unused hostdata->aborted flag Finn Thain
2016-01-03 5:05 ` [PATCH v4 13/78] ncr5380: Remove redundant register writes Finn Thain
2016-01-03 5:05 ` [PATCH v4 14/78] ncr5380: Use return instead of goto in NCR5380_select() Finn Thain
2016-01-03 5:05 ` [PATCH v4 15/78] ncr5380: Always escalate bad target time-out " Finn Thain
2016-01-03 5:05 ` [PATCH v4 16/78] ncr5380: Proceed with next command after NCR5380_select() calls scsi_done Finn Thain
2016-01-03 5:05 ` [PATCH v4 17/78] ncr5380: Keep BSY asserted when entering SELECTION phase Finn Thain
2016-01-03 5:05 ` Finn Thain [this message]
2016-01-03 5:05 ` [PATCH v4 19/78] ncr5380: Cleanup bogus {request, release}_region() calls Finn Thain
2016-01-03 5:05 ` [PATCH v4 20/78] ncr5380: Introduce unbound workqueue Finn Thain
2016-01-03 5:05 ` [PATCH v4 21/78] ncr5380: Sleep when polling, if possible Finn Thain
2016-01-03 5:05 ` [PATCH v4 22/78] ncr5380: Eliminate selecting state Finn Thain
2016-01-03 5:05 ` [PATCH v4 23/78] ncr5380: Always retry arbitration and selection Finn Thain
2016-01-03 5:05 ` [PATCH v4 24/78] ncr5380: Implement NCR5380_dma_xfer_len and remove LIMIT_TRANSFERSIZE macro Finn Thain
2016-01-03 5:05 ` [PATCH v4 25/78] ncr5380: Rework disconnect versus poll logic Finn Thain
2016-01-03 5:05 ` [PATCH v4 26/78] ncr5380: Fix NCR5380_transfer_pio() result Finn Thain
2016-01-03 5:05 ` [PATCH v4 27/78] ncr5380: Add missing lock in eh_abort_handler Finn Thain
2016-01-03 5:05 ` [PATCH v4 28/78] ncr5380: Drop DEF_SCSI_QCMD macro Finn Thain
2016-01-03 5:05 ` [PATCH v4 29/78] ncr5380: Remove references to linked commands Finn Thain
2016-01-03 5:05 ` [PATCH v4 30/78] ncr5380: Add missing break after case MESSAGE_REJECT Finn Thain
2016-01-03 5:05 ` [PATCH v4 31/78] ncr5380: Fix !REQ timeout in do_abort() Finn Thain
2016-01-03 5:05 ` [PATCH v4 32/78] ncr5380: Fix bus phase " Finn Thain
2016-01-03 5:05 ` [PATCH v4 33/78] atari_NCR5380: Set do_abort() timeouts Finn Thain
2016-01-03 5:05 ` [PATCH v4 34/78] atari_NCR5380: Use arbitration timeout Finn Thain
2016-01-24 10:38 ` Geert Uytterhoeven
2016-01-25 2:45 ` Finn Thain
2016-01-25 8:05 ` Geert Uytterhoeven
2016-01-25 18:28 ` Andreas Schwab
2016-01-25 19:56 ` Michael Schmitz
2016-01-26 0:18 ` Finn Thain
2016-01-26 11:13 ` Geert Uytterhoeven
2016-01-03 5:05 ` [PATCH v4 35/78] ncr5380: Dont wait for BUS FREE after disconnect Finn Thain
2016-01-03 5:05 ` [PATCH v4 36/78] ncr5380: Use work_struct instead of delayed_work Finn Thain
2016-01-03 5:05 ` [PATCH v4 37/78] ncr5380: Standardize work queueing algorithm Finn Thain
2016-01-03 5:05 ` [PATCH v4 38/78] ncr5380: Remove UNSAFE macro Finn Thain
2016-01-03 5:05 ` [PATCH v4 39/78] ncr5380: Standardize interrupt handling Finn Thain
2016-01-03 5:05 ` [PATCH v4 40/78] ncr5380: Introduce NCR5380_poll_politely2 Finn Thain
2016-01-03 5:05 ` [PATCH v4 41/78] ncr5380: Replace redundant flags with FLAG_NO_DMA_FIXUP Finn Thain
2016-01-03 5:05 ` [PATCH v4 42/78] ncr5380: Replace READ_OVERRUNS macro with FLAG_NO_DMA_FIXUPS Finn Thain
2016-01-03 5:05 ` [PATCH v4 43/78] ncr5380: Standardize reselection handling Finn Thain
2016-01-03 5:05 ` [PATCH v4 44/78] ncr5380: Fix off-by-one bug in extended_msg[] bounds check Finn Thain
2016-01-03 5:05 ` [PATCH v4 45/78] ncr5380: Cleanup #include directives Finn Thain
2016-01-03 5:05 ` [PATCH v4 46/78] ncr5380: Fix NDEBUG_NO_DATAOUT flag Finn Thain
2016-01-03 5:05 ` [PATCH v4 47/78] ncr5380: Fix and cleanup scsi_host_template initializers Finn Thain
2016-01-03 5:05 ` [PATCH v4 48/78] atari_NCR5380: Fix queue_size limit Finn Thain
2016-01-03 5:05 ` [PATCH v4 49/78] ncr5380: Remove redundant ICR_ARBITRATION_LOST test and eliminate FLAG_DTC3181E Finn Thain
2016-01-03 5:05 ` [PATCH v4 50/78] ncr5380: Change instance->host_lock to hostdata->lock Finn Thain
2016-01-03 5:05 ` [PATCH v4 51/78] ncr5380: Remove command list debug code Finn Thain
2016-01-03 5:05 ` [PATCH v4 52/78] ncr5380: Remove H_NO macro and introduce dsprintk Finn Thain
2016-01-03 5:05 ` [PATCH v4 53/78] ncr5380: Use shost_priv helper Finn Thain
2016-01-03 5:05 ` [PATCH v4 54/78] ncr5380: Use dsprintk() for queue debugging Finn Thain
2016-01-03 5:05 ` [PATCH v4 55/78] ncr5380: Remove LIST and REMOVE macros Finn Thain
2016-01-03 5:05 ` [PATCH v4 56/78] ncr5380: Remove redundant volatile qualifiers Finn Thain
2016-01-03 5:05 ` [PATCH v4 57/78] ncr5380: Use standard list data structure Finn Thain
2016-01-03 5:05 ` [PATCH v4 58/78] ncr5380: Refactor command completion Finn Thain
2016-01-03 5:06 ` [PATCH v4 59/78] ncr5380: Fix autosense bugs Finn Thain
2016-01-03 5:06 ` [PATCH v4 60/78] ncr5380: Implement new eh_abort_handler Finn Thain
2016-01-03 5:06 ` [PATCH v4 61/78] ncr5380: Fix EH during arbitration and selection Finn Thain
2016-01-03 5:06 ` [PATCH v4 62/78] ncr5380: Implement new eh_bus_reset_handler Finn Thain
2016-01-03 5:06 ` [PATCH v4 63/78] atari_NCR5380: Remove HOSTNO macro from printk() and seq_printf() calls Finn Thain
2016-01-03 5:06 ` [PATCH v4 64/78] atari_NCR5380: Eliminate HOSTNO macro Finn Thain
2016-01-03 5:06 ` [PATCH v4 65/78] atari_scsi, sun3_scsi: Remove global Scsi_Host pointer Finn Thain
2016-01-03 5:06 ` [PATCH v4 66/78] ncr5380: Fix soft lockups Finn Thain
2016-01-03 5:06 ` [PATCH v4 67/78] ncr5380: Cleanup comments Finn Thain
2016-01-03 5:06 ` [PATCH v4 68/78] ncr5380: Fix trailing whitespace using regexp Finn Thain
2016-01-03 5:06 ` [PATCH v4 69/78] ncr5380: Fix whitespace in comments " Finn Thain
2016-01-03 7:54 ` Joe Perches
2016-01-03 8:17 ` Finn Thain
2016-01-03 14:10 ` One Thousand Gnomes
2016-01-03 18:20 ` Joe Perches
2016-01-03 21:29 ` One Thousand Gnomes
2016-01-03 21:46 ` Joe Perches
[not found] ` <20160103220505.7734a879@lxorguk.ukuu.org.uk>
2016-01-03 22:13 ` Joe Perches
2016-01-03 5:06 ` [PATCH v4 70/78] ncr5380: Merge changes from atari_NCR5380.c Finn Thain
2016-01-03 5:06 ` [PATCH v4 71/78] atari_NCR5380: Merge changes from NCR5380.c Finn Thain
2016-01-03 5:06 ` [PATCH v4 72/78] ncr5380: Cleanup whitespace and parentheses Finn Thain
2016-01-03 5:06 ` [PATCH v4 73/78] ncr5380: Fix pseudo DMA transfers on 53C400 Finn Thain
2016-01-03 5:06 ` [PATCH v4 74/78] ncr5380: Use runtime register mapping Finn Thain
2016-01-03 5:06 ` [PATCH v4 75/78] ncr5380: Enable PDMA for NCR53C400A Finn Thain
2016-01-03 5:06 ` [PATCH v4 76/78] ncr5380: Enable PDMA for DTC chips Finn Thain
2016-01-03 5:06 ` [PATCH v4 77/78] ncr5380: Fix wait for 53C80 registers registers after PDMA Finn Thain
2016-01-03 5:06 ` [PATCH v4 78/78] ncr5380: Add support for HP C2502 Finn Thain
2016-01-07 2:53 ` [PATCH v4 00/78] More fixes, cleanup and modernization for NCR5380 drivers Martin K. Petersen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160103050505.811790912@telegraphics.com.au \
--to=fthain@telegraphics.com.au \
--cc=JBottomley@odin.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-m68k@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=schmitzmic@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox