linux/drivers/gpu/drm/radeon/radeon_fb.c

837 lines
22 KiB
C
Raw Normal View History

drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
/*
* Copyright © 2007 David Airlie
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*
* Authors:
* David Airlie
*/
/*
* Modularization
*/
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/tty.h>
#include <linux/slab.h>
#include <linux/delay.h>
#include <linux/fb.h>
#include <linux/init.h>
#include "drmP.h"
#include "drm.h"
#include "drm_crtc.h"
#include "drm_crtc_helper.h"
#include "radeon_drm.h"
#include "radeon.h"
struct radeon_fb_device {
struct radeon_device *rdev;
struct drm_display_mode *mode;
struct radeon_framebuffer *rfb;
int crtc_count;
/* crtc currently bound to this */
uint32_t crtc_ids[2];
};
static int radeonfb_setcolreg(unsigned regno,
unsigned red,
unsigned green,
unsigned blue,
unsigned transp,
struct fb_info *info)
{
struct radeon_fb_device *rfbdev = info->par;
struct drm_device *dev = rfbdev->rdev->ddev;
struct drm_crtc *crtc;
int i;
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
struct drm_mode_set *modeset = &radeon_crtc->mode_set;
struct drm_framebuffer *fb = modeset->fb;
for (i = 0; i < rfbdev->crtc_count; i++) {
if (crtc->base.id == rfbdev->crtc_ids[i]) {
break;
}
}
if (i == rfbdev->crtc_count) {
continue;
}
if (regno > 255) {
return 1;
}
if (fb->depth == 8) {
radeon_crtc_fb_gamma_set(crtc, red, green, blue, regno);
return 0;
}
if (regno < 16) {
switch (fb->depth) {
case 15:
fb->pseudo_palette[regno] = ((red & 0xf800) >> 1) |
((green & 0xf800) >> 6) |
((blue & 0xf800) >> 11);
break;
case 16:
fb->pseudo_palette[regno] = (red & 0xf800) |
((green & 0xfc00) >> 5) |
((blue & 0xf800) >> 11);
break;
case 24:
case 32:
fb->pseudo_palette[regno] = ((red & 0xff00) << 8) |
(green & 0xff00) |
((blue & 0xff00) >> 8);
break;
}
}
}
return 0;
}
static int radeonfb_check_var(struct fb_var_screeninfo *var,
struct fb_info *info)
{
struct radeon_fb_device *rfbdev = info->par;
struct radeon_framebuffer *rfb = rfbdev->rfb;
struct drm_framebuffer *fb = &rfb->base;
int depth;
if (var->pixclock == -1 || !var->pixclock) {
return -EINVAL;
}
/* Need to resize the fb object !!! */
if (var->xres > fb->width || var->yres > fb->height) {
DRM_ERROR("Requested width/height is greater than current fb "
"object %dx%d > %dx%d\n", var->xres, var->yres,
fb->width, fb->height);
DRM_ERROR("Need resizing code.\n");
return -EINVAL;
}
switch (var->bits_per_pixel) {
case 16:
depth = (var->green.length == 6) ? 16 : 15;
break;
case 32:
depth = (var->transp.length > 0) ? 32 : 24;
break;
default:
depth = var->bits_per_pixel;
break;
}
switch (depth) {
case 8:
var->red.offset = 0;
var->green.offset = 0;
var->blue.offset = 0;
var->red.length = 8;
var->green.length = 8;
var->blue.length = 8;
var->transp.length = 0;
var->transp.offset = 0;
break;
case 15:
var->red.offset = 10;
var->green.offset = 5;
var->blue.offset = 0;
var->red.length = 5;
var->green.length = 5;
var->blue.length = 5;
var->transp.length = 1;
var->transp.offset = 15;
break;
case 16:
var->red.offset = 11;
var->green.offset = 5;
var->blue.offset = 0;
var->red.length = 5;
var->green.length = 6;
var->blue.length = 5;
var->transp.length = 0;
var->transp.offset = 0;
break;
case 24:
var->red.offset = 16;
var->green.offset = 8;
var->blue.offset = 0;
var->red.length = 8;
var->green.length = 8;
var->blue.length = 8;
var->transp.length = 0;
var->transp.offset = 0;
break;
case 32:
var->red.offset = 16;
var->green.offset = 8;
var->blue.offset = 0;
var->red.length = 8;
var->green.length = 8;
var->blue.length = 8;
var->transp.length = 8;
var->transp.offset = 24;
break;
default:
return -EINVAL;
}
return 0;
}
/* this will let fbcon do the mode init */
static int radeonfb_set_par(struct fb_info *info)
{
struct radeon_fb_device *rfbdev = info->par;
struct drm_device *dev = rfbdev->rdev->ddev;
struct fb_var_screeninfo *var = &info->var;
struct drm_crtc *crtc;
int ret;
int i;
if (var->pixclock != -1) {
DRM_ERROR("PIXEL CLCOK SET\n");
return -EINVAL;
}
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
for (i = 0; i < rfbdev->crtc_count; i++) {
if (crtc->base.id == rfbdev->crtc_ids[i]) {
break;
}
}
if (i == rfbdev->crtc_count) {
continue;
}
if (crtc->fb == radeon_crtc->mode_set.fb) {
mutex_lock(&dev->mode_config.mutex);
ret = crtc->funcs->set_config(&radeon_crtc->mode_set);
mutex_unlock(&dev->mode_config.mutex);
if (ret) {
return ret;
}
}
}
return 0;
}
static int radeonfb_pan_display(struct fb_var_screeninfo *var,
struct fb_info *info)
{
struct radeon_fb_device *rfbdev = info->par;
struct drm_device *dev = rfbdev->rdev->ddev;
struct drm_mode_set *modeset;
struct drm_crtc *crtc;
struct radeon_crtc *radeon_crtc;
int ret = 0;
int i;
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
for (i = 0; i < rfbdev->crtc_count; i++) {
if (crtc->base.id == rfbdev->crtc_ids[i]) {
break;
}
}
if (i == rfbdev->crtc_count) {
continue;
}
radeon_crtc = to_radeon_crtc(crtc);
modeset = &radeon_crtc->mode_set;
modeset->x = var->xoffset;
modeset->y = var->yoffset;
if (modeset->num_connectors) {
mutex_lock(&dev->mode_config.mutex);
ret = crtc->funcs->set_config(modeset);
mutex_unlock(&dev->mode_config.mutex);
if (!ret) {
info->var.xoffset = var->xoffset;
info->var.yoffset = var->yoffset;
}
}
}
return ret;
}
static void radeonfb_on(struct fb_info *info)
{
struct radeon_fb_device *rfbdev = info->par;
struct drm_device *dev = rfbdev->rdev->ddev;
struct drm_crtc *crtc;
struct drm_encoder *encoder;
int i;
/*
* For each CRTC in this fb, find all associated encoders
* and turn them off, then turn off the CRTC.
*/
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private;
for (i = 0; i < rfbdev->crtc_count; i++) {
if (crtc->base.id == rfbdev->crtc_ids[i]) {
break;
}
}
mutex_lock(&dev->mode_config.mutex);
crtc_funcs->dpms(crtc, DRM_MODE_DPMS_ON);
mutex_unlock(&dev->mode_config.mutex);
/* Found a CRTC on this fb, now find encoders */
list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
if (encoder->crtc == crtc) {
struct drm_encoder_helper_funcs *encoder_funcs;
encoder_funcs = encoder->helper_private;
mutex_lock(&dev->mode_config.mutex);
encoder_funcs->dpms(encoder, DRM_MODE_DPMS_ON);
mutex_unlock(&dev->mode_config.mutex);
}
}
}
}
static void radeonfb_off(struct fb_info *info, int dpms_mode)
{
struct radeon_fb_device *rfbdev = info->par;
struct drm_device *dev = rfbdev->rdev->ddev;
struct drm_crtc *crtc;
struct drm_encoder *encoder;
int i;
/*
* For each CRTC in this fb, find all associated encoders
* and turn them off, then turn off the CRTC.
*/
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private;
for (i = 0; i < rfbdev->crtc_count; i++) {
if (crtc->base.id == rfbdev->crtc_ids[i]) {
break;
}
}
/* Found a CRTC on this fb, now find encoders */
list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
if (encoder->crtc == crtc) {
struct drm_encoder_helper_funcs *encoder_funcs;
encoder_funcs = encoder->helper_private;
mutex_lock(&dev->mode_config.mutex);
encoder_funcs->dpms(encoder, dpms_mode);
mutex_unlock(&dev->mode_config.mutex);
}
}
if (dpms_mode == DRM_MODE_DPMS_OFF) {
mutex_lock(&dev->mode_config.mutex);
crtc_funcs->dpms(crtc, dpms_mode);
mutex_unlock(&dev->mode_config.mutex);
}
}
}
int radeonfb_blank(int blank, struct fb_info *info)
{
switch (blank) {
case FB_BLANK_UNBLANK:
radeonfb_on(info);
break;
case FB_BLANK_NORMAL:
radeonfb_off(info, DRM_MODE_DPMS_STANDBY);
break;
case FB_BLANK_HSYNC_SUSPEND:
radeonfb_off(info, DRM_MODE_DPMS_STANDBY);
break;
case FB_BLANK_VSYNC_SUSPEND:
radeonfb_off(info, DRM_MODE_DPMS_SUSPEND);
break;
case FB_BLANK_POWERDOWN:
radeonfb_off(info, DRM_MODE_DPMS_OFF);
break;
}
return 0;
}
static struct fb_ops radeonfb_ops = {
.owner = THIS_MODULE,
.fb_check_var = radeonfb_check_var,
.fb_set_par = radeonfb_set_par,
.fb_setcolreg = radeonfb_setcolreg,
.fb_fillrect = cfb_fillrect,
.fb_copyarea = cfb_copyarea,
.fb_imageblit = cfb_imageblit,
.fb_pan_display = radeonfb_pan_display,
.fb_blank = radeonfb_blank,
};
/**
* Curretly it is assumed that the old framebuffer is reused.
*
* LOCKING
* caller should hold the mode config lock.
*
*/
int radeonfb_resize(struct drm_device *dev, struct drm_crtc *crtc)
{
struct fb_info *info;
struct drm_framebuffer *fb;
struct drm_display_mode *mode = crtc->desired_mode;
fb = crtc->fb;
if (fb == NULL) {
return 1;
}
info = fb->fbdev;
if (info == NULL) {
return 1;
}
if (mode == NULL) {
return 1;
}
info->var.xres = mode->hdisplay;
info->var.right_margin = mode->hsync_start - mode->hdisplay;
info->var.hsync_len = mode->hsync_end - mode->hsync_start;
info->var.left_margin = mode->htotal - mode->hsync_end;
info->var.yres = mode->vdisplay;
info->var.lower_margin = mode->vsync_start - mode->vdisplay;
info->var.vsync_len = mode->vsync_end - mode->vsync_start;
info->var.upper_margin = mode->vtotal - mode->vsync_end;
info->var.pixclock = 10000000 / mode->htotal * 1000 / mode->vtotal * 100;
/* avoid overflow */
info->var.pixclock = info->var.pixclock * 1000 / mode->vrefresh;
return 0;
}
EXPORT_SYMBOL(radeonfb_resize);
static struct drm_mode_set panic_mode;
int radeonfb_panic(struct notifier_block *n, unsigned long ununsed,
void *panic_str)
{
DRM_ERROR("panic occurred, switching back to text console\n");
drm_crtc_helper_set_config(&panic_mode);
return 0;
}
EXPORT_SYMBOL(radeonfb_panic);
static struct notifier_block paniced = {
.notifier_call = radeonfb_panic,
};
static int radeon_align_pitch(struct radeon_device *rdev, int width, int bpp)
{
int aligned = width;
int align_large = (ASIC_IS_AVIVO(rdev));
int pitch_mask = 0;
switch (bpp / 8) {
case 1:
pitch_mask = align_large ? 255 : 127;
break;
case 2:
pitch_mask = align_large ? 127 : 31;
break;
case 3:
case 4:
pitch_mask = align_large ? 63 : 15;
break;
}
aligned += pitch_mask;
aligned &= ~pitch_mask;
return aligned;
}
int radeonfb_create(struct radeon_device *rdev,
uint32_t fb_width, uint32_t fb_height,
uint32_t surface_width, uint32_t surface_height,
struct radeon_framebuffer **rfb_p)
{
struct fb_info *info;
struct radeon_fb_device *rfbdev;
struct drm_framebuffer *fb = NULL;
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
struct radeon_framebuffer *rfb;
struct drm_mode_fb_cmd mode_cmd;
struct drm_gem_object *gobj = NULL;
struct radeon_object *robj = NULL;
struct device *device = &rdev->pdev->dev;
int size, aligned_size, ret;
u64 fb_gpuaddr;
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
void *fbptr = NULL;
unsigned long tmp;
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
mode_cmd.width = surface_width;
mode_cmd.height = surface_height;
mode_cmd.bpp = 32;
/* need to align pitch with crtc limits */
mode_cmd.pitch = radeon_align_pitch(rdev, mode_cmd.width, mode_cmd.bpp) * ((mode_cmd.bpp + 1) / 8);
mode_cmd.depth = 24;
size = mode_cmd.pitch * mode_cmd.height;
aligned_size = ALIGN(size, PAGE_SIZE);
ret = radeon_gem_object_create(rdev, aligned_size, 0,
RADEON_GEM_DOMAIN_VRAM,
false, ttm_bo_type_kernel,
false, &gobj);
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
if (ret) {
printk(KERN_ERR "failed to allocate framebuffer (%d %d)\n",
surface_width, surface_height);
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
ret = -ENOMEM;
goto out;
}
robj = gobj->driver_private;
mutex_lock(&rdev->ddev->struct_mutex);
fb = radeon_framebuffer_create(rdev->ddev, &mode_cmd, gobj);
if (fb == NULL) {
DRM_ERROR("failed to allocate fb.\n");
ret = -ENOMEM;
goto out_unref;
}
ret = radeon_object_pin(robj, RADEON_GEM_DOMAIN_VRAM, &fb_gpuaddr);
if (ret) {
printk(KERN_ERR "failed to pin framebuffer\n");
ret = -ENOMEM;
goto out_unref;
}
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
list_add(&fb->filp_head, &rdev->ddev->mode_config.fb_kernel_list);
rfb = to_radeon_framebuffer(fb);
*rfb_p = rfb;
rdev->fbdev_rfb = rfb;
rdev->fbdev_robj = robj;
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
info = framebuffer_alloc(sizeof(struct radeon_fb_device), device);
if (info == NULL) {
ret = -ENOMEM;
goto out_unref;
}
rfbdev = info->par;
ret = radeon_object_kmap(robj, &fbptr);
if (ret) {
goto out_unref;
}
strcpy(info->fix.id, "radeondrmfb");
info->fix.type = FB_TYPE_PACKED_PIXELS;
info->fix.visual = FB_VISUAL_TRUECOLOR;
info->fix.type_aux = 0;
info->fix.xpanstep = 1; /* doing it in hw */
info->fix.ypanstep = 1; /* doing it in hw */
info->fix.ywrapstep = 0;
info->fix.accel = FB_ACCEL_I830;
info->fix.type_aux = 0;
info->flags = FBINFO_DEFAULT;
info->fbops = &radeonfb_ops;
info->fix.line_length = fb->pitch;
tmp = fb_gpuaddr - rdev->mc.vram_location;
info->fix.smem_start = rdev->mc.aper_base + tmp;
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
info->fix.smem_len = size;
info->screen_base = fbptr;
info->screen_size = size;
info->pseudo_palette = fb->pseudo_palette;
info->var.xres_virtual = fb->width;
info->var.yres_virtual = fb->height;
info->var.bits_per_pixel = fb->bits_per_pixel;
info->var.xoffset = 0;
info->var.yoffset = 0;
info->var.activate = FB_ACTIVATE_NOW;
info->var.height = -1;
info->var.width = -1;
info->var.xres = fb_width;
info->var.yres = fb_height;
info->fix.mmio_start = pci_resource_start(rdev->pdev, 2);
info->fix.mmio_len = pci_resource_len(rdev->pdev, 2);
info->pixmap.size = 64*1024;
info->pixmap.buf_align = 8;
info->pixmap.access_align = 32;
info->pixmap.flags = FB_PIXMAP_SYSTEM;
info->pixmap.scan_align = 1;
if (info->screen_base == NULL) {
ret = -ENOSPC;
goto out_unref;
}
DRM_INFO("fb mappable at 0x%lX\n", info->fix.smem_start);
DRM_INFO("vram apper at 0x%lX\n", (unsigned long)rdev->mc.aper_base);
DRM_INFO("size %lu\n", (unsigned long)size);
DRM_INFO("fb depth is %d\n", fb->depth);
DRM_INFO(" pitch is %d\n", fb->pitch);
switch (fb->depth) {
case 8:
info->var.red.offset = 0;
info->var.green.offset = 0;
info->var.blue.offset = 0;
info->var.red.length = 8; /* 8bit DAC */
info->var.green.length = 8;
info->var.blue.length = 8;
info->var.transp.offset = 0;
info->var.transp.length = 0;
break;
case 15:
info->var.red.offset = 10;
info->var.green.offset = 5;
info->var.blue.offset = 0;
info->var.red.length = 5;
info->var.green.length = 5;
info->var.blue.length = 5;
info->var.transp.offset = 15;
info->var.transp.length = 1;
break;
case 16:
info->var.red.offset = 11;
info->var.green.offset = 5;
info->var.blue.offset = 0;
info->var.red.length = 5;
info->var.green.length = 6;
info->var.blue.length = 5;
info->var.transp.offset = 0;
break;
case 24:
info->var.red.offset = 16;
info->var.green.offset = 8;
info->var.blue.offset = 0;
info->var.red.length = 8;
info->var.green.length = 8;
info->var.blue.length = 8;
info->var.transp.offset = 0;
info->var.transp.length = 0;
break;
case 32:
info->var.red.offset = 16;
info->var.green.offset = 8;
info->var.blue.offset = 0;
info->var.red.length = 8;
info->var.green.length = 8;
info->var.blue.length = 8;
info->var.transp.offset = 24;
info->var.transp.length = 8;
break;
default:
break;
}
fb->fbdev = info;
rfbdev->rfb = rfb;
rfbdev->rdev = rdev;
mutex_unlock(&rdev->ddev->struct_mutex);
return 0;
out_unref:
if (robj) {
radeon_object_kunmap(robj);
}
if (fb && ret) {
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
list_del(&fb->filp_head);
drm_gem_object_unreference(gobj);
drm_framebuffer_cleanup(fb);
kfree(fb);
}
drm_gem_object_unreference(gobj);
mutex_unlock(&rdev->ddev->struct_mutex);
out:
return ret;
}
static int radeonfb_single_fb_probe(struct radeon_device *rdev)
{
struct drm_crtc *crtc;
struct drm_connector *connector;
unsigned int fb_width = (unsigned)-1, fb_height = (unsigned)-1;
unsigned int surface_width = 0, surface_height = 0;
int new_fb = 0;
int crtc_count = 0;
int ret, i, conn_count = 0;
struct radeon_framebuffer *rfb;
struct fb_info *info;
struct radeon_fb_device *rfbdev;
struct drm_mode_set *modeset = NULL;
/* first up get a count of crtcs now in use and new min/maxes width/heights */
list_for_each_entry(crtc, &rdev->ddev->mode_config.crtc_list, head) {
if (drm_helper_crtc_in_use(crtc)) {
if (crtc->desired_mode) {
if (crtc->desired_mode->hdisplay < fb_width)
fb_width = crtc->desired_mode->hdisplay;
if (crtc->desired_mode->vdisplay < fb_height)
fb_height = crtc->desired_mode->vdisplay;
if (crtc->desired_mode->hdisplay > surface_width)
surface_width = crtc->desired_mode->hdisplay;
if (crtc->desired_mode->vdisplay > surface_height)
surface_height = crtc->desired_mode->vdisplay;
}
crtc_count++;
}
}
if (crtc_count == 0 || fb_width == -1 || fb_height == -1) {
/* hmm everyone went away - assume VGA cable just fell out
and will come back later. */
return 0;
}
/* do we have an fb already? */
if (list_empty(&rdev->ddev->mode_config.fb_kernel_list)) {
/* create an fb if we don't have one */
ret = radeonfb_create(rdev, fb_width, fb_height, surface_width, surface_height, &rfb);
if (ret) {
return -EINVAL;
}
new_fb = 1;
} else {
struct drm_framebuffer *fb;
fb = list_first_entry(&rdev->ddev->mode_config.fb_kernel_list, struct drm_framebuffer, filp_head);
rfb = to_radeon_framebuffer(fb);
/* if someone hotplugs something bigger than we have already allocated, we are pwned.
As really we can't resize an fbdev that is in the wild currently due to fbdev
not really being designed for the lower layers moving stuff around under it.
- so in the grand style of things - punt. */
if ((fb->width < surface_width) || (fb->height < surface_height)) {
DRM_ERROR("Framebuffer not large enough to scale console onto.\n");
return -EINVAL;
}
}
info = rfb->base.fbdev;
rdev->fbdev_info = info;
rfbdev = info->par;
crtc_count = 0;
/* okay we need to setup new connector sets in the crtcs */
list_for_each_entry(crtc, &rdev->ddev->mode_config.crtc_list, head) {
struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
modeset = &radeon_crtc->mode_set;
modeset->fb = &rfb->base;
conn_count = 0;
list_for_each_entry(connector, &rdev->ddev->mode_config.connector_list, head) {
if (connector->encoder)
if (connector->encoder->crtc == modeset->crtc) {
modeset->connectors[conn_count] = connector;
conn_count++;
if (conn_count > RADEONFB_CONN_LIMIT)
BUG();
}
}
for (i = conn_count; i < RADEONFB_CONN_LIMIT; i++)
modeset->connectors[i] = NULL;
rfbdev->crtc_ids[crtc_count++] = crtc->base.id;
modeset->num_connectors = conn_count;
if (modeset->crtc->desired_mode) {
if (modeset->mode) {
drm_mode_destroy(rdev->ddev, modeset->mode);
}
modeset->mode = drm_mode_duplicate(rdev->ddev,
modeset->crtc->desired_mode);
}
}
rfbdev->crtc_count = crtc_count;
if (new_fb) {
info->var.pixclock = -1;
if (register_framebuffer(info) < 0)
return -EINVAL;
} else {
radeonfb_set_par(info);
}
printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
info->fix.id);
/* Switch back to kernel console on panic */
panic_mode = *modeset;
atomic_notifier_chain_register(&panic_notifier_list, &paniced);
printk(KERN_INFO "registered panic notifier\n");
return 0;
}
int radeonfb_probe(struct drm_device *dev)
{
int ret;
/* something has changed in the lower levels of hell - deal with it
here */
/* two modes : a) 1 fb to rule all crtcs.
b) one fb per crtc.
two actions 1) new connected device
2) device removed.
case a/1 : if the fb surface isn't big enough - resize the surface fb.
if the fb size isn't big enough - resize fb into surface.
if everything big enough configure the new crtc/etc.
case a/2 : undo the configuration
possibly resize down the fb to fit the new configuration.
case b/1 : see if it is on a new crtc - setup a new fb and add it.
case b/2 : teardown the new fb.
*/
ret = radeonfb_single_fb_probe(dev->dev_private);
return ret;
}
EXPORT_SYMBOL(radeonfb_probe);
int radeonfb_remove(struct drm_device *dev, struct drm_framebuffer *fb)
{
struct fb_info *info;
struct radeon_framebuffer *rfb = to_radeon_framebuffer(fb);
struct radeon_object *robj;
if (!fb) {
return -EINVAL;
}
info = fb->fbdev;
if (info) {
robj = rfb->obj->driver_private;
unregister_framebuffer(info);
radeon_object_kunmap(robj);
radeon_object_unpin(robj);
drm/radeon: introduce kernel modesetting for radeon hardware Add kernel modesetting support to radeon driver, use the ttm memory manager to manage memory and DRM/GEM to provide userspace API. In order to avoid backward compatibility issue and to allow clean design and code the radeon kernel modesetting use different code path than old radeon/drm driver. When kernel modesetting is enabled the IOCTL of radeon/drm driver are considered as invalid and an error message is printed in the log and they return failure. KMS enabled userspace will use new API to talk with the radeon/drm driver. The new API provide functions to create/destroy/share/mmap buffer object which are then managed by the kernel memory manager (here TTM). In order to submit command to the GPU the userspace provide a buffer holding the command stream, along this buffer userspace have to provide a list of buffer object used by the command stream. The kernel radeon driver will then place buffer in GPU accessible memory and will update command stream to reflect the position of the different buffers. The kernel will also perform security check on command stream provided by the user, we want to catch and forbid any illegal use of the GPU such as DMA into random system memory or into memory not owned by the process supplying the command stream. This part of the code is still incomplete and this why we propose that patch as a staging driver addition, future security might forbid current experimental userspace to run. This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX (radeon up to X1950). Works is underway to provide support for R6XX, R7XX and newer hardware (radeon from HD2XXX to HD4XXX). Authors: Jerome Glisse <jglisse@redhat.com> Dave Airlie <airlied@redhat.com> Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-06-05 20:42:42 +08:00
framebuffer_release(info);
}
printk(KERN_INFO "unregistered panic notifier\n");
atomic_notifier_chain_unregister(&panic_notifier_list, &paniced);
memset(&panic_mode, 0, sizeof(struct drm_mode_set));
return 0;
}
EXPORT_SYMBOL(radeonfb_remove);
MODULE_LICENSE("GPL");