summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2018-11-27Make it work on ppc64le: add conditionals, fix implicitly big-endian ASM code.HEADmasterBen Crocker3-8/+65
Signed-off-by: Ben Crocker <bcrocker@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2016-01-29autogen: remove unrecognized --enable-maintainer-modeOded Gabbay1-1/+1
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2016-01-29Add .gitignore fileOded Gabbay1-0/+31
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2014-07-02radeonreg: add support for dce8Alex Deucher1-0/+19
Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2014-07-02avivotool: fix BAR mapping for newer asicsAlex Deucher1-2/+7
Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2014-07-02radeonreg: add support for SI and CIK partsAlex Deucher3-4/+48
CIK parts use a different PCI BAR for registers. Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2013-11-06avivotool: fix dumping HDMI blocks on DCE2, DCE3 and DCE3.2Rafał Miłecki2-9/+29
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2013-08-02radeontool: update families and pci ids from ati driverDave Airlie2-0/+170
2013-08-02avivotool: dump audio engine registers on DCE6Rafał Miłecki2-0/+11
Tested on HD7750. Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2013-08-02avivotool: dump DCE4/5 specific audio registers conditionallyRafał Miłecki1-8/+10
They don't exist on DCE6+ Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2013-08-02avivotool: switch to audio registers names provided by AMDRafał Miłecki2-22/+48
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-08-22radeonreg: add support for dce6Alex Deucher1-0/+19
Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2012-08-22update chip families for cayman, aruba, SIAlex Deucher1-4/+15
Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Dave Airlie <airlied@gmail.com>
2012-03-21update versionradeontool-1.6.3Dave Airlie1-1/+1
2012-03-21avivotool: dump whole HDMI block on EvergreenRafał Miłecki1-1/+1
Every HDMI block is longer by at least 4 more registers. For example 0x7d28 on HD6320 contains buffer status. Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09radeontool: make card detection and mapping errors non-fatalJonathan Nieder1-23/+51
"radeontool --help" tries to map the control region in order to print dac and light state indicators embedded in the usage message. Unfortunately that means that running "radeontool --help" without sufficient privileges errors out without a usage message that would have hinted at what the tool is for and why it needs root. fatal error: cannot map ctrl region: Permission denied It would be more helpful to write error: cannot map ctrl region: Permission denied usage: radeontool [options] [command] followed by a usage message without the dac and light indicators. Based on a patch by Tormod Volden. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09radeontool: document functions that require mapped control regionJonathan Nieder1-8/+12
Add assertions to all functions that assume the caller has called map_radeon_cntl_mem and radeon_cntl_mem is non-NULL. Use assert() uniformly for this, replacing the more verbose if (!radeon_cntl_mem) die("internal error"); since assert() tends to produce more useful output that precisely pinpoints the call site. The intent is to document the current assumptions before breaking them by changing map_radeon_cntl() to sometimes leave radeon_cntl_mem equal to NULL (for example, when no Radeon is present). Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09include errno string in more messagesJonathan Nieder3-26/+58
Introduce a die_error() helper that includes strerror in a fatal error message and use it where possible. In particular, when running radeontool as non-root, instead of the cryptic fatal error: mapping ctrl region the operator will get a more helpful diagnosis: fatal error: cannot map ctrl region: Permission denied Inspired by a patch by Tormod Volden. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09switch to modern pci_device_map_range APIJonathan Nieder4-42/+35
Ever since it was introduced in libpciaccess 0.10.0 four years ago, pci_device_map_range has been the preferred way to map a PCI region and pci_device_map_region has been a deprecated backward-compatibility shim. Switch over. No functional change intended. Noticed by gcc -Wdeprecated-declarations. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09radeontool, avivotool: remove unnecessary region autodetection magicJonathan Nieder2-27/+5
Trying to guess which region is which based on size is needlessly error-prone and complicated. Region 0 is always the frame buffer and region 2 is always mmio. radeonreg already hard-codes the appropriate region numbers. This patch teaches radeontool and avivotool to do the same. Suggested-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09exit "for each PCI device" loop as soon as Radeon device is foundJonathan Nieder3-50/+48
This just unindents the code some more. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09avivotool, radeonreg: completely skip early cards with --skipJonathan Nieder2-6/+7
If only one card is present but --skip=n was supplied, it is more intuitive to error out than to guess which earlier card the user meant. Compare commit a8ce2f23c589 (radeontool: completely skip early cards with --skip, 2010-03-23). Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-03-09use continue statement to simplify "for each PCI device" loopJonathan Nieder3-69/+75
In the construct for each PCI device: if PCI ID is interesting: long block of code to handle the detected device the "if" with long body leaves the reader in suspense about what will happen when the PCI ID is not interesting. Rewriting as for each PCI device: if PCI ID is uninteresting: continue handle the detected device deals with the easy case right away and decreases nesting for the bulk of the "while" loop body as a nice side effect. No change in functionality intended. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-02-22avivotool, radeonreg: make fatal errors exitJonathan Nieder2-2/+2
It's safer not to continue once in an erroneous situation. This reverses a change made in radeontool-1.6.0~28 ("add more ram dumping stuff") and makes the behavior of avivotool match radeontool again. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-02-22avivotool: make card detection and mapping errors non-fatalJonathan Nieder1-4/+8
After radeontool-1.6.0~28 ("add more ram dumping stuff") avivotool tries to detect a Radeon unconditionally, even for the romtables subcommand that doesn't require one. To keep avivotool useful when there is an error detecting a card, the patch changed the fatal error codepath to keep going rather than exiting. That had some unpleasant side effects, so make lack of a card and other card detection trouble into non-fatal errors as preparation for making die() exit again. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-01-03avivotool: dump enabled HDMI blocks on EvergreenRafał Miłecki2-2/+32
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-01-03avivotool: dump current audio state on EvergreenRafał Miłecki2-0/+9
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-20avivotool: allow reading HDMI related regs on R600Rafał Miłecki2-0/+88
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-20avivotool: allow reading basic audio info on EvergreenRafał Miłecki2-0/+21
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-06avivotool: allow reading regs from given rangeRafał Miłecki1-0/+16
Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-06teach fatal() to write newlineJonathan Nieder3-31/+31
This is more robust: if a caller forgets the final newline, the resulting messages that run together can be hard to read, whereas if a caller adds an extra newline, it just means an extra blank line in the output. Rename the function when changing the interface so any callers that have not been adjusted will break the build. No change in functionality intended. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-06prefix fatal error messages with "fatal error:"Tormod Volden3-3/+3
Let the user know which message corresponds to the error that caused the tool to exit. [jn: using "fatal error:" instead of "Error:" for consistency with "usage:"] Signed-off-by: Tormod Volden <lists.tormod@gmail.com> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-06radeontool: add missing newline to error messageTormod Volden1-1/+1
$ >test $ radeontool romtables test $ PCI ROM signature 0x55 0xaa missing$ Signed-off-by: Tormod Volden <debian.tormod@gmail.com> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-12-05avoid -Wformat-security warningsJonathan Nieder3-3/+3
fatal() is never called with a user-specified argument so it is safe to pass its argument as an fprintf format string, but gcc doesn't know that. Use fprintf(stderr, "%s", arg) to make the intent clearer. This fixes the build in distro setups that pass -Werror=format-security. Based on the patch by Tormod. Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-08-31radeontool: add support for parsing OF ibm,* setup tablesDave Airlie1-9/+103
using tables from rv100 on js22 machine, not sure what 0xa op in init2 table is these tables are found in the open firmware device tree, /proc/device-tree/pci@800000020000202/display@1/ on the js22 Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-05-06radeontool: update pci idsDave Airlie2-1/+22
2011-05-02evergreen : more regsDave Airlie2-25/+35
2011-02-08radeonreg: add radeonreg regs all optionAlex Deucher2-40/+107
detect the chip family and dump appropriately. Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-02-08add radeonreg programAlex Deucher2-1/+427
simple reg dumper for different families. radeonreg regs <set> <set> radeon - r128, r1xx-r4xx avivo - r5xx, rs600/690/740, r600, rv610/630/670 dce3 - rv620/635/770/710/730/740, rs780/880 dce4 - evergreen dce5 - ni Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-02-08add latest chip families from the ddxAlex Deucher2-1/+45
Signed-off-by: Dave Airlie <airlied@redhat.com>
2011-02-02radeontool 1.6.2radeontool-1.6.2Dave Airlie1-1/+1
2010-12-08add support for 256k mmio aperturesAlex Deucher1-0/+2
Signed-off-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-11-19add combios ram reset tableDave Airlie1-0/+33
2010-11-12evergreen: add more regs to all dumpsDave Airlie1-0/+60
2010-11-12evergreen: add support for other commandline regsDave Airlie1-1/+4
2010-11-12evergreen: add initial dac supportDave Airlie2-1/+21
2010-08-16radeontool: fix mc dumpsDave Airlie1-3/+5
2010-08-10avoid crashing on dce3.2 all reg dumpsDave Airlie1-7/+7
2010-08-10evergreen: add initial cursorDave Airlie2-22/+48
2010-08-10evergreen: add initial support using code from DDXDave Airlie3-14/+115
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-13 22:53:15 +0000