x86/bpf: Fixes for the BPF JIT with retbleed=stuff
Subsystems (5)
v2 cover x86/bpf: Fixes for the BPF JIT with retbleed=stuff
From: Joan Bruguera Micó <[email protected]>
Fixes two issues that cause kernels panic when using the BPF JIT with the call depth tracking / stuffing mitigation for Skylake processors (retbleed=stuff). Both issues can be triggered by running simple BPF programs (e.g. running the test suite should trigger both).
The first (resubmit) fixes a trivial issue related to calculating the destination IP for call instructions with call depth tracking.
The second is related to using the correct IP for relocations, related to the recently introduced %rip-relative addressing for PER_CPU_VAR.
---
v2:
Simplify calculation of "ip".
Add more details to the commit message.
Joan Bruguera Micó (1):
x86/bpf: Fix IP for relocating call depth accounting
Uros Bizjak (1):
x86/bpf: Fix IP after emitting call depth accounting
arch/x86/include/asm/alternative.h | 4 ++-- arch/x86/kernel/callthunks.c | 4 ++-- arch/x86/net/bpf_jit_comp.c | 19 ++++++++----------- 3 files changed, 12 insertions(+), 15 deletions(-)
Signature · 1 line
2.42.0
PATCH RESEND v2 1/2 x86/bpf: Fix IP after emitting call depth accounting
Adjust the IP passed to emit_patch so it calculates the correct offset for the CALL instruction if x86_call_depth_emit_accounting emits code. Otherwise we will skip some instructions and most likely crash.
---
v2: Simplify calculation of "ip".
--- arch/x86/net/bpf_jit_comp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/net/bpf_jit_comp.c +1 -1
@@ -480,7 +480,7 @@ static int emit_call(u8 **pprog, void *func, void *ip)static int emit_rsb_call (u8 **pprog,void *func,void *ip) {OPTIMIZER_HIDE_VAR (func); -x86_call_depth_emit_accounting (pprog, func); + ip +=x86_call_depth_emit_accounting (pprog, func);return emit_patch (pprog, func, ip,0xE8 ); }
Signature · 1 line
2.42.0
PATCH RESEND v2 2/2 x86/bpf: Fix IP for relocating call depth accounting
In reply to Uros BizjakFrom: Joan Bruguera Micó <[email protected]>
The commit:
59bec00ace28 ("x86/percpu: Introduce %rip-relative addressing to PER_CPU_VAR()")
made PER_CPU_VAR() to use rip-relative addressing, hence INCREMENT_CALL_DEPTH macro and skl_call_thunk_template got rip-relative asm code inside of it. A follow up commit:
17bce3b2ae2d ("x86/callthunks: Handle %rip-relative relocations in call thunk template")
changed x86_call_depth_emit_accounting() to use apply_relocation(), but mistakenly assumed that the code is being patched in-place (where the destination of the relocation matches the address of the code), using *pprog as the destination ip. This is not true for the call depth accounting, emitted by the BPF JIT, so the calculated address was wrong, JIT-ed BPF progs on kernels with call depth tracking got broken and usually caused a page fault.
Pass the destination IP when the BPF JIT emits call depth accounting.
---
v2: Add more details to the commit message.
--- arch/x86/include/asm/alternative.h | 4 ++-- arch/x86/kernel/callthunks.c | 4 ++-- arch/x86/net/bpf_jit_comp.c | 19 ++++++++----------- 3 files changed, 12 insertions(+), 15 deletions(-)
arch/x86/include/asm/alternative.h +2 -2
@@ -117,7 +117,7 @@ extern void callthunks_patch_builtin_calls(void);extern void callthunks_patch_module_calls (struct callthunk_sites *sites,struct module *mod);extern void *callthunks_translate_call_dest (void *dest); -extern int x86_call_depth_emit_accounting (u8 **pprog,void *func); +extern int x86_call_depth_emit_accounting (u8 **pprog,void *func,void *ip);#else static __always_inline void callthunks_patch_builtin_calls(void) {}static __always_inline void
@@ -128,7 +128,7 @@ static __always_inline void *callthunks_translate_call_dest(void *dest)return dest; } static __always_inline int x86_call_depth_emit_accounting(u8 **pprog, -void *func) +void *func,void *ip) {return 0 ; }
arch/x86/kernel/callthunks.c +2 -2
@@ -314,7 +314,7 @@ static bool is_callthunk(void *addr)return !bcmp (pad, insn_buff, tmpl_size); } -int x86_call_depth_emit_accounting (u8 **pprog,void *func) +int x86_call_depth_emit_accounting (u8 **pprog,void *func,void *ip) {unsigned int tmpl_size =SKL_TMPL_SIZE ;u8 insn_buff[MAX_PATCH_LEN ];
@@ -327,7 +327,7 @@ int x86_call_depth_emit_accounting(u8 **pprog, void *func)return 0 ;memcpy (insn_buff, skl_call_thunk_template, tmpl_size); -apply_relocation (insn_buff, tmpl_size, *pprog, +apply_relocation (insn_buff, tmpl_size, ip, skl_call_thunk_template, tmpl_size);memcpy (*pprog, insn_buff, tmpl_size);
arch/x86/net/bpf_jit_comp.c +8 -11
@@ -480,7 +480,7 @@ static int emit_call(u8 **pprog, void *func, void *ip)static int emit_rsb_call (u8 **pprog,void *func,void *ip) {OPTIMIZER_HIDE_VAR (func); - ip +=x86_call_depth_emit_accounting (pprog, func); + ip +=x86_call_depth_emit_accounting (pprog, func, ip);return emit_patch (pprog, func, ip,0xE8 ); }
@@ -1972,20 +1972,17 @@ st: if (is_imm8(insn->off))/* call */ case BPF_JMP |BPF_CALL : { -int offs; +u8 *ip = image + addrs[i -1 ]; func = (u8 *) __bpf_call_base + imm32;if (tail_call_reachable) {RESTORE_TAIL_CALL_CNT (bpf_prog->aux->stack_depth); -if (!imm32) -return -EINVAL ; - offs =7 +x86_call_depth_emit_accounting (&prog, func); - }else { -if (!imm32) -return -EINVAL ; - offs =x86_call_depth_emit_accounting (&prog, func); + ip +=7 ; } -if (emit_call (&prog, func, image + addrs[i -1 ] + offs)) +if (!imm32) +return -EINVAL ; + ip +=x86_call_depth_emit_accounting (&prog, func, ip); +if (emit_call (&prog, func, ip))return -EINVAL ;break ; }
@@ -2835,7 +2832,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *rw_im* Direct-call fentry stub, as such it needs accounting for the * __fentry__ call. */ -x86_call_depth_emit_accounting (&prog,NULL ); +x86_call_depth_emit_accounting (&prog,NULL , image); }EMIT1 (0x55 );/* push rbp */ EMIT3 (0x48 ,0x89 ,0xE5 );/* mov rbp, rsp */
Signature · 1 line
2.42.0
This series was applied to bpf/bpf.git (master) by Alexei Starovoitov <[email protected]>:
Hello:
This series was applied to bpf/bpf.git (master)
by Alexei Starovoitov <[email protected]>:
Quoted 8 lines · On Mon, 1 Apr 2024 20:55:28 +0200 you wrote:
From: Joan Bruguera Micó <[email protected]>
Fixes two issues that cause kernels panic when using the BPF JIT with the call depth tracking / stuffing mitigation for Skylake processors (
retbleed=stuff). Both issues can be triggered by running simple BPF programs (e.g. running the test suite should trigger both).[...]
Here is the summary with links:
- [RESEND,bpf,v2,1/2] x86/bpf: Fix IP after emitting call depth accounting
https://git.kernel.org/bpf/bpf/c/9d98aa088386 - [RESEND,bpf,v2,2/2] x86/bpf: Fix IP for relocating call depth accounting https://git.kernel.org/bpf/bpf/c/6a537453000a
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html
Quoted 13 lines · * [email protected] <[email protected]> wrote:
Hello:
This series was applied to bpf/bpf.git (master)
by Alexei Starovoitov <[email protected]>:Quoted 7 lines · On Mon, 1 Apr 2024 20:55:28 +0200 you wrote:
From: Joan Bruguera Micó <[email protected]>
Fixes two issues that cause kernels panic when using the BPF JIT with the call depth tracking / stuffing mitigation for Skylake processors (
retbleed=stuff). Both issues can be triggered by running simple BPF programs (e.g. running the test suite should trigger both).
Quoted 7 lines
Quoted 1 line
[...]
Here is the summary with links:
- [RESEND,bpf,v2,1/2] x86/bpf: Fix IP after emitting call depth accountinghttps://git.kernel.org/bpf/bpf/c/9d98aa088386 - [RESEND,bpf,v2,2/2] x86/bpf: Fix IP for relocating call depth accounting https://git.kernel.org/bpf/bpf/c/6a537453000a
Just wondering, which kernel version is this targeted for?
The bug is upstream as well, so a fix needs to be sent to Linus.
I can pick all of this up into tip:x86/urgent, if that accelerates things.
Thanks,
Ingo
Quoted 19 lines · Ingo Molnar
Quoted 15 lines · * [email protected] <[email protected]> wrote:
Hello:
This series was applied to bpf/bpf.git (master)
by Alexei Starovoitov <[email protected]>:Quoted 8 lines · On Mon, 1 Apr 2024 20:55:28 +0200 you wrote:
From: Joan Bruguera Micó <[email protected]>
Fixes two issues that cause kernels panic when using the BPF JIT with the call depth tracking / stuffing mitigation for Skylake processors (
retbleed=stuff). Both issues can be triggered by running simple BPF programs (e.g. running the test suite should trigger both).[...]
Quoted 7 lines
Quoted 5 lines
Here is the summary with links:
- [RESEND,bpf,v2,1/2] x86/bpf: Fix IP after emitting call depth accountinghttps://git.kernel.org/bpf/bpf/c/9d98aa088386 - [RESEND,bpf,v2,2/2] x86/bpf: Fix IP for relocating call depth accounting https://git.kernel.org/bpf/bpf/c/6a537453000aJust wondering, which kernel version is this targeted for?
The whole series is intended for the current mainline (v6.9), this is why it is developed against the bpf (*not* bpf-next) branch. Please note that the kernel panics with retbleed=stuff even without %rip-relative changes (patch 1/2 above) [1], so patch 1/2 should be backported to stable branches.
[1] https://lore.kernel.org/lkml/[email protected]/
Uros.
Quoted 8 lines
The bug is upstream as well, so a fix needs to be sent to Linus.
I can pick all of this up into tip:x86/urgent, if that accelerates things.
Thanks,
Ingo
Quoted 10 lines · Uros Bizjak
Quoted 7 lines
Quoted 5 lines
Here is the summary with links:
- [RESEND,bpf,v2,1/2] x86/bpf: Fix IP after emitting call depth accountinghttps://git.kernel.org/bpf/bpf/c/9d98aa088386 - [RESEND,bpf,v2,2/2] x86/bpf: Fix IP for relocating call depth accounting https://git.kernel.org/bpf/bpf/c/6a537453000aJust wondering, which kernel version is this targeted for?
The whole series is intended for the current mainline (v6.9), this is why it is developed against the bpf (*not* bpf-next) branch.
I see - so bpf.git:master are current-mainline BPF fixes - that's perfect.
BTW., bpf-next is not a branch, but a separate Git tree. (Which is what confused me.)
Thanks,
Ingo
Daniel Borkmann
maintainerQuoted 8 lines · Ingo Molnar
Quoted 6 lines · Uros Bizjak
Quoted 6 lines
Quoted 5 lines
Here is the summary with links:
- [RESEND,bpf,v2,1/2] x86/bpf: Fix IP after emitting call depth accountinghttps://git.kernel.org/bpf/bpf/c/9d98aa088386 - [RESEND,bpf,v2,2/2] x86/bpf: Fix IP for relocating call depth accounting https://git.kernel.org/bpf/bpf/c/6a537453000a
Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Just wondering, which kernel version is this targeted for?
The whole series is intended for the current mainline (v6.9), this is why it is developed against the bpf (*not* bpf-next) branch.
I see - so bpf.git:master are current-mainline BPF fixes - that's perfect.
Yes, correct, that is for mainline (v6.9).
Thanks,
Daniel