[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #42 from Dan Horák --- for the record output of "git describe 38e624a18f9a05b" v4.18-rc1-36-g38e624a18f9a -- You are receiving this mail because: You are the assignee for the bug.___

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 Michel Dänzer changed: What|Removed |Added Resolution|--- |FIXED Status|NEEDINFO

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #40 from Timothy Pearson --- (In reply to foxbat from comment #39) > (In reply to Michel Dänzer from comment #35) > > Created attachment 140257 [details] [review] [review] > > drm/amdgpu: GPU vs CPU page size fixes in

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #39 from fox...@ruin.net --- (In reply to Michel Dänzer from comment #35) > Created attachment 140257 [details] [review] > drm/amdgpu: GPU vs CPU page size fixes in amdgpu_vm_bo_split_mapping > > Does this patch help? Hi Michel,

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #38 from fox...@ruin.net --- (In reply to Michel Dänzer from comment #37) > (In reply to foxbat from comment #36) > > It looks like the patch isn't fully applying (using tree 4.17): > > It should apply to 4.17. Did you revert Ben's

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #37 from Michel Dänzer --- (In reply to foxbat from comment #36) > It looks like the patch isn't fully applying (using tree 4.17): It should apply to 4.17. Did you revert Ben's debugging patches before applying it? -- You are

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #36 from fox...@ruin.net --- (In reply to Michel Dänzer from comment #35) > Created attachment 140257 [details] [review] > drm/amdgpu: GPU vs CPU page size fixes in amdgpu_vm_bo_split_mapping > > Does this patch help? It looks

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #35 from Michel Dänzer --- Created attachment 140257 --> https://bugs.freedesktop.org/attachment.cgi?id=140257=edit drm/amdgpu: GPU vs CPU page size fixes in amdgpu_vm_bo_split_mapping Does this patch help? -- You are

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #34 from Ben Crocker --- Things start to go haywire at 246.305790, which appears to be the first time the do-loop in amdgpu_vm_bo_update_mapping executes more than once. The value of max_entries is obviously absurd; a little later,

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #33 from Ben Crocker --- Created attachment 140253 --> https://bugs.freedesktop.org/attachment.cgi?id=140253=edit Refining the amdgpu_vm.c:amdgpu_vm_bo_split_mapping further Print max_entries in both decimal and hex. -- You are

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #32 from fox...@ruin.net --- Created attachment 140252 --> https://bugs.freedesktop.org/attachment.cgi?id=140252=edit amdgpu dmesg output Crash is at 246.306571 -- You are receiving this mail because: You are the assignee for

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #31 from Ben Crocker --- Created attachment 140239 --> https://bugs.freedesktop.org/attachment.cgi?id=140239=edit More patches (2) for additional debug info Hi foxbat, Could you please apply these two patches on top of the

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #30 from Michel Dänzer --- I suspect there's an issue in amdgpu_vm_bo_split_mapping, e.g. this code looks suspicious: if (pages_addr) { uint64_t count; max_entries =

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #29 from fox...@ruin.net --- (In reply to Ben Crocker from comment #28) Hi Ben, Please see the dmesg output at https://pastebin.com/i7sZhUKr I reproduced the graphics "crash" at around 950. However, one thing I did find out is I

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #28 from Ben Crocker --- Created attachment 140208 --> https://bugs.freedesktop.org/attachment.cgi?id=140208=edit Patches for additional debug info foxbat, could you please apply the three attached patches and provide dmesg

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #27 from Christian König --- (In reply to Ben Crocker from comment #26) > foxbat, thanks for posting this! > > > [ 175.501965] [drm] size=167768376 > sa_manager->size=1048576 > > Well, isn't THAT interesting! > Requested size is

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #26 from Ben Crocker --- foxbat, thanks for posting this! > [ 175.501965] [drm] size=167768376 > sa_manager->size=1048576 Well, isn't THAT interesting! Requested size is ALMOST, but not quite, 160 MB, while what the

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #25 from fox...@ruin.net --- (In reply to Michel Dänzer from comment #24) > Created attachment 140046 [details] [review] > Add some debugging output to amdgpu_sa_bo_new > > This patch should tell us which of the WARN_ON_ONCE in

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #24 from Michel Dänzer --- Created attachment 140046 --> https://bugs.freedesktop.org/attachment.cgi?id=140046=edit Add some debugging output to amdgpu_sa_bo_new This patch should tell us which of the WARN_ON_ONCE in

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #23 from fox...@ruin.net --- (In reply to Michel Dänzer from comment #22) > (In reply to foxbat from comment #21) > > [ 75.701323] NIP [c0081748cd08] amdgpu_sa_bo_new+0x640/0x6c0 [amdgpu] > > What does > > scripts/faddr2line

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #22 from Michel Dänzer --- (In reply to foxbat from comment #21) > [ 75.701323] NIP [c0081748cd08] amdgpu_sa_bo_new+0x640/0x6c0 [amdgpu] What does scripts/faddr2line drivers/gpu/drm/amd/amdgpu/amdgpu.ko

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #21 from fox...@ruin.net --- I am still getting amdgpu crashes on POWER9 (ppc64le) using the latest AMD firmware and on kernel 4.17 (release). My card is a WX5100 - Polaris 10. I can reliably reproduce this by having spectacle

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #20 from Ben Crocker --- Please note that the path to the firmware, /lib/firmware/amdgpu, is correct for Red Hat products (RHEL, Fedora, CentOS) as well. -- You are receiving this mail because: You are the assignee for the

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #19 from Alex Deucher --- (In reply to Timothy Pearson from comment #17) > We have been unable to replicate this with the latest AMD GPU firmware from >

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #18 from Timothy Pearson --- (In reply to Timothy Pearson from comment #17) > We have been unable to replicate this with the latest AMD GPU firmware from >

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 Timothy Pearson changed: What|Removed |Added Status|NEW

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #16 from Timothy Pearson --- (In reply to Ben Crocker from comment #15) > Hello Matt, Timothy, Michel, Andrey, and Dan, > > I have been testing some older cards (FirePro 2270, Embedded Radeon >

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #15 from Ben Crocker --- Hello Matt, Timothy, Michel, Andrey, and Dan, I have been testing some older cards (FirePro 2270, Embedded Radeon E6465) on a Power8 system with no problems of the type you're

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #14 from Timothy Pearson --- (In reply to Andrey Grodzovsky from comment #13) > (In reply to Timothy Pearson from comment #11) > > We could get you direct access to test hardware if that would help.

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #13 from Andrey Grodzovsky --- (In reply to Timothy Pearson from comment #11) > We could get you direct access to test hardware if that would help. > Basically, these systems are completely controllable

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #12 from Andrey Grodzovsky --- (In reply to Timothy Pearson from comment #11) > We could get you direct access to test hardware if that would help. > Basically, these systems are completely controllable

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #11 from Timothy Pearson --- We could get you direct access to test hardware if that would help. Basically, these systems are completely controllable via BMC, so we could give remote access to a

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 Dan Horák changed: What|Removed |Added CC||bcroc...@redhat.com ---

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #9 from Matt Corallo --- Switched the machine over to a 4k-PAGE_SIZE kernel and it works fine, so that does appear to be the issue here. -- You are receiving this mail because: You are the assignee for the

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #8 from Andrey Grodzovsky --- I Wonder is there a way to reproduce this on x86 platform, from looking in kflags there is no CONFIG_x86_4/16/64K_PAGES flags. -- You are receiving this mail because: You

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #7 from Michel Dänzer --- (In reply to Timothy Pearson from comment #6) > This looks like something that should be filed at bugzilla.kernel.org > against the amdgpu driver itself. It will get more visibility by

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #6 from Timothy Pearson --- This looks like something that should be filed at bugzilla.kernel.org against the amdgpu driver itself. It will get more visibility by the right people there. -- You are

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #5 from Matt Corallo --- Just checked with today's linus/master (somwehere between 4.17-rc2 and 4.17-rc3) and the issue is still present. -- You are receiving this mail because: You are the assignee for

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #4 from Matt Corallo --- Oops, looks like we got our wires crossed on what has and hasn't been tested. It only *may* be PAGE_SIZE related, but seems relevant 3d stuff was never tested on PPC64LE with 4K

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #3 from Timothy Pearson --- (In reply to Timothy Pearson from comment #1) > This also affects the WX7100, same symptoms. Shows up as soon as an OpenGL > application tries to use the accelerated

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #2 from Matt Corallo --- Oops, forgot to specify this is on a WX4100. -- You are receiving this mail because: You are the assignee for the bug.___ dri-devel

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 --- Comment #1 from Timothy Pearson --- This also affects the WX7100, same symptoms. Shows up as soon as an OpenGL application tries to use the accelerated graphics. Kernel 4.16. -- You are receiving this

[Bug 106258] AMD Xorg start failes with non-4K page sizes

https://bugs.freedesktop.org/show_bug.cgi?id=106258 Bug ID: 106258 Summary: AMD Xorg start failes with non-4K page sizes Product: DRI Version: unspecified Hardware: PowerPC OS: All Status: NEW Severity: