Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
H. Peter Anvin wrote: > On 11/25/2009 08:44 AM, Jakub Jelinek wrote: >> If you compile kernels 90%+ people out there run with -p on i?86/x86_64, >> then certainly coming up with a new gcc switch and new profiling ABI is >> desirable. -p on i?86/x86_64 e.g. forces -fno-omit-frame-pointer, which >> makes code on these register starved arches significantly worse. >> Making GCC output profiling call before prologue instead of after prologue >> is a 4 liner in generic code and a few lines in target specific code. >> The important thing is that we shouldn't have 100 different profiling ABIs, >> so it is desirable to agree on something that will be generally useful not >> just for the kernel, but perhaps for other purposes. > > There is really just one that makes sense, which is providing the > ABI-defined entry state, which means intercepting at the point of entry. > > Anything else is/was a mistake. Indeed. The problem, though, is that the "naked call" approach, while attractive, requires the back end to be modified and so requires the help of the gcc maintainers for every Linux target. Not that this is a terrible idea, but such co-ordination is going to take time. Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On 11/25/2009 08:44 AM, Jakub Jelinek wrote: > > If you compile kernels 90%+ people out there run with -p on i?86/x86_64, > then certainly coming up with a new gcc switch and new profiling ABI is > desirable. -p on i?86/x86_64 e.g. forces -fno-omit-frame-pointer, which > makes code on these register starved arches significantly worse. > Making GCC output profiling call before prologue instead of after prologue > is a 4 liner in generic code and a few lines in target specific code. > The important thing is that we shouldn't have 100 different profiling ABIs, > so it is desirable to agree on something that will be generally useful not > just for the kernel, but perhaps for other purposes. > There is really just one that makes sense, which is providing the ABI-defined entry state, which means intercepting at the point of entry. Anything else is/was a mistake. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On 11/24/2009 09:30 AM, Steven Rostedt wrote: > > For other archs, Linus showed some examples: > > http://lkml.org/lkml/2009/11/19/349 > Yes; the key here is that the ABI-defined entry state is readily mappable onto the state on entry to the __fentry__ function. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Wed, Nov 25, 2009 at 04:44:52PM +0100, Ingo Molnar wrote: > > * Thomas Gleixner wrote: > > > On Tue, 24 Nov 2009, Jakub Jelinek wrote: > > > > > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > > > > > you should compile your code with -maccumulate-outgoing-args, and > > > > > there's > > > > > no need to use -mtune=generic. Is that right? > > > > > > > > Seems to work. What other side effects has that ? > > > > > > Faster code, significant increase in code size though. Note that on many > > > architectures it is the only supported model. > > > > Just checked on the affected -marchs. The increase in code size is > > about 3% which is not that bad and definitely acceptable for the > > tracing case. Will zap the -mtune=generic patch and use > > -maccumulate-outgoing-args instead. > > hm, 3% sounds quite large :( dyn-ftrace is enabled in distro configs, so > 3% is a big deal IMO. If you compile kernels 90%+ people out there run with -p on i?86/x86_64, then certainly coming up with a new gcc switch and new profiling ABI is desirable. -p on i?86/x86_64 e.g. forces -fno-omit-frame-pointer, which makes code on these register starved arches significantly worse. Making GCC output profiling call before prologue instead of after prologue is a 4 liner in generic code and a few lines in target specific code. The important thing is that we shouldn't have 100 different profiling ABIs, so it is desirable to agree on something that will be generally useful not just for the kernel, but perhaps for other purposes. Jakub
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
* Thomas Gleixner wrote: > On Wed, 25 Nov 2009, Ingo Molnar wrote: > > * Thomas Gleixner wrote: > > > > > On Tue, 24 Nov 2009, Jakub Jelinek wrote: > > > > > > > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > > > > > > you should compile your code with -maccumulate-outgoing-args, and > > > > > > there's > > > > > > no need to use -mtune=generic. Is that right? > > > > > > > > > > Seems to work. What other side effects has that ? > > > > > > > > Faster code, significant increase in code size though. Note that on > > > > many > > > > architectures it is the only supported model. > > > > > > Just checked on the affected -marchs. The increase in code size is > > > about 3% which is not that bad and definitely acceptable for the > > > tracing case. Will zap the -mtune=generic patch and use > > > -maccumulate-outgoing-args instead. > > > > hm, 3% sounds quite large :( dyn-ftrace is enabled in distro configs, so > > 3% is a big deal IMO. > > Distro-configs have -mtune=generic anyway. So it's not changing > anything for them. > > I'm talking about the -march flags which result in that weird code > (pentium-mmx ). ok! Ingo
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Wed, 25 Nov 2009, Ingo Molnar wrote: > * Thomas Gleixner wrote: > > > On Tue, 24 Nov 2009, Jakub Jelinek wrote: > > > > > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > > > > > you should compile your code with -maccumulate-outgoing-args, and > > > > > there's > > > > > no need to use -mtune=generic. Is that right? > > > > > > > > Seems to work. What other side effects has that ? > > > > > > Faster code, significant increase in code size though. Note that on many > > > architectures it is the only supported model. > > > > Just checked on the affected -marchs. The increase in code size is > > about 3% which is not that bad and definitely acceptable for the > > tracing case. Will zap the -mtune=generic patch and use > > -maccumulate-outgoing-args instead. > > hm, 3% sounds quite large :( dyn-ftrace is enabled in distro configs, so > 3% is a big deal IMO. Distro-configs have -mtune=generic anyway. So it's not changing anything for them. I'm talking about the -march flags which result in that weird code (pentium-mmx ). Thanks, tglx
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
* Thomas Gleixner wrote: > On Tue, 24 Nov 2009, Jakub Jelinek wrote: > > > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > > > > you should compile your code with -maccumulate-outgoing-args, and > > > > there's > > > > no need to use -mtune=generic. Is that right? > > > > > > Seems to work. What other side effects has that ? > > > > Faster code, significant increase in code size though. Note that on many > > architectures it is the only supported model. > > Just checked on the affected -marchs. The increase in code size is > about 3% which is not that bad and definitely acceptable for the > tracing case. Will zap the -mtune=generic patch and use > -maccumulate-outgoing-args instead. hm, 3% sounds quite large :( dyn-ftrace is enabled in distro configs, so 3% is a big deal IMO. Ingo
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Tue, 24 Nov 2009, Jakub Jelinek wrote: > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > > > you should compile your code with -maccumulate-outgoing-args, and there's > > > no need to use -mtune=generic. Is that right? > > > > Seems to work. What other side effects has that ? > > Faster code, significant increase in code size though. Note that on many > architectures it is the only supported model. Just checked on the affected -marchs. The increase in code size is about 3% which is not that bad and definitely acceptable for the tracing case. Will zap the -mtune=generic patch and use -maccumulate-outgoing-args instead. Thanks, tglx
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On 11/24/2009 09:12 AM, Andrew Haley wrote: >> >> If we're changing gcc anyway, then let's add the option of intercepting >> the function at the point where the machine state is well-defined by >> ABI, which is before the function stack frame is set up. > > Hmm. On the x86 I suppose we could just inject a naked call instruction, > but not all aeches allow us to call anything before we've saved the return > address. Or are you talking x86 only? > For x86, we should use a naked call. For architectures where that is not possible, we should use a minimal sequence such that the ABI state at the invocation point is 100% derivable. On MIPS, for example, we could use a sequence such as: mov at, ra jal __fentry__ It would be up to __fentry__ to save the value in at and to restore it back into ra before resuming, meaning that __fentry__ has a nonstandard calling convention. -hpa
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Tue, 2009-11-24 at 17:12 +, Andrew Haley wrote: > H. Peter Anvin wrote: > > If we're changing gcc anyway, then let's add the option of intercepting > > the function at the point where the machine state is well-defined by > > ABI, which is before the function stack frame is set up. > > Hmm. On the x86 I suppose we could just inject a naked call instruction, > but not all aeches allow us to call anything before we've saved the return > address. Or are you talking x86 only? Earlier in the GCC BUG thread we talked about this. Adding a __fentry__ call at the beginning of the function. This could be done for other archs as well, but yes, the return address must be stored. For x86 it is the easiest because it automatically stores the return address on the stack (Andi already has a working patch I believe). For other archs, Linus showed some examples: http://lkml.org/lkml/2009/11/19/349 -- Steve
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
H. Peter Anvin wrote: > On 11/24/2009 07:46 AM, Andrew Haley wrote: >>> Yes, a lot. The difference is that -maccumulate-outgoing-args allocates >>> space for arguments of the callee with most arguments in the prologue, using >>> subtraction from sp, then to pass arguments uses movl XXX, 4(%esp) etc. >>> and the stack pointer doesn't usually change within the function (except for >>> alloca/VLAs). >>> With -mno-accumulate-outgoing-args args are pushed using push instructions >>> and stack pointer is constantly changing. >> Alright. So, it is possible in theory for gcc to generate code that >> only uses -maccumulate-outgoing-args when it needs to realign SP. >> And, therefore, we could have a nice option for the kernel: one with >> (mostly) good code density and never generates the bizarre code >> sequence in the prologue. > > If we're changing gcc anyway, then let's add the option of intercepting > the function at the point where the machine state is well-defined by > ABI, which is before the function stack frame is set up. Hmm. On the x86 I suppose we could just inject a naked call instruction, but not all aeches allow us to call anything before we've saved the return address. Or are you talking x86 only? Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Ross Ridge wrote: > Andrew Haley writes: >> Alright. So, it is possible in theory for gcc to generate code that >> only uses -maccumulate-outgoing-args when it needs to realign SP. >> And, therefore, we could have a nice option for the kernel: one with >> (mostly) good code density and never generates the bizarre code >> sequence in the prologue. > > The best option would be for the Linux people to fix the underlying > problem in their kernel sources. If the code no longer requested > that certain automatic variables be aligned, then not only would this > bizarre code sequence not be emitted, the unnecessary stack alignment > would disapear as well. The kernel would then be free to choose to use > whatever code generation options it felt was appropriate. Well, yeah. But, for my sins, I tend to assume that the Linux kernel people have some kind of reason for the things they do. Working with them over the years has helped us improve gcc, even though at times things get to be a little ill-tempered. Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Andrew Haley writes: >Alright. So, it is possible in theory for gcc to generate code that >only uses -maccumulate-outgoing-args when it needs to realign SP. >And, therefore, we could have a nice option for the kernel: one with >(mostly) good code density and never generates the bizarre code >sequence in the prologue. The best option would be for the Linux people to fix the underlying problem in their kernel sources. If the code no longer requested that certain automatic variables be aligned, then not only would this bizarre code sequence not be emitted, the unnecessary stack alignment would disapear as well. The kernel would then be free to choose to use whatever code generation options it felt was appropriate. Ross Ridge
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On 11/24/2009 07:46 AM, Andrew Haley wrote: >> >> Yes, a lot. The difference is that -maccumulate-outgoing-args allocates >> space for arguments of the callee with most arguments in the prologue, using >> subtraction from sp, then to pass arguments uses movl XXX, 4(%esp) etc. >> and the stack pointer doesn't usually change within the function (except for >> alloca/VLAs). >> With -mno-accumulate-outgoing-args args are pushed using push instructions >> and stack pointer is constantly changing. > > Alright. So, it is possible in theory for gcc to generate code that > only uses -maccumulate-outgoing-args when it needs to realign SP. > And, therefore, we could have a nice option for the kernel: one with > (mostly) good code density and never generates the bizarre code > sequence in the prologue. > If we're changing gcc anyway, then let's add the option of intercepting the function at the point where the machine state is well-defined by ABI, which is before the function stack frame is set up. -maccumulate-outgoing-args sounds like it would be painful on x86 (not using its cheap push/pop instructions), but I guess since it's only when tracing it's less of an issue. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Jakub Jelinek wrote: > On Tue, Nov 24, 2009 at 03:32:20PM +, Andrew Haley wrote: >> Jakub Jelinek wrote: >>> On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > you should compile your code with -maccumulate-outgoing-args, and there's > no need to use -mtune=generic. Is that right? Seems to work. What other side effects has that ? >>> Faster code, significant increase in code size though. >> Does it affect code size when we don't have to realign the stack pointer? > > Yes, a lot. The difference is that -maccumulate-outgoing-args allocates > space for arguments of the callee with most arguments in the prologue, using > subtraction from sp, then to pass arguments uses movl XXX, 4(%esp) etc. > and the stack pointer doesn't usually change within the function (except for > alloca/VLAs). > With -mno-accumulate-outgoing-args args are pushed using push instructions > and stack pointer is constantly changing. Alright. So, it is possible in theory for gcc to generate code that only uses -maccumulate-outgoing-args when it needs to realign SP. And, therefore, we could have a nice option for the kernel: one with (mostly) good code density and never generates the bizarre code sequence in the prologue. Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Tue, Nov 24, 2009 at 03:32:20PM +, Andrew Haley wrote: > Jakub Jelinek wrote: > > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > >>> you should compile your code with -maccumulate-outgoing-args, and there's > >>> no need to use -mtune=generic. Is that right? > >> Seems to work. What other side effects has that ? > > > > Faster code, significant increase in code size though. > > Does it affect code size when we don't have to realign the stack pointer? Yes, a lot. The difference is that -maccumulate-outgoing-args allocates space for arguments of the callee with most arguments in the prologue, using subtraction from sp, then to pass arguments uses movl XXX, 4(%esp) etc. and the stack pointer doesn't usually change within the function (except for alloca/VLAs). With -mno-accumulate-outgoing-args args are pushed using push instructions and stack pointer is constantly changing. Jakub
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Jakub Jelinek wrote: > On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: >>> you should compile your code with -maccumulate-outgoing-args, and there's >>> no need to use -mtune=generic. Is that right? >> Seems to work. What other side effects has that ? > > Faster code, significant increase in code size though. Does it affect code size when we don't have to realign the stack pointer? Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Tue, Nov 24, 2009 at 03:55:49PM +0100, Thomas Gleixner wrote: > > you should compile your code with -maccumulate-outgoing-args, and there's > > no need to use -mtune=generic. Is that right? > > Seems to work. What other side effects has that ? Faster code, significant increase in code size though. Note that on many architectures it is the only supported model. Jakub
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Tue, 24 Nov 2009, Andrew Haley wrote: > H.J. Lu wrote: > > On Sun, Nov 22, 2009 at 9:20 AM, Andrew Haley wrote: > >> H.J. Lu wrote: > >>> On Fri, Nov 20, 2009 at 11:35 AM, Andrew Haley wrote: > Steven Rostedt wrote: > > Ingo, Thomas and Linus, > > > > I know Thomas did a patch to force the -mtune=generic, but just in case > > gcc decides to do something crazy again, this patch will catch it. > > > > Should we try to get this in now? > I'm sure this makes sense, but a gcc test case would be even better. > If this can be detected in the gcc test suite it'll be found and > fixed long before y'all in kernel land get to see it. That's the > only way to guarantee this never bothers you again. > > H.J., who wrote the code in question, is hopefully looking at why > this odd code is being generated. Once he's done I can put a > suitable test case in the gcc test suite. > > >>> See: > >>> > >>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42109#c7 > >> I saw that, but does it mean you're going to investigate? There is > >> no obvious reason why -mtune=generic should affect code generation > >> in this way, but it does. > > > > Why not, there is > > > > static const unsigned int x86_accumulate_outgoing_args > > = m_AMD_MULTIPLE | m_ATOM | m_PENT4 | m_NOCONA | m_PPRO | m_CORE2 > > | m_GENERIC; > > > > -mtune=generic turns on -maccumulate-outgoing-args. > > Alright, so let's at least try to give the kernel people the information > that they need. > > What you're saying is, to avoid this: > > 05f0 : > 5f0: 57 push %edi > 5f1: 8d 7c 24 08 lea0x8(%esp),%edi > 5f5: 83 e4 f0and$0xfff0,%esp > 5f8: ff 77 fcpushl -0x4(%edi) > 5fb: 55 push %ebp > 5fc: 89 e5 mov%esp,%ebp > > you should compile your code with -maccumulate-outgoing-args, and there's > no need to use -mtune=generic. Is that right? Seems to work. What other side effects has that ? Thanks, tglx
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
H.J. Lu wrote: > On Sun, Nov 22, 2009 at 9:20 AM, Andrew Haley wrote: >> H.J. Lu wrote: >>> On Fri, Nov 20, 2009 at 11:35 AM, Andrew Haley wrote: Steven Rostedt wrote: > Ingo, Thomas and Linus, > > I know Thomas did a patch to force the -mtune=generic, but just in case > gcc decides to do something crazy again, this patch will catch it. > > Should we try to get this in now? I'm sure this makes sense, but a gcc test case would be even better. If this can be detected in the gcc test suite it'll be found and fixed long before y'all in kernel land get to see it. That's the only way to guarantee this never bothers you again. H.J., who wrote the code in question, is hopefully looking at why this odd code is being generated. Once he's done I can put a suitable test case in the gcc test suite. >>> See: >>> >>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42109#c7 >> I saw that, but does it mean you're going to investigate? There is >> no obvious reason why -mtune=generic should affect code generation >> in this way, but it does. > > Why not, there is > > static const unsigned int x86_accumulate_outgoing_args > = m_AMD_MULTIPLE | m_ATOM | m_PENT4 | m_NOCONA | m_PPRO | m_CORE2 > | m_GENERIC; > > -mtune=generic turns on -maccumulate-outgoing-args. Alright, so let's at least try to give the kernel people the information that they need. What you're saying is, to avoid this: 05f0 : 5f0: 57 push %edi 5f1: 8d 7c 24 08 lea0x8(%esp),%edi 5f5: 83 e4 f0and$0xfff0,%esp 5f8: ff 77 fcpushl -0x4(%edi) 5fb: 55 push %ebp 5fc: 89 e5 mov%esp,%ebp you should compile your code with -maccumulate-outgoing-args, and there's no need to use -mtune=generic. Is that right? Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Sun, Nov 22, 2009 at 9:20 AM, Andrew Haley wrote: > H.J. Lu wrote: >> On Fri, Nov 20, 2009 at 11:35 AM, Andrew Haley wrote: >>> Steven Rostedt wrote: Ingo, Thomas and Linus, I know Thomas did a patch to force the -mtune=generic, but just in case gcc decides to do something crazy again, this patch will catch it. Should we try to get this in now? >>> I'm sure this makes sense, but a gcc test case would be even better. >>> If this can be detected in the gcc test suite it'll be found and >>> fixed long before y'all in kernel land get to see it. That's the >>> only way to guarantee this never bothers you again. >>> >>> H.J., who wrote the code in question, is hopefully looking at why >>> this odd code is being generated. Once he's done I can put a >>> suitable test case in the gcc test suite. >>> >> >> See: >> >> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42109#c7 > > I saw that, but does it mean you're going to investigate? There is > no obvious reason why -mtune=generic should affect code generation > in this way, but it does. > Why not, there is static const unsigned int x86_accumulate_outgoing_args = m_AMD_MULTIPLE | m_ATOM | m_PENT4 | m_NOCONA | m_PPRO | m_CORE2 | m_GENERIC; -mtune=generic turns on -maccumulate-outgoing-args. -- H.J.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
H.J. Lu wrote: > On Fri, Nov 20, 2009 at 11:35 AM, Andrew Haley wrote: >> Steven Rostedt wrote: >>> Ingo, Thomas and Linus, >>> >>> I know Thomas did a patch to force the -mtune=generic, but just in case >>> gcc decides to do something crazy again, this patch will catch it. >>> >>> Should we try to get this in now? >> I'm sure this makes sense, but a gcc test case would be even better. >> If this can be detected in the gcc test suite it'll be found and >> fixed long before y'all in kernel land get to see it. That's the >> only way to guarantee this never bothers you again. >> >> H.J., who wrote the code in question, is hopefully looking at why >> this odd code is being generated. Once he's done I can put a >> suitable test case in the gcc test suite. >> > > See: > > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42109#c7 I saw that, but does it mean you're going to investigate? There is no obvious reason why -mtune=generic should affect code generation in this way, but it does. Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Fri, Nov 20, 2009 at 11:35 AM, Andrew Haley wrote: > Steven Rostedt wrote: >> Ingo, Thomas and Linus, >> >> I know Thomas did a patch to force the -mtune=generic, but just in case >> gcc decides to do something crazy again, this patch will catch it. >> >> Should we try to get this in now? > > I'm sure this makes sense, but a gcc test case would be even better. > If this can be detected in the gcc test suite it'll be found and > fixed long before y'all in kernel land get to see it. That's the > only way to guarantee this never bothers you again. > > H.J., who wrote the code in question, is hopefully looking at why > this odd code is being generated. Once he's done I can put a > suitable test case in the gcc test suite. > See: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42109#c7 -- H.J.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
* Steven Rostedt wrote: > Ingo, Thomas and Linus, > > I know Thomas did a patch to force the -mtune=generic, but just in > case gcc decides to do something crazy again, this patch will catch > it. > > Should we try to get this in now? Very nice example of defensive coding - i like this. I've queued it up for .33 (unless anyone objects) as i think it's too late for .32. Ingo
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On 11/20/2009 11:46 AM, Steven Rostedt wrote: > > Yes a gcc test suite will help new instances of gcc. But we need to > worry about the instances of gcc that people have on their desktops now. > This test case will catch the discrepancy between gcc and the function > graph tracer. I'm not 100% convince that just adding -mtune=generic will > help in all cases. If we miss another instance, then the function graph > tracer may crash someone's kernel. > Furthermore, for future gcc instances what we really want is the early interception support anyway. -hpa
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On Fri, 2009-11-20 at 19:35 +, Andrew Haley wrote: > Steven Rostedt wrote: > > Ingo, Thomas and Linus, > > > > I know Thomas did a patch to force the -mtune=generic, but just in case > > gcc decides to do something crazy again, this patch will catch it. > > > > Should we try to get this in now? > > I'm sure this makes sense, but a gcc test case would be even better. > If this can be detected in the gcc test suite it'll be found and > fixed long before y'all in kernel land get to see it. That's the > only way to guarantee this never bothers you again. > > H.J., who wrote the code in question, is hopefully looking at why > this odd code is being generated. Once he's done I can put a > suitable test case in the gcc test suite. Yes a gcc test suite will help new instances of gcc. But we need to worry about the instances of gcc that people have on their desktops now. This test case will catch the discrepancy between gcc and the function graph tracer. I'm not 100% convince that just adding -mtune=generic will help in all cases. If we miss another instance, then the function graph tracer may crash someone's kernel. -- Steve
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Steven Rostedt wrote: > Ingo, Thomas and Linus, > > I know Thomas did a patch to force the -mtune=generic, but just in case > gcc decides to do something crazy again, this patch will catch it. > > Should we try to get this in now? I'm sure this makes sense, but a gcc test case would be even better. If this can be detected in the gcc test suite it'll be found and fixed long before y'all in kernel land get to see it. That's the only way to guarantee this never bothers you again. H.J., who wrote the code in question, is hopefully looking at why this odd code is being generated. Once he's done I can put a suitable test case in the gcc test suite. Andrew.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
On 11/20/2009 09:00 AM, Steven Rostedt wrote: > Ingo, Thomas and Linus, > > I know Thomas did a patch to force the -mtune=generic, but just in case > gcc decides to do something crazy again, this patch will catch it. > > Should we try to get this in now? > Sounds like a very good idea to me. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Ingo, Thomas and Linus, I know Thomas did a patch to force the -mtune=generic, but just in case gcc decides to do something crazy again, this patch will catch it. Should we try to get this in now? -- Steve On Fri, 2009-11-20 at 00:23 -0500, Steven Rostedt wrote: > commit c7715fb611c69ac4b7f722a891de08b206fb7686 > Author: Steven Rostedt > Date: Thu Nov 19 23:41:02 2009 -0500 > > tracing/x86: Add check to detect GCC messing with mcount prologue > > Latest versions of GCC create a funny prologue for some functions. > Instead of the typical: > > push %ebp > mov%esp,%ebp > and$0xffe0,%esp > [...] > call mcount > > GCC may try to align the stack before setting up the frame pointer > register: > > push %edi > lea0x8(%esp),%edi > and$0xffe0,%esp > pushl -0x4(%edi) > push %ebp > mov%esp,%ebp > [...] > call mcount > > This crazy code places a copy of the return address into the > frame pointer. The function graph tracer uses this pointer to > save and replace the return address of the calling function to jump > to the function graph tracer's return handler, which will put back > the return address. But instead instead of the typical return: > > mov%ebp,%esp > pop%ebp > ret > > The return of the function performs: > > lea-0x8(%edi),%esp > pop%edi > ret > > The function graph tracer return handler will not be called at the exit > of the function, but the parent function will call it. Because we missed > the return of the child function, the handler will replace the parent's > return address with that of the child. Obviously this will cause a crash > (Note, there is code to detect this case and safely panic the kernel). > > The kicker is that this happens to just a handful of functions. > And only with certain gcc options. > > Compiling with: -march=pentium-mmx > will cause the problem to appear. But if you were to change > pentium-mmx to i686 or add -mtune=generic, then the problem goes away. > > I first saw this problem when compiling with optimize for size. > But it seems that various other options may cause this issue to arise. > > Instead of completely disabling the function graph tracer for i386 builds > this patch adds a check to recordmcount.pl to make sure that all > functions that contain a call to mcount start with "push %ebp". > If not, it will fail the compile and print out the nasty warning: > > CC kernel/time/timer_stats.o > > > Your version of GCC breaks the function graph tracer > Please disable CONFIG_FUNCTION_GRAPH_TRACER > Failed function was "timer_stats_update_stats" > > > The script recordmcount.pl is given a new parameter "do_check". If > this is negative, the script will only perform this check without > creating the mcount caller section. This will be executed for x86_32 > when CONFIG_FUNCTION_GRAPH_TRACER is enabled and CONFIG_DYNAMIC_FTRACE > is not. > > If the arch is x86_32 and $do_check is greater than 1, it will perform > the check while processing the mcount callers. If $do_check is 0, then > no check will be performed. This is for non x86_32 archs and when > compiling without CONFIG_FUNCTION_GRAPH_TRACER enabled, even on x86_32. > > Reported-by: Thomas Gleixner > LKML-Reference: > > Signed-off-by: Steven Rostedt
Re: [PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
This touches the Makefile scripts. I forgot to CC kbuild and Sam. -- Steve On Fri, 2009-11-20 at 00:23 -0500, Steven Rostedt wrote: > Ingo, > > Not sure if this is too much for this late in the -rc game, but it finds > the gcc bug at build time, and we don't need to disable function graph > tracer for all i386 builds. > > This is built on my last urgent repo pull request. > > Please pull the latest tip/tracing/urgent-2 tree, which can be found at: > > git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace.git > tip/tracing/urgent-2 > > > Steven Rostedt (1): > tracing/x86: Add check to detect GCC messing with mcount prologue > > > kernel/trace/Kconfig|1 - > scripts/Makefile.build | 25 +++- > scripts/recordmcount.pl | 74 > +-- > 3 files changed, 95 insertions(+), 5 deletions(-) > --- > commit c7715fb611c69ac4b7f722a891de08b206fb7686 > Author: Steven Rostedt > Date: Thu Nov 19 23:41:02 2009 -0500 > > tracing/x86: Add check to detect GCC messing with mcount prologue > > Latest versions of GCC create a funny prologue for some functions. > Instead of the typical: > > push %ebp > mov%esp,%ebp > and$0xffe0,%esp > [...] > call mcount > > GCC may try to align the stack before setting up the frame pointer > register: > > push %edi > lea0x8(%esp),%edi > and$0xffe0,%esp > pushl -0x4(%edi) > push %ebp > mov%esp,%ebp > [...] > call mcount > > This crazy code places a copy of the return address into the > frame pointer. The function graph tracer uses this pointer to > save and replace the return address of the calling function to jump > to the function graph tracer's return handler, which will put back > the return address. But instead instead of the typical return: > > mov%ebp,%esp > pop%ebp > ret > > The return of the function performs: > > lea-0x8(%edi),%esp > pop%edi > ret > > The function graph tracer return handler will not be called at the exit > of the function, but the parent function will call it. Because we missed > the return of the child function, the handler will replace the parent's > return address with that of the child. Obviously this will cause a crash > (Note, there is code to detect this case and safely panic the kernel). > > The kicker is that this happens to just a handful of functions. > And only with certain gcc options. > > Compiling with: -march=pentium-mmx > will cause the problem to appear. But if you were to change > pentium-mmx to i686 or add -mtune=generic, then the problem goes away. > > I first saw this problem when compiling with optimize for size. > But it seems that various other options may cause this issue to arise. > > Instead of completely disabling the function graph tracer for i386 builds > this patch adds a check to recordmcount.pl to make sure that all > functions that contain a call to mcount start with "push %ebp". > If not, it will fail the compile and print out the nasty warning: > > CC kernel/time/timer_stats.o > > > Your version of GCC breaks the function graph tracer > Please disable CONFIG_FUNCTION_GRAPH_TRACER > Failed function was "timer_stats_update_stats" > > > The script recordmcount.pl is given a new parameter "do_check". If > this is negative, the script will only perform this check without > creating the mcount caller section. This will be executed for x86_32 > when CONFIG_FUNCTION_GRAPH_TRACER is enabled and CONFIG_DYNAMIC_FTRACE > is not. > > If the arch is x86_32 and $do_check is greater than 1, it will perform > the check while processing the mcount callers. If $do_check is 0, then > no check will be performed. This is for non x86_32 archs and when > compiling without CONFIG_FUNCTION_GRAPH_TRACER enabled, even on x86_32. > > Reported-by: Thomas Gleixner > LKML-Reference: > > Signed-off-by: Steven Rostedt > > diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig > index b416512..cd39064 100644 > --- a/kernel/trace/Kconfig > +++ b/kernel/trace/Kconfig > @@ -143,7 +143,6 @@ config FUNCTION_GRAPH_TRACER > bool "Kernel Function Graph Tracer" > depends on HAVE_FUNCTION_GRAPH_TRACER > depends on FUNCTION_TRACER > - depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE > default y > help >
[PATCH][GIT PULL][v2.6.32] tracing/x86: Add check to detect GCC messing with mcount prologue
Ingo, Not sure if this is too much for this late in the -rc game, but it finds the gcc bug at build time, and we don't need to disable function graph tracer for all i386 builds. This is built on my last urgent repo pull request. Please pull the latest tip/tracing/urgent-2 tree, which can be found at: git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace.git tip/tracing/urgent-2 Steven Rostedt (1): tracing/x86: Add check to detect GCC messing with mcount prologue kernel/trace/Kconfig|1 - scripts/Makefile.build | 25 +++- scripts/recordmcount.pl | 74 +-- 3 files changed, 95 insertions(+), 5 deletions(-) --- commit c7715fb611c69ac4b7f722a891de08b206fb7686 Author: Steven Rostedt Date: Thu Nov 19 23:41:02 2009 -0500 tracing/x86: Add check to detect GCC messing with mcount prologue Latest versions of GCC create a funny prologue for some functions. Instead of the typical: push %ebp mov%esp,%ebp and$0xffe0,%esp [...] call mcount GCC may try to align the stack before setting up the frame pointer register: push %edi lea0x8(%esp),%edi and$0xffe0,%esp pushl -0x4(%edi) push %ebp mov%esp,%ebp [...] call mcount This crazy code places a copy of the return address into the frame pointer. The function graph tracer uses this pointer to save and replace the return address of the calling function to jump to the function graph tracer's return handler, which will put back the return address. But instead instead of the typical return: mov%ebp,%esp pop%ebp ret The return of the function performs: lea-0x8(%edi),%esp pop%edi ret The function graph tracer return handler will not be called at the exit of the function, but the parent function will call it. Because we missed the return of the child function, the handler will replace the parent's return address with that of the child. Obviously this will cause a crash (Note, there is code to detect this case and safely panic the kernel). The kicker is that this happens to just a handful of functions. And only with certain gcc options. Compiling with: -march=pentium-mmx will cause the problem to appear. But if you were to change pentium-mmx to i686 or add -mtune=generic, then the problem goes away. I first saw this problem when compiling with optimize for size. But it seems that various other options may cause this issue to arise. Instead of completely disabling the function graph tracer for i386 builds this patch adds a check to recordmcount.pl to make sure that all functions that contain a call to mcount start with "push %ebp". If not, it will fail the compile and print out the nasty warning: CC kernel/time/timer_stats.o Your version of GCC breaks the function graph tracer Please disable CONFIG_FUNCTION_GRAPH_TRACER Failed function was "timer_stats_update_stats" The script recordmcount.pl is given a new parameter "do_check". If this is negative, the script will only perform this check without creating the mcount caller section. This will be executed for x86_32 when CONFIG_FUNCTION_GRAPH_TRACER is enabled and CONFIG_DYNAMIC_FTRACE is not. If the arch is x86_32 and $do_check is greater than 1, it will perform the check while processing the mcount callers. If $do_check is 0, then no check will be performed. This is for non x86_32 archs and when compiling without CONFIG_FUNCTION_GRAPH_TRACER enabled, even on x86_32. Reported-by: Thomas Gleixner LKML-Reference: Signed-off-by: Steven Rostedt diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index b416512..cd39064 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -143,7 +143,6 @@ config FUNCTION_GRAPH_TRACER bool "Kernel Function Graph Tracer" depends on HAVE_FUNCTION_GRAPH_TRACER depends on FUNCTION_TRACER - depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE default y help Enable the kernel to trace a function at both its return diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 341b589..3b897f2 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -206,10 +206,33 @@ cmd_modversions = \ endif ifdef CONFIG_FTR