x86/retbleed: Call depth tracking mitigation
Status from this message: The following commit has been merged into the x86/paravirt branch of tip:
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v3 Latest59 patches 19 repliesApplied
- v259 patches 21 repliesSuperseded
Patches in v3 Applied
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/59 | x86/paravirt: Ensure proper alignmentarch/ arch/ | +2 | 0 replies | |
| 2/59 | x86/cpu: Remove segment load from switch_to_new_gdt()arch/ arch/ | +31 | 0 replies | |
| 3/59 | x86/cpu: Get rid of redundant switch_to_new_gdt() invocationsarch/ arch/ arch/ +2 more | +14 | 0 replies | |
| 4/59 | x86/cpu: Re-enable stackprotectorarch/ arch/ | +3 | 0 replies | |
| 5/59 | x86/modules: Set VM_FLUSH_RESET_PERMS in module_alloc()arch/ arch/ arch/ | +5 | 0 replies | |
| 6/59 | x86/vdso: Ensure all kernel code is seen by objtoolarch/ | +6 | 0 replies | |
| 7/59 | x86: Sanitize linker scriptarch/ | +7 | 1 reply | |
| 8/59 | arch: Introduce CONFIG_FUNCTION_ALIGNMENTMakefile arch/ arch/ +6 more | +36 | 0 replies | |
| 9/59 | x86/asm: Differentiate between code and function alignmentarch/ | +9 | 0 replies | |
| 10/59 | x86/error_inject: Align function properlyarch/ | +1 | 0 replies | |
| 11/59 | x86/paravirt: Properly align PV functionsarch/ arch/ arch/ +1 more | +5 | 1 reply | |
| 12/59 | x86/entry: Align SYM_CODE_START() variantsarch/ arch/ | +10 | 0 replies | |
| 13/59 | crypto: x86/camellia: Remove redundant alignmentsarch/ arch/ | +0 | 0 replies | |
| 14/59 | crypto: x86/cast5: Remove redundant alignmentsarch/ | +0 | 0 replies | |
| 15/59 | crypto: x86/crct10dif-pcl: Remove redundant alignmentsarch/ | +0 | 0 replies | |
| 16/59 | crypto: x86/serpent: Remove redundant alignmentsarch/ arch/ | +0 | 0 replies | |
| 17/59 | crypto: x86/sha1: Remove custom alignmentsarch/ | +0 | 0 replies | |
| 18/59 | crypto: x86/sha256: Remove custom alignmentsarch/ arch/ arch/ +1 more | +0 | 0 replies | |
| 19/59 | crypto: x86/sm[34]: Remove redundant alignmentsarch/ arch/ arch/ | +0 | 0 replies | |
| 20/59 | crypto: twofish: Remove redundant alignmentsarch/ | +0 | 0 replies | |
| 21/59 | crypto: x86/poly1305: Remove custom function alignmentarch/ | +0 | 0 replies | |
| 22/59 | x86: Put hot per CPU variables into a structarch/ arch/ arch/ +2 more | +24 | 0 replies | |
| 23/59 | x86/percpu: Move preempt_count next to current_taskarch/ arch/ arch/ | +16 | 0 replies | |
| 24/59 | x86/percpu: Move cpu_number next to current_taskarch/ arch/ arch/ | +7 | 0 replies | |
| 25/59 | x86/percpu: Move current_top_of_stack next to current_taskarch/ arch/ arch/ +8 more | +19 | 0 replies | |
| 26/59 | x86/percpu: Move irq_stack variables next to current_taskarch/ arch/ arch/ +6 more | +24 | 0 replies | |
| 27/59 | x86/softirq: Move softirq pending next to current taskarch/ arch/ | +3 | 0 replies | |
| 28/59 | objtool: Allow !PC relative relocationstools/ tools/ tools/ | +27 | 0 replies | |
| 29/59 | objtool: Track init sectiontools/ tools/ | +11 | 0 replies | |
| 30/59 | objtool: Add .call_sites sectionarch/ tools/ tools/ +1 more | +60 | 0 replies | |
| 31/59 | objtool: Add --hacks=skylakescripts/ tools/ tools/ +1 more | +15 | 0 replies | |
| 32/59 | objtool: Allow STT_NOTYPE -> STT_FUNC+0 tail-callstools/ | +20 | 6 replies | |
| 33/59 | objtool: Fix find_{symbol,func}_containing()tools/ tools/ tools/ | +229 | 0 replies | |
| 34/59 | objtool: Allow symbol range comparisons for IBT/ENDBRarch/ tools/ | +28 | 0 replies | |
| 35/59 | x86/entry: Make sync_regs() invocation a tail callarch/ | +2 | 0 replies | |
| 36/59 | ftrace: Add HAVE_DYNAMIC_FTRACE_NO_PATCHABLEinclude/ kernel/ tools/ | +18 | 0 replies | |
| 37/59 | x86/putuser: Provide room for paddingarch/ | +49 | 0 replies | |
| 38/59 | x86/Kconfig: Add CONFIG_CALL_THUNKSarch/ | +8 | 0 replies | |
| 39/59 | x86/Kconfig: Introduce function paddingarch/ arch/ arch/ +1 more | +40 | 0 replies | |
| 40/59 | x86/retbleed: Add X86_FEATURE_CALL_DEPTHarch/ arch/ arch/ | +28 | 0 replies | |
| 41/59 | x86/alternatives: Provide text_poke_copy_locked()arch/ arch/ | +23 | 0 replies | |
| 42/59 | x86/entry: Make some entry symbols globalarch/ | +6 | 0 replies | |
| 43/59 | x86/paravirt: Make struct paravirt_call_site unconditionally availablearch/ arch/ | +14 | 0 replies | |
| 44/59 | x86/callthunks: Add call patching for call depth trackingarch/ arch/ arch/ +5 more | +287 | 0 replies | |
| 45/59 | x86/modules: Add call patchingarch/ arch/ arch/ | +43 | 0 replies | |
| 46/59 | x86/returnthunk: Allow different return thunksarch/ arch/ arch/ +2 more | +23 | 0 replies | |
| 47/59 | x86/asm: Provide ALTERNATIVE_3arch/ | +30 | 0 replies | |
| 48/59 | x86/retbleed: Add SKL return thunkarch/ arch/ arch/ +3 more | +156 | 8 replies | |
| 49/59 | x86/retpoline: Add SKL retthunk retpolinesarch/ arch/ arch/ +1 more | +135 | 0 replies | |
| 50/59 | x86/retbleed: Add SKL call thunkarch/ | +20 | 0 replies | |
| 51/59 | x86/calldepth: Add ret/call counting for debugarch/ arch/ arch/ | +91 | 0 replies | |
| 52/59 | static_call: Add call depth tracking supportarch/ arch/ arch/ +2 more | +44 | 0 replies | |
| 53/59 | kallsyms: Take callthunks into accountkernel/ | +40 | 0 replies | |
| 54/59 | x86/orc: Make it callthunk awarearch/ arch/ arch/ | +38 | 0 replies | |
| 55/59 | x86/bpf: Emit call depth accounting if requiredarch/ arch/ arch/ | +48 | 1 reply | |
| 56/59 | x86/ftrace: Remove ftrace_epilogue()arch/ | +6 | 1 reply | |
| 57/59 | x86/ftrace: Rebalance RSBarch/ | +11 | 0 replies | |
| 58/59 | x86/ftrace: Make it call depth tracking awarearch/ arch/ arch/ +8 more | +64 | 0 replies | |
| 58/59 | x86/ftrace: Make it call depth tracking awarearch/ arch/ arch/ +8 more | +68 | 1 reply | |
| 59/59 | x86/retbleed: Add call depth tracking mitigationarch/ | +30 | 0 replies |
Reviews trailers per patch
2 of 60 patches have a Reviewed-by or Acked-by.
| Patch | Juergen Gross | Linus Torvalds | Nathan Chancellor |
|---|---|---|---|
| 1/59 | R-b | ||
| 2/59 | |||
| 3/59 | |||
| 4/59 | |||
| 5/59 | |||
| 6/59 | |||
| 7/59 | |||
| 8/59 | Requested-by | ||
| 9/59 | |||
| 10/59 | |||
| 11/59 | R-b | ||
| 12/59 | |||
| 13/59 | |||
| 14/59 | |||
| 15/59 | |||
| 16/59 | |||
| 17/59 | |||
| 18/59 | |||
| 19/59 | |||
| 20/59 | |||
| 21/59 | |||
| 22/59 | |||
| 23/59 | |||
| 24/59 | |||
| 25/59 | |||
| 26/59 | |||
| 27/59 | |||
| 28/59 | |||
| 29/59 | |||
| 30/59 | |||
| 31/59 | |||
| 32/59 | |||
| 33/59 | |||
| 34/59 | |||
| 35/59 | |||
| 36/59 | |||
| 37/59 | |||
| 38/59 | |||
| 39/59 | |||
| 40/59 | |||
| 41/59 | |||
| 42/59 | |||
| 43/59 | |||
| 44/59 | |||
| 45/59 | |||
| 46/59 | |||
| 47/59 | |||
| 48/59 | T-b | ||
| 49/59 | |||
| 50/59 | |||
| 51/59 | |||
| 52/59 | |||
| 53/59 | |||
| 54/59 | |||
| 55/59 | |||
| 56/59 | |||
| 57/59 | |||
| 58/59 | |||
| 58/59 | |||
| 59/59 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 3 reviewers.
Cover letter [PATCH v3 00/59] x86/retbleed: Call depth tracking mitigation
Hi!
Previous postings:
v2: https://lkml.kernel.org/r/[email protected]
v1: https://lkml.kernel.org/r/[email protected]
Changes since v2 are minimal; I reworked the alignment thing per Linus' request (patch #8) and collected a few tags.
Barring great objections I'm hoping to merge this soon so we can all get on with other things.
--- text from v2 ---
This version is significantly different from the last in that it no longer makes use of external call thunks allocated from the module space. Instead every function gets aligned to 16 bytes and gets 16 bytes of (pre-symbol) padding. (This padding will also come in handy for other things, like the kCFI/FineIBT work.)
Prior to these patches function alignment is basically non-existent, as such any instruction fetch for the first instructions of a function will have (on average) half the fetch window filled with whatever comes before. By pushing the alignment up to 16 bytes this improves matters for chips that happen to have a 16 byte i-fetch window size (Intel) while not making matters worse for chips that have a larger 32 byte i-fetch window (AMD Zen). In fact, it improves the worst case for Zen from 31 bytes of garbage to 16 bytes of garbage.
As such the first many patches of the series fix up lots of alignment quirks.
The second big difference is the introduction of struct pcpu_hot. Because the compiler managed to place two adjacent (in code) DEFINE_PER_CPU() variables in random cachelines (it is absolutely free to do so) the introduction of the per-cpu x86_call_depth variable sometimes introduced significant additional cache pressure, while other times it would sit nicely in the same line with preempt_count and not show up at all.
In order to alleviate this problem; introduce struct pcpu_hot and collect a number of hot per-cpu variables in a way the compiler can't mess up.
Since these changes are 'unconditional', Mel was gracious enough to help test this on his test setup across all the relevant uarchs (very much including both Intel and AMD machines) and found that while these changes cause some very small wins and losses across the board it is mostly noise.
Aside from these changes; the core of the depth tracking is still the same.
- objtool creates a list of functions and a list of function call sites.
- for every function the padding is overwritten with the call accounting thunk; for every call site the call target is adjusted to point to this thunk.
- the retbleed return thunk mechanism is used for a custom return thunk that includes return accounting and does RSB stuffing when required.
This ensures no new compiler is required and avoids almost all overhead for non affected machines. This new option can still be selected using:
"retbleed=stuff"
on the kernel command line.
As a refresher; the theory behind call depth tracking is:
The Return-Stack-Buffer (RSB) is a 16 deep stack that is filled on every call. On the return path speculation will "pop" an entry and takes that as the return target. Once the RSB is empty, the CPU falls back to other predictors, e.g. the Branch History Buffer, which can be mistrained by user space and misguides the (return) speculation path to a disclosure gadget of your choice -- as described in the retbleed paper.
Call depth tracking is designed to break this speculation path by stuffing speculation trap calls into the RSB whenver the RSB is running low. This way the speculation stalls and never falls back to other predictors.
The assumption is that stuffing at the 12th return is sufficient to break the speculation before it hits the underflow and the fallback to the other predictors. Testing confirms that it works. Johannes, one of the retbleed researchers, tried to attack this approach and confirmed that it brings the signal to noise ratio down to the crystal ball level.
Excerpts from IBRS vs stuff from Mel's testing:
perfsyscall
6.0.0-rc1 6.0.0-rc1
tglx-mit-spectre-ibrs tglx-mit-spectre-retpoline-retstuff
Duration User 136.16 69.10
Duration System 100.50 33.04
Duration Elapsed 237.20 102.65That's a massive improvement with a major reduction in system CPU usage.
Kernel compilation is variable. Skylake-X was modest with 2-18% gain depending on degree of parallelisation.
Git checkouts are roughly 14% faster on Skylake-X
Network test were localhost only so are limited but even so, the gain is large. Skylake-X again;
Netperf-TCP
6.0.0-rc1 6.0.0-rc1
tglx-mit-spectre-ibrs tglx-mit-spectre-retpoline-retstuff
Hmean send-64 241.39 ( 0.00%) 298.00 * 23.45%*
Hmean send-128 489.55 ( 0.00%) 610.46 * 24.70%*
Hmean send-256 990.85 ( 0.00%) 1201.73 * 21.28%*
Hmean send-1024 4051.84 ( 0.00%) 5006.19 * 23.55%*
Hmean send-2048 7924.75 ( 0.00%) 9777.14 * 23.37%*
Hmean send-3312 12319.98 ( 0.00%) 15210.07 * 23.46%*
Hmean send-4096 14770.62 ( 0.00%) 17941.32 * 21.47%*
Hmean send-8192 26302.00 ( 0.00%) 30170.04 * 14.71%*
Hmean send-16384 42449.51 ( 0.00%) 48036.45 * 13.16%*While this is UDP_STREAM, TCP_STREAM is similarly impressive.
FIO measurements done by Tim Chen:
read (kIOPs) Mean stdev mitigations=off retbleed=off CPU util
================================================================================
mitigations=off 357.33 3.79 0.00% 6.14% 98.93% retbleed=off 336.67 6.43 -5.78% 0.00% 99.01% retbleed=ibrs 242.00 0.00 -32.28% -28.12% 99.41% retbleed=stuff (pad) 314.33 1.53 -12.03% -6.63% 99.31%
read/write Baseline Baseline
70/30 (kIOPs) Mean stdev mitigations=off retbleed=off CPU util
================================================================================
mitigations=off 349.00 5.29 0.00% 9.06% 96.66% retbleed=off 320.00 5.05 -8.31% 0.00% 95.54% retbleed=ibrs 238.60 0.17 -31.63% -25.44% 98.18% retbleed=stuff (pad) 293.37 0.81 -15.94% -8.32% 97.71%
Baseline Baseline
write (kIOPs) Mean stdev mitigations=off retbleed=off CPU util
================================================================================
mitigations=off 296.33 8.08 0.00% 6.21% 93.96% retbleed=off 279.00 2.65 -5.85% 0.00% 93.63% retbleed=ibrs 230.33 0.58 -22.27% -17.44% 95.92% retbleed=stuff (pad) 266.67 1.53 -10.01% -4.42% 94.75%
---
The patches can also be found in git here:
git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git call-depth-tracking