Message ID | 20210224021603.446274-3-punit1.agrawal@toshiba.co.jp |
---|---|
State | New |
Headers | show |
Series | Make rteval usable on arm, arm64 and i386 | expand |
On Wed, 24 Feb 2021, Punit Agrawal wrote: > Not all kernel archiectures provide the "bzImage" target, e.g., arm64 > provides "Image" which generates an uncompressed kernel binary. > > Instead of going down the path of customizing build targets > per-architecture, let's drop them altogether. The default kernel > target should build the kernel and modules on all architectures > anyways. > > Signed-off-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp> > --- > rteval/modules/loads/kcompile.py | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py > index 326f1ae..e747b9f 100644 > --- a/rteval/modules/loads/kcompile.py > +++ b/rteval/modules/loads/kcompile.py > @@ -58,7 +58,7 @@ class KBuildJob: > else: > self.jobs = self.calc_jobs_per_cpu() * len(self.node) > self.log(Log.DEBUG, "node %d: jobs == %d" % (int(node), self.jobs)) > - self.runcmd = "%s make O=%s -C %s -j%d bzImage modules" \ > + self.runcmd = "%s make O=%s -C %s -j%d" \ > % (self.binder, self.objdir, self.kdir, self.jobs) > self.cleancmd = "%s make O=%s -C %s clean allmodconfig" \ > % (self.binder, self.objdir, self.kdir) > -- > 2.30.0 > > Signed-off-by: John Kacur <jkacur@redhat.com>
John Kacur <jkacur@redhat.com> writes: > On Wed, 24 Feb 2021, Punit Agrawal wrote: > >> Not all kernel archiectures provide the "bzImage" target, e.g., arm64 >> provides "Image" which generates an uncompressed kernel binary. >> >> Instead of going down the path of customizing build targets >> per-architecture, let's drop them altogether. The default kernel >> target should build the kernel and modules on all architectures >> anyways. >> >> Signed-off-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp> >> --- >> rteval/modules/loads/kcompile.py | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py >> index 326f1ae..e747b9f 100644 >> --- a/rteval/modules/loads/kcompile.py >> +++ b/rteval/modules/loads/kcompile.py >> @@ -58,7 +58,7 @@ class KBuildJob: >> else: >> self.jobs = self.calc_jobs_per_cpu() * len(self.node) >> self.log(Log.DEBUG, "node %d: jobs == %d" % (int(node), self.jobs)) >> - self.runcmd = "%s make O=%s -C %s -j%d bzImage modules" \ >> + self.runcmd = "%s make O=%s -C %s -j%d" \ >> % (self.binder, self.objdir, self.kdir, self.jobs) >> self.cleancmd = "%s make O=%s -C %s clean allmodconfig" \ >> % (self.binder, self.objdir, self.kdir) >> -- >> 2.30.0 >> >> > > Signed-off-by: John Kacur <jkacur@redhat.com> Hi John, Thanks for picking up this patch. Just wanted to make sure about the status of the other 2 patches in the set - 1/3 and 3/3, I am assuming they're on your TODO but haven't had a chance to confirm yet. Let me know if that's not the case for some reason. Thanks, Punit
On Thu, 4 Mar 2021, Punit Agrawal wrote: > John Kacur <jkacur@redhat.com> writes: > > > On Wed, 24 Feb 2021, Punit Agrawal wrote: > > > >> Not all kernel archiectures provide the "bzImage" target, e.g., arm64 > >> provides "Image" which generates an uncompressed kernel binary. > >> > >> Instead of going down the path of customizing build targets > >> per-architecture, let's drop them altogether. The default kernel > >> target should build the kernel and modules on all architectures > >> anyways. > >> > >> Signed-off-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp> > >> --- > >> rteval/modules/loads/kcompile.py | 2 +- > >> 1 file changed, 1 insertion(+), 1 deletion(-) > >> > >> diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py > >> index 326f1ae..e747b9f 100644 > >> --- a/rteval/modules/loads/kcompile.py > >> +++ b/rteval/modules/loads/kcompile.py > >> @@ -58,7 +58,7 @@ class KBuildJob: > >> else: > >> self.jobs = self.calc_jobs_per_cpu() * len(self.node) > >> self.log(Log.DEBUG, "node %d: jobs == %d" % (int(node), self.jobs)) > >> - self.runcmd = "%s make O=%s -C %s -j%d bzImage modules" \ > >> + self.runcmd = "%s make O=%s -C %s -j%d" \ > >> % (self.binder, self.objdir, self.kdir, self.jobs) > >> self.cleancmd = "%s make O=%s -C %s clean allmodconfig" \ > >> % (self.binder, self.objdir, self.kdir) > >> -- > >> 2.30.0 > >> > >> > > > > Signed-off-by: John Kacur <jkacur@redhat.com> > > Hi John, > > Thanks for picking up this patch. > > Just wanted to make sure about the status of the other 2 patches in the > set - 1/3 and 3/3, I am assuming they're on your TODO but haven't had a > chance to confirm yet. > > Let me know if that's not the case for some reason. > > Thanks, > Punit > Yes, they are on my TODO list, I will get to them. I will probably take the patch that deal with the model name, after I convince myself that it won't break anything else. I don't really like the idea of a fake NUMA node to deal with rteval on non-NUMA machines, but I am open to ideas to get rteval working on such machines. On my list to give you feedback. Thanks John
John Kacur <jkacur@redhat.com> writes: > On Thu, 4 Mar 2021, Punit Agrawal wrote: > >> John Kacur <jkacur@redhat.com> writes: >> >> > On Wed, 24 Feb 2021, Punit Agrawal wrote: >> > >> >> Not all kernel archiectures provide the "bzImage" target, e.g., arm64 >> >> provides "Image" which generates an uncompressed kernel binary. >> >> >> >> Instead of going down the path of customizing build targets >> >> per-architecture, let's drop them altogether. The default kernel >> >> target should build the kernel and modules on all architectures >> >> anyways. >> >> >> >> Signed-off-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp> >> >> --- >> >> rteval/modules/loads/kcompile.py | 2 +- >> >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> >> >> diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py >> >> index 326f1ae..e747b9f 100644 >> >> --- a/rteval/modules/loads/kcompile.py >> >> +++ b/rteval/modules/loads/kcompile.py >> >> @@ -58,7 +58,7 @@ class KBuildJob: >> >> else: >> >> self.jobs = self.calc_jobs_per_cpu() * len(self.node) >> >> self.log(Log.DEBUG, "node %d: jobs == %d" % (int(node), self.jobs)) >> >> - self.runcmd = "%s make O=%s -C %s -j%d bzImage modules" \ >> >> + self.runcmd = "%s make O=%s -C %s -j%d" \ >> >> % (self.binder, self.objdir, self.kdir, self.jobs) >> >> self.cleancmd = "%s make O=%s -C %s clean allmodconfig" \ >> >> % (self.binder, self.objdir, self.kdir) >> >> -- >> >> 2.30.0 >> >> >> >> >> > >> > Signed-off-by: John Kacur <jkacur@redhat.com> >> >> Hi John, >> >> Thanks for picking up this patch. >> >> Just wanted to make sure about the status of the other 2 patches in the >> set - 1/3 and 3/3, I am assuming they're on your TODO but haven't had a >> chance to confirm yet. >> >> Let me know if that's not the case for some reason. >> >> Thanks, >> Punit >> > > Yes, they are on my TODO list, I will get to them. > I will probably take the patch that deal with the model name, after I > convince myself that it won't break anything else. > > I don't really like the idea of a fake NUMA node to deal with rteval > on non-NUMA machines, but I am open to ideas to get rteval working > on such machines. On my list to give you feedback. I am not too attached to the fake NUMA node either. The resulting patch seemed like the least invasive way to make rteval more inclusive on non-NUMA architectures. I am open to alternate ways to enable this functionality. I'll look out for your feedback once you've taken a closer look. [...]
diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py index 326f1ae..e747b9f 100644 --- a/rteval/modules/loads/kcompile.py +++ b/rteval/modules/loads/kcompile.py @@ -58,7 +58,7 @@ class KBuildJob: else: self.jobs = self.calc_jobs_per_cpu() * len(self.node) self.log(Log.DEBUG, "node %d: jobs == %d" % (int(node), self.jobs)) - self.runcmd = "%s make O=%s -C %s -j%d bzImage modules" \ + self.runcmd = "%s make O=%s -C %s -j%d" \ % (self.binder, self.objdir, self.kdir, self.jobs) self.cleancmd = "%s make O=%s -C %s clean allmodconfig" \ % (self.binder, self.objdir, self.kdir)
Not all kernel archiectures provide the "bzImage" target, e.g., arm64 provides "Image" which generates an uncompressed kernel binary. Instead of going down the path of customizing build targets per-architecture, let's drop them altogether. The default kernel target should build the kernel and modules on all architectures anyways. Signed-off-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp> --- rteval/modules/loads/kcompile.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)