Minki/linux
1
0
Fork 1
mirror of https://github.com/torvalds/linux.git synced 2024-11-20 02:51:44 +00:00
Code Issues Packages Projects Releases Wiki Activity
cedb655a3a
linux/drivers/gpu/drm/radeon/radeon_clocks.c

913 lines
26 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 12:42:42 +00:00
/*
* 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
*/
UAPI: (Scripted) Convert #include "..." to #include <path/...> in drivers/gpu/ Convert #include "..." to #include <path/...> in drivers/gpu/. Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Dave Airlie <airlied@redhat.com> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Dave Jones <davej@redhat.com>
2012-10-02 17:01:07 +00:00
#include <drm/drmP.h>
#include <drm/radeon_drm.h>
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 12:42:42 +00:00
#include "radeon_reg.h"
#include "radeon.h"
#include "atom.h"
/* 10 khz */
drm/radeon/kms: add debugfs for power management for AtomBIOS devices Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-11-02 23:53:02 +00:00
uint32_t radeon_legacy_get_engine_clock(struct radeon_device *rdev)
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 12:42:42 +00:00
{
struct radeon_pll *spll = &rdev->clock.spll;
uint32_t fb_div, ref_div, post_div, sclk;
fb_div = RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV);
fb_div = (fb_div >> RADEON_SPLL_FB_DIV_SHIFT) & RADEON_SPLL_FB_DIV_MASK;
fb_div <<= 1;
fb_div *= spll->reference_freq;
ref_div =
RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV) & RADEON_M_SPLL_REF_DIV_MASK;
drm/radeon/kms: fix divide by 0 in clocks code If the chip isn't initialised properly this can happen. also fix return value in combios clocks function. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-11-30 23:13:40 +00:00
if (ref_div == 0)
return 0;
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 12:42:42 +00:00
sclk = fb_div / ref_div;
post_div = RREG32_PLL(RADEON_SCLK_CNTL) & RADEON_SCLK_SRC_SEL_MASK;
if (post_div == 2)
sclk >>= 1;
else if (post_div == 3)
sclk >>= 2;
else if (post_div == 4)
drm/radeon/kms: fix legacy get_engine/memory clock Fix a bad shift in the post div. Should fix fdo bug 26145 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-01-22 00:28:18 +00:00
sclk >>= 3;
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 12:42:42 +00:00
return sclk;
}
/* 10 khz */
drm/radeon/kms: enable memory clock reading on legacy (V2) V2: detect IGP cards (which don't have own memory) Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Acked-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-12-17 12:50:09 +00:00
uint32_t radeon_legacy_get_memory_clock(struct radeon_device *rdev)
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 12:42:42 +00:00
{
struct radeon_pll *mpll = &rdev->clock.mpll;
uint32_t fb_div, ref_div, post_div, mclk;
fb_div = RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV);
fb_div = (fb_div >> RADEON_MPLL_FB_DIV_SHIFT) & RADEON_MPLL_FB_DIV_MASK;
fb_div <<= 1;
fb_div *= mpll->reference_freq;
ref_div =
RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV) & RADEON_M_SPLL_REF_DIV_MASK;
drm/radeon/kms: fix divide by 0 in clocks code If the chip isn't initialised properly this can happen. also fix return value in combios clocks function. Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-11-30 23:13:40 +00:00
if (ref_div == 0)
return 0;
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 12:42:42 +00:00
mclk = fb_div / ref_div;
post_div = RREG32_PLL(RADEON_MCLK_CNTL) & 0x7;
if (post_div == 2)
mclk >>= 1;
else if (post_div == 3)
mclk >>= 2;
else if (post_div == 4)
drm/radeon/kms: fix legacy get_engine/memory clock Fix a bad shift in the post div. Should fix fdo bug 26145 Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-01-22 00:28:18 +00:00
mclk >>= 3;
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 12:42:42 +00:00
return mclk;
}
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
#ifdef CONFIG_OF
/*
* Read XTAL (ref clock), SCLK and MCLK from Open Firmware device
* tree. Hopefully, ATI OF driver is kind enough to fill these
*/
DRM: Radeon: Fix section mismatch. WARNING: drivers/gpu/drm/radeon/radeon.o(.text+0x5d1fc): Section mismatch in reference from the function radeon_get_clock_info() to the function .devinit.text:radeon_read_clocks_OF() The function radeon_get_clock_info() references the function __devinit radeon_read_clocks_OF(). This is often because radeon_get_clock_info lacks a __devinit annotation or the annotation of radeon_read_clocks_OF is wrong. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Cc: dri-devel@lists.freedesktop.org Cc: linux-kernel@vger.kernel.org Cc: linux-mips@linux-mips.org Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-06-27 13:40:35 +00:00
static bool radeon_read_clocks_OF(struct drm_device *dev)
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
{
struct radeon_device *rdev = dev->dev_private;
struct device_node *dp = rdev->pdev->dev.of_node;
const u32 *val;
struct radeon_pll *p1pll = &rdev->clock.p1pll;
struct radeon_pll *p2pll = &rdev->clock.p2pll;
struct radeon_pll *spll = &rdev->clock.spll;
struct radeon_pll *mpll = &rdev->clock.mpll;
if (dp == NULL)
return false;
val = of_get_property(dp, "ATY,RefCLK", NULL);
if (!val || !*val) {
printk(KERN_WARNING "radeonfb: No ATY,RefCLK property !\n");
return false;
}
p1pll->reference_freq = p2pll->reference_freq = (*val) / 10;
p1pll->reference_div = RREG32_PLL(RADEON_PPLL_REF_DIV) & 0x3ff;
if (p1pll->reference_div < 2)
p1pll->reference_div = 12;
drm/radeon/kms: check modes against max pixel clock Filter out modes that are higher than the max pixel clock. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-06-08 17:01:11 +00:00
p2pll->reference_div = p1pll->reference_div;
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
/* These aren't in the device-tree */
if (rdev->family >= CHIP_R420) {
p1pll->pll_in_min = 100;
p1pll->pll_in_max = 1350;
p1pll->pll_out_min = 20000;
p1pll->pll_out_max = 50000;
p2pll->pll_in_min = 100;
p2pll->pll_in_max = 1350;
p2pll->pll_out_min = 20000;
p2pll->pll_out_max = 50000;
} else {
p1pll->pll_in_min = 40;
p1pll->pll_in_max = 500;
p1pll->pll_out_min = 12500;
p1pll->pll_out_max = 35000;
p2pll->pll_in_min = 40;
p2pll->pll_in_max = 500;
p2pll->pll_out_min = 12500;
p2pll->pll_out_max = 35000;
}
drm/radeon/kms: check modes against max pixel clock Filter out modes that are higher than the max pixel clock. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-06-08 17:01:11 +00:00
/* not sure what the max should be in all cases */
rdev->clock.max_pixel_clock = 35000;
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
spll->reference_freq = mpll->reference_freq = p1pll->reference_freq;
spll->reference_div = mpll->reference_div =
RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV) &
RADEON_M_SPLL_REF_DIV_MASK;
val = of_get_property(dp, "ATY,SCLK", NULL);
if (val && *val)
rdev->clock.default_sclk = (*val) / 10;
else
rdev->clock.default_sclk =
radeon_legacy_get_engine_clock(rdev);
drm/radeon/kms: check modes against max pixel clock Filter out modes that are higher than the max pixel clock. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-06-08 17:01:11 +00:00
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
val = of_get_property(dp, "ATY,MCLK", NULL);
if (val && *val)
rdev->clock.default_mclk = (*val) / 10;
else
rdev->clock.default_mclk =
radeon_legacy_get_memory_clock(rdev);
DRM_INFO("Using device-tree clock info\n");
drm/radeon/kms: check modes against max pixel clock Filter out modes that are higher than the max pixel clock. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-06-08 17:01:11 +00:00
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
return true;
}
#else
DRM: Radeon: Fix section mismatch. WARNING: drivers/gpu/drm/radeon/radeon.o(.text+0x5d1fc): Section mismatch in reference from the function radeon_get_clock_info() to the function .devinit.text:radeon_read_clocks_OF() The function radeon_get_clock_info() references the function __devinit radeon_read_clocks_OF(). This is often because radeon_get_clock_info lacks a __devinit annotation or the annotation of radeon_read_clocks_OF is wrong. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Cc: dri-devel@lists.freedesktop.org Cc: linux-kernel@vger.kernel.org Cc: linux-mips@linux-mips.org Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-06-27 13:40:35 +00:00
static bool radeon_read_clocks_OF(struct drm_device *dev)
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
{
return false;
}
#endif /* CONFIG_OF */
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 12:42:42 +00:00
void radeon_get_clock_info(struct drm_device *dev)
{
struct radeon_device *rdev = dev->dev_private;
struct radeon_pll *p1pll = &rdev->clock.p1pll;
struct radeon_pll *p2pll = &rdev->clock.p2pll;
drm/radeon/kms: add initial Evergreen support (Radeon HD 5xxx) This adds initial Evergreen KMS support, it doesn't include any acceleration features or interrupt handling yet. Major changes are DCE4 handling for PLLs for the > 2 crtcs. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-01-12 22:54:34 +00:00
struct radeon_pll *dcpll = &rdev->clock.dcpll;
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 12:42:42 +00:00
struct radeon_pll *spll = &rdev->clock.spll;
struct radeon_pll *mpll = &rdev->clock.mpll;
int ret;
if (rdev->is_atom_bios)
ret = radeon_atom_get_clock_info(dev);
else
ret = radeon_combios_get_clock_info(dev);
drm/radeon: Add probing of clocks from device-tree When we find no ROM we understand and a device-tree is present, see if we can retreive clock info from there. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-09 06:16:36 +00:00
if (!ret)
ret = radeon_read_clocks_OF(dev);
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 12:42:42 +00:00
if (ret) {
drm/radeon/kms: more r4xx lvds fixes Grab pll ref div from regs at driver init. r4xx seems very picky about the dividers for the pll driving lvds. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-12-04 21:35:57 +00:00
if (p1pll->reference_div < 2) {
if (!ASIC_IS_AVIVO(rdev)) {
u32 tmp = RREG32_PLL(RADEON_PPLL_REF_DIV);
if (ASIC_IS_R300(rdev))
p1pll->reference_div =
(tmp & R300_PPLL_REF_DIV_ACC_MASK) >> R300_PPLL_REF_DIV_ACC_SHIFT;
else
p1pll->reference_div = tmp & RADEON_PPLL_REF_DIV_MASK;
if (p1pll->reference_div < 2)
p1pll->reference_div = 12;
} else
p1pll->reference_div = 12;
}
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 12:42:42 +00:00
if (p2pll->reference_div < 2)
p2pll->reference_div = 12;
drm/radeon/kms: add r600 KMS support This adds the r600 KMS + CS support to the Linux kernel. The r600 TTM support is quite basic and still needs more work esp around using interrupts, but the polled fencing should work okay for now. Also currently TTM is using memcpy to do VRAM moves, the code is here to use a 3D blit to do this, but isn't fully debugged yet. Authors: Alex Deucher <alexdeucher@gmail.com> Dave Airlie <airlied@redhat.com> Jerome Glisse <jglisse@redhat.com> Signed-off-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-09-08 00:10:24 +00:00
if (rdev->family < CHIP_RS600) {
if (spll->reference_div < 2)
spll->reference_div =
RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV) &
RADEON_M_SPLL_REF_DIV_MASK;
}
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 12:42:42 +00:00
if (mpll->reference_div < 2)
mpll->reference_div = spll->reference_div;
} else {
if (ASIC_IS_AVIVO(rdev)) {
/* TODO FALLBACK */
} else {
DRM_INFO("Using generic clock info\n");
drm/radeon/kms: set a default max_pixel_clock On some Power rv100 cards, we have no ATY OF table, but we have no combios table either, and hence we refuse all modes on VGA-0 since we end up with a 0 max pixel clock. Signed-off-by: Dave Airlie <airlied@redhat.com> Cc: stable@kernel.org Reviewed-by: Alex Deucher <alexdeucher@gmail.com> Reviewed-by: Jerome Glisse <jglisse@redhat.com>
2011-08-30 19:22:04 +00:00
/* may need to be per card */
rdev->clock.max_pixel_clock = 35000;
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 12:42:42 +00:00
if (rdev->flags & RADEON_IS_IGP) {
p1pll->reference_freq = 1432;
p2pll->reference_freq = 1432;
spll->reference_freq = 1432;
mpll->reference_freq = 1432;
} else {
p1pll->reference_freq = 2700;
p2pll->reference_freq = 2700;
spll->reference_freq = 2700;
mpll->reference_freq = 2700;
}
p1pll->reference_div =
RREG32_PLL(RADEON_PPLL_REF_DIV) & 0x3ff;
if (p1pll->reference_div < 2)
p1pll->reference_div = 12;
p2pll->reference_div = p1pll->reference_div;
if (rdev->family >= CHIP_R420) {
p1pll->pll_in_min = 100;
p1pll->pll_in_max = 1350;
p1pll->pll_out_min = 20000;
p1pll->pll_out_max = 50000;
p2pll->pll_in_min = 100;
p2pll->pll_in_max = 1350;
p2pll->pll_out_min = 20000;
p2pll->pll_out_max = 50000;
} else {
p1pll->pll_in_min = 40;
p1pll->pll_in_max = 500;
p1pll->pll_out_min = 12500;
p1pll->pll_out_max = 35000;
p2pll->pll_in_min = 40;
p2pll->pll_in_max = 500;
p2pll->pll_out_min = 12500;
p2pll->pll_out_max = 35000;
}
spll->reference_div =
RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV) &
RADEON_M_SPLL_REF_DIV_MASK;
mpll->reference_div = spll->reference_div;
rdev->clock.default_sclk =
radeon_legacy_get_engine_clock(rdev);
rdev->clock.default_mclk =
radeon_legacy_get_memory_clock(rdev);
}
}
/* pixel clocks */
if (ASIC_IS_AVIVO(rdev)) {
p1pll->min_post_div = 2;
p1pll->max_post_div = 0x7f;
p1pll->min_frac_feedback_div = 0;
p1pll->max_frac_feedback_div = 9;
p2pll->min_post_div = 2;
p2pll->max_post_div = 0x7f;
p2pll->min_frac_feedback_div = 0;
p2pll->max_frac_feedback_div = 9;
} else {
p1pll->min_post_div = 1;
p1pll->max_post_div = 16;
p1pll->min_frac_feedback_div = 0;
p1pll->max_frac_feedback_div = 0;
p2pll->min_post_div = 1;
p2pll->max_post_div = 12;
p2pll->min_frac_feedback_div = 0;
p2pll->max_frac_feedback_div = 0;
}
drm/radeon/kms: add initial Evergreen support (Radeon HD 5xxx) This adds initial Evergreen KMS support, it doesn't include any acceleration features or interrupt handling yet. Major changes are DCE4 handling for PLLs for the > 2 crtcs. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-01-12 22:54:34 +00:00
/* dcpll is DCE4 only */
dcpll->min_post_div = 2;
dcpll->max_post_div = 0x7f;
dcpll->min_frac_feedback_div = 0;
dcpll->max_frac_feedback_div = 9;
dcpll->min_ref_div = 2;
dcpll->max_ref_div = 0x3ff;
dcpll->min_feedback_div = 4;
dcpll->max_feedback_div = 0xfff;
dcpll->best_vco = 0;
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 12:42:42 +00:00
p1pll->min_ref_div = 2;
p1pll->max_ref_div = 0x3ff;
p1pll->min_feedback_div = 4;
p1pll->max_feedback_div = 0x7ff;
p1pll->best_vco = 0;
p2pll->min_ref_div = 2;
p2pll->max_ref_div = 0x3ff;
p2pll->min_feedback_div = 4;
p2pll->max_feedback_div = 0x7ff;
p2pll->best_vco = 0;
/* system clock */
spll->min_post_div = 1;
spll->max_post_div = 1;
spll->min_ref_div = 2;
spll->max_ref_div = 0xff;
spll->min_feedback_div = 4;
spll->max_feedback_div = 0xff;
spll->best_vco = 0;
/* memory clock */
mpll->min_post_div = 1;
mpll->max_post_div = 1;
mpll->min_ref_div = 2;
mpll->max_ref_div = 0xff;
mpll->min_feedback_div = 4;
mpll->max_feedback_div = 0xff;
mpll->best_vco = 0;
drm/radeon/kms: use tracked values for sclk and mclk Rather than calling get_memory_clock and get_engine_clock, used the tracked values from the pm code. Calling the tables adds additional latency in the modesetting and pm paths. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-10 16:33:20 +00:00
if (!rdev->clock.default_sclk)
rdev->clock.default_sclk = radeon_get_engine_clock(rdev);
drm/radeon/kms: move clock/pcie setting callbacks into pm struct tidy up radeon_asic struct. Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König<christian.koenig@amd.com> Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Jerome Glisse <jglisse@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-02-23 22:53:48 +00:00
if ((!rdev->clock.default_mclk) && rdev->asic->pm.get_memory_clock)
drm/radeon/kms: use tracked values for sclk and mclk Rather than calling get_memory_clock and get_engine_clock, used the tracked values from the pm code. Calling the tables adds additional latency in the modesetting and pm paths. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-08-10 16:33:20 +00:00
rdev->clock.default_mclk = radeon_get_memory_clock(rdev);
rdev->pm.current_sclk = rdev->clock.default_sclk;
rdev->pm.current_mclk = rdev->clock.default_mclk;
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 12:42:42 +00:00
}
/* 10 khz */
static uint32_t calc_eng_mem_clock(struct radeon_device *rdev,
uint32_t req_clock,
int *fb_div, int *post_div)
{
struct radeon_pll *spll = &rdev->clock.spll;
int ref_div = spll->reference_div;
if (!ref_div)
ref_div =
RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV) &
RADEON_M_SPLL_REF_DIV_MASK;
if (req_clock < 15000) {
*post_div = 8;
req_clock *= 8;
} else if (req_clock < 30000) {
*post_div = 4;
req_clock *= 4;
} else if (req_clock < 60000) {
*post_div = 2;
req_clock *= 2;
} else
*post_div = 1;
req_clock *= ref_div;
req_clock += spll->reference_freq;
req_clock /= (2 * spll->reference_freq);
*fb_div = req_clock & 0xff;
req_clock = (req_clock & 0xffff) << 1;
req_clock *= spll->reference_freq;
req_clock /= ref_div;
req_clock /= *post_div;
return req_clock;
}
/* 10 khz */
void radeon_legacy_set_engine_clock(struct radeon_device *rdev,
uint32_t eng_clock)
{
uint32_t tmp;
int fb_div, post_div;
/* XXX: wait for idle */
eng_clock = calc_eng_mem_clock(rdev, eng_clock, &fb_div, &post_div);
tmp = RREG32_PLL(RADEON_CLK_PIN_CNTL);
tmp &= ~RADEON_DONT_USE_XTALIN;
WREG32_PLL(RADEON_CLK_PIN_CNTL, tmp);
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp &= ~RADEON_SCLK_SRC_SEL_MASK;
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
udelay(10);
tmp = RREG32_PLL(RADEON_SPLL_CNTL);
tmp |= RADEON_SPLL_SLEEP;
WREG32_PLL(RADEON_SPLL_CNTL, tmp);
udelay(2);
tmp = RREG32_PLL(RADEON_SPLL_CNTL);
tmp |= RADEON_SPLL_RESET;
WREG32_PLL(RADEON_SPLL_CNTL, tmp);
udelay(200);
tmp = RREG32_PLL(RADEON_M_SPLL_REF_FB_DIV);
tmp &= ~(RADEON_SPLL_FB_DIV_MASK << RADEON_SPLL_FB_DIV_SHIFT);
tmp |= (fb_div & RADEON_SPLL_FB_DIV_MASK) << RADEON_SPLL_FB_DIV_SHIFT;
WREG32_PLL(RADEON_M_SPLL_REF_FB_DIV, tmp);
/* XXX: verify on different asics */
tmp = RREG32_PLL(RADEON_SPLL_CNTL);
tmp &= ~RADEON_SPLL_PVG_MASK;
if ((eng_clock * post_div) >= 90000)
tmp |= (0x7 << RADEON_SPLL_PVG_SHIFT);
else
tmp |= (0x4 << RADEON_SPLL_PVG_SHIFT);
WREG32_PLL(RADEON_SPLL_CNTL, tmp);
tmp = RREG32_PLL(RADEON_SPLL_CNTL);
tmp &= ~RADEON_SPLL_SLEEP;
WREG32_PLL(RADEON_SPLL_CNTL, tmp);
udelay(2);
tmp = RREG32_PLL(RADEON_SPLL_CNTL);
tmp &= ~RADEON_SPLL_RESET;
WREG32_PLL(RADEON_SPLL_CNTL, tmp);
udelay(200);
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp &= ~RADEON_SCLK_SRC_SEL_MASK;
switch (post_div) {
case 1:
default:
tmp |= 1;
break;
case 2:
tmp |= 2;
break;
case 4:
tmp |= 3;
break;
case 8:
tmp |= 4;
break;
}
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
udelay(20);
tmp = RREG32_PLL(RADEON_CLK_PIN_CNTL);
tmp |= RADEON_DONT_USE_XTALIN;
WREG32_PLL(RADEON_CLK_PIN_CNTL, tmp);
udelay(10);
}
void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable)
{
uint32_t tmp;
if (enable) {
if (rdev->flags & RADEON_SINGLE_CRTC) {
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
if ((RREG32(RADEON_CONFIG_CNTL) &
RADEON_CFG_ATI_REV_ID_MASK) >
RADEON_CFG_ATI_REV_A13) {
tmp &=
~(RADEON_SCLK_FORCE_CP |
RADEON_SCLK_FORCE_RB);
}
tmp &=
~(RADEON_SCLK_FORCE_HDP | RADEON_SCLK_FORCE_DISP1 |
RADEON_SCLK_FORCE_TOP | RADEON_SCLK_FORCE_SE |
RADEON_SCLK_FORCE_IDCT | RADEON_SCLK_FORCE_RE |
RADEON_SCLK_FORCE_PB | RADEON_SCLK_FORCE_TAM |
RADEON_SCLK_FORCE_TDM);
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
} else if (ASIC_IS_R300(rdev)) {
if ((rdev->family == CHIP_RS400) ||
(rdev->family == CHIP_RS480)) {
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp &=
~(RADEON_SCLK_FORCE_DISP2 |
RADEON_SCLK_FORCE_CP |
RADEON_SCLK_FORCE_HDP |
RADEON_SCLK_FORCE_DISP1 |
RADEON_SCLK_FORCE_TOP |
RADEON_SCLK_FORCE_E2 | R300_SCLK_FORCE_VAP
| RADEON_SCLK_FORCE_IDCT |
RADEON_SCLK_FORCE_VIP | R300_SCLK_FORCE_SR
| R300_SCLK_FORCE_PX | R300_SCLK_FORCE_TX
| R300_SCLK_FORCE_US |
RADEON_SCLK_FORCE_TV_SCLK |
R300_SCLK_FORCE_SU |
RADEON_SCLK_FORCE_OV0);
tmp |= RADEON_DYN_STOP_LAT_MASK;
tmp |=
RADEON_SCLK_FORCE_TOP |
RADEON_SCLK_FORCE_VIP;
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL);
tmp &= ~RADEON_SCLK_MORE_FORCEON;
tmp |= RADEON_SCLK_MORE_MAX_DYN_STOP_LAT;
WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp);
tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL);
tmp |= (RADEON_PIXCLK_ALWAYS_ONb |
RADEON_PIXCLK_DAC_ALWAYS_ONb);
WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp);
tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL);
tmp |= (RADEON_PIX2CLK_ALWAYS_ONb |
RADEON_PIX2CLK_DAC_ALWAYS_ONb |
RADEON_DISP_TVOUT_PIXCLK_TV_ALWAYS_ONb |
R300_DVOCLK_ALWAYS_ONb |
RADEON_PIXCLK_BLEND_ALWAYS_ONb |
RADEON_PIXCLK_GV_ALWAYS_ONb |
R300_PIXCLK_DVO_ALWAYS_ONb |
RADEON_PIXCLK_LVDS_ALWAYS_ONb |
RADEON_PIXCLK_TMDS_ALWAYS_ONb |
R300_PIXCLK_TRANS_ALWAYS_ONb |
R300_PIXCLK_TVO_ALWAYS_ONb |
R300_P2G2CLK_ALWAYS_ONb |
drm/radeon: Fix setting of bits Duplicate bits set Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-10-06 19:48:40 +00:00
R300_P2G2CLK_DAC_ALWAYS_ONb);
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 12:42:42 +00:00
WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp);
} else if (rdev->family >= CHIP_RV350) {
tmp = RREG32_PLL(R300_SCLK_CNTL2);
tmp &= ~(R300_SCLK_FORCE_TCL |
R300_SCLK_FORCE_GA |
R300_SCLK_FORCE_CBA);
tmp |= (R300_SCLK_TCL_MAX_DYN_STOP_LAT |
R300_SCLK_GA_MAX_DYN_STOP_LAT |
R300_SCLK_CBA_MAX_DYN_STOP_LAT);
WREG32_PLL(R300_SCLK_CNTL2, tmp);
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp &=
~(RADEON_SCLK_FORCE_DISP2 |
RADEON_SCLK_FORCE_CP |
RADEON_SCLK_FORCE_HDP |
RADEON_SCLK_FORCE_DISP1 |
RADEON_SCLK_FORCE_TOP |
RADEON_SCLK_FORCE_E2 | R300_SCLK_FORCE_VAP
| RADEON_SCLK_FORCE_IDCT |
RADEON_SCLK_FORCE_VIP | R300_SCLK_FORCE_SR
| R300_SCLK_FORCE_PX | R300_SCLK_FORCE_TX
| R300_SCLK_FORCE_US |
RADEON_SCLK_FORCE_TV_SCLK |
R300_SCLK_FORCE_SU |
RADEON_SCLK_FORCE_OV0);
tmp |= RADEON_DYN_STOP_LAT_MASK;
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL);
tmp &= ~RADEON_SCLK_MORE_FORCEON;
tmp |= RADEON_SCLK_MORE_MAX_DYN_STOP_LAT;
WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp);
tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL);
tmp |= (RADEON_PIXCLK_ALWAYS_ONb |
RADEON_PIXCLK_DAC_ALWAYS_ONb);
WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp);
tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL);
tmp |= (RADEON_PIX2CLK_ALWAYS_ONb |
RADEON_PIX2CLK_DAC_ALWAYS_ONb |
RADEON_DISP_TVOUT_PIXCLK_TV_ALWAYS_ONb |
R300_DVOCLK_ALWAYS_ONb |
RADEON_PIXCLK_BLEND_ALWAYS_ONb |
RADEON_PIXCLK_GV_ALWAYS_ONb |
R300_PIXCLK_DVO_ALWAYS_ONb |
RADEON_PIXCLK_LVDS_ALWAYS_ONb |
RADEON_PIXCLK_TMDS_ALWAYS_ONb |
R300_PIXCLK_TRANS_ALWAYS_ONb |
R300_PIXCLK_TVO_ALWAYS_ONb |
R300_P2G2CLK_ALWAYS_ONb |
drm/radeon: Fix setting of bits Duplicate bits set Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-10-06 19:48:40 +00:00
R300_P2G2CLK_DAC_ALWAYS_ONb);
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 12:42:42 +00:00
WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp);
tmp = RREG32_PLL(RADEON_MCLK_MISC);
tmp |= (RADEON_MC_MCLK_DYN_ENABLE |
RADEON_IO_MCLK_DYN_ENABLE);
WREG32_PLL(RADEON_MCLK_MISC, tmp);
tmp = RREG32_PLL(RADEON_MCLK_CNTL);
tmp |= (RADEON_FORCEON_MCLKA |
RADEON_FORCEON_MCLKB);
tmp &= ~(RADEON_FORCEON_YCLKA |
RADEON_FORCEON_YCLKB |
RADEON_FORCEON_MC);
/* Some releases of vbios have set DISABLE_MC_MCLKA
and DISABLE_MC_MCLKB bits in the vbios table. Setting these
bits will cause H/W hang when reading video memory with dynamic clocking
enabled. */
if ((tmp & R300_DISABLE_MC_MCLKA) &&
(tmp & R300_DISABLE_MC_MCLKB)) {
/* If both bits are set, then check the active channels */
tmp = RREG32_PLL(RADEON_MCLK_CNTL);
if (rdev->mc.vram_width == 64) {
if (RREG32(RADEON_MEM_CNTL) &
R300_MEM_USE_CD_CH_ONLY)
tmp &=
~R300_DISABLE_MC_MCLKB;
else
tmp &=
~R300_DISABLE_MC_MCLKA;
} else {
tmp &= ~(R300_DISABLE_MC_MCLKA |
R300_DISABLE_MC_MCLKB);
}
}
WREG32_PLL(RADEON_MCLK_CNTL, tmp);
} else {
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp &= ~(R300_SCLK_FORCE_VAP);
tmp |= RADEON_SCLK_FORCE_CP;
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
tmp = RREG32_PLL(R300_SCLK_CNTL2);
tmp &= ~(R300_SCLK_FORCE_TCL |
R300_SCLK_FORCE_GA |
R300_SCLK_FORCE_CBA);
WREG32_PLL(R300_SCLK_CNTL2, tmp);
}
} else {
tmp = RREG32_PLL(RADEON_CLK_PWRMGT_CNTL);
tmp &= ~(RADEON_ACTIVE_HILO_LAT_MASK |
RADEON_DISP_DYN_STOP_LAT_MASK |
RADEON_DYN_STOP_MODE_MASK);
tmp |= (RADEON_ENGIN_DYNCLK_MODE |
(0x01 << RADEON_ACTIVE_HILO_LAT_SHIFT));
WREG32_PLL(RADEON_CLK_PWRMGT_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
tmp = RREG32_PLL(RADEON_CLK_PIN_CNTL);
tmp |= RADEON_SCLK_DYN_START_CNTL;
WREG32_PLL(RADEON_CLK_PIN_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
/* When DRI is enabled, setting DYN_STOP_LAT to zero can cause some R200
to lockup randomly, leave them as set by BIOS.
*/
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
/*tmp &= RADEON_SCLK_SRC_SEL_MASK; */
tmp &= ~RADEON_SCLK_FORCEON_MASK;
/*RAGE_6::A11 A12 A12N1 A13, RV250::A11 A12, R300 */
if (((rdev->family == CHIP_RV250) &&
((RREG32(RADEON_CONFIG_CNTL) &
RADEON_CFG_ATI_REV_ID_MASK) <
RADEON_CFG_ATI_REV_A13))
|| ((rdev->family == CHIP_RV100)
&&
((RREG32(RADEON_CONFIG_CNTL) &
RADEON_CFG_ATI_REV_ID_MASK) <=
RADEON_CFG_ATI_REV_A13))) {
tmp |= RADEON_SCLK_FORCE_CP;
tmp |= RADEON_SCLK_FORCE_VIP;
}
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
if ((rdev->family == CHIP_RV200) ||
(rdev->family == CHIP_RV250) ||
(rdev->family == CHIP_RV280)) {
tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL);
tmp &= ~RADEON_SCLK_MORE_FORCEON;
/* RV200::A11 A12 RV250::A11 A12 */
if (((rdev->family == CHIP_RV200) ||
(rdev->family == CHIP_RV250)) &&
((RREG32(RADEON_CONFIG_CNTL) &
RADEON_CFG_ATI_REV_ID_MASK) <
RADEON_CFG_ATI_REV_A13)) {
tmp |= RADEON_SCLK_MORE_FORCEON;
}
WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
}
/* RV200::A11 A12, RV250::A11 A12 */
if (((rdev->family == CHIP_RV200) ||
(rdev->family == CHIP_RV250)) &&
((RREG32(RADEON_CONFIG_CNTL) &
RADEON_CFG_ATI_REV_ID_MASK) <
RADEON_CFG_ATI_REV_A13)) {
tmp = RREG32_PLL(RADEON_PLL_PWRMGT_CNTL);
tmp |= RADEON_TCL_BYPASS_DISABLE;
WREG32_PLL(RADEON_PLL_PWRMGT_CNTL, tmp);
}
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
/*enable dynamic mode for display clocks (PIXCLK and PIX2CLK) */
tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL);
tmp |= (RADEON_PIX2CLK_ALWAYS_ONb |
RADEON_PIX2CLK_DAC_ALWAYS_ONb |
RADEON_PIXCLK_BLEND_ALWAYS_ONb |
RADEON_PIXCLK_GV_ALWAYS_ONb |
RADEON_PIXCLK_DIG_TMDS_ALWAYS_ONb |
RADEON_PIXCLK_LVDS_ALWAYS_ONb |
RADEON_PIXCLK_TMDS_ALWAYS_ONb);
WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL);
tmp |= (RADEON_PIXCLK_ALWAYS_ONb |
RADEON_PIXCLK_DAC_ALWAYS_ONb);
WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(15);
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 12:42:42 +00:00
}
} else {
/* Turn everything OFF (ForceON to everything) */
if (rdev->flags & RADEON_SINGLE_CRTC) {
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp |= (RADEON_SCLK_FORCE_CP | RADEON_SCLK_FORCE_HDP |
RADEON_SCLK_FORCE_DISP1 | RADEON_SCLK_FORCE_TOP
| RADEON_SCLK_FORCE_E2 | RADEON_SCLK_FORCE_SE |
RADEON_SCLK_FORCE_IDCT | RADEON_SCLK_FORCE_VIP |
RADEON_SCLK_FORCE_RE | RADEON_SCLK_FORCE_PB |
RADEON_SCLK_FORCE_TAM | RADEON_SCLK_FORCE_TDM |
RADEON_SCLK_FORCE_RB);
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
} else if ((rdev->family == CHIP_RS400) ||
(rdev->family == CHIP_RS480)) {
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp |= (RADEON_SCLK_FORCE_DISP2 | RADEON_SCLK_FORCE_CP |
RADEON_SCLK_FORCE_HDP | RADEON_SCLK_FORCE_DISP1
| RADEON_SCLK_FORCE_TOP | RADEON_SCLK_FORCE_E2 |
R300_SCLK_FORCE_VAP | RADEON_SCLK_FORCE_IDCT |
RADEON_SCLK_FORCE_VIP | R300_SCLK_FORCE_SR |
R300_SCLK_FORCE_PX | R300_SCLK_FORCE_TX |
R300_SCLK_FORCE_US | RADEON_SCLK_FORCE_TV_SCLK |
R300_SCLK_FORCE_SU | RADEON_SCLK_FORCE_OV0);
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL);
tmp |= RADEON_SCLK_MORE_FORCEON;
WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp);
tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL);
tmp &= ~(RADEON_PIXCLK_ALWAYS_ONb |
RADEON_PIXCLK_DAC_ALWAYS_ONb |
R300_DISP_DAC_PIXCLK_DAC_BLANK_OFF);
WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp);
tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL);
tmp &= ~(RADEON_PIX2CLK_ALWAYS_ONb |
RADEON_PIX2CLK_DAC_ALWAYS_ONb |
RADEON_DISP_TVOUT_PIXCLK_TV_ALWAYS_ONb |
R300_DVOCLK_ALWAYS_ONb |
RADEON_PIXCLK_BLEND_ALWAYS_ONb |
RADEON_PIXCLK_GV_ALWAYS_ONb |
R300_PIXCLK_DVO_ALWAYS_ONb |
RADEON_PIXCLK_LVDS_ALWAYS_ONb |
RADEON_PIXCLK_TMDS_ALWAYS_ONb |
R300_PIXCLK_TRANS_ALWAYS_ONb |
R300_PIXCLK_TVO_ALWAYS_ONb |
R300_P2G2CLK_ALWAYS_ONb |
drm/radeon: Fix setting of bits Duplicate bits set Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-10-06 19:48:40 +00:00
R300_P2G2CLK_DAC_ALWAYS_ONb |
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 12:42:42 +00:00
R300_DISP_DAC_PIXCLK_DAC2_BLANK_OFF);
WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp);
} else if (rdev->family >= CHIP_RV350) {
/* for RV350/M10, no delays are required. */
tmp = RREG32_PLL(R300_SCLK_CNTL2);
tmp |= (R300_SCLK_FORCE_TCL |
R300_SCLK_FORCE_GA | R300_SCLK_FORCE_CBA);
WREG32_PLL(R300_SCLK_CNTL2, tmp);
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp |= (RADEON_SCLK_FORCE_DISP2 | RADEON_SCLK_FORCE_CP |
RADEON_SCLK_FORCE_HDP | RADEON_SCLK_FORCE_DISP1
| RADEON_SCLK_FORCE_TOP | RADEON_SCLK_FORCE_E2 |
R300_SCLK_FORCE_VAP | RADEON_SCLK_FORCE_IDCT |
RADEON_SCLK_FORCE_VIP | R300_SCLK_FORCE_SR |
R300_SCLK_FORCE_PX | R300_SCLK_FORCE_TX |
R300_SCLK_FORCE_US | RADEON_SCLK_FORCE_TV_SCLK |
R300_SCLK_FORCE_SU | RADEON_SCLK_FORCE_OV0);
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL);
tmp |= RADEON_SCLK_MORE_FORCEON;
WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp);
tmp = RREG32_PLL(RADEON_MCLK_CNTL);
tmp |= (RADEON_FORCEON_MCLKA |
RADEON_FORCEON_MCLKB |
RADEON_FORCEON_YCLKA |
RADEON_FORCEON_YCLKB | RADEON_FORCEON_MC);
WREG32_PLL(RADEON_MCLK_CNTL, tmp);
tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL);
tmp &= ~(RADEON_PIXCLK_ALWAYS_ONb |
RADEON_PIXCLK_DAC_ALWAYS_ONb |
R300_DISP_DAC_PIXCLK_DAC_BLANK_OFF);
WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp);
tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL);
tmp &= ~(RADEON_PIX2CLK_ALWAYS_ONb |
RADEON_PIX2CLK_DAC_ALWAYS_ONb |
RADEON_DISP_TVOUT_PIXCLK_TV_ALWAYS_ONb |
R300_DVOCLK_ALWAYS_ONb |
RADEON_PIXCLK_BLEND_ALWAYS_ONb |
RADEON_PIXCLK_GV_ALWAYS_ONb |
R300_PIXCLK_DVO_ALWAYS_ONb |
RADEON_PIXCLK_LVDS_ALWAYS_ONb |
RADEON_PIXCLK_TMDS_ALWAYS_ONb |
R300_PIXCLK_TRANS_ALWAYS_ONb |
R300_PIXCLK_TVO_ALWAYS_ONb |
R300_P2G2CLK_ALWAYS_ONb |
drm/radeon: Fix setting of bits Duplicate bits set Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2009-10-06 19:48:40 +00:00
R300_P2G2CLK_DAC_ALWAYS_ONb |
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 12:42:42 +00:00
R300_DISP_DAC_PIXCLK_DAC2_BLANK_OFF);
WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp);
} else {
tmp = RREG32_PLL(RADEON_SCLK_CNTL);
tmp |= (RADEON_SCLK_FORCE_CP | RADEON_SCLK_FORCE_E2);
tmp |= RADEON_SCLK_FORCE_SE;
if (rdev->flags & RADEON_SINGLE_CRTC) {
tmp |= (RADEON_SCLK_FORCE_RB |
RADEON_SCLK_FORCE_TDM |
RADEON_SCLK_FORCE_TAM |
RADEON_SCLK_FORCE_PB |
RADEON_SCLK_FORCE_RE |
RADEON_SCLK_FORCE_VIP |
RADEON_SCLK_FORCE_IDCT |
RADEON_SCLK_FORCE_TOP |
RADEON_SCLK_FORCE_DISP1 |
RADEON_SCLK_FORCE_DISP2 |
RADEON_SCLK_FORCE_HDP);
} else if ((rdev->family == CHIP_R300) ||
(rdev->family == CHIP_R350)) {
tmp |= (RADEON_SCLK_FORCE_HDP |
RADEON_SCLK_FORCE_DISP1 |
RADEON_SCLK_FORCE_DISP2 |
RADEON_SCLK_FORCE_TOP |
RADEON_SCLK_FORCE_IDCT |
RADEON_SCLK_FORCE_VIP);
}
WREG32_PLL(RADEON_SCLK_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(16);
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 12:42:42 +00:00
if ((rdev->family == CHIP_R300) ||
(rdev->family == CHIP_R350)) {
tmp = RREG32_PLL(R300_SCLK_CNTL2);
tmp |= (R300_SCLK_FORCE_TCL |
R300_SCLK_FORCE_GA |
R300_SCLK_FORCE_CBA);
WREG32_PLL(R300_SCLK_CNTL2, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(16);
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 12:42:42 +00:00
}
if (rdev->flags & RADEON_IS_IGP) {
tmp = RREG32_PLL(RADEON_MCLK_CNTL);
tmp &= ~(RADEON_FORCEON_MCLKA |
RADEON_FORCEON_YCLKA);
WREG32_PLL(RADEON_MCLK_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(16);
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 12:42:42 +00:00
}
if ((rdev->family == CHIP_RV200) ||
(rdev->family == CHIP_RV250) ||
(rdev->family == CHIP_RV280)) {
tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL);
tmp |= RADEON_SCLK_MORE_FORCEON;
WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(16);
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 12:42:42 +00:00
}
tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL);
tmp &= ~(RADEON_PIX2CLK_ALWAYS_ONb |
RADEON_PIX2CLK_DAC_ALWAYS_ONb |
RADEON_PIXCLK_BLEND_ALWAYS_ONb |
RADEON_PIXCLK_GV_ALWAYS_ONb |
RADEON_PIXCLK_DIG_TMDS_ALWAYS_ONb |
RADEON_PIXCLK_LVDS_ALWAYS_ONb |
RADEON_PIXCLK_TMDS_ALWAYS_ONb);
WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp);
drm/radeon: replace udelay with mdelay for long timeouts Some architectures require that delays longer than a few miliseconds are called through mdelay. This was triggered on ARM randconfig builds. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-04-05 18:58:22 +00:00
mdelay(16);
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 12:42:42 +00:00
tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL);
tmp &= ~(RADEON_PIXCLK_ALWAYS_ONb |
RADEON_PIXCLK_DAC_ALWAYS_ONb);
WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp);
}
}
}
Reference in New Issue Copy Permalink