4 ms·

> for example, maybe it would have just been easier to flat-out introduce the concept of transactions, if you're already going to write a verifier to check ever

> for example, maybe it would have just been easier to flat-out introduce the concept of transactions, if you're already going to write a verifier to check every incoming program Could you elaborate on this? Loading the program is a one time operation. The actual evocation is dependent on the program type and its location in the kernel. I do not believe there is a blanket way to describe how eBPF interacts with the VM or its concurrency model. To your second point about verification, yes this is something that is actively being worked on. with BTF and CO-RE we've made huge strides over module loading, however, and eBPF already has many advantages over use cases where people were forced to load bespoke kernel bypass modules.

aseipp 5y ago
I mean that, instead of exposing atomic operations in the programming model (which are imperative constructs from the CPU to write to a memory location and require a notion of a memory model), the BPF virtual ISA could have just given you an ability to execute a series of high-level actions atomically or have them fail. It is basically a very small implementation of the ACI in ACID. This is only practical in this case because an eBPF program would need to be verified (by a trusted component; in this case the kernel verifier) before it can be executed anyway; any normal language with unverified, unrestricted semantics is too difficult to bolt transactions onto. You can just define a statically-determinable subset of programs that work and are accepted. This approach would have both upsides and downsides. The downsides include the need for more precise semantics, and an increase in verifier complexity (a lesser downside is an increase in compiler complexity to compile transactions to the underlying ISA; but I consider this 'lesser' because the verifier is more important, overall.) The upsides include a nicer programming model for users and independence from the underlying memory model in the processor, which gives you a lot of degrees of freedom to attack the problem. I don't know what you would do in the case of like, livelock though. But I also don't know what the failure mode is if, say, the program I ran today always ran and hit some dynamic error, either e.g. in the case of livelock you have to eventually just kill the program to make any further forward progress, but this seems no different than if an eBPF program simply always failed due to dynamic constraints. If the program just keeps exploding you eventually have to back off... Another downside I guess is that it's probably hard to implement transactional models in raw RTL, like this project :) But this is all just an idea. The thrust of it is to turn small exposures of the underlying processor semantics into higher-level compiler problems. I think this would work because you don't need to solve every compiler problem perfectly, just a small subset of them for a small set of domains. It doesn't really matter for Linux today because it just picked a common subset of the memory model for eBPF, but if anyone was to design a newer language for things like this in some new system, maybe some ideas like that are worthwhile. After all, CO-RE and related projects are about bringing "portability" to BPF programs. This is just an (somewhat logical) extension of that idea to the concurrent programming model. I don't need to worry about processor consistency models when I type "COMMIT TRANSACTION" in a SQL prompt, after all. > To your second point about verification, yes this is something that is actively being worked on. It's sort of hard to solve right now in the "Will this arbitrary C program work or not" sense, because the underlying toolchain and kernel implementation is evolving. I know there's a userspace verifier. What would be excellent is something like the ability to compile the kernel implementation of the verifier, but in userspace, so you could use it in CI. Does this already work, maybe?
- ilovecaching 5y ago
  Ah, that makes more sense. The byte code is an evolution and not a clean room design, and the focus has always been on remaining as close to the host ISA as possible. We've had several academics work on the verifier and most of the time the solutions allow for more programs to pass, but at the cost of far, far slower program loading (upwards of 30 minutes). > I don't know what you would do in the case of like, livelock though. The idea is we never allow this to happen by enforcing your use of locks and memory through helpers. Helpers are really the secret sauce to how we are able to enforce memory barriers, etc. without having the author think about them too much. Helpers are essentially the new syscall layer for the kernel with the advantage that they are far less expensive to add, and there's less complaining when it's later decided they were a bad decision... :) Of course, there will be bugs in eBPF, just as their are bugs in any other part of the kernel that will cause unexpected behavior. You are right in that a different design might have been easier to verify and less error prone, but I think that eBPF overall has been very robust, and is still far better than what we had with loading kernel modules and letting people do whatever they want, or trying to create really rigid policies through syscalls and exposing internals through VFS. It's sort of hard to solve right now in the "Will this arbitrary C program work or not" sense, because the underlying toolchain and kernel implementation is evolving. I know there's a userspace verifier. What would be excellent is something like the ability to compile the kernel implementation of the verifier, but in userspace, so you could use it in CI. Does this already work, maybe? There are two issues with loading - the first is as you described, which is that the verifier sometimes reject programs that clang can build. The second is that sometimes there are issues resolving BTF and fixing up programs to run against the target ABI. Both of these problems are hard to solve and I have personally seen users run into this as a major pain point. It needs improvement. Verification in userland is definitely one way to improve this. No matter what though, the nature of sitting inside the kernel is that it's very hard to keep any sort of userland step in lockstep with the kernel, and so I think this will always be a hard problem to solve.

aseipp 5y ago

I mean that, instead of exposing atomic operations in the programming model (which are imperative constructs from the CPU to write to a memory location and require a notion of a memory model), the BPF virtual ISA could have just given you an ability to execute a series of high-level actions atomically or have them fail. It is basically a very small implementation of the ACI in ACID. This is only practical in this case because an eBPF program would need to be verified (by a trusted component; in this case the kernel verifier) before it can be executed anyway; any normal language with unverified, unrestricted semantics is too difficult to bolt transactions onto. You can just define a statically-determinable subset of programs that work and are accepted. This approach would have both upsides and downsides. The downsides include the need for more precise semantics, and an increase in verifier complexity (a lesser downside is an increase in compiler complexity to compile transactions to the underlying ISA; but I consider this 'lesser' because the verifier is more important, overall.) The upsides include a nicer programming model for users and independence from the underlying memory model in the processor, which gives you a lot of degrees of freedom to attack the problem. I don't know what you would do in the case of like, livelock though. But I also don't know what the failure mode is if, say, the program I ran today always ran and hit some dynamic error, either e.g. in the case of livelock you have to eventually just kill the program to make any further forward progress, but this seems no different than if an eBPF program simply always failed due to dynamic constraints. If the program just keeps exploding you eventually have to back off... Another downside I guess is that it's probably hard to implement transactional models in raw RTL, like this project :) But this is all just an idea. The thrust of it is to turn small exposures of the underlying processor semantics into higher-level compiler problems. I think this would work because you don't need to solve every compiler problem perfectly, just a small subset of them for a small set of domains. It doesn't really matter for Linux today because it just picked a common subset of the memory model for eBPF, but if anyone was to design a newer language for things like this in some new system, maybe some ideas like that are worthwhile. After all, CO-RE and related projects are about bringing "portability" to BPF programs. This is just an (somewhat logical) extension of that idea to the concurrent programming model. I don't need to worry about processor consistency models when I type "COMMIT TRANSACTION" in a SQL prompt, after all. > To your second point about verification, yes this is something that is actively being worked on. It's sort of hard to solve right now in the "Will this arbitrary C program work or not" sense, because the underlying toolchain and kernel implementation is evolving. I know there's a userspace verifier. What would be excellent is something like the ability to compile the kernel implementation of the verifier, but in userspace, so you could use it in CI. Does this already work, maybe?

ilovecaching 5y ago
Ah, that makes more sense. The byte code is an evolution and not a clean room design, and the focus has always been on remaining as close to the host ISA as possible. We've had several academics work on the verifier and most of the time the solutions allow for more programs to pass, but at the cost of far, far slower program loading (upwards of 30 minutes). > I don't know what you would do in the case of like, livelock though. The idea is we never allow this to happen by enforcing your use of locks and memory through helpers. Helpers are really the secret sauce to how we are able to enforce memory barriers, etc. without having the author think about them too much. Helpers are essentially the new syscall layer for the kernel with the advantage that they are far less expensive to add, and there's less complaining when it's later decided they were a bad decision... :) Of course, there will be bugs in eBPF, just as their are bugs in any other part of the kernel that will cause unexpected behavior. You are right in that a different design might have been easier to verify and less error prone, but I think that eBPF overall has been very robust, and is still far better than what we had with loading kernel modules and letting people do whatever they want, or trying to create really rigid policies through syscalls and exposing internals through VFS. It's sort of hard to solve right now in the "Will this arbitrary C program work or not" sense, because the underlying toolchain and kernel implementation is evolving. I know there's a userspace verifier. What would be excellent is something like the ability to compile the kernel implementation of the verifier, but in userspace, so you could use it in CI. Does this already work, maybe? There are two issues with loading - the first is as you described, which is that the verifier sometimes reject programs that clang can build. The second is that sometimes there are issues resolving BTF and fixing up programs to run against the target ABI. Both of these problems are hard to solve and I have personally seen users run into this as a major pain point. It needs improvement. Verification in userland is definitely one way to improve this. No matter what though, the nature of sitting inside the kernel is that it's very hard to keep any sort of userland step in lockstep with the kernel, and so I think this will always be a hard problem to solve.