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>

1 files changed, 324 insertions, 0 deletions

diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
new file mode 100644
index 000000000000..ab0c967553e6
--- /dev/null
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -0,0 +1,324 @@
+/*
+ * Copyright 2008 Advanced Micro Devices, Inc.
+ * Copyright 2008 Red Hat Inc.
+ * Copyright 2009 Jerome Glisse.
+ *
+ * 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 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 COPYRIGHT HOLDER(S) OR AUTHOR(S) 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: Dave Airlie
+ *          Alex Deucher
+ *          Jerome Glisse
+ */
+#include "drmP.h"
+#include "radeon_reg.h"
+#include "radeon.h"
+
+/* rs600 depends on : */
+void r100_hdp_reset(struct radeon_device *rdev);
+int r100_gui_wait_for_idle(struct radeon_device *rdev);
+int r300_mc_wait_for_idle(struct radeon_device *rdev);
+void r420_pipes_init(struct radeon_device *rdev);
+
+/* This files gather functions specifics to :
+ * rs600
+ *
+ * Some of these functions might be used by newer ASICs.
+ */
+void rs600_gpu_init(struct radeon_device *rdev);
+int rs600_mc_wait_for_idle(struct radeon_device *rdev);
+void rs600_disable_vga(struct radeon_device *rdev);
+
+
+/*
+ * GART.
+ */
+void rs600_gart_tlb_flush(struct radeon_device *rdev)
+{
+	uint32_t tmp;
+
+	tmp = RREG32_MC(RS600_MC_PT0_CNTL);
+	tmp &= ~(RS600_INVALIDATE_ALL_L1_TLBS | RS600_INVALIDATE_L2_CACHE);
+	WREG32_MC(RS600_MC_PT0_CNTL, tmp);
+
+	tmp = RREG32_MC(RS600_MC_PT0_CNTL);
+	tmp |= RS600_INVALIDATE_ALL_L1_TLBS | RS600_INVALIDATE_L2_CACHE;
+	WREG32_MC(RS600_MC_PT0_CNTL, tmp);
+
+	tmp = RREG32_MC(RS600_MC_PT0_CNTL);
+	tmp &= ~(RS600_INVALIDATE_ALL_L1_TLBS | RS600_INVALIDATE_L2_CACHE);
+	WREG32_MC(RS600_MC_PT0_CNTL, tmp);
+	tmp = RREG32_MC(RS600_MC_PT0_CNTL);
+}
+
+int rs600_gart_enable(struct radeon_device *rdev)
+{
+	uint32_t tmp;
+	int i;
+	int r;
+
+	/* Initialize common gart structure */
+	r = radeon_gart_init(rdev);
+	if (r) {
+		return r;
+	}
+	rdev->gart.table_size = rdev->gart.num_gpu_pages * 8;
+	r = radeon_gart_table_vram_alloc(rdev);
+	if (r) {
+		return r;
+	}
+	/* FIXME: setup default page */
+	WREG32_MC(RS600_MC_PT0_CNTL,
+		 (RS600_EFFECTIVE_L2_CACHE_SIZE(6) |
+		  RS600_EFFECTIVE_L2_QUEUE_SIZE(6)));
+	for (i = 0; i < 19; i++) {
+		WREG32_MC(RS600_MC_PT0_CLIENT0_CNTL + i,
+			 (RS600_ENABLE_TRANSLATION_MODE_OVERRIDE |
+			  RS600_SYSTEM_ACCESS_MODE_IN_SYS |
+			  RS600_SYSTEM_APERTURE_UNMAPPED_ACCESS_DEFAULT_PAGE |
+			  RS600_EFFECTIVE_L1_CACHE_SIZE(3) |
+			  RS600_ENABLE_FRAGMENT_PROCESSING |
+			  RS600_EFFECTIVE_L1_QUEUE_SIZE(3)));
+	}
+
+	/* System context map to GART space */
+	WREG32_MC(RS600_MC_PT0_SYSTEM_APERTURE_LOW_ADDR, rdev->mc.gtt_location);
+	tmp = rdev->mc.gtt_location + rdev->mc.gtt_size - 1;
+	WREG32_MC(RS600_MC_PT0_SYSTEM_APERTURE_HIGH_ADDR, tmp);
+
+	/* enable first context */
+	WREG32_MC(RS600_MC_PT0_CONTEXT0_FLAT_START_ADDR, rdev->mc.gtt_location);
+	tmp = rdev->mc.gtt_location + rdev->mc.gtt_size - 1;
+	WREG32_MC(RS600_MC_PT0_CONTEXT0_FLAT_END_ADDR, tmp);
+	WREG32_MC(RS600_MC_PT0_CONTEXT0_CNTL,
+		 (RS600_ENABLE_PAGE_TABLE | RS600_PAGE_TABLE_TYPE_FLAT));
+	/* disable all other contexts */
+	for (i = 1; i < 8; i++) {
+		WREG32_MC(RS600_MC_PT0_CONTEXT0_CNTL + i, 0);
+	}
+
+	/* setup the page table */
+	WREG32_MC(RS600_MC_PT0_CONTEXT0_FLAT_BASE_ADDR,
+		 rdev->gart.table_addr);
+	WREG32_MC(RS600_MC_PT0_CONTEXT0_DEFAULT_READ_ADDR, 0);
+
+	/* enable page tables */
+	tmp = RREG32_MC(RS600_MC_PT0_CNTL);
+	WREG32_MC(RS600_MC_PT0_CNTL, (tmp | RS600_ENABLE_PT));
+	tmp = RREG32_MC(RS600_MC_CNTL1);
+	WREG32_MC(RS600_MC_CNTL1, (tmp | RS600_ENABLE_PAGE_TABLES));
+	rs600_gart_tlb_flush(rdev);
+	rdev->gart.ready = true;
+	return 0;
+}
+
+void rs600_gart_disable(struct radeon_device *rdev)
+{
+	uint32_t tmp;
+
+	/* FIXME: disable out of gart access */
+	WREG32_MC(RS600_MC_PT0_CNTL, 0);
+	tmp = RREG32_MC(RS600_MC_CNTL1);
+	tmp &= ~RS600_ENABLE_PAGE_TABLES;
+	WREG32_MC(RS600_MC_CNTL1, tmp);
+	radeon_object_kunmap(rdev->gart.table.vram.robj);
+	radeon_object_unpin(rdev->gart.table.vram.robj);
+}
+
+#define R600_PTE_VALID     (1 << 0)
+#define R600_PTE_SYSTEM    (1 << 1)
+#define R600_PTE_SNOOPED   (1 << 2)
+#define R600_PTE_READABLE  (1 << 5)
+#define R600_PTE_WRITEABLE (1 << 6)
+
+int rs600_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+{
+	void __iomem *ptr = (void *)rdev->gart.table.vram.ptr;
+
+	if (i < 0 || i > rdev->gart.num_gpu_pages) {
+		return -EINVAL;
+	}
+	addr = addr & 0xFFFFFFFFFFFFF000ULL;
+	addr |= R600_PTE_VALID | R600_PTE_SYSTEM | R600_PTE_SNOOPED;
+	addr |= R600_PTE_READABLE | R600_PTE_WRITEABLE;
+	writeq(addr, ((void __iomem *)ptr) + (i * 8));
+	return 0;
+}
+
+
+/*
+ * MC.
+ */
+void rs600_mc_disable_clients(struct radeon_device *rdev)
+{
+	unsigned tmp;
+
+	if (r100_gui_wait_for_idle(rdev)) {
+		printk(KERN_WARNING "Failed to wait GUI idle while "
+		       "programming pipes. Bad things might happen.\n");
+	}
+
+	tmp = RREG32(AVIVO_D1VGA_CONTROL);
+	WREG32(AVIVO_D1VGA_CONTROL, tmp & ~AVIVO_DVGA_CONTROL_MODE_ENABLE);
+	tmp = RREG32(AVIVO_D2VGA_CONTROL);
+	WREG32(AVIVO_D2VGA_CONTROL, tmp & ~AVIVO_DVGA_CONTROL_MODE_ENABLE);
+
+	tmp = RREG32(AVIVO_D1CRTC_CONTROL);
+	WREG32(AVIVO_D1CRTC_CONTROL, tmp & ~AVIVO_CRTC_EN);
+	tmp = RREG32(AVIVO_D2CRTC_CONTROL);
+	WREG32(AVIVO_D2CRTC_CONTROL, tmp & ~AVIVO_CRTC_EN);
+
+	/* make sure all previous write got through */
+	tmp = RREG32(AVIVO_D2CRTC_CONTROL);
+
+	mdelay(1);
+}
+
+int rs600_mc_init(struct radeon_device *rdev)
+{
+	uint32_t tmp;
+	int r;
+
+	if (r100_debugfs_rbbm_init(rdev)) {
+		DRM_ERROR("Failed to register debugfs file for RBBM !\n");
+	}
+
+	rs600_gpu_init(rdev);
+	rs600_gart_disable(rdev);
+
+	/* Setup GPU memory space */
+	rdev->mc.vram_location = 0xFFFFFFFFUL;
+	rdev->mc.gtt_location = 0xFFFFFFFFUL;
+	r = radeon_mc_setup(rdev);
+	if (r) {
+		return r;
+	}
+
+	/* Program GPU memory space */
+	/* Enable bus master */
+	tmp = RREG32(RADEON_BUS_CNTL) & ~RS600_BUS_MASTER_DIS;
+	WREG32(RADEON_BUS_CNTL, tmp);
+	/* FIXME: What does AGP means for such chipset ? */
+	WREG32_MC(RS600_MC_AGP_LOCATION, 0x0FFFFFFF);
+	/* FIXME: are this AGP reg in indirect MC range ? */
+	WREG32_MC(RS600_MC_AGP_BASE, 0);
+	WREG32_MC(RS600_MC_AGP_BASE_2, 0);
+	rs600_mc_disable_clients(rdev);
+	if (rs600_mc_wait_for_idle(rdev)) {
+		printk(KERN_WARNING "Failed to wait MC idle while "
+		       "programming pipes. Bad things might happen.\n");
+	}
+	tmp = rdev->mc.vram_location + rdev->mc.vram_size - 1;
+	tmp = REG_SET(RS600_MC_FB_TOP, tmp >> 16);
+	tmp |= REG_SET(RS600_MC_FB_START, rdev->mc.vram_location >> 16);
+	WREG32_MC(RS600_MC_FB_LOCATION, tmp);
+	WREG32(RS690_HDP_FB_LOCATION, rdev->mc.vram_location >> 16);
+	return 0;
+}
+
+void rs600_mc_fini(struct radeon_device *rdev)
+{
+	rs600_gart_disable(rdev);
+	radeon_gart_table_vram_free(rdev);
+	radeon_gart_fini(rdev);
+}
+
+
+/*
+ * Global GPU functions
+ */
+void rs600_disable_vga(struct radeon_device *rdev)
+{
+	unsigned tmp;
+
+	WREG32(0x330, 0);
+	WREG32(0x338, 0);
+	tmp = RREG32(0x300);
+	tmp &= ~(3 << 16);
+	WREG32(0x300, tmp);
+	WREG32(0x308, (1 << 8));
+	WREG32(0x310, rdev->mc.vram_location);
+	WREG32(0x594, 0);
+}
+
+int rs600_mc_wait_for_idle(struct radeon_device *rdev)
+{
+	unsigned i;
+	uint32_t tmp;
+
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		/* read MC_STATUS */
+		tmp = RREG32_MC(RS600_MC_STATUS);
+		if (tmp & RS600_MC_STATUS_IDLE) {
+			return 0;
+		}
+		DRM_UDELAY(1);
+	}
+	return -1;
+}
+
+void rs600_errata(struct radeon_device *rdev)
+{
+	rdev->pll_errata = 0;
+}
+
+void rs600_gpu_init(struct radeon_device *rdev)
+{
+	/* FIXME: HDP same place on rs600 ? */
+	r100_hdp_reset(rdev);
+	rs600_disable_vga(rdev);
+	/* FIXME: is this correct ? */
+	r420_pipes_init(rdev);
+	if (rs600_mc_wait_for_idle(rdev)) {
+		printk(KERN_WARNING "Failed to wait MC idle while "
+		       "programming pipes. Bad things might happen.\n");
+	}
+}
+
+
+/*
+ * VRAM info.
+ */
+void rs600_vram_info(struct radeon_device *rdev)
+{
+	/* FIXME: to do or is these values sane ? */
+	rdev->mc.vram_is_ddr = true;
+	rdev->mc.vram_width = 128;
+}
+
+
+/*
+ * Indirect registers accessor
+ */
+uint32_t rs600_mc_rreg(struct radeon_device *rdev, uint32_t reg)
+{
+	uint32_t r;
+
+	WREG32(RS600_MC_INDEX,
+	       ((reg & RS600_MC_ADDR_MASK) | RS600_MC_IND_CITF_ARB0));
+	r = RREG32(RS600_MC_DATA);
+	return r;
+}
+
+void rs600_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
+{
+	WREG32(RS600_MC_INDEX,
+		RS600_MC_IND_WR_EN | RS600_MC_IND_CITF_ARB0 |
+		((reg) & RS600_MC_ADDR_MASK));
+	WREG32(RS600_MC_DATA, v);
+}