/* * Initialisation code for Cyrix/NatSemi VSA1 softaudio * * (C) Copyright 2003 Red Hat Inc * * XpressAudio(tm) is used on the Cyrix MediaGX (now NatSemi Geode) systems. * The older version (VSA1) provides fairly good soundblaster emulation * although there are a couple of bugs: large DMA buffers break record, * and the MPU event handling seems suspect. VSA2 allows the native driver * to control the AC97 audio engine directly and requires a different driver. * * Thanks to National Semiconductor for providing the needed information * on the XpressAudio(tm) internals. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the * Free Software Foundation; either version 2, or (at your option) any * later version. * * This program is distributed in the hope that it will be useful, but * WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * General Public License for more details. * * TO DO: * Investigate whether we can portably support Cognac (5520) in the * same manner. */ #include #include #include #include #include #include "sound_config.h" #include "sb.h" /* * Read a soundblaster compatible mixer register. * In this case we are actually reading an SMI trap * not real hardware. */ static u8 mixer_read(unsigned long io, u8 reg) { outb(reg, io + 4); udelay(20); reg = inb(io + 5); udelay(20); return reg; } static int probe_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct address_info *hw_config; unsigned long base; void __iomem *mem; unsigned long io; u16 map; u8 irq, dma8, dma16; int oldquiet; extern int sb_be_quiet; base = pci_resource_start(pdev, 0); if(base == 0UL) return 1; mem = ioremap(base, 128); if (!mem) return 1; map = readw(mem + 0x18); /* Read the SMI enables */ iounmap(mem); /* Map bits 0:1 * 0x20 + 0x200 = sb base 2 sb enable 3 adlib enable 5 MPU enable 0x330 6 MPU enable 0x300 The other bits may be used internally so must be masked */ io = 0x220 + 0x20 * (map & 3); if(map & (1<<2)) printk(KERN_INFO "kahlua: XpressAudio at 0x%lx\n", io); else return 1; if(map & (1<<5)) printk(KERN_INFO "kahlua: MPU at 0x300\n"); else if(map & (1<<6)) printk(KERN_INFO "kahlua: MPU at 0x330\n"); irq = mixer_read(io, 0x80) & 0x0F; dma8 = mixer_read(io, 0x81); // printk("IRQ=%x MAP=%x DMA=%x\n", irq, map, dma8); if(dma8 & 0x20) dma16 = 5; else if(dma8 & 0x40) dma16 = 6; else if(dma8 & 0x80) dma16 = 7; else { printk(KERN_ERR "kahlua: No 16bit DMA enabled.\n"); return 1; } if(dma8 & 0x01) dma8 = 0; else if(dma8 & 0x02) dma8 = 1; else if(dma8 & 0x08) dma8 = 3; else { printk(KERN_ERR "kahlua: No 8bit DMA enabled.\n"); return 1; } if(irq & 1) irq = 9; else if(irq & 2) irq = 5; else if(irq & 4) irq = 7; else if(irq & 8) irq = 10; else { printk(KERN_ERR "kahlua: SB IRQ not set.\n"); return 1; } printk(KERN_INFO "kahlua: XpressAudio on IRQ %d, DMA %d, %d\n", irq, dma8, dma16); hw_config = kzalloc(sizeof(struct address_info), GFP_KERNEL); if(hw_config == NULL) { printk(KERN_ERR "kahlua: out of memory.\n"); return 1; } pci_set_drvdata(pdev, hw_config); hw_config->io_base = io; hw_config->irq = irq; hw_config->dma = dma8; hw_config->dma2 = dma16; hw_config->name = "Cyrix XpressAudio"; hw_config->driver_use_1 = SB_NO_MIDI | SB_PCI_IRQ; if (!request_region(io, 16, "soundblaster")) goto err_out_free; if(sb_dsp_detect(hw_config, 0, 0, NULL)==0) { printk(KERN_ERR "kahlua: audio not responding.\n"); release_region(io, 16); goto err_out_free; } oldquiet = sb_be_quiet; sb_be_quiet = 1; if(sb_dsp_init(hw_config, THIS_MODULE)) { sb_be_quiet = oldquiet; goto err_out_free; } sb_be_quiet = oldquiet; return 0; err_out_free: kfree(hw_config); return 1; } static void remove_one(struct pci_dev *pdev) { struct address_info *hw_config = pci_get_drvdata(pdev); sb_dsp_unload(hw_config, 0); kfree(hw_config); } MODULE_AUTHOR("Alan Cox"); MODULE_DESCRIPTION("Kahlua VSA1 PCI Audio"); MODULE_LICENSE("GPL"); /* * 5530 only. The 5510/5520 decode is different. */ static const struct pci_device_id id_tbl[] = { { PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5530_AUDIO), 0 }, { } }; MODULE_DEVICE_TABLE(pci, id_tbl); static struct pci_driver kahlua_driver = { .name = "kahlua", .id_table = id_tbl, .probe = probe_one, .remove = remove_one, }; static int __init kahlua_init_module(void) { printk(KERN_INFO "Cyrix Kahlua VSA1 XpressAudio support (c) Copyright 2003 Red Hat Inc\n"); return pci_register_driver(&kahlua_driver); } static void kahlua_cleanup_module(void) { pci_unregister_driver(&kahlua_driver); } module_init(kahlua_init_module); module_exit(kahlua_cleanup_module); -0800 committerVinod Koul <vinod.koul@intel.com>2017-01-25 11:29:11 +0530 commitae4a3e028bb8b59e7cfeb0cc9ef03d885182ce8b (patch) tree139fc7e29f97d6bb6c4dca2a97be2dc3f824bd51 /tools/leds/Makefile parent49def1853334396f948dcb4cedb9347abb318df5 (diff)
dmaengine: cppi41: Fix runtime PM timeouts with USB mass storage
Commit fdea2d09b997 ("dmaengine: cppi41: Add basic PM runtime support") added runtime PM support for cppi41, but had corner case issues. Some of the issues were fixed with commit 098de42ad670 ("dmaengine: cppi41: Fix unpaired pm runtime when only a USB hub is connected"). That fix however caused a new regression where we can get error -115 messages with USB on BeagleBone when connecting a USB mass storage device to a hub. This is because when connecting a USB mass storage device to a hub, the initial DMA transfers can take over 200ms to complete and cppi41 autosuspend delay times out. To fix the issue, we want to implement refcounting for chan_busy array that contains the active dma transfers. Increasing the autosuspend delay won't help as that the delay could be potentially seconds, and it's best to let the USB subsystem to deal with the timeouts on errors. The earlier attempt for runtime PM was buggy as the pm_runtime_get/put() calls could get unpaired easily as they did not follow the state of the chan_busy array as described in commit 098de42ad670 ("dmaengine: cppi41: Fix unpaired pm runtime when only a USB hub is connected". Let's fix the issue by adding pm_runtime_get() to where a new transfer is added to the chan_busy array, and calls to pm_runtime_put() where chan_busy array entry is cleared. This prevents any autosuspend timeouts from happening while dma transfers are active. Fixes: 098de42ad670 ("dmaengine: cppi41: Fix unpaired pm runtime when only a USB hub is connected") Fixes: fdea2d09b997 ("dmaengine: cppi41: Add basic PM runtime support") Cc: Andy Shevchenko <andy.shevchenko@gmail.com> Cc: Bin Liu <b-liu@ti.com> Cc: Grygorii Strashko <grygorii.strashko@ti.com> Cc: Kevin Hilman <khilman@baylibre.com> Cc: Patrick Titiano <ptitiano@baylibre.com> Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Tested-by: Bin Liu <b-liu@ti.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Diffstat (limited to 'tools/leds/Makefile')