September 2023 - Linux-stable-mirror

Re: [PATCH] drm/bridge: ti-sn65dsi83: Do not generate HFP/HBP/HSA and EOT packet

by Francesco Dolcini

On Mon, Apr 03, 2023 at 09:02:42PM +0200, Marek Vasut wrote: > Do not generate the HS front and back porch gaps, the HSA gap and > EOT packet, as per "SN65DSI83 datasheet SLLSEC1I - SEPTEMBER 2012 > - REVISED OCTOBER 2020", page 22, these packets are not required. > This makes the TI SN65DSI83 bridge work with Samsung DSIM on i.MX8MN. > > Signed-off-by: Marek Vasut <marex(a)denx.de> Hello, can you please queue up this to kernel v6.1-stable ? commit ca161b259cc84fe1f4a2ce4c73c3832cf6f713f1 Author: Marek Vasut <marex(a)denx.de> Commit: Robert Foss <rfoss(a)kernel.org> drm/bridge: ti-sn65dsi83: Do not generate HFP/HBP/HSA and EOT packet Do not generate the HS front and back porch gaps, the HSA gap and EOT packet, as per "SN65DSI83 datasheet SLLSEC1I - SEPTEMBER 2012 - REVISED OCTOBER 2020", page 22, these packets are not required. This makes the TI SN65DSI83 bridge work with Samsung DSIM on i.MX8MN. Signed-off-by: Marek Vasut <marex(a)denx.de> Reviewed-by: Laurent Pinchart <laurent.pinchart(a)ideasonboard.com> Signed-off-by: Robert Foss <rfoss(a)kernel.org> Link: https://patchwork.freedesktop.org/patch/msgid/20230403190242.224490-1-marex… It solves a real issue with some displays not working without it. Thanks, Francesco

1 year, 2 months

1
0
0 0

[PATCH RFC] binfmt_elf: fully allocate bss pages

by Thomas Weißschuh

When allocating the pages for bss the start address needs to be rounded down instead of up. Otherwise the start of the bss segment may be unmapped. The was reported to happen on Aarch64: Memory allocated by set_brk(): Before: start=0x420000 end=0x420000 After: start=0x41f000 end=0x420000 The triggering binary looks like this: Elf file type is EXEC (Executable file) Entry point 0x400144 There are 4 program headers, starting at offset 64 Program Headers: Type Offset VirtAddr PhysAddr FileSiz MemSiz Flags Align LOAD 0x0000000000000000 0x0000000000400000 0x0000000000400000 0x0000000000000178 0x0000000000000178 R E 0x10000 LOAD 0x000000000000ffe8 0x000000000041ffe8 0x000000000041ffe8 0x0000000000000000 0x0000000000000008 RW 0x10000 NOTE 0x0000000000000120 0x0000000000400120 0x0000000000400120 0x0000000000000024 0x0000000000000024 R 0x4 GNU_STACK 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000 RW 0x10 Section to Segment mapping: Segment Sections... 00 .note.gnu.build-id .text .eh_frame 01 .bss 02 .note.gnu.build-id 03 Reported-by: Sebastian Ott <sebott(a)redhat.com> Closes: https://lore.kernel.org/lkml/5d49767a-fbdc-fbe7-5fb2-d99ece3168cb@redhat.co… Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable(a)vger.kernel.org Signed-off-by: Thomas Weißschuh <linux(a)weissschuh.net> --- I'm not really familiar with the ELF loading process, so putting this out as RFC. A example binary compiled with aarch64-linux-gnu-gcc 13.2.0 is available at https://test.t-8ch.de/binfmt-bss-repro.bin --- fs/binfmt_elf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 7b3d2d491407..4008a57d388b 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -112,7 +112,7 @@ static struct linux_binfmt elf_format = { static int set_brk(unsigned long start, unsigned long end, int prot) { - start = ELF_PAGEALIGN(start); + start = ELF_PAGESTART(start); end = ELF_PAGEALIGN(end); if (end > start) { /* --- base-commit: aed8aee11130a954356200afa3f1b8753e8a9482 change-id: 20230914-bss-alloc-f523fa61718c Best regards, -- Thomas Weißschuh <linux(a)weissschuh.net>

1 year, 2 months

8
18
0 0

[PATCH] ACPI: FPDT: break out of the loop if record length is zero

by Vasily Khoruzhick

Buggy BIOSes may have zero-length records in FPDT, table, as a result fpdt_process_subtable() spins in eternal loop. Break out of the loop if record length is zero. Fixes: d1eb86e59be0 ("ACPI: tables: introduce support for FPDT table") Cc: stable(a)vger.kernel.org Signed-off-by: Vasily Khoruzhick <anarsoul(a)gmail.com> --- drivers/acpi/acpi_fpdt.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/acpi/acpi_fpdt.c b/drivers/acpi/acpi_fpdt.c index a2056c4c8cb7..53d8f9601a55 100644 --- a/drivers/acpi/acpi_fpdt.c +++ b/drivers/acpi/acpi_fpdt.c @@ -194,6 +194,11 @@ static int fpdt_process_subtable(u64 address, u32 subtable_type) record_header = (void *)subtable_header + offset; offset += record_header->length; + if (!record_header->length) { + pr_info(FW_BUG "Zero-length record found.\n"); + break; + } + switch (record_header->type) { case RECORD_S3_RESUME: if (subtable_type != SUBTABLE_S3PT) { -- 2.42.0

1 year, 2 months

2
2
0 0

[PATCH] LoongArch: numa: Fix high_memory calculation

by Huacai Chen

high_memory is the virtual address of the 'highest physical address' in the system. But __va(get_num_physpages() << PAGE_SHIFT) is not what we want because there may be holes in the physical address space. On the other hand, max_low_pfn is calculated from memblock_end_of_DRAM(), which is exactly corresponding to the highest physical address, so use it for high_memory calculation. Cc: <stable(a)vger.kernel.org> Signed-off-by: Chong Qiao <qiaochong(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> --- arch/loongarch/kernel/numa.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/loongarch/kernel/numa.c b/arch/loongarch/kernel/numa.c index c7d33c489e04..6e65ff12d5c7 100644 --- a/arch/loongarch/kernel/numa.c +++ b/arch/loongarch/kernel/numa.c @@ -436,7 +436,7 @@ void __init paging_init(void) void __init mem_init(void) { - high_memory = (void *) __va(get_num_physpages() << PAGE_SHIFT); + high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT); memblock_free_all(); } -- 2.39.3

1 year, 2 months

3
2
0 0

Re: [PATCH v4 0/2] Avoid spurious freezer wakeups

by Carlos Llamas

On Tue, Sep 26, 2023 at 10:02:38PM +0200, Peter Zijlstra wrote: > On Tue, Sep 26, 2023 at 04:17:33PM +0000, Carlos Llamas wrote: > > > > This issue is hurting the performance of our stable 6.1 releases. Does > > it make sense to backport these patches into stable branches once they > > land in mainline? I would assume we want to fix the perf regression > > there too? > > Note that these patches are in tip/sched/core, slated for the next merge > window. We can wait, no problem. I just wanted to make sure we also patch stable if needed. Elliot, would you be able to send a backport of your patches to stable once they land in mainline on the next merge window? Cc: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: stable(a)vger.kernel.org -- Carlos Llamas

1 year, 2 months

2
1
0 0

[PATCH 04/14] drm/amd/display: apply edge-case DISPCLK WDIVIDER changes to master OTG pipes only

by Aurabindo Pillai

From: Samson Tam <samson.tam(a)amd.com> [Why] The edge-case DISPCLK WDIVIDER changes call stream_enc functions. But with MPC pipes, downstream pipes have null stream_enc and will cause crash. [How] Only call stream_enc functions for pipes that are OTG master. Reviewed-by: Alvin Lee <alvin.lee2(a)amd.com> Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Acked-by: Aurabindo Pillai <aurabindo.pillai(a)amd.com> Signed-off-by: Samson Tam <samson.tam(a)amd.com> --- drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c | 4 ++-- drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index c435f7632e8e..5ee87965a078 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -157,7 +157,7 @@ void dcn20_update_clocks_update_dentist(struct clk_mgr_internal *clk_mgr, struct int32_t N; int32_t j; - if (!pipe_ctx->stream) + if (!resource_is_pipe_type(pipe_ctx, OTG_MASTER)) continue; /* Virtual encoders don't have this function */ if (!stream_enc->funcs->get_fifo_cal_average_level) @@ -188,7 +188,7 @@ void dcn20_update_clocks_update_dentist(struct clk_mgr_internal *clk_mgr, struct int32_t N; int32_t j; - if (!pipe_ctx->stream) + if (!resource_is_pipe_type(pipe_ctx, OTG_MASTER)) continue; /* Virtual encoders don't have this function */ if (!stream_enc->funcs->get_fifo_cal_average_level) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c index 37ffa0050e60..a496930b1f9c 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c @@ -363,7 +363,7 @@ static void dcn32_update_clocks_update_dentist( int32_t N; int32_t j; - if (!pipe_ctx->stream) + if (!resource_is_pipe_type(pipe_ctx, OTG_MASTER)) continue; /* Virtual encoders don't have this function */ if (!stream_enc->funcs->get_fifo_cal_average_level) @@ -409,7 +409,7 @@ static void dcn32_update_clocks_update_dentist( int32_t N; int32_t j; - if (!pipe_ctx->stream) + if (!resource_is_pipe_type(pipe_ctx, OTG_MASTER)) continue; /* Virtual encoders don't have this function */ if (!stream_enc->funcs->get_fifo_cal_average_level) -- 2.42.0

1 year, 2 months

1
0
0 0

[PATCH 03/14] drm/amd/display: enable dsc_clk even if dsc_pg disabled

by Aurabindo Pillai

From: Muhammad Ahmed <ahmed.ahmed(a)amd.com> [why] need to enable dsc_clk regardless dsc_pg Reviewed-by: Charlene Liu <charlene.liu(a)amd.com> Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Acked-by: Aurabindo Pillai <aurabindo.pillai(a)amd.com> Signed-off-by: Muhammad Ahmed <ahmed.ahmed(a)amd.com> --- drivers/gpu/drm/amd/display/dc/core/dc.c | 8 ++++---- drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c | 3 +++ drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c | 2 +- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index cb3cb2db90ee..6f2300f71ca8 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1853,7 +1853,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c if (dc->hwss.subvp_pipe_control_lock) dc->hwss.subvp_pipe_control_lock(dc, context, true, true, NULL, subvp_prev_use); - if (dc->debug.enable_double_buffered_dsc_pg_support) + if (dc->hwss.update_dsc_pg) dc->hwss.update_dsc_pg(dc, context, false); disable_dangling_plane(dc, context); @@ -1960,7 +1960,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c dc->hwss.optimize_bandwidth(dc, context); } - if (dc->debug.enable_double_buffered_dsc_pg_support) + if (dc->hwss.update_dsc_pg) dc->hwss.update_dsc_pg(dc, context, true); if (dc->ctx->dce_version >= DCE_VERSION_MAX) @@ -2207,7 +2207,7 @@ void dc_post_update_surfaces_to_stream(struct dc *dc) dc->hwss.optimize_bandwidth(dc, context); - if (dc->debug.enable_double_buffered_dsc_pg_support) + if (dc->hwss.update_dsc_pg) dc->hwss.update_dsc_pg(dc, context, true); } @@ -3565,7 +3565,7 @@ static void commit_planes_for_stream(struct dc *dc, if (get_seamless_boot_stream_count(context) == 0) dc->hwss.prepare_bandwidth(dc, context); - if (dc->debug.enable_double_buffered_dsc_pg_support) + if (dc->hwss.update_dsc_pg) dc->hwss.update_dsc_pg(dc, context, false); context_clock_trace(dc, context); diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c index 76fd7a41bdbf..45b557d8e089 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c @@ -77,6 +77,9 @@ void dcn32_dsc_pg_control( if (hws->ctx->dc->debug.disable_dsc_power_gate) return; + if (!hws->ctx->dc->debug.enable_double_buffered_dsc_pg_support) + return; + REG_GET(DC_IP_REQUEST_CNTL, IP_REQUEST_EN, &org_ip_request_cntl); if (org_ip_request_cntl == 0) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c index 10ae1b3da751..6214866916c7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c @@ -742,7 +742,7 @@ static const struct dc_debug_options debug_defaults_drv = { .disable_mem_low_power = false, .enable_hpo_pg_support = false, //must match enable_single_display_2to1_odm_policy to support dynamic ODM transitions - .enable_double_buffered_dsc_pg_support = false, + .enable_double_buffered_dsc_pg_support = true, .enable_dp_dig_pixel_rate_div_policy = 1, .disable_z10 = false, .ignore_pg = true, -- 2.42.0

1 year, 2 months

1
0
0 0

[PATCH 02/14] drm/amd/display: Add Null check for DPP resource

by Aurabindo Pillai

From: Gabe Teeger <gabe.teeger(a)amd.com> [what and why] Check whether dpp resource pointer is null in advance and return early if so. Reviewed-by: Charlene Liu <charlene.liu(a)amd.com> Reviewed-by: Martin Leung <martin.leung(a)amd.com> Signed-off-by: Gabe Teeger <gabe.teeger(a)amd.com> Cc: Mario Limonciello <mario.limonciello(a)amd.com> Cc: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org Acked-by: Aurabindo Pillai <aurabindo.pillai(a)amd.com> --- drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 628b902a4cec..aa7b5db83644 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -945,7 +945,7 @@ static void adjust_recout_for_visual_confirm(struct rect *recout, struct dc *dc = pipe_ctx->stream->ctx->dc; int dpp_offset, base_offset; - if (dc->debug.visual_confirm == VISUAL_CONFIRM_DISABLE) + if (dc->debug.visual_confirm == VISUAL_CONFIRM_DISABLE || !pipe_ctx->plane_res.dpp) return; dpp_offset = pipe_ctx->stream->timing.v_addressable / VISUAL_CONFIRM_DPP_OFFSET_DENO; -- 2.42.0

1 year, 2 months

1
0
0 0

+ selftests-mm-fix-awk-usage-in-charge_reserved_hugetlbsh-and-hugetlb_reparenting_testsh-that-may-cause-error.patch added to mm-hotfixes-unstable branch

by Andrew Morton

The patch titled Subject: selftests/mm: fix awk usage in charge_reserved_hugetlb.sh and hugetlb_reparenting_test.sh that may cause error has been added to the -mm mm-hotfixes-unstable branch. Its filename is selftests-mm-fix-awk-usage-in-charge_reserved_hugetlbsh-and-hugetlb_reparenting_testsh-that-may-cause-error.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patche… This patch will later appear in the mm-hotfixes-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days ------------------------------------------------------ From: Juntong Deng <juntong.deng(a)outlook.com> Subject: selftests/mm: fix awk usage in charge_reserved_hugetlb.sh and hugetlb_reparenting_test.sh that may cause error Date: Wed, 27 Sep 2023 02:19:44 +0800 According to the awk manual, the -e option does not need to be specified in front of 'program' (unless you need to mix program-file). The redundant -e option can cause error when users use awk tools other than gawk (for example, mawk does not support the -e option). Error Example: awk: not an option: -e Link: https://lkml.kernel.org/r/VI1P193MB075228810591AF2FDD7D42C599C3A@VI1P193MB0… Signed-off-by: Juntong Deng <juntong.deng(a)outlook.com> Cc: Shuah Khan <shuah(a)kernel.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- tools/testing/selftests/mm/charge_reserved_hugetlb.sh | 4 ++-- tools/testing/selftests/mm/hugetlb_reparenting_test.sh | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) --- a/tools/testing/selftests/mm/charge_reserved_hugetlb.sh~selftests-mm-fix-awk-usage-in-charge_reserved_hugetlbsh-and-hugetlb_reparenting_testsh-that-may-cause-error +++ a/tools/testing/selftests/mm/charge_reserved_hugetlb.sh @@ -25,7 +25,7 @@ if [[ "$1" == "-cgroup-v2" ]]; then fi if [[ $cgroup2 ]]; then - cgroup_path=$(mount -t cgroup2 | head -1 | awk -e '{print $3}') + cgroup_path=$(mount -t cgroup2 | head -1 | awk '{print $3}') if [[ -z "$cgroup_path" ]]; then cgroup_path=/dev/cgroup/memory mount -t cgroup2 none $cgroup_path @@ -33,7 +33,7 @@ if [[ $cgroup2 ]]; then fi echo "+hugetlb" >$cgroup_path/cgroup.subtree_control else - cgroup_path=$(mount -t cgroup | grep ",hugetlb" | awk -e '{print $3}') + cgroup_path=$(mount -t cgroup | grep ",hugetlb" | awk '{print $3}') if [[ -z "$cgroup_path" ]]; then cgroup_path=/dev/cgroup/memory mount -t cgroup memory,hugetlb $cgroup_path --- a/tools/testing/selftests/mm/hugetlb_reparenting_test.sh~selftests-mm-fix-awk-usage-in-charge_reserved_hugetlbsh-and-hugetlb_reparenting_testsh-that-may-cause-error +++ a/tools/testing/selftests/mm/hugetlb_reparenting_test.sh @@ -20,7 +20,7 @@ fi if [[ $cgroup2 ]]; then - CGROUP_ROOT=$(mount -t cgroup2 | head -1 | awk -e '{print $3}') + CGROUP_ROOT=$(mount -t cgroup2 | head -1 | awk '{print $3}') if [[ -z "$CGROUP_ROOT" ]]; then CGROUP_ROOT=/dev/cgroup/memory mount -t cgroup2 none $CGROUP_ROOT @@ -28,7 +28,7 @@ if [[ $cgroup2 ]]; then fi echo "+hugetlb +memory" >$CGROUP_ROOT/cgroup.subtree_control else - CGROUP_ROOT=$(mount -t cgroup | grep ",hugetlb" | awk -e '{print $3}') + CGROUP_ROOT=$(mount -t cgroup | grep ",hugetlb" | awk '{print $3}') if [[ -z "$CGROUP_ROOT" ]]; then CGROUP_ROOT=/dev/cgroup/memory mount -t cgroup memory,hugetlb $CGROUP_ROOT _ Patches currently in -mm which might be from juntong.deng(a)outlook.com are selftests-mm-fix-awk-usage-in-charge_reserved_hugetlbsh-and-hugetlb_reparenting_testsh-that-may-cause-error.patch

1 year, 2 months

1
0
0 0

[PATCH] x86/kvm: Account for fpstate->user_xfeatures changes

by Tyler Stachecki

Live-migrations under qemu result in guest corruption when the following three conditions are all met: * The source host CPU has capabilities that itself extend that of the guest CPU fpstate->user_xfeatures * The source kernel emits guest_fpu->user_xfeatures with respect to the host CPU (i.e. it *does not* have the "Fixes:" commit) * The destination kernel enforces that the xfeatures in the buffer given to KVM_SET_IOCTL are compatible with the guest CPU (i.e., it *does* have the "Fixes:" commit) When these conditions are met, the semantical changes to fpstate->user_features trigger a subtle bug in qemu that results in qemu failing to put the XSAVE architectural state into KVM. qemu then both ceases to put the remaining (non-XSAVE) x86 architectural state into KVM and makes the fateful mistake of resuming the guest anyways. This usually results in immediate guest corruption, silent or not. Due to the grave nature of this qemu bug, attempt to retain behavior of old kernels by clamping the xfeatures specified in the buffer given to KVM_SET_IOCTL such that it aligns with the guests fpstate->user_xfeatures instead of returning an error. Fixes: ad856280ddea ("x86/kvm/fpu: Limit guest user_xfeatures to supported bits of XCR0") Cc: stable(a)vger.kernel.org Cc: Leonardo Bras <leobras(a)redhat.com> Signed-off-by: Tyler Stachecki <tstachecki(a)bloomberg.net> --- arch/x86/kvm/x86.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 6c9c81e82e65..baad160b592f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5407,11 +5407,21 @@ static void kvm_vcpu_ioctl_x86_get_xsave2(struct kvm_vcpu *vcpu, static int kvm_vcpu_ioctl_x86_set_xsave(struct kvm_vcpu *vcpu, struct kvm_xsave *guest_xsave) { + union fpregs_state *ustate = (union fpregs_state *) guest_xsave->region; + u64 user_xfeatures = vcpu->arch.guest_fpu.fpstate->user_xfeatures; + if (fpstate_is_confidential(&vcpu->arch.guest_fpu)) return 0; - return fpu_copy_uabi_to_guest_fpstate(&vcpu->arch.guest_fpu, - guest_xsave->region, + /* + * In previous kernels, kvm_arch_vcpu_create() set the guest's fpstate + * based on what the host CPU supported. Recent kernels changed this + * and only accept ustate containing xfeatures that the guest CPU is + * capable of supporting. + */ + ustate->xsave.header.xfeatures &= user_xfeatures; + + return fpu_copy_uabi_to_guest_fpstate(&vcpu->arch.guest_fpu, ustate, kvm_caps.supported_xcr0, &vcpu->arch.pkru); } -- 2.30.2

1 year, 2 months

4
15
0 0

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror September 2023