xserver-xorg-video-nouveau: Changes to 'refs/tags/debian/0.0.10-git+20090519+9656762-1'
Tag 'debian/0.0.10-git+20090519+9656762-1' created by Chris Lamb <lamby@debian.org> at 2009-05-20 09:18 +0000
Uploading 0.0.10-git+20090519+9656762-1 to experimental.
Changes since xserver-xorg-video-nouveau-0.0.10-git+20090205+4dfd0b1-1:
Ben Skeggs (74):
keep resources around after leavevt, far too problematic currently
kms: update drmmode_display
exa: remove notifier waits in UTS/DFS
xv: remove notifier usage
exa: don't need exaMarkSync on DFS, the operation will be done already
exa: remove exaWaitSync() call in nouveau_exa_pixmap_map()
exa: preparation for reintroducing driver-controlled pixmaps
exa: reimplement driver-controlled pixmaps
default to autodetecting whether to enable the driver's kms paths
nv50/exa: properly align offset for mem_tile ioctl
clear the scanout buffer on statup to avoid seeing garbage on-screen
kms: fix resize func
kms: fix check for xf86drmMode.h
move memset of fb after NVSave(), hopefully fix lost VGA fonts
nv30/xv: restore original viewport/clipping after putimage
kms: implement dpms
kms: work around some bong hits with dpms
kms: remove early-exit from gamma_set, the kernel side works now
kms: work around some xserver stupidity with DPMS
kms: get current values for output properties on startup
kms: cleanup set_property
improve drm/dri initialisation code
exa: fix multiple prepare/finish_access on a pixmap
Call NVEnterVT from NVScreenInit rather than duplicating it.
kms: copy old fb to new fb on modeset
kms: support framebuffer resize if driver pixmaps enabled
dri2: initial implementation, only with driver pixmaps
dri: pass object handle instead of offset if using real mm
fix ppc build
more ppc..
another ppc fix
device close doesn't belong in DRI close, can have DRM without DRI
dri: fail harder
kms: drm_mode_modeinfo struct changed names at some point
kms: implement AdjustFrame, should fix crash in fdo#24236
nv50: call NVSync() in CloseScreen() before restoring video mode
nv50/xv: correct rendering to partically obscured windows
kms: check for mm_enabled as an additional test for kms presence
randr: fix crash when rotation requested
kms: fix bug which prevented getting edid from the kernel
nv50: add default modes to mode pool for LVDS panel
kms: small cleanup
xv: post damage after we draw (rh#492239)
xv: oops
kms: CONNECTORn -> CONNECTOR-n
bios: support for pll limits table v3.0
bios: xf86DrvMsg->NV_ERROR
bios: fix chips with pll limits table version 0
nv50: fix i2c port addresses
consistent connector naming across <nv50, nv50 and kms
bios: add exception for chipset 0x73 for lack of fp table
bios: some G8x don't have PBUS at 0x1800, us 0x88000 unconditionally there
bios: fix typo
kms: fix setting the randr edid property
nv50: modify ddc<->nvreg assignments again..
bios: logic typo from earlier commit
do e-edid for legacy modesetting path
kms: fix displayWidth in resize
Remove pNv->GART conditional for selecting CB_LOCATION in AGP/PCI
nv50: use vbios-programmed lvds mode as native mode if no ddc
nv50: missed a WAIT_RING
bios: parse BIT U table
bios: dcb location is only 2 bits
bios: some nv50 init scripts have flags in reg values, deal with (1<<30)
bios/nv50: initial parsing of display script tables, not quite complete
nv50: merge NV50SorSetClockMode and NV50SorSetClockModeLVDS
bios: modify get_fp_strap() for g80
bios: remove dodgy mode-table search for g80
nv50: group encoders into connectors by i2c port
nv50: remove reading back vbios-programmed lvds native mode
bios: use NV_ARCH_50 to match G8x chips
nv50: return immediately in GetDDCModes if no DDC on connector...
exa: use exaDriverAlloc() to prevent issues across exa changes
nv50: fix multiple-display hangs when encoders swap crtcs
Chris Lamb (6):
Merge branch 'upstream-experimental' into debian-experimental
New upstream snapshot (Closes: #526144)
Bump Standards-Version to 3.8.1.
Bump build dependency on libdrm-dev to 2.4.11.
Add bug #524456 which is fixed by rebuilding.
releasing version 1:0.0.10~git+20090519+9656762-1
Francisco Jerez (1):
Fix nouveau_hw_decode_pll on NV30/35.
Maarten Maathuis (16):
Fix some cursor fallout, mostly in legacy code.
Fix a few warnings.
Always allocate 2 hw cursors.
exa: add nouveau_exa_pixmap_is_tiled
bios: implement opcode 0x76 (INIT_IO_CONDITION).
bios: fix create_i2c_device for g80
bios: add some more valid regs
nv50: correct a poorly written comment
shadowfb: clip copied regions to frontbuffer size.
nv04: unbreak copy
exa: support major version 3
Revert "exa: support major version 3"
exa: we support PREPARE_AUX
exa: forgot to check if EXA_SUPPORTS_PREPARE_AUX was defined
nv50: some minor modesetting changes
nv50: use E-EDID when available + apply edid quirks
Pekka Paalanen (1):
nv50: fix xf86GetDefaultModes() call for older servers
Peter Hjalmarsson (1):
Unused variable warning fixes
Pierre Pronchery (1):
Fix crash with Xinerama enabled
Stuart Bennett (106):
bios/randr12: misc (mainly lvds related) clarifications and cleanups
Simplify FP strapping semantics and trust bios parsing of use_straps_for_mode
bios/randr12: be strict about when to use modes from bios vs modes from ddc
Allow reading of PROM and PCI rom on PPC (for people using x86 cards)
randr12: bodge-in continuous timings when the scaler is in use (#19203)
Make PROM score badly, out of available bios images, if checksum bad
Miscellaneous fixes to bios PLL code
Pull out ramdac580 setting and powerctrl_1 shift code to separate functions
Pull in extras from CRTC PLL code to bios routines
Readability improvements for nv_hw.c and nvreg.h
randr12: use bios PLL setting routines for modesetting
randr12: anti-crash measures for #19854
Revert "bios: fix create_i2c_device for g80"
Variety of DCB enhancements
Attempt to remove pNv and bios_t use in parsing the DCB
Split parsing and init into separate functions
Parse DCB before running init
Return of "bios: fix create_i2c_device for g80"
bios/randr12: fix dual link TMDS (#20006)
Check whether NV_RAMDAC_580 has changed when setting PLLs
New common struct for pll values
Rework PLL decoding in terms of pll_vals, save/restore pll_vals on vt switch
Trivial bios convenience variable changes
Remove never updated bios opcodes 0x31, 0x5A and 0x73
Split bios struct into struct for nv_bios.c, and exported public info struct
Move parsed DCB to bios structs, split into public and private parts
Detypedef bios_t, rename bios io functions, group various helper functions
Read embedded bios edid only on demand
Variety of LVDS script changes
Call parse_lvds_manufacturer_table during modesetting only (#19986 partial fix)
Get dual_link and dithering bios info through lvds table parsing function
Parse bios fp mode during modesetting, removing extra alloc
randr12: make treatment of CR4B more closely resemble blob's behaviour
Don't rely on the bios mobile feature bit for BMP biosen (#19986)
Allow parse_fp_mode_table for x86 cards that find themselves on ppc
randr12: no need to re-detect edid for lvds, it's not going away
Remove VBIOS access in parse_dcb_entry
Remove duplicated DCB parsing
xv: return BadAlloc if destination pixmap is not put into EXA offscreen (#15792)
Handle failure to calculate pll values better
bios/randr12: mode validation fixes (includes fix for #20298)
More robust PLL upclocking for old cards
randr12: make props code slightly less awful to behold
randr12: functionate output properties creation
randr12: add output properties for digital vibrance and image sharpening
Another DCB 1.5 entry (seen in rh#455194)
Name some regs/values and reduce crtc reg struct size
Remove bios lvds parsing pxclk test; ddc_permitted is a sufficient condition
Fix xserver shutdown for randr12 off
Hack around missing display table in rdivacky's GeForce 7050 PV/nForce 630a
Remove calls to load ddc and i2c modules -- they're built-in since server 1.3
Kill pre-randr12 hwcursor code, and tidy randr12 hwcursor code
randr12: store nv0x hw cursor image in VRAM
randr12: use nv0x cursor colour expansion funcs on later cards, unify argb paths
randr12: nv10 hw cursor fixes/changes
Don't pointlessly overallocate cursor storage
randr12: cull old unused pll code
Switch to using pll_vals struct for passing pll stuff around
Wrap bios endian casting
Add DCB 1.5 entry from a GeForce2 Go reported by Martin Ketzer on the ML
Separate maximum limits for encoder devices and i2c records, DCB_ prefixes
pre-nv50: use nvidia's names for PCRTC regs where known
pre-nv50: use nvidia's names for PRAMDAC regs where known
randr12: name some FP flags, move some unrelated stuff out from nvreg.h
randr12: rename reg state members to follow reg names
randr12: name VGA CRTC fields
Some laptop mode finding fixes resulting from rh#487456
Put arbitration and mnp calcs in separate file
Turn the smaller mmio wrappers into static inline
randr12: move a load of hw mode programming into nouveau_hw.c
randr12: permit vga outputs to be force enabled in xorg.conf
randr12: fix LVDS legitimately disconnected case
rh#492511 has a CRTC_OWNER of 0x7 for some reason
randr12: avoid nv11 chip lockup when saving palette regs
The dcb prior to v1.5 is pretty hopeless, just add a crt.
Split DCB 1.5 parsing from 2.0+, get closer to 80 cols
randr12: FP_TG_CONTROL 1: turn off digital path when using VGA encoder
randr12: FP_TG_CONTROL 2: dpms improvements
nForce DIMM check is now in DRM
Abstract logging in nv_bios.c
Restore call to NVDRIGetVersion, to ensure the dri module is loaded
randr12: fix/improve a load of version/feature tests
randr12: remove nv30 special case
randr12: reorder ramdac reg access
randr12: enable tmds reg access on pre-nv17
Every card since nv5 can do panels
randr12: make nv11 dvi work for both crtcs
randr12: fix restore for cards where CR21 is not left unlocked at POST
randr12: pre-nv17 digital fixes
randr12: fix hw cursor for fully transparent pixels
randr12: fall back to standard timings when finding native mode (rh#492819)
randr12: unlock CR21 *after* turning off digital output
randr12: set 1 on cr59 for off-chip digital, 0 otherwise (#21023, rh#492399)
plls: max_log2p_bias is likely actually max_log2p
Partial fix for nv28 lvds bios parsing (part of rh#487456)
randr12: fix for off-chip dual link digital (enable extra wide interface?)
randr12: more use of logging abstraction in modesetting code
randr12: simplify nv11 digital encoder-crtc binding
randr12: solve off-chip encoder crtc exclusion in prepare, rather than dpms
randr12: fix fp_control again again again
randr12: pre-nv17 load detection
randr12: de-magic ramdac general control values
'drmCheckModesettingSupported' implicit function declaration fix
randr12: off-chip lvds for nv28 (rh#487456)
Fix server regeneration again
randr12: fix digital dpms regression
Younes Manton (1):
Fix mouse cursor disappearing when near top/left edge of screen.
---
configure.ac | 9
debian/changelog | 260 ++++
debian/control | 4
src/Makefile.am | 5
src/drmmode_display.c | 925 +++++++++------
src/drmmode_display.h | 67 -
src/nouveau_bios.h | 76 -
src/nouveau_calc.c | 612 ++++++++++
src/nouveau_crtc.h | 1
src/nouveau_dri2.c | 213 +++
src/nouveau_exa.c | 326 ++++-
src/nouveau_hw.c | 970 ++++++++++++++++
src/nouveau_hw.h | 335 +++++
src/nouveau_local.h | 26
src/nouveau_xv.c | 212 ---
src/nv04_exa.c | 20
src/nv04_xv_blit.c | 6
src/nv10_exa.c | 10
src/nv10_xv_ovl.c | 6
src/nv30_exa.c | 8
src/nv30_xv_tex.c | 15
src/nv40_exa.c | 22
src/nv40_xv_tex.c | 4
src/nv50_connector.c | 20
src/nv50_crtc.c | 23
src/nv50_dac.c | 4
src/nv50_display.c | 1
src/nv50_exa.c | 32
src/nv50_output.c | 153 +-
src/nv50_randr.c | 70 -
src/nv50_sor.c | 97 -
src/nv50_xv.c | 21
src/nv50reg.h | 13
src/nv_bios.c | 3004 +++++++++++++++++++++++---------------------------
src/nv_const.h | 6
src/nv_crtc.c | 1093 ++++--------------
src/nv_cursor.c | 473 +------
src/nv_dac.c | 17
src/nv_dma.c | 11
src/nv_dri.c | 141 --
src/nv_driver.c | 492 ++++----
src/nv_hw.c | 947 ---------------
src/nv_i2c.c | 37
src/nv_include.h | 4
src/nv_output.c | 815 +++++++++----
src/nv_proto.h | 80 -
src/nv_setup.c | 92 -
src/nv_shadow.c | 50
src/nv_type.h | 181 +--
src/nvreg.h | 539 ++++----
50 files changed, 6786 insertions(+), 5762 deletions(-)
---
Reply to: