/* * udlfb.c -- Framebuffer driver for DisplayLink USB controller * * Copyright (C) 2009 Roberto De Ioris * Copyright (C) 2009 Jaya Kumar * Copyright (C) 2009 Bernie Thompson * * This file is subject to the terms and conditions of the GNU General Public * License v2. See the file COPYING in the main directory of this archive for * more details. * * Layout is based on skeletonfb by James Simmons and Geert Uytterhoeven, * usb-skeleton by GregKH. * * Device-specific portions based on information from Displaylink, with work * from Florian Echtler, Henrik Bjerregaard Pedersen, and others. */ #include #include #include #include #include #include #include #include #include #include "udlfb.h" static struct fb_fix_screeninfo dlfb_fix = { .id = "udlfb", .type = FB_TYPE_PACKED_PIXELS, .visual = FB_VISUAL_TRUECOLOR, .xpanstep = 0, .ypanstep = 0, .ywrapstep = 0, .accel = FB_ACCEL_NONE, }; static const u32 udlfb_info_flags = FBINFO_DEFAULT | FBINFO_READS_FAST | #ifdef FBINFO_VIRTFB FBINFO_VIRTFB | #endif FBINFO_HWACCEL_IMAGEBLIT | FBINFO_HWACCEL_FILLRECT | FBINFO_HWACCEL_COPYAREA | FBINFO_MISC_ALWAYS_SETPAR; /* * There are many DisplayLink-based products, all with unique PIDs. We are able * to support all volume ones (circa 2009) with a single driver, so we match * globally on VID. TODO: Probe() needs to detect when we might be running * "future" chips, and bail on those, so a compatible driver can match. */ static struct usb_device_id id_table[] = { {.idVendor = 0x17e9, .match_flags = USB_DEVICE_ID_MATCH_VENDOR,}, {}, }; MODULE_DEVICE_TABLE(usb, id_table); #ifndef CONFIG_FB_DEFERRED_IO #warning Please set CONFIG_FB_DEFFERRED_IO option to support generic fbdev apps #endif #ifndef CONFIG_FB_SYS_IMAGEBLIT #ifndef CONFIG_FB_SYS_IMAGEBLIT_MODULE #warning Please set CONFIG_FB_SYS_IMAGEBLIT option to support fb console #endif #endif #ifndef CONFIG_FB_MODE_HELPERS #warning CONFIG_FB_MODE_HELPERS required. Expect build break #endif /* dlfb keeps a list of urbs for efficient bulk transfers */ static void dlfb_urb_completion(struct urb *urb); static struct urb *dlfb_get_urb(struct dlfb_data *dev); static int dlfb_submit_urb(struct dlfb_data *dev, struct urb * urb, size_t len); static int dlfb_alloc_urb_list(struct dlfb_data *dev, int count, size_t size); static void dlfb_free_urb_list(struct dlfb_data *dev); /* other symbols with dependents */ #ifdef CONFIG_FB_DEFERRED_IO static struct fb_deferred_io dlfb_defio; #endif /* * All DisplayLink bulk operations start with 0xAF, followed by specific code * All operations are written to buffers which then later get sent to device */ static char *dlfb_set_register(char *buf, u8 reg, u8 val) { *buf++ = 0xAF; *buf++ = 0x20; *buf++ = reg; *buf++ = val; return buf; } static char *dlfb_vidreg_lock(char *buf) { return dlfb_set_register(buf, 0xFF, 0x00); } static char *dlfb_vidreg_unlock(char *buf) { return dlfb_set_register(buf, 0xFF, 0xFF); } /* * On/Off for driving the DisplayLink framebuffer to the display */ static char *dlfb_enable_hvsync(char *buf, bool enable) { if (enable) return dlfb_set_register(buf, 0x1F, 0x00); else return dlfb_set_register(buf, 0x1F, 0x01); } static char *dlfb_set_color_depth(char *buf, u8 selection) { return dlfb_set_register(buf, 0x00, selection); } static char *dlfb_set_base16bpp(char *wrptr, u32 base) { /* the base pointer is 16 bits wide, 0x20 is hi byte. */ wrptr = dlfb_set_register(wrptr, 0x20, base >> 16); wrptr = dlfb_set_register(wrptr, 0x21, base >> 8); return dlfb_set_register(wrptr, 0x22, base); } /* * DisplayLink HW has separate 16bpp and 8bpp framebuffers. * In 24bpp modes, the low 323 RGB bits go in the 8bpp framebuffer */ static char *dlfb_set_base8bpp(char *wrptr, u32 base) { wrptr = dlfb_set_register(wrptr, 0x26, base >> 16); wrptr = dlfb_set_register(wrptr, 0x27, base >> 8); return dlfb_set_register(wrptr, 0x28, base); } static char *dlfb_set_register_16(char *wrptr, u8 reg, u16 value) { wrptr = dlfb_set_register(wrptr, reg, value >> 8); return dlfb_set_register(wrptr, reg+1, value); } /* * This is kind of weird because the controller takes some * register values in a different byte order than other registers. */ static char *dlfb_set_register_16be(char *wrptr, u8 reg, u16 value) { wrptr = dlfb_set_register(wrptr, reg, value); return dlfb_set_register(wrptr, reg+1, value >> 8); } /* * LFSR is linear feedback shift register. The reason we have this is * because the display controller needs to minimize the clock depth of * various counters used in the display path. So this code reverses the * provided value into the lfsr16 value by counting backwards to get * the value that needs to be set in the hardware comparator to get the * same actual count. This makes sense once you read above a couple of * times and think about it from a hardware perspective. */ static u16 dlfb_lfsr16(u16 actual_count) { u32 lv = 0xFFFF; /* This is the lfsr value that the hw starts with */ while (actual_count--) { lv = ((lv << 1) | (((lv >> 15) ^ (lv >> 4) ^ (lv >> 2) ^ (lv >> 1)) & 1)) & 0xFFFF; } return (u16) lv; } /* * This does LFSR conversion on the value that is to be written. * See LFSR explanation above for more detail. */ static char *dlfb_set_register_lfsr16(char *wrptr, u8 reg, u16 value) { return dlfb_set_register_16(wrptr, reg, dlfb_lfsr16(value)); } /* * This takes a standard fbdev screeninfo struct and all of its monitor mode * details and converts them into the DisplayLink equivalent register commands. */ static char *dlfb_set_vid_cmds(char *wrptr, struct fb_var_screeninfo *var) { u16 xds, yds; u16 xde, yde; u16 yec; /* x display start */ xds = var->left_margin + var->hsync_len; wrptr = dlfb_set_register_lfsr16(wrptr, 0x01, xds); /* x display end */ xde = xds + var->xres; wrptr = dlfb_set_register_lfsr16(wrptr, 0x03, xde); /* y display start */ yds = var->upper_margin + var->vsync_len; wrptr = dlfb_set_register_lfsr16(wrptr, 0x05, yds); /* y display end */ yde = yds + var->yres; wrptr = dlfb_set_register_lfsr16(wrptr, 0x07, yde); /* x end count is active + blanking - 1 */ wrptr = dlfb_set_register_lfsr16(wrptr, 0x09, xde + var->right_margin - 1); /* libdlo hardcodes hsync start to 1 */ wrptr = dlfb_set_register_lfsr16(wrptr, 0x0B, 1); /* hsync end is width of sync pulse + 1 */ wrptr = dlfb_set_register_lfsr16(wrptr, 0x0D, var->hsync_len + 1); /* hpixels is active pixels */ wrptr = dlfb_set_register_16(wrptr, 0x0F, var->xres); /* yendcount is vertical active + vertical blanking */ yec = var->yres + var->upper_margin + var->lower_margin + var->vsync_len; wrptr = dlfb_set_register_lfsr16(wrptr, 0x11, yec); /* libdlo hardcodes vsync start to 0 */ wrptr = dlfb_set_register_lfsr16(wrptr, 0x13, 0); /* vsync end is width of vsync pulse */ wrptr = dlfb_set_register_lfsr16(wrptr, 0x15, var->vsync_len); /* vpixels is active pixels */ wrptr = dlfb_set_register_16(wrptr, 0x17, var->yres); /* convert picoseconds to 5kHz multiple for pclk5k = x * 1E12/5k */ wrptr = dlfb_set_register_16be(wrptr, 0x1B, 200*1000*1000/var->pixclock); return wrptr; } /* * This takes a standard fbdev screeninfo struct that was fetched or prepared * and then generates the appropriate command sequence that then drives the * display controller. */ static int dlfb_set_video_mode(struct dlfb_data *dev, struct fb_var_screeninfo *var) { char *buf; char *wrptr; int retval = 0; int writesize; struct urb *urb; if (!atomic_read(&dev->usb_active)) return -EPERM; urb = dlfb_get_urb(dev); if (!urb) return -ENOMEM; buf = (char *) urb->transfer_buffer; /* * This first section has to do with setting the base address on the * controller * associated with the display. There are 2 base * pointers, currently, we only * use the 16 bpp segment. */ wrptr = dlfb_vidreg_lock(buf); wrptr = dlfb_set_color_depth(wrptr, 0x00); /* set base for 16bpp segment to 0 */ wrptr = dlfb_set_base16bpp(wrptr, 0); /* set base for 8bpp segment to end of fb */ wrptr = dlfb_set_base8bpp(wrptr, dev->info->fix.smem_len); wrptr = dlfb_set_vid_cmds(wrptr, var); wrptr = dlfb_enable_hvsync(wrptr, true); wrptr = dlfb_vidreg_unlock(wrptr); writesize = wrptr - buf; retval = dlfb_submit_urb(dev, urb, writesize); return retval; } static int dlfb_ops_mmap(struct fb_info *info, struct vm_area_struct *vma) { unsigned long start = vma->vm_start; unsigned long size = vma->vm_end - vma->vm_start; unsigned long offset = vma->vm_pgoff << PAGE_SHIFT; unsigned long page, pos; struct dlfb_data *dev = info->par; dl_notice("MMAP: %lu %u\n", offset + size, info->fix.smem_len); if (offset + size > info->fix.smem_len) return -EINVAL; pos = (unsigned long)info->fix.smem_start + offset; while (size > 0) { page = vmalloc_to_pfn((void *)pos); if (remap_pfn_range(vma, start, page, PAGE_SIZE, PAGE_SHARED)) return -EAGAIN; start += PAGE_SIZE; pos += PAGE_SIZE; if (size > PAGE_SIZE) size -= PAGE_SIZE; else size = 0; } vma->vm_flags |= VM_RESERVED; /* avoid to swap out this VMA */ return 0; } /* * Trims identical data from front and back of line * Sets new front buffer address and width * And returns byte count of identical pixels * Assumes CPU natural alignment (unsigned long) * for back and front buffer ptrs and width */ static int dlfb_trim_hline(const u8 *bback, const u8 **bfront, int *width_bytes) { int j, k; const unsigned long *back = (const unsigned long *) bback; const unsigned long *front = (const unsigned long *) *bfront; const int width = *width_bytes / sizeof(unsigned long); int identical = width; int start = width; int end = width; prefetch((void *) front); prefetch((void *) back); for (j = 0; j < width; j++) { if (back[j] != front[j]) { start = j; break; } } for (k = width - 1; k > j; k--) { if (back[k] != front[k]) { end = k+1; break; } } identical = start + (width - end); *bfront = (u8 *) &front[start]; *width_bytes = (end - start) * sizeof(unsigned long); return identical * sizeof(unsigned long); } /* * Render a command stream for an encoded horizontal line segment of pixels. * * A command buffer holds several commands. * It always begins with a fresh command header * (the protocol doesn't require this, but we enforce it to allow * multiple buffers to be potentially encoded and sent in parallel). * A single command encodes one contiguous horizontal line of pixels * * The function relies on the client to do all allocation, so that * rendering can be done directly to output buffers (e.g. USB URBs). * The function fills the supplied command buffer, providing information * on where it left off, so the client may call in again with additional * buffers if the line will take several buffers to complete. * * A single command can transmit a maximum of 256 pixels, * regardless of the compression ratio (protocol design limit). * To the hardware, 0 for a size byte means 256 * * Rather than 256 pixel commands which are either rl or raw encoded, * the rlx command simply assumes alternating raw and rl spans within one cmd. * This has a slightly larger header overhead, but produces more even results. * It also processes all data (read and write) in a single pass. * Performance benchmarks of common cases show it having just slightly better * compression than 256 pixel raw -or- rle commands, with similar CPU consumpion. * But for very rl friendly data, will compress not quite as well. */ static void dlfb_compress_hline( const uint16_t **pixel_start_ptr, const uint16_t *const pixel_end, uint32_t *device_address_ptr, uint8_t **command_buffer_ptr, const uint8_t *const cmd_buffer_end) { const uint16_t *pixel = *pixel_start_ptr; uint32_t dev_addr = *device_address_ptr; uint8_t *cmd = *command_buffer_ptr; const int bpp = 2; while ((pixel_end > pixel) && (cmd_buffer_end - MIN_RLX_CMD_BYTES > cmd)) { uint8_t *raw_pixels_count_byte = 0; uint8_t *cmd_pixels_count_byte = 0; const uint16_t *raw_pixel_start = 0; const uint16_t *cmd_pixel_start, *cmd_pixel_end = 0; const uint32_t be_dev_addr = cpu_to_be32(dev_addr); prefetchw((void *) cmd); /* pull in one cache line at least */ *cmd++ = 0xAF; *cmd++ = 0x6B; *cmd++ = (uint8_t) ((be_dev_addr >> 8) & 0xFF); *cmd++ = (uint8_t) ((be_dev_addr >> 16) & 0xFF); *cmd++ = (uint8_t) ((be_dev_addr >> 24) & 0xFF); cmd_pixels_count_byte = cmd++; /* we'll know this later */ cmd_pixel_start = pixel; raw_pixels_count_byte = cmd++; /* we'll know this later */ raw_pixel_start = pixel; cmd_pixel_end = pixel + min(MAX_CMD_PIXELS + 1, min((int)(pixel_end - pixel), (int)(cmd_buffer_end - cmd) / bpp)); prefetch_range((void *) pixel, (cmd_pixel_end - pixel) * bpp); while (pixel < cmd_pixel_end) { const uint16_t * const repeating_pixel = pixel; *(uint16_t *)cmd = cpu_to_be16p(pixel); cmd += 2; pixel++; if (unlikely((pixel < cmd_pixel_end) && (*pixel == *repeating_pixel))) { /* go back and fill in raw pixel count */ *raw_pixels_count_byte = ((repeating_pixel - raw_pixel_start) + 1) & 0xFF; while ((pixel < cmd_pixel_end) && (*pixel == *repeating_pixel)) { pixel++; } /* immediately after raw data is repeat byte */ *cmd++ = ((pixel - repeating_pixel) - 1) & 0xFF; /* Then start another raw pixel span */ raw_pixel_start = pixel; raw_pixels_count_byte = cmd++; } } if (pixel > raw_pixel_start) { /* finalize last RAW span */ *raw_pixels_count_byte = (pixel-raw_pixel_start) & 0xFF; } *cmd_pixels_count_byte = (pixel - cmd_pixel_start) & 0xFF; dev_addr += (pixel - cmd_pixel_start) * bpp; } if (cmd_buffer_end <= MIN_RLX_CMD_BYTES + cmd) { /* Fill leftover bytes with no-ops */ if (cmd_buffer_end > cmd) memset(cmd, 0xAF, cmd_buffer_end - cmd); cmd = (uint8_t *) cmd_buffer_end; } *command_buffer_ptr = cmd; *pixel_start_ptr = pixel; *device_address_ptr = dev_addr; return; } /* * There are 3 copies of every pixel: The front buffer that the fbdev * client renders to, the actual framebuffer across the USB bus in hardware * (that we can only write to, slowly, and can never read), and (optionally) * our shadow copy that tracks what's been sent to that hardware buffer. */ static void dlfb_render_hline(struct dlfb_data *dev, struct urb **urb_ptr, const char *front, char **urb_buf_ptr, u32 byte_offset, u32 byte_width, int *ident_ptr, int *sent_ptr) { const u8 *line_start, *line_end, *next_pixel; u32 dev_addr = dev->base16 + byte_offset; struct urb *urb = *urb_ptr; u8 *cmd = *urb_buf_ptr; u8 *cmd_end = (u8 *) urb->transfer_buffer + urb->transfer_buffer_length; line_start = (u8 *) (front + byte_offset); next_pixel = line_start; line_end = next_pixel + byte_width; if (dev->backing_buffer) { int offset; const u8 *back_start = (u8 *) (dev->backing_buffer + byte_offset); *ident_ptr += dlfb_trim_hline(back_start, &next_pixel, &byte_width); offset = next_pixel - line_start; line_end = next_pixel + byte_width; dev_addr += offset; back_start += offset; line_start += offset; memcpy((char *)back_start, (char *) line_start, byte_width); } while (next_pixel < line_end) { dlfb_compress_hline((const uint16_t **) &next_pixel, (const uint16_t *) line_end, &dev_addr, (u8 **) &cmd, (u8 *) cmd_end); if (cmd >= cmd_end) { int len = cmd - (u8 *) urb->transfer_buffer; if (dlfb_submit_urb(dev, urb, len)) return; /* lost pixels is set */ *sent_ptr += len; urb = dlfb_get_urb(dev); if (!urb) return; /* lost_pixels is set */ *urb_ptr = urb; cmd = urb->transfer_buffer; cmd_end = &cmd[urb->transfer_buffer_length]; } } *urb_buf_ptr = cmd; } int dlfb_handle_damage(struct dlfb_data *dev, int x, int y, int width, int height, char *data) { int i, ret; char *cmd; cycles_t start_cycles, end_cycles; int bytes_sent = 0; int bytes_identical = 0; struct urb *urb; int aligned_x; start_cycles = get_cycles(); aligned_x = DL_ALIGN_DOWN(x, sizeof(unsigned long)); width = DL_ALIGN_UP(width + (x-aligned_x), sizeof(unsigned long)); x = aligned_x; if ((width <= 0) || (x + width > dev->info->var.xres) || (y + height > dev->info->var.yres)) return -EINVAL; if (!atomic_read(&dev->usb_active)) return 0; urb = dlfb_get_urb(dev); if (!urb) return 0; cmd = urb->transfer_buffer; for (i = y; i < y + height ; i++) { const int line_offset = dev->info->fix.line_length * i; const int byte_offset = line_offset + (x * BPP); dlfb_render_hline(dev, &urb, (char *) dev->info->fix.smem_start, &cmd, byte_offset, width * BPP, &bytes_identical, &bytes_sent); } if (cmd > (char *) urb->transfer_buffer) { /* Send partial buffer remaining before exiting */ int len = cmd - (char *) urb->transfer_buffer; ret = dlfb_submit_urb(dev, urb, len); bytes_sent += len; } else dlfb_urb_completion(urb); atomic_add(bytes_sent, &dev->bytes_sent); atomic_add(bytes_identical, &dev->bytes_identical); atomic_add(width*height*2, &dev->bytes_rendered); end_cycles = get_cycles(); atomic_add(((unsigned int) ((end_cycles - start_cycles) >> 10)), /* Kcycles */ &dev->cpu_kcycles_used); return 0; } /* hardware has native COPY command (see libdlo), but not worth it for fbcon */ static void dlfb_ops_copyarea(struct fb_info *info, const struct fb_copyarea *area) { struct dlfb_data *dev = info->par; #if defined CONFIG_FB_SYS_COPYAREA || defined CONFIG_FB_SYS_COPYAREA_MODULE sys_copyarea(info, area); dlfb_handle_damage(dev, area->dx, area->dy, area->width, area->height, info->screen_base); #endif atomic_inc(&dev->copy_count); } static void dlfb_ops_imageblit(struct fb_info *info, const struct fb_image *image) { struct dlfb_data *dev = info->par; #if defined CONFIG_FB_SYS_IMAGEBLIT || defined CONFIG_FB_SYS_IMAGEBLIT_MODULE sys_imageblit(info, image); dlfb_handle_damage(dev, image->dx, image->dy, image->width, image->height, info->screen_base); #endif atomic_inc(&dev->blit_count); } static void dlfb_ops_fillrect(struct fb_info *info, const struct fb_fillrect *rect) { struct dlfb_data *dev = info->par; #if defined CONFIG_FB_SYS_FILLRECT || defined CONFIG_FB_SYS_FILLRECT_MODULE sys_fillrect(info, rect); dlfb_handle_damage(dev, rect->dx, rect->dy, rect->width, rect->height, info->screen_base); #endif atomic_inc(&dev->fill_count); } static void dlfb_get_edid(struct dlfb_data *dev) { int i; int ret; char rbuf[2]; for (i = 0; i < sizeof(dev->edid); i++) { ret = usb_control_msg(dev->udev, usb_rcvctrlpipe(dev->udev, 0), (0x02), (0x80 | (0x02 << 5)), i << 8, 0xA1, rbuf, 2, 0); dev->edid[i] = rbuf[1]; } } static int dlfb_ops_ioctl(struct fb_info *info, unsigned int cmd, unsigned long arg) { struct dlfb_data *dev = info->par; struct dloarea *area = NULL; if (!atomic_read(&dev->usb_active)) return 0; /* TODO: Update X server to get this from sysfs instead */ if (cmd == DLFB_IOCTL_RETURN_EDID) { char *edid = (char *)arg; dlfb_get_edid(dev); if (copy_to_user(edid, dev->edid, sizeof(dev->edid))) return -EFAULT; return 0; } /* TODO: Help propose a standard fb.h ioctl to report mmap damage */ if (cmd == DLFB_IOCTL_REPORT_DAMAGE) { area = (struct dloarea *)arg; if (area->x < 0) area->x = 0; if (area->x > info->var.xres) area->x = info->var.xres; if (area->y < 0) area->y = 0; if (area->y > info->var.yres) area->y = info->var.yres; atomic_set(&dev->use_defio, 0); dlfb_handle_damage(dev, area->x, area->y, area->w, area->h, info->screen_base); atomic_inc(&dev->damage_count); } return 0; } /* taken from vesafb */ static int dlfb_ops_setcolreg(unsigned regno, unsigned red, unsigned green, unsigned blue, unsigned transp, struct fb_info *info) { int err = 0; if (regno >= info->cmap.len) return 1; if (regno < 16) { if (info->var.red.offset == 10) { /* 1:5:5:5 */ ((u32 *) (info->pseudo_palette))[regno] = ((red & 0xf800) >> 1) | ((green & 0xf800) >> 6) | ((blue & 0xf800) >> 11); } else { /* 0:5:6:5 */ ((u32 *) (info->pseudo_palette))[regno] = ((red & 0xf800)) | ((green & 0xfc00) >> 5) | ((blue & 0xf800) >> 11); } } return err; } /* * It's common for several clients to have framebuffer open simultaneously. * e.g. both fbcon and X. Makes things interesting. */ static int dlfb_ops_open(struct fb_info *info, int user) { struct dlfb_data *dev = info->par; /* if (user == 0) * We could special case kernel mode clients (fbcon) here */ mutex_lock(&dev->fb_open_lock); dev->fb_count++; #ifdef CONFIG_FB_DEFERRED_IO if ((atomic_read(&dev->use_defio)) && (info->fbdefio == NULL)) { /* enable defio */ info->fbdefio = &dlfb_defio; fb_deferred_io_init(info); } #endif dl_notice("open /dev/fb%d user=%d fb_info=%p count=%d\n", info->node, user, info, dev->fb_count); mutex_unlock(&dev->fb_open_lock); return 0; } static int dlfb_ops_release(struct fb_info *info, int user) { struct dlfb_data *dev = info->par; mutex_lock(&dev->fb_open_lock); dev->fb_count--; #ifdef CONFIG_FB_DEFERRED_IO if ((dev->fb_count == 0) && (info->fbdefio)) { fb_deferred_io_cleanup(info); info->fbdefio = NULL; info->fbops->fb_mmap = dlfb_ops_mmap; } #endif dl_notice("release /dev/fb%d user=%d count=%d\n", info->node, user, dev->fb_count); mutex_unlock(&dev->fb_open_lock); return 0; } /* * Called when all client interfaces to start transactions have been disabled, * and all references to our device instance (dlfb_data) are released. * Every transaction must have a reference, so we know are fully spun down */ static void dlfb_delete(struct kref *kref) { struct dlfb_data *dev = container_of(kref, struct dlfb_data, kref); if (dev->backing_buffer) vfree(dev->backing_buffer); mutex_destroy(&dev->fb_open_lock); kfree(dev); } /* * Called by fbdev as last part of unregister_framebuffer() process * No new clients can open connections. Deallocate everything fb_info. */ static void dlfb_ops_destroy(struct fb_info *info) { struct dlfb_data *dev = info->par; if (info->cmap.len != 0) fb_dealloc_cmap(&info->cmap); if (info->monspecs.modedb) fb_destroy_modedb(info->monspecs.modedb); if (info->screen_base) vfree(info->screen_base); fb_destroy_modelist(&info->modelist); framebuffer_release(info); /* ref taken before register_framebuffer() for dlfb_data clients */ kref_put(&dev->kref, dlfb_delete); } /* * Check whether a video mode is supported by the DisplayLink chip * We start from monitor's modes, so don't need to filter that here */ static int dlfb_is_valid_mode(struct fb_videomode *mode, struct fb_info *info) { struct dlfb_data *dev = info->par; if (mode->xres * mode->yres > dev->sku_pixel_limit) return 0; return 1; } static void dlfb_var_color_format(struct fb_var_screeninfo *var) { const struct fb_bitfield red = { 11, 5, 0 }; const struct fb_bitfield green = { 5, 6, 0 }; const struct fb_bitfield blue = { 0, 5, 0 }; var->bits_per_pixel = 16; var->red = red; var->green = green; var->blue = blue; } static int dlfb_ops_check_var(struct fb_var_screeninfo *var, struct fb_info *info) { struct fb_videomode mode; /* TODO: support dynamically changing framebuffer size */ if ((var->xres * var->yres * 2) > info->fix.smem_len) return -EINVAL; /* set device-specific elements of var unrelated to mode */ dlfb_var_color_format(var); fb_var_to_videomode(&mode, var); if (!dlfb_is_valid_mode(&mode, info)) return -EINVAL; return 0; } static int dlfb_ops_set_par(struct fb_info *info) { struct dlfb_data *dev = info->par; dl_notice("set_par mode %dx%d\n", info->var.xres, info->var.yres); return dlfb_set_video_mode(dev, &info->var); } static int dlfb_ops_blank(int blank_mode, struct fb_info *info) { struct dlfb_data *dev = info->par; char *bufptr; struct urb *urb; urb = dlfb_get_urb(dev); if (!urb) return 0; bufptr = (char *) urb->transfer_buffer; /* overloading usb_active. UNBLANK can conflict with teardown */ bufptr = dlfb_vidreg_lock(bufptr); if (blank_mode != FB_BLANK_UNBLANK) { atomic_set(&dev->usb_active, 0); bufptr = dlfb_enable_hvsync(bufptr, false); } else { atomic_set(&dev->usb_active, 1); bufptr = dlfb_enable_hvsync(bufptr, true); } bufptr = dlfb_vidreg_unlock(bufptr); dlfb_submit_urb(dev, urb, bufptr - (char *) urb->transfer_buffer); return 0; } static struct fb_ops dlfb_ops = { .owner = THIS_MODULE, .fb_setcolreg = dlfb_ops_setcolreg, .fb_fillrect = dlfb_ops_fillrect, .fb_copyarea = dlfb_ops_copyarea, .fb_imageblit = dlfb_ops_imageblit, .fb_mmap = dlfb_ops_mmap, .fb_ioctl = dlfb_ops_ioctl, .fb_open = dlfb_ops_open, .fb_release = dlfb_ops_release, .fb_blank = dlfb_ops_blank, .fb_check_var = dlfb_ops_check_var, .fb_set_par = dlfb_ops_set_par, }; /* * Calls dlfb_get_edid() to query the EDID of attached monitor via usb cmds * Then parses EDID into three places used by various parts of fbdev: * fb_var_screeninfo contains the timing of the monitor's preferred mode * fb_info.monspecs is full parsed EDID info, including monspecs.modedb * fb_info.modelist is a linked list of all monitor & VESA modes which work * * If EDID is not readable/valid, then modelist is all VESA modes, * monspecs is NULL, and fb_var_screeninfo is set to safe VESA mode * Returns 0 if EDID parses successfully */ static int dlfb_parse_edid(struct dlfb_data *dev, struct fb_var_screeninfo *var, struct fb_info *info) { int i; const struct fb_videomode *default_vmode = NULL; int result = 0; fb_destroy_modelist(&info->modelist); memset(&info->monspecs, 0, sizeof(info->monspecs)); dlfb_get_edid(dev); fb_edid_to_monspecs(dev->edid, &info->monspecs); if (info->monspecs.modedb_len > 0) { for (i = 0; i < info->monspecs.modedb_len; i++) { if (dlfb_is_valid_mode(&info->monspecs.modedb[i], info)) fb_add_videomode(&info->monspecs.modedb[i], &info->modelist); } default_vmode = fb_find_best_display(&info->monspecs, &info->modelist); } else { struct fb_videomode fb_vmode = {0}; dl_err("Unable to get valid EDID from device/display\n"); result = 1; /* * Add the standard VESA modes to our modelist * Since we don't have EDID, there may be modes that * overspec monitor and/or are incorrect aspect ratio, etc. * But at least the user has a chance to choose */ for (i = 0; i < VESA_MODEDB_SIZE; i++) { if (dlfb_is_valid_mode((struct fb_videomode *) &vesa_modes[i], info)) fb_add_videomode(&vesa_modes[i], &info->modelist); } /* * default to resolution safe for projectors * (since they are most common case without EDID) */ fb_vmode.xres = 800; fb_vmode.yres = 600; fb_vmode.refresh = 60; default_vmode = fb_find_nearest_mode(&fb_vmode, &info->modelist); } fb_videomode_to_var(var, default_vmode); dlfb_var_color_format(var); return result; } static ssize_t metrics_bytes_rendered_show(struct device *fbdev, struct device_attribute *a, char *buf) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; return snprintf(buf, PAGE_SIZE, "%u\n", atomic_read(&dev->bytes_rendered)); } static ssize_t metrics_bytes_identical_show(struct device *fbdev, struct device_attribute *a, char *buf) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; return snprintf(buf, PAGE_SIZE, "%u\n", atomic_read(&dev->bytes_identical)); } static ssize_t metrics_bytes_sent_show(struct device *fbdev, struct device_attribute *a, char *buf) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; return snprintf(buf, PAGE_SIZE, "%u\n", atomic_read(&dev->bytes_sent)); } static ssize_t metrics_cpu_kcycles_used_show(struct device *fbdev, struct device_attribute *a, char *buf) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; return snprintf(buf, PAGE_SIZE, "%u\n", atomic_read(&dev->cpu_kcycles_used)); } static ssize_t metrics_misc_show(struct device *fbdev, struct device_attribute *a, char *buf) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; return snprintf(buf, PAGE_SIZE, "Calls to\ndamage: %u\nblit: %u\n" "defio faults: %u\ncopy: %u\n" "fill: %u\n\n" "active framebuffer clients: %d\n" "urbs available %d(%d)\n" "Shadow framebuffer in use? %s\n" "Any lost pixels? %s\n", atomic_read(&dev->damage_count), atomic_read(&dev->blit_count), atomic_read(&dev->defio_fault_count), atomic_read(&dev->copy_count), atomic_read(&dev->fill_count), dev->fb_count, dev->urbs.available, dev->urbs.limit_sem.count, (dev->backing_buffer) ? "yes" : "no", atomic_read(&dev->lost_pixels) ? "yes" : "no"); } static ssize_t edid_show(struct file *filp, struct kobject *kobj, struct bin_attribute *a, char *buf, loff_t off, size_t count) { struct device *fbdev = container_of(kobj, struct device, kobj); struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; char *edid = &dev->edid[0]; const size_t size = sizeof(dev->edid); if (dlfb_parse_edid(dev, &fb_info->var, fb_info)) return 0; if (off >= size) return 0; if (off + count > size) count = size - off; memcpy(buf, edid + off, count); return count; } static ssize_t metrics_reset_store(struct device *fbdev, struct device_attribute *attr, const char *buf, size_t count) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; atomic_set(&dev->bytes_rendered, 0); atomic_set(&dev->bytes_identical, 0); atomic_set(&dev->bytes_sent, 0); atomic_set(&dev->cpu_kcycles_used, 0); atomic_set(&dev->blit_count, 0); atomic_set(&dev->copy_count, 0); atomic_set(&dev->fill_count, 0); atomic_set(&dev->defio_fault_count, 0); atomic_set(&dev->damage_count, 0); return count; } static ssize_t use_defio_show(struct device *fbdev, struct device_attribute *a, char *buf) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; return snprintf(buf, PAGE_SIZE, "%d\n", atomic_read(&dev->use_defio)); } static ssize_t use_defio_store(struct device *fbdev, struct device_attribute *attr, const char *buf, size_t count) { struct fb_info *fb_info = dev_get_drvdata(fbdev); struct dlfb_data *dev = fb_info->par; if (count > 0) { if (buf[0] == '0') atomic_set(&dev->use_defio, 0); if (buf[0] == '1') atomic_set(&dev->use_defio, 1); } return count; } static struct bin_attribute edid_attr = { .attr.name = "edid", .attr.mode = 0444, .size = 128, .read = edid_show, }; static struct device_attribute fb_device_attrs[] = { __ATTR_RO(metrics_bytes_rendered), __ATTR_RO(metrics_bytes_identical), __ATTR_RO(metrics_bytes_sent), __ATTR_RO(metrics_cpu_kcycles_used), __ATTR_RO(metrics_misc), __ATTR(metrics_reset, S_IWUGO, NULL, metrics_reset_store), __ATTR_RW(use_defio), }; #ifdef CONFIG_FB_DEFERRED_IO static void dlfb_dpy_deferred_io(struct fb_info *info, struct list_head *pagelist) { struct page *cur; struct fb_deferred_io *fbdefio = info->fbdefio; struct dlfb_data *dev = info->par; struct urb *urb; char *cmd; cycles_t start_cycles, end_cycles; int bytes_sent = 0; int bytes_identical = 0; int bytes_rendered = 0; int fault_count = 0; if (!atomic_read(&dev->use_defio)) return; if (!atomic_read(&dev->usb_active)) return; start_cycles = get_cycles(); urb = dlfb_get_urb(dev); if (!urb) return; cmd = urb->transfer_buffer; /* walk the written page list and render each to device */ list_for_each_entry(cur, &fbdefio->pagelist, lru) { dlfb_render_hline(dev, &urb, (char *) info->fix.smem_start, &cmd, cur->index << PAGE_SHIFT, PAGE_SIZE, &bytes_identical, &bytes_sent); bytes_rendered += PAGE_SIZE; fault_count++; } if (cmd > (char *) urb->transfer_buffer) { /* Send partial buffer remaining before exiting */ int len = cmd - (char *) urb->transfer_buffer; dlfb_submit_urb(dev, urb, len); bytes_sent += len; } else dlfb_urb_completion(urb); atomic_add(fault_count, &dev->defio_fault_count); atomic_add(bytes_sent, &dev->bytes_sent); atomic_add(bytes_identical, &dev->bytes_identical); atomic_add(bytes_rendered, &dev->bytes_rendered); end_cycles = get_cycles(); atomic_add(((unsigned int) ((end_cycles - start_cycles) >> 10)), /* Kcycles */ &dev->cpu_kcycles_used); } static struct fb_deferred_io dlfb_defio = { .delay = 5, .deferred_io = dlfb_dpy_deferred_io, }; #endif /* * This is necessary before we can communicate with the display controller. */ static int dlfb_select_std_channel(struct dlfb_data *dev) { int ret; u8 set_def_chn[] = { 0x57, 0xCD, 0xDC, 0xA7, 0x1C, 0x88, 0x5E, 0x15, 0x60, 0xFE, 0xC6, 0x97, 0x16, 0x3D, 0x47, 0xF2 }; ret = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0), NR_USB_REQUEST_CHANNEL, (USB_DIR_OUT | USB_TYPE_VENDOR), 0, 0, set_def_chn, sizeof(set_def_chn), USB_CTRL_SET_TIMEOUT); return ret; } static int dlfb_usb_probe(struct usb_interface *interface, const struct usb_device_id *id) { struct usb_device *usbdev; struct dlfb_data *dev; struct fb_info *info; int videomemorysize; int i; unsigned char *videomemory; int retval = -ENOMEM; struct fb_var_screeninfo *var; int registered = 0; u16 *pix_framebuffer; /* usb initialization */ usbdev = interface_to_usbdev(interface); dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (dev == NULL) { err("dlfb_usb_probe: failed alloc of dev struct\n"); goto error; } /* we need to wait for both usb and fbdev to spin down on disconnect */ kref_init(&dev->kref); /* matching kref_put in usb .disconnect fn */ kref_get(&dev->kref); /* matching kref_put in .fb_destroy function*/ dev->udev = usbdev; dev->gdev = &usbdev->dev; /* our generic struct device * */ usb_set_intfdata(interface, dev); if (!dlfb_alloc_urb_list(dev, WRITES_IN_FLIGHT, MAX_TRANSFER)) { retval = -ENOMEM; dl_err("dlfb_alloc_urb_list failed\n"); goto error; } mutex_init(&dev->fb_open_lock); /* We don't register a new USB class. Our client interface is fbdev */ /* allocates framebuffer driver structure, not framebuffer memory */ info = framebuffer_alloc(0, &usbdev->dev); if (!info) { retval = -ENOMEM; dl_err("framebuffer_alloc failed\n"); goto error; } dev->info = info; info->par = dev; info->pseudo_palette = dev->pseudo_palette; info->fbops = &dlfb_ops; var = &info->var; /* TODO set limit based on actual SKU detection */ dev->sku_pixel_limit = 2048 * 1152; INIT_LIST_HEAD(&info->modelist); dlfb_parse_edid(dev, var, info); /* * ok, now that we've got the size info, we can alloc our framebuffer. */ info->fix = dlfb_fix; info->fix.line_length = var->xres * (var->bits_per_pixel / 8); videomemorysize = info->fix.line_length * var->yres; /* * The big chunk of system memory we use as a virtual framebuffer. * TODO: Handle fbcon cursor code calling blit in interrupt context */ videomemory = vmalloc(videomemorysize); if (!videomemory) { retval = -ENOMEM; dl_err("Virtual framebuffer alloc failed\n"); goto error; } info->screen_base = videomemory; info->fix.smem_len = PAGE_ALIGN(videomemorysize); info->fix.smem_start = (unsigned long) videomemory; info->flags = udlfb_info_flags; /* * Second framebuffer copy, mirroring the state of the framebuffer * on the physical USB device. We can function without this. * But with imperfect damage info we may end up sending pixels over USB * that were, in fact, unchanged -- wasting limited USB bandwidth */ dev->backing_buffer = vmalloc(videomemorysize); if (!dev->backing_buffer) dl_warn("No shadow/backing buffer allcoated\n"); else memset(dev->backing_buffer, 0, videomemorysize); retval = fb_alloc_cmap(&info->cmap, 256, 0); if (retval < 0) { dl_err("fb_alloc_cmap failed %x\n", retval); goto error; } /* ready to begin using device */ #ifdef CONFIG_FB_DEFERRED_IO atomic_set(&dev->use_defio, 1); #endif atomic_set(&dev->usb_active, 1); dlfb_select_std_channel(dev); dlfb_ops_check_var(var, info); dlfb_ops_set_par(info); /* paint greenscreen */ pix_framebuffer = (u16 *) videomemory; for (i = 0; i < videomemorysize / 2; i++) pix_framebuffer[i] = 0x37e6; dlfb_handle_damage(dev, 0, 0, info->var.xres, info->var.yres, videomemory); retval = register_framebuffer(info); if (retval < 0) { dl_err("register_framebuffer failed %d\n", retval); goto error; } registered = 1; for (i = 0; i < ARRAY_SIZE(fb_device_attrs); i++) device_create_file(info->dev, &fb_device_attrs[i]); device_create_bin_file(info->dev, &edid_attr); dl_err("DisplayLink USB device /dev/fb%d attached. %dx%d resolution." " Using %dK framebuffer memory\n", info->node, var->xres, var->yres, ((dev->backing_buffer) ? videomemorysize * 2 : videomemorysize) >> 10); return 0; error: if (dev) { if (registered) { unregister_framebuffer(info); dlfb_ops_destroy(info); } else kref_put(&dev->kref, dlfb_delete); if (dev->urbs.count > 0) dlfb_free_urb_list(dev); kref_put(&dev->kref, dlfb_delete); /* last ref from kref_init */ /* dev has been deallocated. Do not dereference */ } return retval; } static void dlfb_usb_disconnect(struct usb_interface *interface) { struct dlfb_data *dev; struct fb_info *info; int i; dev = usb_get_intfdata(interface); info = dev->info; /* when non-active we'll update virtual framebuffer, but no new urbs */ atomic_set(&dev->usb_active, 0); usb_set_intfdata(interface, NULL); for (i = 0; i < ARRAY_SIZE(fb_device_attrs); i++) device_remove_file(info->dev, &fb_device_attrs[i]); device_remove_bin_file(info->dev, &edid_attr); /* this function will wait for all in-flight urbs to complete */ dlfb_free_urb_list(dev); if (info) { dl_notice("Detaching /dev/fb%d\n", info->node); unregister_framebuffer(info); dlfb_ops_destroy(info); } /* release reference taken by kref_init in probe() */ kref_put(&dev->kref, dlfb_delete); /* consider dlfb_data freed */ return; } static struct usb_driver dlfb_driver = { .name = "udlfb", .probe = dlfb_usb_probe, .disconnect = dlfb_usb_disconnect, .id_table = id_table, }; static int __init dlfb_module_init(void) { int res; res = usb_register(&dlfb_driver); if (res) err("usb_register failed. Error number %d", res); printk(KERN_INFO "VMODES initialized\n"); return res; } static void __exit dlfb_module_exit(void) { usb_deregister(&dlfb_driver); } module_init(dlfb_module_init); module_exit(dlfb_module_exit); static void dlfb_urb_completion(struct urb *urb) { struct urb_node *unode = urb->context; struct dlfb_data *dev = unode->dev; unsigned long flags; /* sync/async unlink faults aren't errors */ if (urb->status) { if (!(urb->status == -ENOENT || urb->status == -ECONNRESET || urb->status == -ESHUTDOWN)) { dl_err("%s - nonzero write bulk status received: %d\n", __func__, urb->status); atomic_set(&dev->lost_pixels, 1); } } urb->transfer_buffer_length = dev->urbs.size; /* reset to actual */ spin_lock_irqsave(&dev->urbs.lock, flags); list_add_tail(&unode->entry, &dev->urbs.list); dev->urbs.available++; spin_unlock_irqrestore(&dev->urbs.lock, flags); up(&dev->urbs.limit_sem); } static void dlfb_free_urb_list(struct dlfb_data *dev) { int count = dev->urbs.count; struct list_head *node; struct urb_node *unode; struct urb *urb; int ret; unsigned long flags; dl_notice("Waiting for completes and freeing all render urbs\n"); /* keep waiting and freeing, until we've got 'em all */ while (count--) { /* Timeout means a memory leak and/or fault */ ret = down_timeout(&dev->urbs.limit_sem, FREE_URB_TIMEOUT); if (ret) { BUG_ON(ret); break; } spin_lock_irqsave(&dev->urbs.lock, flags); node = dev->urbs.list.next; /* have reserved one with sem */ list_del_init(node); spin_unlock_irqrestore(&dev->urbs.lock, flags); unode = list_entry(node, struct urb_node, entry); urb = unode->urb; /* Free each separately allocated piece */ usb_free_coherent(urb->dev, dev->urbs.size, urb->transfer_buffer, urb->transfer_dma); usb_free_urb(urb); kfree(node); } kref_put(&dev->kref, dlfb_delete); } static int dlfb_alloc_urb_list(struct dlfb_data *dev, int count, size_t size) { int i = 0; struct urb *urb; struct urb_node *unode; char *buf; spin_lock_init(&dev->urbs.lock); dev->urbs.size = size; INIT_LIST_HEAD(&dev->urbs.list); while (i < count) { unode = kzalloc(sizeof(struct urb_node), GFP_KERNEL); if (!unode) break; unode->dev = dev; urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { kfree(unode); break; } unode->urb = urb; buf = usb_alloc_coherent(dev->udev, MAX_TRANSFER, GFP_KERNEL, &urb->transfer_dma); if (!buf) { kfree(unode); usb_free_urb(urb); break; } /* urb->transfer_buffer_length set to actual before submit */ usb_fill_bulk_urb(urb, dev->udev, usb_sndbulkpipe(dev->udev, 1), buf, size, dlfb_urb_completion, unode); urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; list_add_tail(&unode->entry, &dev->urbs.list); i++; } sema_init(&dev->urbs.limit_sem, i); dev->urbs.count = i; dev->urbs.available = i; kref_get(&dev->kref); /* released in free_render_urbs() */ dl_notice("allocated %d %d byte urbs\n", i, (int) size); return i; } static struct urb *dlfb_get_urb(struct dlfb_data *dev) { int ret = 0; struct list_head *entry; struct urb_node *unode; struct urb *urb = NULL; unsigned long flags; /* Wait for an in-flight buffer to complete and get re-queued */ ret = down_timeout(&dev->urbs.limit_sem, GET_URB_TIMEOUT); if (ret) { atomic_set(&dev->lost_pixels, 1); dl_err("wait for urb interrupted: %x\n", ret); goto error; } spin_lock_irqsave(&dev->urbs.lock, flags); BUG_ON(list_empty(&dev->urbs.list)); /* reserved one with limit_sem */ entry = dev->urbs.list.next; list_del_init(entry); dev->urbs.available--; spin_unlock_irqrestore(&dev->urbs.lock, flags); unode = list_entry(entry, struct urb_node, entry); urb = unode->urb; error: return urb; } static int dlfb_submit_urb(struct dlfb_data *dev, struct urb *urb, size_t len) { int ret; BUG_ON(len > dev->urbs.size); urb->transfer_buffer_length = len; /* set to actual payload len */ ret = usb_submit_urb(urb, GFP_KERNEL); if (ret) { dlfb_urb_completion(urb); /* because no one else will */ atomic_set(&dev->lost_pixels, 1); dl_err("usb_submit_urb error %x\n", ret); } return ret; } MODULE_AUTHOR("Roberto De Ioris , " "Jaya Kumar , " "Bernie Thompson "); MODULE_DESCRIPTION("DisplayLink kernel framebuffer driver"); MODULE_LICENSE("GPL");